Pearce-Simpson supercheetah MKIII 27mhz squealing
Trev 999
untill one day i went and turned it on and it just squeals. It still works
fine in PA mode, but in recieve its buggered and when i turn the volume up
it gets louder and changes pitch. Sounds like its feeding back somewheres.
It has been sitting outside under the porch and out of the weather and was
working a couple of days before the fault developed. It was powered by a
5amp variable power supply set at 14v and had a desk mic connected, also the
stationmaster antenna was always connected.
Any ideas and/or a schematic would be great, as i don't know if its worth
taking to a shop for repair and the missus won't let be buy a replacement
SideBand
Test for blown capacitors in the audio chain.. you may be getting dc on
the chain due to a shorted cap.
Also, try pulling the mic off and see if the squeal is still there. If
it goes away when the mic is disconnected, then (if it's a power mic)
the mic may be the problem.
-SSB
Frank Gilliland
in <johHd.123718$K7.9...@news-server.bigpond.net.au>:
First, try turning down the voltage on the power supply. Many radios
behave that way when the power supply voltage is too high. Also, check
your power supply to make sure the regulator hasn't failed.
If the radio is modified for talkback, cover the mic when you key up.
It may also be a fault in the mic. Take it apart to see if any wires
have come loose.
If that doesn't fix it you will probably have to take it to a shop.
----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! >100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---
Trev 999
reduced voltage. I will check the caps next. Please keep the ideas coming.
73's
"Trev 999" <tr...@hotmail.com> wrote in message
news:johHd.123718$K7.9...@news-server.bigpond.net.au...
Frank Gilliland
in <x2lHd.124146$K7.6...@news-server.bigpond.net.au>:
>Thanks guys, i've tried with the mic off and a different one and also with
>reduced voltage. I will check the caps next. Please keep the ideas coming.
I don't think it's a cap. Your radio has an internal voltage regulator
and that's likely the culprit. Those regulators are responsible for
the 'buzzing' of some radios when they are fed with a supply voltage
that's too high, and it sounds like your's is permanently damaged. The
reason your PA still works is because it's powered directly from the
power line, not the internal regulator.
SideBand
... unless a bad cap is biasing his balanced mixer, in which case the
squeal would be produced. I saw this in a Cobra 2000 GTL once. One of
the electrolytics shorted, causing similar behavior.
In my case, however, the power mic I was using (a Turner +2 SSB) had a
bad cap in it as well, and was sending DC voltage down the audio chain,
which caused the other cap to short, which caused the squeal.
Mine didn't change pitch with the volume control, however, but this
might just be a symptom of different radio design.
These radios are designed to run at 13.8 VDC, +/- 5% or so.. 14V isn't
going to hurt the radio or its regulators.
-SSB
Frank Gilliland
in <SWlHd.19407$by5....@newssvr19.news.prodigy.com>:
<snip>
>... unless a bad cap is biasing his balanced mixer, in which case the
>squeal would be produced. I saw this in a Cobra 2000 GTL once. One of
>the electrolytics shorted, causing similar behavior.
I've never seen that but it's possible. The reason I'm sure it's the
regulator is because I have seen the same fault lots of times in
similar radios. In fact, I'd be willing to bet that his radio uses the
same Uniden board as a TRC-453/PC-122.
>In my case, however, the power mic I was using (a Turner +2 SSB) had a
>bad cap in it as well, and was sending DC voltage down the audio chain,
>which caused the other cap to short, which caused the squeal.
>
>Mine didn't change pitch with the volume control, however, but this
>might just be a symptom of different radio design.
The only problem with that scenario is that his PA still works. So the
shorted cap would have to be much further down the chain from the mic
and inside the area fed by the internal regulator.
>These radios are designed to run at 13.8 VDC, +/- 5% or so.. 14V isn't
>going to hurt the radio or its regulators.
I have a TRC-435 that squeals at 14 volts and a Colt 357A that won't
even power up at that voltage. Drop it down to 13.6 (or even as low as
11.5 volts) and both work just fine.
Trev 999
7217) 5.8w amp sure gets hot when not in PA mode, theres a 7808 reg beside
this IC and the voltage is spot on 8v.
Is the input voltage reg normally a 3 pin jobby? If so I can't see it
offhand.
"Frank Gilliland" <wīrenut@NOSPAMīcehouse.net> wrote in message
news:a8rru0dji3fqf9843...@4ax.com...
SideBand
Frank Gilliland wrote:
> On Wed, 19 Jan 2005 05:08:02 GMT, SideBand <AINO8...@cac.net> wrote
> in <SWlHd.19407$by5....@newssvr19.news.prodigy.com>:
>
> <snip>
>
>>... unless a bad cap is biasing his balanced mixer, in which case the
>>squeal would be produced. I saw this in a Cobra 2000 GTL once. One of
>>the electrolytics shorted, causing similar behavior.
>
>
>
> I've never seen that but it's possible. The reason I'm sure it's the
> regulator is because I have seen the same fault lots of times in
> similar radios. In fact, I'd be willing to bet that his radio uses the
> same Uniden board as a TRC-453/PC-122.
>
>
Entirely possible. I'm not familiar with that particular rig.
>
>>In my case, however, the power mic I was using (a Turner +2 SSB) had a
>>bad cap in it as well, and was sending DC voltage down the audio chain,
>>which caused the other cap to short, which caused the squeal.
>>
>>Mine didn't change pitch with the volume control, however, but this
>>might just be a symptom of different radio design.
>
>
>
> The only problem with that scenario is that his PA still works. So the
> shorted cap would have to be much further down the chain from the mic
> and inside the area fed by the internal regulator.
>
>
PA and SSB were clean on the audio on the 2000, too. It only showed up
on AM. It was odd, but replacing that cap (actually, I replaced all the
caps in the audio and mixer chains, just to be safe, after I replaced
the one and tested the rig) fixed the problem.
>
>>These radios are designed to run at 13.8 VDC, +/- 5% or so.. 14V isn't
>>going to hurt the radio or its regulators.
>
>
>
> I have a TRC-435 that squeals at 14 volts and a Colt 357A that won't
> even power up at that voltage. Drop it down to 13.6 (or even as low as
> 11.5 volts) and both work just fine.
>
Odd. Never seen that before. Ok. I'll buy it. Just never seen it before.
-SSB
Frank Gilliland
in <homHd.124207$K7.8...@news-server.bigpond.net.au>:
>Can't see anything that looks suspect visually (caps etc), but the IC 5 (kia
>7217) 5.8w amp sure gets hot when not in PA mode, theres a 7808 reg beside
>this IC and the voltage is spot on 8v.
>Is the input voltage reg normally a 3 pin jobby? If so I can't see it
>offhand.
That would be it. Put the voltmeter on AC (yes, AC) and measure it.
Frank Gilliland
in <2xmHd.19420$by5....@newssvr19.news.prodigy.com>:
<snip>
>> The only problem with that scenario is that his PA still works. So the
>> shorted cap would have to be much further down the chain from the mic
>> and inside the area fed by the internal regulator.
>>
>>
>PA and SSB were clean on the audio on the 2000, too. It only showed up
>on AM. It was odd, but replacing that cap (actually, I replaced all the
>caps in the audio and mixer chains, just to be safe, after I replaced
>the one and tested the rig) fixed the problem.
Yeah, that's pretty strange alright! Was it a DC block or a supply
bypass cap?
>>>These radios are designed to run at 13.8 VDC, +/- 5% or so.. 14V isn't
>>>going to hurt the radio or its regulators.
>>
>>
>>
>> I have a TRC-435 that squeals at 14 volts and a Colt 357A that won't
>> even power up at that voltage. Drop it down to 13.6 (or even as low as
>> 11.5 volts) and both work just fine.
>>
>
>Odd. Never seen that before. Ok. I'll buy it. Just never seen it before.
A lot of radios do it. Just turn up the juice and start running them
through the bench. You'll see what I mean.
SideBand
> On Wed, 19 Jan 2005 05:48:46 GMT, SideBand <AINO8...@cac.net> wrote
> in <2xmHd.19420$by5....@newssvr19.news.prodigy.com>:
>
> <snip>
>
>>>The only problem with that scenario is that his PA still works. So the
>>>shorted cap would have to be much further down the chain from the mic
>>>and inside the area fed by the internal regulator.
>>>
>>>
>>
>>PA and SSB were clean on the audio on the 2000, too. It only showed up
>>on AM. It was odd, but replacing that cap (actually, I replaced all the
>>caps in the audio and mixer chains, just to be safe, after I replaced
>>the one and tested the rig) fixed the problem.
>
>
>
> Yeah, that's pretty strange alright! Was it a DC block or a supply
> bypass cap?
>
>
to ground.
>
>>>>These radios are designed to run at 13.8 VDC, +/- 5% or so.. 14V isn't
>>>>going to hurt the radio or its regulators.
>>>
>>>
>>>
>>>I have a TRC-435 that squeals at 14 volts and a Colt 357A that won't
>>>even power up at that voltage. Drop it down to 13.6 (or even as low as
>>>11.5 volts) and both work just fine.
>>>
>>
>>Odd. Never seen that before. Ok. I'll buy it. Just never seen it before.
>
>
>
> A lot of radios do it. Just turn up the juice and start running them
> through the bench. You'll see what I mean.
>
>
>
soon as I fix the 'scope, so I can fix the ICOM 751A and the Swan 350D.
-SSB
Frank Gilliland
in <hJmHd.19423$by5....@newssvr19.news.prodigy.com>:
<snip>
>> Yeah, that's pretty strange alright! Was it a DC block or a supply
>> bypass cap?
>>
>>
>It was a DC Block, put in place to bypass excess/out of band AC signal
>to ground.
.....huh?
<snip>
>I'll give it a shot on a spare 148 I have lying around here, just as
>soon as I fix the 'scope, so I can fix the ICOM 751A and the Swan 350D.
I've never seen a 148 do it although you could give it a try.
Trev 999
Also the schematic says at the bottom "Hatadi super cheetah", I could post
this (the schematic) on my website or email to anyone who's interested in
taking a look.
Thanks your help so far too guys!
"SideBand" <AINO8...@cac.net> wrote in message
news:2xmHd.19420$by5....@newssvr19.news.prodigy.com...
SideBand
One of a couple of things is happening here..
Either your power supply is going to pot (disconnect the radio and
measure the AC voltage across + and -) or the regulator (or one of its
filter caps) is going/gone bad.
And that's for starters.
Check your power supply first. If the AC voltage across the outputs
isn't VERY close to 0 volts, then it's time for a new power supply.
If it's at or close to 0 volts, the power supply is probably OK, and the
problem is in your radio. How handy are you with a soldering iron? Got
any test equipment besides a VOM?
-SSB
SideBand
Caps pass AC, but not all AC.. like any LC or RC circuit, some
frequencies are filtered and/or blocked, and others are allowed to pass.
Just because there isn't an inductor or a resistor in a circuit doesn't
mean it lacks L or R...
Sorry, didn't mean to be so vague.
-SSB
Frank Gilliland
in <m2qHd.16916$_X7.1...@newssvr33.news.prodigy.com>:
<snip>
>One of a couple of things is happening here..
>
>Either your power supply is going to pot (disconnect the radio and
>measure the AC voltage across + and -) or the regulator (or one of its
>filter caps) is going/gone bad.
>
>And that's for starters.
>
>Check your power supply first. If the AC voltage across the outputs
>isn't VERY close to 0 volts, then it's time for a new power supply.
>
>If it's at or close to 0 volts, the power supply is probably OK, and the
>problem is in your radio. How handy are you with a soldering iron? Got
>any test equipment besides a VOM?
Right on the money. There should be no more than a few (less than 10)
millivolts AC at the output of that regulator. You will need a scope
to find the source of the fault.
Frank Gilliland
in <kYmHd.124244$K7.7...@news-server.bigpond.net.au>:
>Ok, i've measured 1Vac across the output of that 7808 reg, is that cool?
NO! It's very UN-cool! Something is definitely fucked up!
>Also the schematic says at the bottom "Hatadi super cheetah", I could post
>this (the schematic) on my website or email to anyone who's interested in
>taking a look.
You can email it to me and I'll post it from my webspace.
Trev 999
"Frank Gilliland" <wīrenut@NOSPAMīcehouse.net> wrote in message
> On Wed, 19 Jan 2005 09:49:06 GMT, SideBand <AINO8...@cac.net> wrote
> in <m2qHd.16916$_X7.1...@newssvr33.news.prodigy.com>:
>
> <snip>
>>One of a couple of things is happening here..
>>
>>Either your power supply is going to pot (disconnect the radio and
>>measure the AC voltage across + and -) or the regulator (or one of its
>>filter caps) is going/gone bad.
Righto, The power supply is about a month old, but I checked it anyway, VOM
says millivolts AC ripple.
I will check filter caps tomorrow once its daylight.
>>And that's for starters.
>>
>>Check your power supply first. If the AC voltage across the outputs
>>isn't VERY close to 0 volts, then it's time for a new power supply.
>>
>>If it's at or close to 0 volts, the power supply is probably OK, and the
>>problem is in your radio. How handy are you with a soldering iron? Got
>>any test equipment besides a VOM?
I'm fairly handy with a soldering iron, have made a few electronic gadets
over the years (cystral set at age 8 :) )
I only have a FET VOM and a digi VOM, no scope unfortunately.
> Right on the money. There should be no more than a few (less than 10)
> millivolts AC at the output of that regulator. You will need a scope
> to find the source of the fault.
For testing the caps, am I just looking for short or open on them?
Trev 999
sorry SSB
Cheers Guys
Frank Gilliland
in <bgrHd.124638$K7.3...@news-server.bigpond.net.au>:
>Frank, what is your email? I sent the schematic to Sideband by mistake,
>sorry SSB
I see you found my email. Excuse my grouchiness but I've been drinking
and listening to pre-disco BeeGees through a Sony amp. That being
said......
Damn dude, that copy sucks! Try scanning it in grayscale (256 colors),
300 dpi, then reduce it to 16 colors and save it as a .gif file.
Now. Yes it looks like the regulator is the place to start. Disconnect
the output and measure AC ripple. If it's more than a few millivolts
then the regulator is toast. If it's smooth then you have an
oscillation somewhere else and you will -NEED- a scope to find out
where it's coming from.
BTW, make sure that all your power devices have good heatsinks. Those
gray plastic insulators are crap, and quite a few manufacturers
mistakenly use the "more is better" rule for heatsink grease (and
solder, too). I only mention this because of your observation that the
AF amp gets hot. It shouldn't. Not unless you are "peaked & tweaked"
for overmodulation.
Dave Hall
>> I don't think it's a cap. Your radio has an internal voltage regulator
>> and that's likely the culprit. Those regulators are responsible for
>> the 'buzzing' of some radios when they are fed with a supply voltage
>> that's too high, and it sounds like your's is permanently damaged. The
>> reason your PA still works is because it's powered directly from the
>> power line, not the internal regulator.
>... unless a bad cap is biasing his balanced mixer, in which case the
>squeal would be produced. I saw this in a Cobra 2000 GTL once. One of
>the electrolytics shorted, causing similar behavior.
>
>In my case, however, the power mic I was using (a Turner +2 SSB) had a
>bad cap in it as well, and was sending DC voltage down the audio chain,
>which caused the other cap to short, which caused the squeal.
>
>Mine didn't change pitch with the volume control, however, but this
>might just be a symptom of different radio design.
>
>These radios are designed to run at 13.8 VDC, +/- 5% or so.. 14V isn't
>going to hurt the radio or its regulators.
>
I'm not familiar with the radio mentioned. Does Pierce Simpson still
make radios? I'm assuming it's an old 23 channel classic.
With old radios like that, the problem is almost always related to
caps drying out. I once had an old 23 which would receive, but along
with the receive audio was a squeal. The squeal would change pitch and
volume when the volume control was advanced. The problem turned out to
be a bad cap in the audio stage just after the detector.
Dave
"Sandbagger"
http://home.ptd.net/~n3cvj
Frank Gilliland
wrote in <d7ksu0p1to2ikc0bs...@4ax.com>:
>On Wed, 19 Jan 2005 05:08:02 GMT, SideBand <AINO8...@cac.net> wrote:
>
>>> I don't think it's a cap. Your radio has an internal voltage regulator
>>> and that's likely the culprit. Those regulators are responsible for
>>> the 'buzzing' of some radios when they are fed with a supply voltage
>>> that's too high, and it sounds like your's is permanently damaged. The
>>> reason your PA still works is because it's powered directly from the
>>> power line, not the internal regulator.
>
>>... unless a bad cap is biasing his balanced mixer, in which case the
>>squeal would be produced. I saw this in a Cobra 2000 GTL once. One of
>>the electrolytics shorted, causing similar behavior.
>>
>>In my case, however, the power mic I was using (a Turner +2 SSB) had a
>>bad cap in it as well, and was sending DC voltage down the audio chain,
>>which caused the other cap to short, which caused the squeal.
>>
>>Mine didn't change pitch with the volume control, however, but this
>>might just be a symptom of different radio design.
>>
>>These radios are designed to run at 13.8 VDC, +/- 5% or so.. 14V isn't
>>going to hurt the radio or its regulators.
>>
>
>I'm not familiar with the radio mentioned. Does Pierce Simpson still
>make radios? I'm assuming it's an old 23 channel classic.
Dave, when will you learn to just keep your trap shut when you don't
know what you are talking about? I have antique radios from the 1920's
and up that still have good capacitors. They start having problems
when they are pushed over their rated voltage or physically leak; and
most electrolytics made since the 50's don't use liquid electrolyte.
Frank Gilliland
in <zcrHd.124634$K7.3...@news-server.bigpond.net.au>:
<snip>
>For testing the caps, am I just looking for short or open on them?
The only thing you can check is if they are shorted, and that doesn't
always work if they are still in the circuit. If you really want to go
that route then unsolder all electrolytic caps and rig up a charging
circuit (battery + high-value resistor). Trust me, it's a bigger pain
than taking it to the shop.
Twistedhed
in <bgrHd.124638$K7.3...@news-server.bigpond.net.au>:
Frank, what is your email? I sent the schematic to Sideband by mistake,
sorry SSB
>I see you found my email. Excuse my
>grouchiness but I've been drinking and
>listening to pre-disco BeeGees through a
>Sony amp. That being said......
Didn't the Bee Gees actually ROCK back in the late sixties and early
70s?
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>>I'm not familiar with the radio mentioned. Does Pierce Simpson still
>>make radios? I'm assuming it's an old 23 channel classic.
>
>
>Dave, when will you learn to just keep your trap shut when you don't
>know what you are talking about? I have antique radios from the 1920's
>and up that still have good capacitors.
Frank, I'm getting a little tired of your tone.
I've had tube radios where the filter caps exploded with the force of
a firecracker, after not being powered for 10 years or more. In fact,
if you check with the pros who restore old radios, they recommend
replacing aging caps since they are known to change value. They also
recommend that you power up an old radio gradually with a Variac, so
as to no "shock" an old cap into shorting.
> They start having problems
>when they are pushed over their rated voltage or physically leak; and
>most electrolytics made since the 50's don't use liquid electrolyte.
I'm telling you what I've personally experienced in close to 30 years
of repairing CB radios. It is COMMON for small 1uF and such
electrolytic caps to open up from age. I just got done retelling an
exact problem that I once fixed. Now, either you are calling me a
liar, or you need to modify your statement, since my examples are
proof that maybe you should keep your trap shut, since CB radios don't
seem to be your high point in experience.
Dave
"Sandbagger"
I AM NOT N8WWM !
> On Wed, 19 Jan 2005 04:07:57 GMT, "Trev 999" <tr...@hotmail.com>
wrote
> in <x2lHd.124146$K7.6...@news-server.bigpond.net.au>:
>
>
>
POSTS IN ALL THE NEWSGROUPS, TO SPREAD HIS SILLY SHIT. TO SPREAD HIS
SILLY SHIT.
HIS FIRST NAME IS FRANKIE, A KEYCLOWN ASSHOLE! HE SAT DOWN AT HIS
KEYBOARD, SO HE COULD TROLL. SO HE COULD TROLL.
THE NEWSGROUP STARTED GETTING ROUGH, HE WANTED STEVEO PARKS! IF NOT FOR
HELP FROM HIS KEYCLOWN QUEEN, 'OLE FRANKIE WOULD BE LOST. 'OLE FRANKIE
WOULD BE LOST.
STEVEO MET 'OLE FRANKIE AT A PETRO TRUCK STOP........WITH TWISTIE, AND
U KNOW WHO........ LANCER,,,WHO'S STEVEO'S WIFE! THE LANDFAG,, THE GRIM
REAPER,,,,,,,, YES THEY'RE ALL THERE..........THERE ON GILLILAND'S
ISLE!!!!!!!!
SideBand
Or, if you suspect it's a cap, just replace all the caps in that
circuit. A little more trouble, but allot cheaper than sending it to a shop.
If you're going to replace the caps in your regulator circuit, replace
the regulator as well.
One thing you can do to test a cap is to LOOK at it. If the top is bowed
out like a dome, or there's some brownish crusty ichor on the top or
bottom of the cap, it's bad.
Barring that, replace the regulator, and follow the schematic to find
the caps in that circuit to replace them all.
-SSB
SideBand
I don't use my real email address on Usenet. Those who NEED to get ahold
of me that way can deduce my callsign from information posted here, and
get my email from other sources.
-SSB
SideBand
Or if the AF amp is going bad.. An internal short could cause that kind
of heat, too.
-SSB
SideBand
Frank:
Not to jump into this fray, too, but the reason my Swan 350D isn't
functional right now is specifically because the caps have "dried out"..
it DOES happen. It's not as rare as some people think.
-SSB
ron
make the squeeling sound varying with the af gain control. this low voltage
could be caused by a resistor that supplies bias to the af amp breaking down
under increasing current drain. a common problem.
ron
"Trev 999" <tr...@hotmail.com> wrote in message
news:johHd.123718$K7.9...@news-server.bigpond.net.au...
> Hi all, I have a Pearce-Simpson supercheetah MKIII that was working fine,
> untill one day i went and turned it on and it just squeals. It still works
> fine in PA mode, but in recieve its buggered and when i turn the volume up
> it gets louder and changes pitch. Sounds like its feeding back somewheres.
> It has been sitting outside under the porch and out of the weather and was
> working a couple of days before the fault developed. It was powered by a
> 5amp variable power supply set at 14v and had a desk mic connected, also
the
> stationmaster antenna was always connected.
> Any ideas and/or a schematic would be great, as i don't know if its worth
> taking to a shop for repair and the missus won't let be buy a replacement
>
>
Trev 999
Dave, please forgive Frank, I think he had been drinking when he made that
remark ;)
Cheers's and 73's, Trev
Frank Gilliland
wrote in <kt6tu0lcvnm4bonrk...@4ax.com>:
>On Wed, 19 Jan 2005 04:43:25 -0800, Frank Gilliland
><wīrenut@NOSPAMīcehouse.net> wrote:
>
>
>>>I'm not familiar with the radio mentioned. Does Pierce Simpson still
>>>make radios? I'm assuming it's an old 23 channel classic.
>>
>>
>>Dave, when will you learn to just keep your trap shut when you don't
>>know what you are talking about? I have antique radios from the 1920's
>>and up that still have good capacitors.
>
>Frank, I'm getting a little tired of your tone.
I really don't care.
>I've had tube radios where the filter caps exploded with the force of
>a firecracker, after not being powered for 10 years or more. In fact,
>if you check with the pros who restore old radios, they recommend
>replacing aging caps since they are known to change value. They also
>recommend that you power up an old radio gradually with a Variac, so
>as to no "shock" an old cap into shorting.
Caps wear out, that's a fact. And caps do explode on occasion. But in
every case where I have filled the shop with that unique and pungent
aroma there has -always- been some underlying cause of the failure,
such as too much AC on a filter cap, voltage ratings exceeded or
reversed, etc. I've never had one blow up just out of spite.
Also, I -am- one of those "pros who restore old radios", which has
been both a hobby and a business since I restored my first Philco 20
when I was a teenager. And after being in the electronics business
professionally for 30+ years I can say, without any doubt whatsoever,
that capacitor failure rate (i.e, the capacitor being the -primary-
cause of the failure) is no higher than for any other component. I do
-not- recommend blanket replacement of aged capacitors -- on the
contrary, I recommend -testing- aged capacitors. If they are still
within tolerance and show no DC leakage, leave them ALONE!
BTW, for anyone planning on testing caps, you don't need anything
fancy. Just find an old R/C tester like the Eico 950. I have used an
old tube-type Knight R/C tester for 20 years and it hasn't failed me
yet.
>> They start having problems
>>when they are pushed over their rated voltage or physically leak; and
>>most electrolytics made since the 50's don't use liquid electrolyte.
>
> I'm telling you what I've personally experienced in close to 30 years
>of repairing CB radios. It is COMMON for small 1uF and such
>electrolytic caps to open up from age. I just got done retelling an
>exact problem that I once fixed. Now, either you are calling me a
>liar, or you need to modify your statement, since my examples are
>proof that maybe you should keep your trap shut, since CB radios don't
>seem to be your high point in experience.
Your examples are proof of nothing except your lack of experience in
the broad field of electronics and your inability to find the
underlying cause of the failure. Cap failure as the -primary- cause of
failure is no more common than a weak transistor, bad IC, blown
diode/zener, or bad RF/IF can, or even poor manufacturing standards.
Frank Gilliland
in <r1zHd.19518$by5....@newssvr19.news.prodigy.com>:
<snip>
>Frank:
>
>Not to jump into this fray, too, but the reason my Swan 350D isn't
>functional right now is specifically because the caps have "dried out"..
>it DOES happen. It's not as rare as some people think.
Swan was notorious for using cheap PS caps. Haven't seen one yet that
wasn't bad or on the edge. I had a 500 and I replaced the cap with an
-older- cap -- still works great according to the guy who has it now.
Frank Gilliland
wrote in <12354-41...@storefull-3258.bay.webtv.net>:
Absolutely! Have you checked out some of their more recent work?
Sometimes a little retro but with the same unique 'sound'.
Frank Gilliland
in <x%yHd.19514$by5....@newssvr19.news.prodigy.com>:
<snip>
>Or if the AF amp is going bad.. An internal short could cause that kind
>of heat, too.
True. I looked at the schematic again and both the input and output
are DC blocked. If those were bad then the amp wouldn't work at all.
There is another cap on the side that may either be for offset or
compensation, and if that's shorted it may cause the chip to overheat.
That would be my first guess. But it still doesn't explain the squeal.
What I -didn't- see before is that the preamp before the PA is powered
directly from the regulator, so the regulator is good. The oscillation
must be coming from somewhere else.
Frank Gilliland
in <pRAHd.124973$K7.5...@news-server.bigpond.net.au>:
>Thanks all for your input!!
>Dave, please forgive Frank, I think he had been drinking when he made that
>remark ;)
Forget about Dave. Go hit some yard sales and find yourself a scope! I
stopped by a sale last fall and picked up an immaculate Heathkit 4550
for..... are you sitting down?..... $10. They are out there, you just
have to look.
Trev 999
"Frank Gilliland" <wīrenut@NOSPAMīcehouse.net> wrote in message
Twistedhed
wrote in <12354-41...@storefull-3258.bay.webtv.net>:
in <bgrHd.124638$K7.3...@news-server.bigpond.net.au>: Frank, what is
your email? I sent the schematic to Sideband by mistake, sorry SSB
(I see you found my email. Excuse my
listening to pre-disco BeeGees through a
_
Didn't the Bee Gees actually ROCK back in the late sixties and early
70s?
>Absolutely! Have you checked out some of
>their more recent work? Sometimes a little
>retro but with the same unique 'sound'.
No, but I will since you mentioned it. Last thing I heard concerning the
group was Maurice dying. I remember hearing some of their pre-disco
stuff back in the day, like on the King Biscuit Flower Hour on Sunday
nights and Don Kirschner's (sp?) Rock Concert on Saturday night. I
remember it because when the soundtrack came out for Saturday Night
Fever, I thought it was pure shit compared to what I originally heard
from the group. Then I found out they wrote the soundtrack tunes without
having any idea what the movie was about.
Frank Gilliland
wrote in <28053-41E...@storefull-3255.bay.webtv.net>:
Check out the tune, "This is where I came in". Recorded in 2000. I
think you'll like it.
Frank Gilliland
wrote in <28053-41E...@storefull-3255.bay.webtv.net>:
<snip>
>... I thought it was pure shit
Speaking of shitty music, turn on the TV and listen to the song
written by Ashcroft..... blech!!!
Twistedhed
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>On Wed, 19 Jan 2005 20:00:29 GMT, SideBand <AINO8...@cac.net> wrote
>in <x%yHd.19514$by5....@newssvr19.news.prodigy.com>:
>
><snip>
>>Or if the AF amp is going bad.. An internal short could cause that kind
>>of heat, too.
>
>
>True. I looked at the schematic again and both the input and output
>are DC blocked. If those were bad then the amp wouldn't work at all.
>There is another cap on the side that may either be for offset or
>compensation, and if that's shorted it may cause the chip to overheat.
>That would be my first guess. But it still doesn't explain the squeal.
Gee, I thought you said that caps don't go bad?
>What I -didn't- see before is that the preamp before the PA is powered
>directly from the regulator, so the regulator is good. The oscillation
>must be coming from somewhere else.
How about a little gentleman's wager? Nothing monetary. How about just
a simple acknowledgement?
I say it's a bad cap. What say you?
Dave
"Sandbagger"
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>On Wed, 19 Jan 2005 12:47:55 -0500, Dave Hall <nojunk...@ptd.net>
>wrote in <kt6tu0lcvnm4bonrk...@4ax.com>:
>
>>On Wed, 19 Jan 2005 04:43:25 -0800, Frank Gilliland
>><wīrenut@NOSPAMīcehouse.net> wrote:
>>
>>
>>>>I'm not familiar with the radio mentioned. Does Pierce Simpson still
>>>>make radios? I'm assuming it's an old 23 channel classic.
>>>
>>>
>>>Dave, when will you learn to just keep your trap shut when you don't
>>>know what you are talking about? I have antique radios from the 1920's
>>>and up that still have good capacitors.
>>
>>Frank, I'm getting a little tired of your tone.
>
>
>I really don't care.
That much is obvious.
>>I've had tube radios where the filter caps exploded with the force of
>>a firecracker, after not being powered for 10 years or more. In fact,
>>if you check with the pros who restore old radios, they recommend
>>replacing aging caps since they are known to change value. They also
>>recommend that you power up an old radio gradually with a Variac, so
>>as to no "shock" an old cap into shorting.
>
>
>Caps wear out, that's a fact. And caps do explode on occasion. But in
>every case where I have filled the shop with that unique and pungent
>aroma there has -always- been some underlying cause of the failure,
>such as too much AC on a filter cap, voltage ratings exceeded or
>reversed, etc. I've never had one blow up just out of spite.
I have. Usually they're in sweep tube CB amplifiers. For no other
reason than they just became "leaky", and start drawing current until
they heat up and explode. In my Lafayette Comstat 25 a resistor once
lit up like a candle due to a shorted cap in series with it.
The Uniden UPD-858 SSB boards (Realistic TRC-457/458, Cobra 138/139
XLR) had a pesky blue 10 uF tantalum cap that would short out and pull
the B+ feeding the driver and final stages to ground. I've seen that
on at least 3 different radios. I'd call that a statistic.
>Also, I -am- one of those "pros who restore old radios", which has
>been both a hobby and a business since I restored my first Philco 20
>when I was a teenager. And after being in the electronics business
>professionally for 30+ years I can say, without any doubt whatsoever,
>that capacitor failure rate (i.e, the capacitor being the -primary-
>cause of the failure) is no higher than for any other component.
But no lower either. They certainly fail more often than resistors.
>I do
>-not- recommend blanket replacement of aged capacitors -- on the
>contrary, I recommend -testing- aged capacitors. If they are still
>within tolerance and show no DC leakage, leave them ALONE!
I don't either as a rule. It's just a waste of money. But for you to
discard, as idiotic, my suggestion of a small electrolytic cap
changing value causing the problem in that Pierce Simpson radio is
nothing but pure arrogance Frank. I have had a similar problem that
was the fault of a cap.
>BTW, for anyone planning on testing caps, you don't need anything
>fancy. Just find an old R/C tester like the Eico 950. I have used an
>old tube-type Knight R/C tester for 20 years and it hasn't failed me
>yet.
I have a digital capacitance tester. Not only can I tell if the cap is
open or shorted, I can tell if it drifted appreciably from its
original value.
>>> They start having problems
>>>when they are pushed over their rated voltage or physically leak; and
>>>most electrolytics made since the 50's don't use liquid electrolyte.
>>
>> I'm telling you what I've personally experienced in close to 30 years
>>of repairing CB radios. It is COMMON for small 1uF and such
>>electrolytic caps to open up from age. I just got done retelling an
>>exact problem that I once fixed. Now, either you are calling me a
>>liar, or you need to modify your statement, since my examples are
>>proof that maybe you should keep your trap shut, since CB radios don't
>>seem to be your high point in experience.
>
>
>Your examples are proof of nothing except your lack of experience in
>the broad field of electronics and your inability to find the
>underlying cause of the failure.
The primary cause of the failure WAS the cap. When a cap in an audio
chain opens up so that audio no longer passes through, and there are
no voltages to speak of across it (Other than bias level), and when
the cap is replaced, normal function is restored, that's a pretty good
indication of what the cause of the failure was.
>Cap failure as the -primary- cause of
>failure is no more common than a weak transistor, bad IC, blown
>diode/zener, or bad RF/IF can, or even poor manufacturing standards.
Ah! So at least now you admit that it is equally likely that a cap can
fail as often as any particular active device. Very good, you're
making progress.
Dave
"Sandbagger"
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>On Wed, 19 Jan 2005 20:02:31 GMT, SideBand <AINO8...@cac.net> wrote
>in <r1zHd.19518$by5....@newssvr19.news.prodigy.com>:
>
><snip>
>>Frank:
>>
>>Not to jump into this fray, too, but the reason my Swan 350D isn't
>>functional right now is specifically because the caps have "dried out"..
>>it DOES happen. It's not as rare as some people think.
>
>
>Swan was notorious for using cheap PS caps.
And you don't think that cheap, Malaysian made CB radios aren't doing
the same?
Dave
"Sandbagger"
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>On Wed, 19 Jan 2005 22:06:13 GMT, "Trev 999" <tr...@hotmail.com> wrote
>in <pRAHd.124973$K7.5...@news-server.bigpond.net.au>:
>
>>Thanks all for your input!!
>>Dave, please forgive Frank, I think he had been drinking when he made that
>>remark ;)
>
>
>Forget about Dave. Go hit some yard sales and find yourself a scope! I
>stopped by a sale last fall and picked up an immaculate Heathkit 4550
>for..... are you sitting down?..... $10. They are out there, you just
>have to look.
I just sold a Tektronix 2465 for $800...
But I'm hoping to find a nice Tram Titan 3 or D-201 for cheap at a
sale.....
Dave
"Sandbagger"
Frank Gilliland
wrote in <16svu0pt7baomhngk...@4ax.com>:
>On Wed, 19 Jan 2005 21:14:20 -0800, Frank Gilliland
><wīrenut@NOSPAMīcehouse.net> wrote:
>
>>On Wed, 19 Jan 2005 20:00:29 GMT, SideBand <AINO8...@cac.net> wrote
>>in <x%yHd.19514$by5....@newssvr19.news.prodigy.com>:
>>
>><snip>
>>>Or if the AF amp is going bad.. An internal short could cause that kind
>>>of heat, too.
>>
>>
>>True. I looked at the schematic again and both the input and output
>>are DC blocked. If those were bad then the amp wouldn't work at all.
>>There is another cap on the side that may either be for offset or
>>compensation, and if that's shorted it may cause the chip to overheat.
>>That would be my first guess. But it still doesn't explain the squeal.
>
>Gee, I thought you said that caps don't go bad?
The problems with your thought processes are now well-known. I never
said that caps don't go bad. I said they aren't the primary cause of
failure any more than any other component.
>>What I -didn't- see before is that the preamp before the PA is powered
>>directly from the regulator, so the regulator is good. The oscillation
>>must be coming from somewhere else.
>
>How about a little gentleman's wager? Nothing monetary. How about just
>a simple acknowledgement?
>
>I say it's a bad cap. What say you?
Sorry, I won't speculate.
Frank Gilliland
wrote in <tdsvu0dsvos26iqso...@4ax.com>:
<snip>
>> I've never had one blow up just out of spite.
>
>I have. Usually they're in sweep tube CB amplifiers. For no other
>reason than they just became "leaky", and start drawing current until
>they heat up and explode.
I don't believe that for a minute. I have worked on literally hundreds
of pieces tube gear of all shapes and sizes and -NEVER- had a cap blow
up for no reason, let alone overheating caused by leakage. But it's
not my experience that tells me your story is bullshit, it's the fact
that if the cap is leaking that much current in parallel with the
normal load then other components will fail first, such as the filter
resistor or choke, rectifier(s), transformer, etc. Usually it's the
fuse that goes first, and if there's no fuse then the transformer
overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
blow when subjected to too much AC, and even a few volts of ripple can
be enough to pop a can on a tube rig. This happens when the load is
too heavy for the power supply, and the first cap to blow is the one
on the load side of the pi filter (which is a very good indication of
why it blew up). In the case of sweep-tube amps, I find that the most
likely cause of your capacitor mishap.
> In my Lafayette Comstat 25 a resistor once
>lit up like a candle due to a shorted cap in series with it.
Now -that- I can believe.
>The Uniden UPD-858 SSB boards (Realistic TRC-457/458, Cobra 138/139
>XLR) had a pesky blue 10 uF tantalum cap that would short out and pull
>the B+ feeding the driver and final stages to ground. I've seen that
>on at least 3 different radios. I'd call that a statistic.
I would call that a design flaw. Tantalum caps have often been used in
applications where they were not a very good choice. They have certain
limitations, but many engineers haven't learned what those limitations
are, hence their reputation for failure. They are high-density caps
but are VERY sensitive to damage from overvoltage. A digital circuit
with a tightly regulated power supply is an appropriate application
for tantalum caps -- a CB is not.
>>Also, I -am- one of those "pros who restore old radios", which has
>>been both a hobby and a business since I restored my first Philco 20
>>when I was a teenager. And after being in the electronics business
>>professionally for 30+ years I can say, without any doubt whatsoever,
>>that capacitor failure rate (i.e, the capacitor being the -primary-
>>cause of the failure) is no higher than for any other component.
>
>But no lower either. They certainly fail more often than resistors.
Not when you realize that a light bulb is nothing more than a
resistor.
>>I do
>>-not- recommend blanket replacement of aged capacitors -- on the
>>contrary, I recommend -testing- aged capacitors. If they are still
>>within tolerance and show no DC leakage, leave them ALONE!
>
>I don't either as a rule. It's just a waste of money. But for you to
>discard, as idiotic, my suggestion of a small electrolytic cap
>changing value causing the problem in that Pierce Simpson radio is
>nothing but pure arrogance Frank. I have had a similar problem that
>was the fault of a cap.
I didn't discard your example as idiotic. I discarded your presumption
that a capacitor is responsible for the OP's problem. You once found a
radio that had a bad cap. Well, so have I, and so has just about
anyone that has fixed a few radios. But after you work with
electrionics for 30 years you eventually learn that you can't jump to
any conclusions. You have to locate the faulty component(s) AND you
must determine why it failed. A lot of caps do indeed go bad, but most
of the time it's a secondary fault. Finding the primary fault is what
seperates the professionals from the amateurs.
>>BTW, for anyone planning on testing caps, you don't need anything
>>fancy. Just find an old R/C tester like the Eico 950. I have used an
>>old tube-type Knight R/C tester for 20 years and it hasn't failed me
>>yet.
>
>I have a digital capacitance tester. Not only can I tell if the cap is
>open or shorted, I can tell if it drifted appreciably from its
>original value.
ftp://bama.sbc.edu/downloads/knight/rc/
ftp://bama.sbc.edu/downloads/eico/950b/
>>>> They start having problems
>>>>when they are pushed over their rated voltage or physically leak; and
>>>>most electrolytics made since the 50's don't use liquid electrolyte.
>>>
>>> I'm telling you what I've personally experienced in close to 30 years
>>>of repairing CB radios. It is COMMON for small 1uF and such
>>>electrolytic caps to open up from age. I just got done retelling an
>>>exact problem that I once fixed. Now, either you are calling me a
>>>liar, or you need to modify your statement, since my examples are
>>>proof that maybe you should keep your trap shut, since CB radios don't
>>>seem to be your high point in experience.
>>
>>
>>Your examples are proof of nothing except your lack of experience in
>>the broad field of electronics and your inability to find the
>>underlying cause of the failure.
>
>The primary cause of the failure WAS the cap. When a cap in an audio
>chain opens up so that audio no longer passes through, and there are
>no voltages to speak of across it (Other than bias level), and when
>the cap is replaced, normal function is restored, that's a pretty good
>indication of what the cause of the failure was.
No it's not. After replacing the cap it would have been a good idea to
push the audio level rail-to-rail to see if the amp locks up, change
freqs to see if there's any resonance that could swing the AC above
the cap's rating, heat-run the radio for a few hours, and then -maybe-
you could safely conclude that the cap was the primary failure.
>>Cap failure as the -primary- cause of
>>failure is no more common than a weak transistor, bad IC, blown
>>diode/zener, or bad RF/IF can, or even poor manufacturing standards.
>
>Ah! So at least now you admit that it is equally likely that a cap can
>fail as often as any particular active device. Very good, you're
>making progress.
I never denied it. I just didn't jump to the conclusion that it's a
cap like you did.
Frank Gilliland
wrote in <tdsvu0dsvos26iqso...@4ax.com>:
<snip>
>> I've never had one blow up just out of spite.
>
>I have. Usually they're in sweep tube CB amplifiers. For no other
>reason than they just became "leaky", and start drawing current until
>they heat up and explode.
I don't believe that for a minute. I have worked on literally hundreds
of pieces tube gear of all shapes and sizes and -NEVER- had a cap blow
up for no reason, let alone overheating caused by leakage. But it's
not my experience that tells me your story is bullshit, it's the fact
that if the cap is leaking that much current in parallel with the
normal load then other components will fail first, such as the filter
resistor or choke, rectifier(s), transformer, etc. Usually it's the
fuse that goes first, and if there's no fuse then the transformer
overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
blow when subjected to too much AC, and even a few volts of ripple can
be enough to pop a can on a tube rig. This happens when the load is
too heavy for the power supply, and the first cap to blow is the one
on the load side of the pi filter (which is a very good indication of
why it blew up). In the case of sweep-tube amps, I find that the most
likely cause of your capacitor mishap.
> In my Lafayette Comstat 25 a resistor once
>lit up like a candle due to a shorted cap in series with it.
Now -that- I can believe.
>The Uniden UPD-858 SSB boards (Realistic TRC-457/458, Cobra 138/139
>XLR) had a pesky blue 10 uF tantalum cap that would short out and pull
>the B+ feeding the driver and final stages to ground. I've seen that
>on at least 3 different radios. I'd call that a statistic.
I would call that a design flaw. Tantalum caps have often been used in
applications where they were not a very good choice. They have certain
limitations, but many engineers haven't learned what those limitations
are, hence their reputation for failure. They are high-density caps
but are VERY sensitive to damage from overvoltage. A digital circuit
with a tightly regulated power supply is an appropriate application
for tantalum caps -- a CB is not.
>>Also, I -am- one of those "pros who restore old radios", which has
>>been both a hobby and a business since I restored my first Philco 20
>>when I was a teenager. And after being in the electronics business
>>professionally for 30+ years I can say, without any doubt whatsoever,
>>that capacitor failure rate (i.e, the capacitor being the -primary-
>>cause of the failure) is no higher than for any other component.
>
>But no lower either. They certainly fail more often than resistors.
Not when you realize that a light bulb is nothing more than a
resistor.
>>I do
>>-not- recommend blanket replacement of aged capacitors -- on the
>>contrary, I recommend -testing- aged capacitors. If they are still
>>within tolerance and show no DC leakage, leave them ALONE!
>
>I don't either as a rule. It's just a waste of money. But for you to
>discard, as idiotic, my suggestion of a small electrolytic cap
>changing value causing the problem in that Pierce Simpson radio is
>nothing but pure arrogance Frank. I have had a similar problem that
>was the fault of a cap.
I didn't discard your example as idiotic. I discarded your presumption
that a capacitor is responsible for the OP's problem. You once found a
radio that had a bad cap. Well, so have I, and so has just about
anyone that has fixed a few radios. But after you work with
electrionics for 30 years you eventually learn that you can't jump to
any conclusions. You have to locate the faulty component(s) AND you
must determine why it failed. A lot of caps do indeed go bad, but most
of the time it's a secondary fault. Finding the primary fault is what
seperates the professionals from the amateurs.
>>BTW, for anyone planning on testing caps, you don't need anything
>>fancy. Just find an old R/C tester like the Eico 950. I have used an
>>old tube-type Knight R/C tester for 20 years and it hasn't failed me
>>yet.
>
>I have a digital capacitance tester. Not only can I tell if the cap is
>open or shorted, I can tell if it drifted appreciably from its
>original value.
ftp://bama.sbc.edu/downloads/knight/rc/
ftp://bama.sbc.edu/downloads/eico/950b/
>>>> They start having problems
>>>>when they are pushed over their rated voltage or physically leak; and
>>>>most electrolytics made since the 50's don't use liquid electrolyte.
>>>
>>> I'm telling you what I've personally experienced in close to 30 years
>>>of repairing CB radios. It is COMMON for small 1uF and such
>>>electrolytic caps to open up from age. I just got done retelling an
>>>exact problem that I once fixed. Now, either you are calling me a
>>>liar, or you need to modify your statement, since my examples are
>>>proof that maybe you should keep your trap shut, since CB radios don't
>>>seem to be your high point in experience.
>>
>>
>>Your examples are proof of nothing except your lack of experience in
>>the broad field of electronics and your inability to find the
>>underlying cause of the failure.
>
>The primary cause of the failure WAS the cap. When a cap in an audio
>chain opens up so that audio no longer passes through, and there are
>no voltages to speak of across it (Other than bias level), and when
>the cap is replaced, normal function is restored, that's a pretty good
>indication of what the cause of the failure was.
No it's not. After replacing the cap it would have been a good idea to
push the audio level rail-to-rail to see if the amp locks up, change
freqs to see if there's any resonance that could swing the AC above
the cap's rating, heat-run the radio for a few hours, and then -maybe-
you could safely conclude that the cap was the primary failure.
>>Cap failure as the -primary- cause of
>>failure is no more common than a weak transistor, bad IC, blown
>>diode/zener, or bad RF/IF can, or even poor manufacturing standards.
>
>Ah! So at least now you admit that it is equally likely that a cap can
>fail as often as any particular active device. Very good, you're
>making progress.
I never denied it. I just didn't jump to the conclusion that it's a
cap like you did.
----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
moparholic
"Lancer" <lancer@rock..com> wrote in message
news:jcvcu09uaqj1v0hsd...@news.easynews.com...
> On Wed, 19 Jan 2005 11:12:07 GMT, "Trev 999" <tr...@hotmail.com>
> wrote:
>
>>Frank, what is your email? I sent the schematic to Sideband by mistake,
>>sorry SSB
>>
>>
>
> Trev;
> Can you send me a copy of that schematic?
Lancer!!!!! Where's your fucking manners. Trev will only send it to you
after you suck-off Frank Silliland.
Thats how the club works,,,,,,remember sweetie??
_________________________________________
Usenet Zone Free Binaries Usenet Server
More than 120,000 groups
Unlimited download
http://www.usenetzone.com to open account
Steveo
>> >Frank, I'm getting a little tired of your tone.
>
> I really don't care.
>
Hi Frank. I propose me renting the boxing ring time in Toledo, and Doug
and I settling his -hooked on gay- in the squared circle. Like you, I
really don't care, except the part where no one talks that way to
my face without trouble. He only lives 3 exit's down the Ohio green-stamp
from me.
I know the pat answer is to ignore him. Most times I do, but the urge
to rub his deputy dogie jowls in my spit is hard to resist.
Figure a fair boxing match right down the street from his house, with
me paying the rental is a fair way to settle it?
I value your opinion now, only reason I ask.
Steveo
> On Wed, 19 Jan 2005 11:15:34 -0500, twist...@webtv.net (Twistedhed)
> wrote in <12354-41...@storefull-3258.bay.webtv.net>:
>
> >On Wed, 19 Jan 2005 11:12:07 GMT, "Trev 999" <tr...@hotmail.com> wrote
> >in <bgrHd.124638$K7.3...@news-server.bigpond.net.au>:
> >Frank, what is your email? I sent the schematic to Sideband by mistake,
> >sorry SSB
> >
> >>I see you found my email. Excuse my
> >
> >>grouchiness but I've been drinking and
> >
> >>listening to pre-disco BeeGees through a
> >
> >>Sony amp. That being said......
> >
> >
> >Didn't the Bee Gees actually ROCK back in the late sixties and early
> >70s?
>
> Absolutely! Have you checked out some of their more recent work?
> Sometimes a little retro but with the same unique 'sound'.
>
portable E-150 fuck palace back in the day. (mirror-ball?)
Steveo
Frank Gilliland
<jcvcu09uaqj1v0hsd...@news.easynews.com>:
>On Wed, 19 Jan 2005 11:12:07 GMT, "Trev 999" <tr...@hotmail.com>
>wrote:
>
>>Frank, what is your email? I sent the schematic to Sideband by mistake,
>>sorry SSB
>>
>>Cheers Guys
>>
>
>Trev;
> Can you send me a copy of that schematic?
I just sent it to you. Trev's file was a 2.2M jpeg so I reduced it to
a 2-color gif of about 200K. Quality is no worse than the original,
which was pretty bad. I think he scanned a xerox copy.
moparholic
"Steveo" <mopar...@hotmail-nospam.com> wrote in message
news:20050120222501.205$kg...@newsreader.com...
HOW'S YOUR CRACK HEAD WHORE-MAW DOING?
Steveo
wrote:
> Frank you really are suckinfg twistys dick hard these days,
>
There's no call for that. You can disagree with someone without using
gay references.
moparholic
Steveo's right. when you make gay references it makes Steveo feel unwelcome
here. Just because Steveo is homosexual doesn't mean that he doesn't have as
much right as you to post here.
Plus you bring up memories of Steveo's father leaving the home when Steveo
was very young for another man.We don't need to remind him of this over and
over again.
Steveo
wrote:
> Steveo <mopar...@hotmail-nospam.com> wrote in news:20050120232903.592
> $G...@newsreader.com:
> they are pen pals. give me a break.
>
Disagree.. hooptie whooptie. There's way yonder too many gay references
in this NG over such nonsense.
Frank Gilliland
<itoldyouiamno...@yahoo.comSpam> wrote in
<Xns95E4AB7E83...@216.196.97.136>:
>Frank Gilliland <wīrenut@NOSPAMīcehouse.net> wrote in
>news:jgmvu0dsqec804snk...@4ax.com:
>Frank you really are suckinfg twistys dick hard these days, are you going
>to post his name again off ebay, or are you befriendung him for other
>reasons we dont know about yet, either way your make me sick
I'm not responsible for your state of ill health.
However, in the past couple months I have noticed something very
interesting: Say you find a strange piece of equipment and have no
idea what it does. So you start pushing buttons until something
happens. Well, I pushed a button marked "Dave Hall" and all of a
sudden I'm being treated like a keyclown by you and your other
sock-puppets, complete with forgeries and homo-dribble.
Do you have an explanation for this?
Frank Gilliland
"Frank Gilliland" <wīrenut@NOSPAMīcehouse.net> wrote in message
news:vb61v0te367gjo546...@4ax.com...
Kensock d`Fehrett
wrote:
>Do they only allow gay guys to post!?
Are you thinking of opening an account?
Trev 999
then please take your shit to another thread.
THANKYOU
"Dave Hall" <nojunk...@ptd.net> wrote in message
news:sftvu0p410ua9rcq1...@4ax.com...
Trev 999
tossers...get a life!
"Trev 999" <tr...@hotmail.com> wrote in message
news:K07Id.127847$K7.1...@news-server.bigpond.net.au...
U Know Who
"itoldyouiamnotiamnotgeorge" <itoldyouiamno...@yahoo.comSpam>
wrote in message news:Xns95E552ED61...@216.196.97.136...
> Frank Gilliland <wīrenut@NOSPAMīcehouse.net> wrote in
Exactly. You hang with Dung Adair. You get treated like shit.
Frank Gilliland
<itoldyouiamno...@yahoo.comSpam> wrote in
<Xns95E552ED61...@216.196.97.136>:
>Frank Gilliland <wīrenut@NOSPAMīcehouse.net> wrote in
>news:vb61v0te367gjo546...@4ax.com:
>Frank the company you keep determines the way you are treated.
The company I keep is my business, not yours.
Steveo
at 2 am. The same time the forges were occurring. What's really sick
is he signed up for newsfeeds in order to do that.
Frank Gilliland
<itoldyouiamno...@yahoo.comSpam> wrote in
<Xns95E564ADBF...@216.196.97.136>:
>Frank Gilliland <wïrenut@NOSPAMïcehouse.net> wrote in
>news:b102v0tugj66vcstd...@4ax.com:
>
>> On Fri, 21 Jan 2005 07:07:29 -0600, itoldyouiamnotiamnotgeorge
>> <itoldyouiamno...@yahoo.comSpam> wrote in
>> <Xns95E552ED61...@216.196.97.136>:
>>
>>>Frank Gilliland <wïrenut@NOSPAMïcehouse.net> wrote in
>>>news:vb61v0te367gjo546...@4ax.com:
>>>
>>>> On Thu, 20 Jan 2005 15:49:53 -0600, itoldyouiamnotiamnotgeorge
>>>> <itoldyouiamno...@yahoo.comSpam> wrote in
>>>> <Xns95E4AB7E83...@216.196.97.136>:
>>>>
>>>>>Frank Gilliland <wïrenut@NOSPAMïcehouse.net> wrote in
>>>>>news:jgmvu0dsqec804snk...@4ax.com:
>>>>>
>>>>>> On Thu, 20 Jan 2005 10:22:02 -0500, twist...@webtv.net
>>>>>> (Twistedhed) wrote in
>>>>>> <28053-41E...@storefull-3255.bay.webtv.net>:
>>>>>>
>>>>>> <snip>
>>>>>>>... I thought it was pure shit
>>>>>>
>>>>>>
>>>>>> Speaking of shitty music, turn on the TV and listen to the song
>>>>>> written by Ashcroft..... blech!!!
>>>>>>
>>>>>
>>>>>Frank you really are suckinfg twistys dick hard these days, are you
>>>>>going to post his name again off ebay, or are you befriendung him for
>>>>>other reasons we dont know about yet, either way your make me sick
>>>>
>>>>
>>>> I'm not responsible for your state of ill health.
>>>>
>>>> However, in the past couple months I have noticed something very
>>>> interesting: Say you find a strange piece of equipment and have no
>>>> idea what it does. So you start pushing buttons until something
>>>> happens. Well, I pushed a button marked "Dave Hall" and all of a
>>>> sudden I'm being treated like a keyclown by you and your other
>>>> sock-puppets, complete with forgeries and homo-dribble.
>>>>
>>>> Do you have an explanation for this?
>>>
>>>Frank the company you keep determines the way you are treated.
>
>
>
>
>> The company I keep is my business, not yours.
>
>Well that's funny Frank, cause you posted all that transexual info about
>Landshark, so why did the company he kept bother you into posting that
>info?
The company Landshark keeps is none of your business, either. It's
also none of my business. I simply cited the posts and never mentioned
it again.
> And if the company you keep makes you act like a jackass towards
>other then yes it is our business.
If I acted like a jackass towards you then THAT'S your business. But
the fact is that I generally ignore you. And I haven't changed my
position regarding illegal radio. So what are you whining about? My
outspoken opposition to Bush? My grouchy attitude? A technical issue
while you were posting under a different name? I can't address your
personal problems with me if all you do is fuss about who I talk with
on a newsgroup (and it's pretty sad if that gets your panties is such
a bunch). So what's your beef?
I AmnotGeorgeBush
On Wed, 19 Jan 2005 11:15:34 -0500, twist...@webtv.net (Twistedhed)
wrote in <12354-41...@storefull-3258.bay.webtv.net>:
From: wïrenut@NOSPAMïcehouse.net (Frank Gilliland)
On Wed, 19 Jan 2005 11:12:07 GMT, "Trev 999" <tr...@hotmail.com> wrote
in <bgrHd.124638$K7.3...@news-server.bigpond.net.au>: Frank, what is
your email? I sent the schematic to Sideband by mistake, sorry SSB
I see you found my email. Excuse my
grouchiness but I've been drinking and
listening to pre-disco BeeGees through a
Sony amp. That being said......
>Didn't the Bee Gees actually ROCK back in
>the late sixties and early 70s?
Absolutely! Have you checked out some of their more recent work?
Sometimes a little retro but with the same unique 'sound'.
-
(We're showing our age here, guys. I had Bee Gee 8-track tapes in my
portable E-150 fuck palace back in the day. (mirror-ball?) )
Yea,,,you know the pathetic part? I still have a ton of 8 tracks in an
old footlocker..I have way more albums, but I still have a good
selection of 8 tracks. As of two years ago, my portable 8 track player
still worked. Last thing I listened to on 8 track was UFO "Light's Out".
Among my older tapes: Jefferson Airplane's "Bless it's Pointed Little
Head" and "Bathing at Baxters", Santana's "Abraxas", and a bootleg
Captain Beefheart.
Steveo
> On Wed, 19 Jan 2005 11:15:34 -0500, twist...@webtv.net (Twistedhed)
> wrote in <12354-41...@storefull-3258.bay.webtv.net>:
> On Wed, 19 Jan 2005 11:12:07 GMT, "Trev 999" <tr...@hotmail.com> wrote
> in <bgrHd.124638$K7.3...@news-server.bigpond.net.au>: Frank, what is
> your email? I sent the schematic to Sideband by mistake, sorry SSB
> _
> I see you found my email. Excuse my
> grouchiness but I've been drinking and
> listening to pre-disco BeeGees through a
> Sony amp. That being said......
>
>
> >Didn't the Bee Gees actually ROCK back in
>
> >the late sixties and early 70s?
>
> Absolutely! Have you checked out some of their more recent work?
> Sometimes a little retro but with the same unique 'sound'.
> -
> (We're showing our age here, guys. I had Bee Gee 8-track tapes in my
> portable E-150 fuck palace back in the day. (mirror-ball?) )
>
> Yea,,,you know the pathetic part? I still have a ton of 8 tracks in an
> old footlocker..I have way more albums, but I still have a good
> selection of 8 tracks. As of two years ago, my portable 8 track player
> still worked. Last thing I listened to on 8 track was UFO "Light's Out".
> Among my older tapes: Jefferson Airplane's "Bless it's Pointed Little
> Head" and "Bathing at Baxters", Santana's "Abraxas", and a bootleg
> Captain Beefheart.
>
Steveo
<snip you asshole>
I'VE HEARD ENOUGH ABOUT YOUR FUCKING RADIO........SHOVE IT UP YOUR FUCKING
ASS, FAG-BOI!!!!!!!!!!!!!!!!!
Steveo
Survey
dug88
what in the world is insane and what is OUTSANE
what?
so the virgin thought a blow job was his duty.
i never considered myself as a blowup rubber dolly.
shut up
i don't need no smart remarks
"Kensock d`Fehrett" <slappin...@ndhersocks.nert> wrote in message
news:m9r1v0lgmichc29f8...@4ax.com...
dug88
hey steveo
i KNOW you can think fast, & like that idea.
little violin concepts.
not even fundamental
can add terror of memories to any man on this planet.
some worse some less.
About the most powerfull and greatest thing that ever happened to you?
well, maybe some of the readers will ask
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>On Thu, 20 Jan 2005 13:15:35 -0500, Dave Hall <nojunk...@ptd.net>
>wrote in <tdsvu0dsvos26iqso...@4ax.com>:
>
><snip>
>>> I've never had one blow up just out of spite.
>>
>>I have. Usually they're in sweep tube CB amplifiers. For no other
>>reason than they just became "leaky", and start drawing current until
>>they heat up and explode.
>
>
>I don't believe that for a minute. I have worked on literally hundreds
>of pieces tube gear of all shapes and sizes and -NEVER- had a cap blow
>up for no reason, let alone overheating caused by leakage.
So, like you chastised me for in another thread, because you've never
witnessed it, it never happened?
You must not have worked on as many tube circuits as you've claimed.
The law of averages and statistical probability would conclude that
you should have seen at least one exploding cap had you worked on
enough equipment.
>But it's
>not my experience that tells me your story is bullshit, it's the fact
>that if the cap is leaking that much current in parallel with the
>normal load then other components will fail first, such as the filter
>resistor or choke, rectifier(s), transformer, etc.
Leaky doesn't mean "dead short". A cap could have just enough
resistance to heat it up and cause it to out gas or explode, but not
cause other components to fail.
Also, if you recall how the HV power supplies were designed in a
typical sweep tube "CB" grade amplifier, the plate supply voltage was
usually designed at about 800V give or take. The filter caps were
usually (2) 40 uF, 450V caps wired in series across the DC output.
Sometimes there will be (2) 100K or 1M ohm resistors in parallel with
the filter caps as well. What happens is that one of the two caps
becomes leaky. It starts to draw current, and it raises the voltage
potential across the other cap, until it exceeds the voltage rating
and it's usually THAT cap which explodes.
> Usually it's the
>fuse that goes first, and if there's no fuse then the transformer
>overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
>blow when subjected to too much AC, and even a few volts of ripple can
>be enough to pop a can on a tube rig.
It's the filter cap's job to smooth out ripple. Of course they will be
subjected to it. If what you say were true, then there would be caps
popping like popcorn. Ripple is not A.C., as the polarity never
reverses. it only "pulses" to some degree. The charge-discharge action
of the filter caps is what smooths these pulses out.
>This happens when the load is
>too heavy for the power supply, and the first cap to blow is the one
>on the load side of the pi filter (which is a very good indication of
>why it blew up). In the case of sweep-tube amps, I find that the most
>likely cause of your capacitor mishap.
I've never seen that happen for the reason's you've stated. An
excessive load on the power supply will draw the DC voltage down far
below the cap's rating. Since caps don't pass DC, that will not
affect them.
>
>
>> In my Lafayette Comstat 25 a resistor once
>>lit up like a candle due to a shorted cap in series with it.
>
>
>Now -that- I can believe.
Wow, alert the media......
>>The Uniden UPD-858 SSB boards (Realistic TRC-457/458, Cobra 138/139
>>XLR) had a pesky blue 10 uF tantalum cap that would short out and pull
>>the B+ feeding the driver and final stages to ground. I've seen that
>>on at least 3 different radios. I'd call that a statistic.
>
>
>I would call that a design flaw. Tantalum caps have often been used in
>applications where they were not a very good choice. They have certain
>limitations, but many engineers haven't learned what those limitations
>are, hence their reputation for failure. They are high-density caps
>but are VERY sensitive to damage from overvoltage. A digital circuit
>with a tightly regulated power supply is an appropriate application
>for tantalum caps -- a CB is not.
I'm sure the engineers at Uniden had their reasons. The MTBF for those
parts would seem to indicate that they were under rated for the job
though.
>>>Also, I -am- one of those "pros who restore old radios", which has
>>>been both a hobby and a business since I restored my first Philco 20
>>>when I was a teenager. And after being in the electronics business
>>>professionally for 30+ years I can say, without any doubt whatsoever,
>>>that capacitor failure rate (i.e, the capacitor being the -primary-
>>>cause of the failure) is no higher than for any other component.
>>
>>But no lower either. They certainly fail more often than resistors.
>
>
>Not when you realize that a light bulb is nothing more than a
>resistor
A light bulb is a resistor as a consequence of it's intended use. A
part designed as a resistor rarely fails, unless too much current is
drawn through it. In which case, the failure is quite obvious.
>>>I do
>>>-not- recommend blanket replacement of aged capacitors -- on the
>>>contrary, I recommend -testing- aged capacitors. If they are still
>>>within tolerance and show no DC leakage, leave them ALONE!
>>
>>I don't either as a rule. It's just a waste of money. But for you to
>>discard, as idiotic, my suggestion of a small electrolytic cap
>>changing value causing the problem in that Pierce Simpson radio is
>>nothing but pure arrogance Frank. I have had a similar problem that
>>was the fault of a cap.
>
>
>I didn't discard your example as idiotic. I discarded your presumption
>that a capacitor is responsible for the OP's problem. You once found a
>radio that had a bad cap. Well, so have I, and so has just about
>anyone that has fixed a few radios. But after you work with
>electrionics for 30 years you eventually learn that you can't jump to
>any conclusions.
I didn't jump to any conclusions. I offered another possible cause for
the failure. I offered my experience with a VERY similar failure as a
reason to consider it.
Yet you told me to "shut my trap" as if I said something that was
totally incapable of happening. Something which you have now softened
to at least acknowledging that it can happen, even if you believe that
it's not likely.
>You have to locate the faulty component(s) AND you
>must determine why it failed. A lot of caps do indeed go bad, but most
>of the time it's a secondary fault. Finding the primary fault is what
>seperates the professionals from the amateurs.
But what you fail to consider, and my experience has shown me to be
true, is that caps, especially electrolytic and tantalum, can and do
fail for no other reason than age. They ARE the primary failure.
Caps will change value with age, even if they don't completely open or
short. Sometimes a change in value is all that it takes to make an
audio amp into an oscillator.
>>>>> They start having problems
>>>>>when they are pushed over their rated voltage or physically leak; and
>>>>>most electrolytics made since the 50's don't use liquid electrolyte.
>>>>
>>>> I'm telling you what I've personally experienced in close to 30 years
>>>>of repairing CB radios. It is COMMON for small 1uF and such
>>>>electrolytic caps to open up from age. I just got done retelling an
>>>>exact problem that I once fixed. Now, either you are calling me a
>>>>liar, or you need to modify your statement, since my examples are
>>>>proof that maybe you should keep your trap shut, since CB radios don't
>>>>seem to be your high point in experience.
>>>
>>>
>>>Your examples are proof of nothing except your lack of experience in
>>>the broad field of electronics and your inability to find the
>>>underlying cause of the failure.
>>
>>The primary cause of the failure WAS the cap. When a cap in an audio
>>chain opens up so that audio no longer passes through, and there are
>>no voltages to speak of across it (Other than bias level), and when
>>the cap is replaced, normal function is restored, that's a pretty good
>>indication of what the cause of the failure was.
>No it's not. After replacing the cap it would have been a good idea to
>push the audio level rail-to-rail to see if the amp locks up, change
>freqs to see if there's any resonance that could swing the AC above
>the cap's rating, heat-run the radio for a few hours, and then -maybe-
>you could safely conclude that the cap was the primary failure.
That's normal procedure for any of my repairs. I always exercise the
radios for several days to make sure there are no hidden "gotcha's".
Sometimes I get a surprise and another intermittent component shows
its face. Usually though, what I replace is what went bad.
There is nothing more strenuous in the audio chain of a radio, than
the clipped and peaked operator yelling "auuuuuuuuudiooooo". If it
survives that, then I'd say the problem was fixed. It took 20 years
for the first cap to fail. I'm sure that radio hit the bone pile
before the cap opened up again.
>>>Cap failure as the -primary- cause of
>>>failure is no more common than a weak transistor, bad IC, blown
>>>diode/zener, or bad RF/IF can, or even poor manufacturing standards.
>>
>>Ah! So at least now you admit that it is equally likely that a cap can
>>fail as often as any particular active device. Very good, you're
>>making progress.
>
>
>I never denied it. I just didn't jump to the conclusion that it's a
>cap like you did.
You seem to have trouble differentiating between a conclusion and a
possibility.
Dave
"Sandbagger"
http://home.ptd.net/~n3cvj
Frank Gilliland
wrote in <0ucav0dpa3u9ajvq2...@4ax.com>:
>On Thu, 20 Jan 2005 12:49:50 -0800, Frank Gilliland
><wīrenut@NOSPAMīcehouse.net> wrote:
>
>>On Thu, 20 Jan 2005 13:15:35 -0500, Dave Hall <nojunk...@ptd.net>
>>wrote in <tdsvu0dsvos26iqso...@4ax.com>:
>>
>><snip>
>>>> I've never had one blow up just out of spite.
>>>
>>>I have. Usually they're in sweep tube CB amplifiers. For no other
>>>reason than they just became "leaky", and start drawing current until
>>>they heat up and explode.
>>
>>
>>I don't believe that for a minute. I have worked on literally hundreds
>>of pieces tube gear of all shapes and sizes and -NEVER- had a cap blow
>>up for no reason, let alone overheating caused by leakage.
>
>So, like you chastised me for in another thread, because you've never
>witnessed it, it never happened?
>
>You must not have worked on as many tube circuits as you've claimed.
>The law of averages and statistical probability would conclude that
>you should have seen at least one exploding cap had you worked on
>enough equipment.
I have seen, heard and smelled plenty of caps blow up. What -I- said,
and what you can't seem to comprehend, is that 1) they don't blow up
for no reason at all, and 2) that I have never had one blow up just
because it was leaking.
>>But it's
>>not my experience that tells me your story is bullshit, it's the fact
>>that if the cap is leaking that much current in parallel with the
>>normal load then other components will fail first, such as the filter
>>resistor or choke, rectifier(s), transformer, etc.
>
>Leaky doesn't mean "dead short". A cap could have just enough
>resistance to heat it up and cause it to out gas or explode, but not
>cause other components to fail.
Never seen it happen. It's easy to understand why it doesn't happen
when you realize that heat requires power; i.e, watts.
Begin with the power supply. It is designed to supply a limited amount
of power, usually a little more than for the intended use. Next, your
cap leaks. This creates a shunt in parallel with the intended load,
and therefore draws more power from the power supply. In fact, your
leaky cap draws so much -additional- power that it evaporates the
contents faster than the gasses can be vented by the vent hole -AND-
builds up enough pressure to literally explode the metal case. I don't
think so, Dave -- that's just too much power to not expect an earlier
failure in the power supply.
And like I said before, if the supply isn't fused then the failure is
probably going to be a shorted transformer or choke, blown rectifier,
or some other device that is less robust than the capacitor. Often the
caps themselves will fuse internally when subjected to too much
current, which is something many techs fail to consider when replacing
an open cap.
You might also realize that electrolytics blow up when subjected to a
reverse current -not- because of any heat buildup or dissipation, but
because the dielectric undergoes electrolysis and one of the products
is a gas. This takes much less power than brute-force heating of a
cap. If a low-voltage electrolytic blows up, the reason is almost
always due to reverse current. High-voltage electrolytics don't blow
up nearly as often because they are rarely subjected to reverse
current.
>Also, if you recall how the HV power supplies were designed in a
>typical sweep tube "CB" grade amplifier, the plate supply voltage was
>usually designed at about 800V give or take. The filter caps were
>usually (2) 40 uF, 450V caps wired in series across the DC output.
>Sometimes there will be (2) 100K or 1M ohm resistors in parallel with
>the filter caps as well. What happens is that one of the two caps
>becomes leaky. It starts to draw current, and it raises the voltage
>potential across the other cap, until it exceeds the voltage rating
>and it's usually THAT cap which explodes.
You're reaching, Dave. If the voltage across the 'other' cap is
exceeded what usually happens is the dielectric will puncture (at the
dielectric's puncture voltage) and cause a brief short as the cap
discharges. If the first discharge doesn't pop the fuse, or fuse the
cap internally, it will continue to charge and discharge just like a
relaxation oscillator, shunting (not dissipating) huge amounts of
power until something else fails, which usually happens in less than a
couple seconds. That's why you sometimes hear a short 'whine' before
the radio goes dead and smoke billows from the transformer.
>> Usually it's the
>>fuse that goes first, and if there's no fuse then the transformer
>>overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
>>blow when subjected to too much AC, and even a few volts of ripple can
>>be enough to pop a can on a tube rig.
>
>It's the filter cap's job to smooth out ripple. Of course they will be
>subjected to it. If what you say were true, then there would be caps
>popping like popcorn. Ripple is not A.C., as the polarity never
>reverses. it only "pulses" to some degree. The charge-discharge action
>of the filter caps is what smooths these pulses out.
Ripple is the AC component of a DC current. But call it whatever
tickles your fancy because the fact is that if a filter cap is doing
it's job of smoothing the ripple then it's not being subjected to very
much ripple! If there -is- much ripple then the cap isn't doing it's
job, the reason being either the cap is weak or the load is too big.
Whatever the cause, the cap sees that ripple as AC current which
causes heat dissipation in the cap due mostly to hysteresis (high
hysteresis being a characteristic of electrolytic caps).
Then capacitor go 'hiss'.
>>This happens when the load is
>>too heavy for the power supply, and the first cap to blow is the one
>>on the load side of the pi filter (which is a very good indication of
>>why it blew up). In the case of sweep-tube amps, I find that the most
>>likely cause of your capacitor mishap.
>
>I've never seen that happen for the reason's you've stated. An
>excessive load on the power supply will draw the DC voltage down far
>below the cap's rating. Since caps don't pass DC, that will not
>affect them.
You missed capacitor lab, didn't you? Excessive load will draw down
the DC voltage but -increase- the ripple.
>>> In my Lafayette Comstat 25 a resistor once
>>>lit up like a candle due to a shorted cap in series with it.
>>
>>
>>Now -that- I can believe.
>
>Wow, alert the media......
You mean that left-wing liberal media that failed to cover over 20
bloody years of civil war in Angola?
>>>The Uniden UPD-858 SSB boards (Realistic TRC-457/458, Cobra 138/139
>>>XLR) had a pesky blue 10 uF tantalum cap that would short out and pull
>>>the B+ feeding the driver and final stages to ground. I've seen that
>>>on at least 3 different radios. I'd call that a statistic.
>>
>>
>>I would call that a design flaw. Tantalum caps have often been used in
>>applications where they were not a very good choice. They have certain
>>limitations, but many engineers haven't learned what those limitations
>>are, hence their reputation for failure. They are high-density caps
>>but are VERY sensitive to damage from overvoltage. A digital circuit
>>with a tightly regulated power supply is an appropriate application
>>for tantalum caps -- a CB is not.
>
>I'm sure the engineers at Uniden had their reasons.
Blind faith.
> The MTBF for those
>parts would seem to indicate that they were under rated for the job
>though.
>
>>>>Also, I -am- one of those "pros who restore old radios", which has
>>>>been both a hobby and a business since I restored my first Philco 20
>>>>when I was a teenager. And after being in the electronics business
>>>>professionally for 30+ years I can say, without any doubt whatsoever,
>>>>that capacitor failure rate (i.e, the capacitor being the -primary-
>>>>cause of the failure) is no higher than for any other component.
>>>
>>>But no lower either. They certainly fail more often than resistors.
>>
>>
>>Not when you realize that a light bulb is nothing more than a
>>resistor
>
>A light bulb is a resistor as a consequence of it's intended use. A
>part designed as a resistor rarely fails, unless too much current is
>drawn through it. In which case, the failure is quite obvious.
A light bulb is most certainly designed as a resistor. If it didn't
have resistance it wouldn't work. A resistor necessarily dissipates
power, and in a light bulb that power is dissipated as light (and
heat). The only real question here is why household light bulbs fail
regularly while automotive light bulbs last for a much, much longer
time.
>>>>I do
>>>>-not- recommend blanket replacement of aged capacitors -- on the
>>>>contrary, I recommend -testing- aged capacitors. If they are still
>>>>within tolerance and show no DC leakage, leave them ALONE!
>>>
>>>I don't either as a rule. It's just a waste of money. But for you to
>>>discard, as idiotic, my suggestion of a small electrolytic cap
>>>changing value causing the problem in that Pierce Simpson radio is
>>>nothing but pure arrogance Frank. I have had a similar problem that
>>>was the fault of a cap.
>>
>>
>>I didn't discard your example as idiotic. I discarded your presumption
>>that a capacitor is responsible for the OP's problem. You once found a
>>radio that had a bad cap. Well, so have I, and so has just about
>>anyone that has fixed a few radios. But after you work with
>>electrionics for 30 years you eventually learn that you can't jump to
>>any conclusions.
>
>I didn't jump to any conclusions. I offered another possible cause for
>the failure. I offered my experience with a VERY similar failure as a
>reason to consider it.
You said, "I'm assuming it's an old 23 channel classic. With old
radios like that, the problem is almost always related to caps drying
out," which demonstrated your ignorance about several things. First,
the radio is a more modern 40 channel PLL model, contemporary with the
TRC-453 and PC-122. Second, low voltage electrolytics don't "dry out"
because they use a solid dielectric. Third, you went on to describe
your one single experience from which you inferred, "the problem is
almost always....". Well Dave, suppose you once burnt yourself with a
soldering iron. Using your logic, the reason other people have burns
"is almost always" because they burn themselves with soldering irons.
It doesn't work.
>Yet you told me to "shut my trap" as if I said something that was
>totally incapable of happening. Something which you have now softened
>to at least acknowledging that it can happen, even if you believe that
>it's not likely.
Ok, how do I say this in a language you can maybe understand.....
You inferred from your single experience that the fault is most likely
a failed capacitor. Yet because I rejected your conclusion (or, if you
prefer, your 'declaration of high probability'), you assert that I am
claiming the fault is -not- a capacitor. That is a perfect example of
the logical fallacy 'Argumentum ad Ignorantiam' which, by definition,
is ignorance.
>>You have to locate the faulty component(s) AND you
>>must determine why it failed. A lot of caps do indeed go bad, but most
>>of the time it's a secondary fault. Finding the primary fault is what
>>seperates the professionals from the amateurs.
>
>But what you fail to consider, and my experience has shown me to be
>true, is that caps, especially electrolytic and tantalum, can and do
>fail for no other reason than age. They ARE the primary failure.
A bad cap CAN BE a primary failure. But more often they are only a
secondary failure. That is what MY experience has shown ME, and part
of that experience includes work in failure analysis at HP. So shall
we compare resumes, Dave?
>Caps will change value with age, even if they don't completely open or
>short. Sometimes a change in value is all that it takes to make an
>audio amp into an oscillator.
Just about everything changes value with age; especially crystals,
semiconductors, and paper/oil caps. All three are known to change more
significantly with age than any other type of component, including
electrolytic capacitors (but excluding disposable devices, of course).
I'm sure you have seen sockets for ICs, relays, crystals, and maybe
even transistors; but when was the last time you saw a socket for a
capacitor?
Maybe you should take a look at alignment procedures sometime. A
frequently used test involves driving the audio hard enough to slam it
rail-to-rail at every stage from mic to speaker, usually by feeding
the mic input with 500 mV or more; i.e, significantly more than you
can generate by screaming into the mic. One of the purposes of the
test is to see if any stages lock up, a problem that is much more
common with high-gain OP-amps produced since the 70's.
Resonance is also an issue that needs to be appreciated. As you noted
before, components change value over time. So a circuit that was not
resonant at the time of manufacture may be resonant 20 years later.
It's not a common cause of failure but I have seen it a few times. The
scary part is that a few volts feeding a series resonant circuit can
generate a hundred volts or more in the middle -- not a good thing for
the high-impedance and voltage-sensitive IC's and FET's produced these
days. As these components become more common and equipment is designed
with tighter voltage tolerances, I suspect resonance will become an
increasing problem in the future.
> If it
>survives that, then I'd say the problem was fixed. It took 20 years
>for the first cap to fail. I'm sure that radio hit the bone pile
>before the cap opened up again.
>
>
>
>>>>Cap failure as the -primary- cause of
>>>>failure is no more common than a weak transistor, bad IC, blown
>>>>diode/zener, or bad RF/IF can, or even poor manufacturing standards.
>>>
>>>Ah! So at least now you admit that it is equally likely that a cap can
>>>fail as often as any particular active device. Very good, you're
>>>making progress.
>>
>>
>>I never denied it. I just didn't jump to the conclusion that it's a
>>cap like you did.
>
>You seem to have trouble differentiating between a conclusion and a
>possibility.
You seem to have trouble with the possibility that the fault could
just as easily be something other than what you learned from your one
and only experience with a similar problem.
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>>>I don't believe that for a minute. I have worked on literally hundreds
>>>of pieces tube gear of all shapes and sizes and -NEVER- had a cap blow
>>>up for no reason, let alone overheating caused by leakage.
>>
>>So, like you chastised me for in another thread, because you've never
>>witnessed it, it never happened?
>>
>>You must not have worked on as many tube circuits as you've claimed.
>>The law of averages and statistical probability would conclude that
>>you should have seen at least one exploding cap had you worked on
>>enough equipment.
>
>
>I have seen, heard and smelled plenty of caps blow up. What -I- said,
>and what you can't seem to comprehend, is that 1) they don't blow up
>for no reason at all, and 2) that I have never had one blow up just
>because it was leaking.
1. Of course they don't blow up for no reason. The reason they blow is
due to several factors, including high resistance leakage.
2. I HAVE seen them blow due to leakage. You can come up with all
sorts of ripple theories and other such to explain the exact reason
why it goes, but the primary event which set the chain into motion was
the cap going leaky.
When you open up a tube rig or an amplifier and have to clean out
shards of aluminum foil and paper wadding, you don't really care about
the exact chain of events. You just know that the cap blew. When you
replace it (Once you determine that the rest of the supply is fine),
and the radio plays fine, it's a safe assumption that the cap was bad.
>>>But it's
>>>not my experience that tells me your story is bullshit, it's the fact
>>>that if the cap is leaking that much current in parallel with the
>>>normal load then other components will fail first, such as the filter
>>>resistor or choke, rectifier(s), transformer, etc.
>>
>>Leaky doesn't mean "dead short". A cap could have just enough
>>resistance to heat it up and cause it to out gas or explode, but not
>>cause other components to fail.
>
>
>Never seen it happen. It's easy to understand why it doesn't happen
>when you realize that heat requires power; i.e, watts.
800V X .25 amps = 200 watts. More than enough to heat up a leaky cap.
>Begin with the power supply. It is designed to supply a limited amount
>of power, usually a little more than for the intended use.
And if the "intended use" is intermittent, (Like only when
transmitting) the supply is usually sitting there idle, supplying
little to no current.
>Next, your
>cap leaks. This creates a shunt in parallel with the intended load,
>and therefore draws more power from the power supply.
Not necessarily. If the supply is capably of supplying .5 amps at 800V
when operating, but a leaky cap only draws .2 amps while on standby,
that's more than enough to heat the cap without overtaxing the supply.
> In fact, your
>leaky cap draws so much -additional- power that it evaporates the
>contents faster than the gasses can be vented by the vent hole -AND-
>builds up enough pressure to literally explode the metal case.
That's exactly what happens.
> I don't
>think so, Dave -- that's just too much power to not expect an earlier
>failure in the power supply.
Not at all, considering the scenario I just presented. I can't count
the number of D&A Mavericks and Thunderbolt 305's and such that I've
replaced blown caps in, and not had any other fault.
>And like I said before, if the supply isn't fused then the failure is
>probably going to be a shorted transformer or choke, blown rectifier,
>or some other device that is less robust than the capacitor.
Sometimes, if the cap hangs on long enough and the resistance drops
further, the rectifiers blow too. But that's more rare.
> Often the
>caps themselves will fuse internally when subjected to too much
>current, which is something many techs fail to consider when replacing
>an open cap.
Yes, they do sometimes open up. The symptom of that is rather obvious,
when the receiver hums louder than the audio. Old caps dry out and
open up too. Not all of them leak and blow.
>You might also realize that electrolytics blow up when subjected to a
>reverse current -not- because of any heat buildup or dissipation, but
>because the dielectric undergoes electrolysis and one of the products
>is a gas.
Yes, I am fully aware of that. When I was a young pup on my first full
time job, I worked 2nd shift and the techs sometimes got bored when
their work was done, so we used to make "capacitor cannons". We'd
solder a set of leads to a radial lead cap (Usually 10 uF at 50V)
place it in a hunk of PVC pipe, hook it up to a power supply
backwards, aim it at our "enemy" and crank up the voltage. It made
quite a projectile. I used to play with the voltage sometimes and I
would often crank it up slowly just to see where the point of
avalanche current draw first started.
But I digress......
Answer me this, how does a functioning DC power supply reverse current
under typical operating conditions?
> This takes much less power than brute-force heating of a
>cap. If a low-voltage electrolytic blows up, the reason is almost
>always due to reverse current. High-voltage electrolytics don't blow
>up nearly as often because they are rarely subjected to reverse
>current.
Which makes that an unlikely reason for cap failure.
>>Also, if you recall how the HV power supplies were designed in a
>>typical sweep tube "CB" grade amplifier, the plate supply voltage was
>>usually designed at about 800V give or take. The filter caps were
>>usually (2) 40 uF, 450V caps wired in series across the DC output.
>>Sometimes there will be (2) 100K or 1M ohm resistors in parallel with
>>the filter caps as well. What happens is that one of the two caps
>>becomes leaky. It starts to draw current, and it raises the voltage
>>potential across the other cap, until it exceeds the voltage rating
>>and it's usually THAT cap which explodes.
>
>
>You're reaching, Dave.
Not at all. This is what happened on at least 2 other amps that I
worked on. One cap had exploded, and the other one was leaky (As
measured with a simple VOM). I concluded, based on the evidence, that
the leaky cap had caused the other cap to blow. The first time this
happened to me, I replaced the "exploded" cap only, and when I fired
up the amp, a short time later, the new cap exploded as well. Only
then did I learn to check the remaining caps for leakage. Once the
leaky caps were replaced, there was no further problem.
>If the voltage across the 'other' cap is
>exceeded what usually happens is the dielectric will puncture (at the
>dielectric's puncture voltage) and cause a brief short as the cap
>discharges. If the first discharge doesn't pop the fuse, or fuse the
>cap internally, it will continue to charge and discharge just like a
>relaxation oscillator, shunting (not dissipating) huge amounts of
>power until something else fails, which usually happens in less than a
>couple seconds. That's why you sometimes hear a short 'whine' before
>the radio goes dead and smoke billows from the transformer.
That "whine" is usually from a 12V to HV multi vibrator which is
forced to stop from excessive current draw. I've seen that on a couple
of Lafayette HA-250's.
>>> Usually it's the
>>>fuse that goes first, and if there's no fuse then the transformer
>>>overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
>>>blow when subjected to too much AC, and even a few volts of ripple can
>>>be enough to pop a can on a tube rig.
>>
>>It's the filter cap's job to smooth out ripple. Of course they will be
>>subjected to it. If what you say were true, then there would be caps
>>popping like popcorn. Ripple is not A.C., as the polarity never
>>reverses. it only "pulses" to some degree. The charge-discharge action
>>of the filter caps is what smooths these pulses out.
>
>
>Ripple is the AC component of a DC current.
Not exactly true. It is the pulsating D.C. current component. There is
no A.C. after a bridge rectifier.
> But call it whatever
>tickles your fancy because the fact is that if a filter cap is doing
>it's job of smoothing the ripple then it's not being subjected to very
>much ripple!
Exactly my point. What you've described is a classic chicken and egg
scenario. The cap can't be subjected to too much ripple if it's doing
it's job. If there is ripple, then the cap is usually bad.
> If there -is- much ripple then the cap isn't doing it's
>job, the reason being either the cap is weak or the load is too big.
Right! But if the voltage is normal, then you can assume the current
draw is not excessive.
>Whatever the cause, the cap sees that ripple as AC current which
>causes heat dissipation in the cap due mostly to hysteresis (high
>hysteresis being a characteristic of electrolytic caps).
Now it's my turn. I've never seen a cap blow from a supply being
dragged down to the point of excessive ripple. Over voltage yes,
reverse voltage yes, leaky cap yes. Too much ripple, no.
>>>This happens when the load is
>>>too heavy for the power supply, and the first cap to blow is the one
>>>on the load side of the pi filter (which is a very good indication of
>>>why it blew up). In the case of sweep-tube amps, I find that the most
>>>likely cause of your capacitor mishap.
>>
>>I've never seen that happen for the reason's you've stated. An
>>excessive load on the power supply will draw the DC voltage down far
>>below the cap's rating. Since caps don't pass DC, that will not
>>affect them.
>
>
>You missed capacitor lab, didn't you? Excessive load will draw down
>the DC voltage but -increase- the ripple.
I've never seen a cap blow from ripple. Never.
>>>>The Uniden UPD-858 SSB boards (Realistic TRC-457/458, Cobra 138/139
>>>>XLR) had a pesky blue 10 uF tantalum cap that would short out and pull
>>>>the B+ feeding the driver and final stages to ground. I've seen that
>>>>on at least 3 different radios. I'd call that a statistic.
>>>
>>>
>>>I would call that a design flaw. Tantalum caps have often been used in
>>>applications where they were not a very good choice. They have certain
>>>limitations, but many engineers haven't learned what those limitations
>>>are, hence their reputation for failure. They are high-density caps
>>>but are VERY sensitive to damage from overvoltage. A digital circuit
>>>with a tightly regulated power supply is an appropriate application
>>>for tantalum caps -- a CB is not.
>>
>>I'm sure the engineers at Uniden had their reasons.
>
>
>Blind faith.
Cost cutting.
>>>>But no lower either. They certainly fail more often than resistors.
>>>
>>>
>>>Not when you realize that a light bulb is nothing more than a
>>>resistor
>>
>>A light bulb is a resistor as a consequence of it's intended use. A
>>part designed as a resistor rarely fails, unless too much current is
>>drawn through it. In which case, the failure is quite obvious.
>
>
>A light bulb is most certainly designed as a resistor. If it didn't
>have resistance it wouldn't work.
Yes resistance is necessary for it's function. But it's not the
function of a light bulb to act as a resistor in a circuit. Light
bulbs are far too fragile for that use.
> A resistor necessarily dissipates
>power, and in a light bulb that power is dissipated as light (and
>heat). The only real question here is why household light bulbs fail
>regularly while automotive light bulbs last for a much, much longer
>time.
Higher voltage, and a smaller ,thinner and more shock prone filament.
>>>I didn't discard your example as idiotic. I discarded your presumption
>>>that a capacitor is responsible for the OP's problem. You once found a
>>>radio that had a bad cap. Well, so have I, and so has just about
>>>anyone that has fixed a few radios. But after you work with
>>>electrionics for 30 years you eventually learn that you can't jump to
>>>any conclusions.
>>
>>I didn't jump to any conclusions. I offered another possible cause for
>>the failure. I offered my experience with a VERY similar failure as a
>>reason to consider it.
>
>
>You said, "I'm assuming it's an old 23 channel classic.
Yes, I was more or less asking for confirmation.
> With old
>radios like that, the problem is almost always related to caps drying
>out," which demonstrated your ignorance about several things. First,
>the radio is a more modern 40 channel PLL model, contemporary with the
>TRC-453 and PC-122.
Ok, you could have just said that in the beginning.
> Second, low voltage electrolytics don't "dry out"
>because they use a solid dielectric.
That may be true, but they do change value and open up due to the
affects of humidity and age. What the process is that causes that to
happen, I'm not sure of the exact name. But "drying out" is a common
term that we use to describe it. If I am technically in error, then so
be it.
> Third, you went on to describe
>your one single experience from which you inferred, "the problem is
>almost always....". Well Dave, suppose you once burnt yourself with a
>soldering iron. Using your logic, the reason other people have burns
>"is almost always" because they burn themselves with soldering irons.
>It doesn't work.
Ok, I'll admit that using the term "almost always" was a bit strong.
But you erred when you made the claim that it was a "single"
experience. I have had a few experiences with older radios where this
has happened. I also had an old SBE were a noisy receiver was traced
to a germanium transistor going bad. There is no hard fast rules, and
troubleshooting a radio over the internet is an exercise in throwing
darts.
All I wanted to do was to provide other alternatives, based on my
extensive experience in this business. The bottom line is to help
other people who have problems. It's not supposed to be a chest
pounding competition on who's had the most experience.
>>Yet you told me to "shut my trap" as if I said something that was
>>totally incapable of happening. Something which you have now softened
>>to at least acknowledging that it can happen, even if you believe that
>>it's not likely.
>
>
>Ok, how do I say this in a language you can maybe understand.....
>
>You inferred from your single experience that the fault is most likely
>a failed capacitor.
Not single. And the symptoms were VERY similar.
>Yet because I rejected your conclusion (or, if you
>prefer, your 'declaration of high probability'), you assert that I am
>claiming the fault is -not- a capacitor. That is a perfect example of
>the logical fallacy 'Argumentum ad Ignorantiam' which, by definition,
>is ignorance.
Not at all. If I claim that there is a high probability of component
"A" failing and you disagree. Then you are saying, in essence, that
there is a very low probability that component "A" is the failure.
If you consider component "A" to have a low probability of failure it
stands to reason that you place a higher probability on another
component failing.
Inductive reasoning.
>>>You have to locate the faulty component(s) AND you
>>>must determine why it failed. A lot of caps do indeed go bad, but most
>>>of the time it's a secondary fault. Finding the primary fault is what
>>>seperates the professionals from the amateurs.
>>
>>But what you fail to consider, and my experience has shown me to be
>>true, is that caps, especially electrolytic and tantalum, can and do
>>fail for no other reason than age. They ARE the primary failure.
>
>
>A bad cap CAN BE a primary failure. But more often they are only a
>secondary failure. That is what MY experience has shown ME, and part
>of that experience includes work in failure analysis at HP. So shall
>we compare resumes, Dave?
More chest thumping? What you've done at HP has no bearing on what
I've seen in 30+ years of CB radio repair.
>>Caps will change value with age, even if they don't completely open or
>>short. Sometimes a change in value is all that it takes to make an
>>audio amp into an oscillator.
>>
>>>No it's not. After replacing the cap it would have been a good idea to
>>>push the audio level rail-to-rail to see if the amp locks up, change
>>>freqs to see if there's any resonance that could swing the AC above
>>>the cap's rating, heat-run the radio for a few hours, and then -maybe-
>>>you could safely conclude that the cap was the primary failure.
>>
>>That's normal procedure for any of my repairs. I always exercise the
>>radios for several days to make sure there are no hidden "gotcha's".
>>Sometimes I get a surprise and another intermittent component shows
>>its face. Usually though, what I replace is what went bad.
>>
>>There is nothing more strenuous in the audio chain of a radio, than
>>the clipped and peaked operator yelling "auuuuuuuuudiooooo".
>Maybe you should take a look at alignment procedures sometime. A
>frequently used test involves driving the audio hard enough to slam it
>rail-to-rail at every stage from mic to speaker, usually by feeding
>the mic input with 500 mV or more; i.e, significantly more than you
>can generate by screaming into the mic.
While I can understand why you might want to do that, I've never seen
it in any of the factory alignment procedures or in the SAM's
photofacts.
Often the only reason given in the procedures to increase the audio
input (And I have a nice Tektronix generator) is to verify proper
operation of the modulation limiter.
> One of the purposes of the
>test is to see if any stages lock up, a problem that is much more
>common with high-gain OP-amps produced since the 70's.
Most CB radios made in the 1960- 1980 time frame do not use Op-amps.
Some Uniden radios used them more recently, but not as audio amps.
They were used as AGC amps and T/R switching.
I have never seen an audio stage on a CB radio "lock up". The Hy-Gain
623 used to have an interestingly annoying "design flaw" where if the
modulator was driven too hard, it would shut down for a few seconds.
You would loose receive audio until it recovered.
>Resonance is also an issue that needs to be appreciated. As you noted
>before, components change value over time. So a circuit that was not
>resonant at the time of manufacture may be resonant 20 years later.
>It's not a common cause of failure but I have seen it a few times.
You should see what happens when a 7812 regulator goes into
oscillation due to bad/missing bypass caps.
> The
>scary part is that a few volts feeding a series resonant circuit can
>generate a hundred volts or more in the middle -- not a good thing for
>the high-impedance and voltage-sensitive IC's and FET's produced these
>days. As these components become more common and equipment is designed
>with tighter voltage tolerances, I suspect resonance will become an
>increasing problem in the future.
By then CB (and probably ham) radio will be a memory. Maybe not. It
would seem that many CB radio designs have not changed all that much
from 1980. While the rest of the industry has embraced surface mount
technology and LSI chips, CB radio is still using leaded components
and discrete parts. From a repair standpoint, this is a good thing.
But I'm sure the radios could be made much cheaper (and smaller) if
they went with newer technologies.
>>>>Ah! So at least now you admit that it is equally likely that a cap can
>>>>fail as often as any particular active device. Very good, you're
>>>>making progress.
>>>
>>>
>>>I never denied it. I just didn't jump to the conclusion that it's a
>>>cap like you did.
>>
>>You seem to have trouble differentiating between a conclusion and a
>>possibility.
>
>
>You seem to have trouble with the possibility that the fault could
>just as easily be something other than what you learned from your one
>and only experience with a similar problem.
Once again, it was not "one and only". And I never said that a cap
was THE only possible failure. I offered it as a suggested course of
action. Something you jumped on me for in a very unprofessional and
crass manner.
Dave
"Sandbagger"
Frank Gilliland
wrote in <5uvcv0haj7ml7kre6...@4ax.com>:
Maybe that's happened to you but never on my bench, not unless the cap
was too small for the load. Like I said before, I have never seen a
cap that blew up on it's own accord, leaky or not (and I don't know
how you can determine it was leakage that caused your cap to blow
unless you have a time machine in your shop).
>>>>But it's
>>>>not my experience that tells me your story is bullshit, it's the fact
>>>>that if the cap is leaking that much current in parallel with the
>>>>normal load then other components will fail first, such as the filter
>>>>resistor or choke, rectifier(s), transformer, etc.
>>>
>>>Leaky doesn't mean "dead short". A cap could have just enough
>>>resistance to heat it up and cause it to out gas or explode, but not
>>>cause other components to fail.
>>
>>
>>Never seen it happen. It's easy to understand why it doesn't happen
>>when you realize that heat requires power; i.e, watts.
>
>800V X .25 amps = 200 watts. More than enough to heat up a leaky cap.
Heat it? Yes. Will it pass that much current without fusing
internally? Doubtful. If it -could- pass that much current would that
be enough power to make it explode? Hardly, since a lot of that power
is shunted to ground and not dissipated in the cap. It will just vent
until it's empty or shorted.
>>Begin with the power supply. It is designed to supply a limited amount
>>of power, usually a little more than for the intended use.
>
>And if the "intended use" is intermittent, (Like only when
>transmitting) the supply is usually sitting there idle, supplying
>little to no current.
>
>>Next, your
>>cap leaks. This creates a shunt in parallel with the intended load,
>>and therefore draws more power from the power supply.
>
>Not necessarily. If the supply is capably of supplying .5 amps at 800V
>when operating, but a leaky cap only draws .2 amps while on standby,
>that's more than enough to heat the cap without overtaxing the supply.
And you are forgetting that the bigger the load the bigger the cap,
and therefore the more power it would take for a leak to heat that cap
to the point of exploding.
>> In fact, your
>>leaky cap draws so much -additional- power that it evaporates the
>>contents faster than the gasses can be vented by the vent hole -AND-
>>builds up enough pressure to literally explode the metal case.
>
>That's exactly what happens.
Uh-huh.
>> I don't
>>think so, Dave -- that's just too much power to not expect an earlier
>>failure in the power supply.
>
>Not at all, considering the scenario I just presented. I can't count
>the number of D&A Mavericks and Thunderbolt 305's and such that I've
>replaced blown caps in, and not had any other fault.
Did you ever think that maybe the cap wasn't large enough for the
application?
>>And like I said before, if the supply isn't fused then the failure is
>>probably going to be a shorted transformer or choke, blown rectifier,
>>or some other device that is less robust than the capacitor.
>
>Sometimes, if the cap hangs on long enough and the resistance drops
>further, the rectifiers blow too. But that's more rare.
It -is- rare because the caps usually fuse internally before anything
else goes.
>> Often the
>>caps themselves will fuse internally when subjected to too much
>>current, which is something many techs fail to consider when replacing
>>an open cap.
>
>Yes, they do sometimes open up. The symptom of that is rather obvious,
>when the receiver hums louder than the audio. Old caps dry out and
>open up too. Not all of them leak and blow.
..........
>>You might also realize that electrolytics blow up when subjected to a
>>reverse current -not- because of any heat buildup or dissipation, but
>>because the dielectric undergoes electrolysis and one of the products
>>is a gas.
>
>Yes, I am fully aware of that. When I was a young pup on my first full
>time job, I worked 2nd shift and the techs sometimes got bored when
>their work was done, so we used to make "capacitor cannons". We'd
>solder a set of leads to a radial lead cap (Usually 10 uF at 50V)
>place it in a hunk of PVC pipe, hook it up to a power supply
>backwards, aim it at our "enemy" and crank up the voltage. It made
>quite a projectile. I used to play with the voltage sometimes and I
>would often crank it up slowly just to see where the point of
>avalanche current draw first started.
>
>
>But I digress......
>
>Answer me this, how does a functioning DC power supply reverse current
>under typical operating conditions?
If the cap is subjected to reverse current then the operating
conditions are not typical, are they? But if you're asking how a low
voltage cap can be subjected to reverse current then there are a
myriad of possibilities, the most obvious and most common is that the
power supply was connected backwards. Also, electrolytics are not
always fixed at ground potential, such as when they are used as a DC
block in an audio line. If the offset fails, the cap pops. There are
also instances where an abnormal AC component swings the normally
positive side of the cap negative. The cap voltage can get reversed by
rectification when a pull-up resistor fails. Etc, etc. Geez, Dave,
you've been doing this for 30 years and you didn't know this stuff?
>> This takes much less power than brute-force heating of a
>>cap. If a low-voltage electrolytic blows up, the reason is almost
>>always due to reverse current. High-voltage electrolytics don't blow
>>up nearly as often because they are rarely subjected to reverse
>>current.
>
>Which makes that an unlikely reason for cap failure.
It means high-voltage electrolytics are less likely to fail than
low-voltage electrolytics. Nothing more.
>>>Also, if you recall how the HV power supplies were designed in a
>>>typical sweep tube "CB" grade amplifier, the plate supply voltage was
>>>usually designed at about 800V give or take. The filter caps were
>>>usually (2) 40 uF, 450V caps wired in series across the DC output.
>>>Sometimes there will be (2) 100K or 1M ohm resistors in parallel with
>>>the filter caps as well. What happens is that one of the two caps
>>>becomes leaky. It starts to draw current, and it raises the voltage
>>>potential across the other cap, until it exceeds the voltage rating
>>>and it's usually THAT cap which explodes.
>>
>>
>>You're reaching, Dave.
>
>Not at all. This is what happened on at least 2 other amps that I
>worked on. One cap had exploded, and the other one was leaky (As
>measured with a simple VOM). I concluded, based on the evidence,
.....LOL!
> that
>the leaky cap had caused the other cap to blow. The first time this
>happened to me, I replaced the "exploded" cap only, and when I fired
>up the amp, a short time later, the new cap exploded as well. Only
>then did I learn to check the remaining caps for leakage. Once the
>leaky caps were replaced, there was no further problem.
I think you are making that up, Dave.
>>If the voltage across the 'other' cap is
>>exceeded what usually happens is the dielectric will puncture (at the
>>dielectric's puncture voltage) and cause a brief short as the cap
>>discharges. If the first discharge doesn't pop the fuse, or fuse the
>>cap internally, it will continue to charge and discharge just like a
>>relaxation oscillator, shunting (not dissipating) huge amounts of
>>power until something else fails, which usually happens in less than a
>>couple seconds. That's why you sometimes hear a short 'whine' before
>>the radio goes dead and smoke billows from the transformer.
>
>That "whine" is usually from a 12V to HV multi vibrator which is
>forced to stop from excessive current draw. I've seen that on a couple
>of Lafayette HA-250's.
I have heard that "whine" from equipment that had no vibrator,
mechanical -or- solid-state. It comes from the cap. If you don't
believe me, do the following experiment: Take a good cap, put it in
series with a resistor of, say, 100 to 1k ohms, then slowly push the
voltage up above the rated voltage. I would suggest using a cap of
only 100-150 volts because some of them won't puncture until they get
up around 400-500 volts. Anyway, at some point you hear the capacitor
begin to whine, and if the whine doesn't stop right away the resistor
will commence smoking (unless your power supply blows first). Try it.
It's fun!
>>>> Usually it's the
>>>>fuse that goes first, and if there's no fuse then the transformer
>>>>overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
>>>>blow when subjected to too much AC, and even a few volts of ripple can
>>>>be enough to pop a can on a tube rig.
>>>
>>>It's the filter cap's job to smooth out ripple. Of course they will be
>>>subjected to it. If what you say were true, then there would be caps
>>>popping like popcorn. Ripple is not A.C., as the polarity never
>>>reverses. it only "pulses" to some degree. The charge-discharge action
>>>of the filter caps is what smooths these pulses out.
>>
>>
>>Ripple is the AC component of a DC current.
>
>Not exactly true. It is the pulsating D.C. current component. There is
>no A.C. after a bridge rectifier.
Like I said, call it whatever you like.
>> But call it whatever
>>tickles your fancy because the fact is that if a filter cap is doing
>>it's job of smoothing the ripple then it's not being subjected to very
>>much ripple!
>
>Exactly my point. What you've described is a classic chicken and egg
>scenario.
The rooster came first.
> The cap can't be subjected to too much ripple if it's doing
>it's job. If there is ripple, then the cap is usually bad.
Not necessarily. A heavy load will increase the ripple.
>> If there -is- much ripple then the cap isn't doing it's
>>job, the reason being either the cap is weak or the load is too big.
>
>Right! But if the voltage is normal, then you can assume the current
>draw is not excessive.
Again, not necessarily. The DC component may stay within norms but the
AC component (ripple) will increase under load. That's just a fact of
life. Haven't you ever studied the basics of power supplies?
>>Whatever the cause, the cap sees that ripple as AC current which
>>causes heat dissipation in the cap due mostly to hysteresis (high
>>hysteresis being a characteristic of electrolytic caps).
>
>Now it's my turn. I've never seen a cap blow from a supply being
>dragged down to the point of excessive ripple. Over voltage yes,
>reverse voltage yes, leaky cap yes. Too much ripple, no.
Yet you claim that you can definitively determine if a completely
destroyed capacitor blew up because it was leaking. So how do you do
that? Do you reconstruct the capacitor like the FAA reconstructs
planes? Because that's what -they- need to do in order to determine
the cause of failure when there's no other obvious cause. Maybe they
should just hire you and your crystal ball, huh?
>>>>This happens when the load is
>>>>too heavy for the power supply, and the first cap to blow is the one
>>>>on the load side of the pi filter (which is a very good indication of
>>>>why it blew up). In the case of sweep-tube amps, I find that the most
>>>>likely cause of your capacitor mishap.
>>>
>>>I've never seen that happen for the reason's you've stated. An
>>>excessive load on the power supply will draw the DC voltage down far
>>>below the cap's rating. Since caps don't pass DC, that will not
>>>affect them.
>>
>>
>>You missed capacitor lab, didn't you? Excessive load will draw down
>>the DC voltage but -increase- the ripple.
>
>I've never seen a cap blow from ripple. Never.
You probably have but didn't know it. You can create and measure a
controlled failure with a couple good chokes for AC isolation, a
couple large motor-run capacitors for DC isolation, and an audio amp
to supply the AC. If you need a schematic just let me know.
Wanna bet? I use them all the time for dummy loads while messing with
my generator experiments. They also make great regulators because of
their non-linear characteristics. And for the same reason, miniature
light bulbs make pretty decent audio compressors. So you still say
that light bulbs aren't resistors? Hogwash. If it's useful as a
resistor then it's a resistor.
>> A resistor necessarily dissipates
>>power, and in a light bulb that power is dissipated as light (and
>>heat). The only real question here is why household light bulbs fail
>>regularly while automotive light bulbs last for a much, much longer
>>time.
>
>Higher voltage, and a smaller ,thinner and more shock prone filament.
Tungsten, one of the strongest metals on the periodic chart, is "shock
prone"? And what difference does the voltage make? It's the current
that causes the filament to heat up. Which is true regardless of it's
thickness, too.
Anyway, it was a rhetorical question.
>>>>I didn't discard your example as idiotic. I discarded your presumption
>>>>that a capacitor is responsible for the OP's problem. You once found a
>>>>radio that had a bad cap. Well, so have I, and so has just about
>>>>anyone that has fixed a few radios. But after you work with
>>>>electrionics for 30 years you eventually learn that you can't jump to
>>>>any conclusions.
>>>
>>>I didn't jump to any conclusions. I offered another possible cause for
>>>the failure. I offered my experience with a VERY similar failure as a
>>>reason to consider it.
>>
>>
>>You said, "I'm assuming it's an old 23 channel classic.
>
>Yes, I was more or less asking for confirmation.
You don't have a web browser?
>> With old
>>radios like that, the problem is almost always related to caps drying
>>out," which demonstrated your ignorance about several things. First,
>>the radio is a more modern 40 channel PLL model, contemporary with the
>>TRC-453 and PC-122.
>
>Ok, you could have just said that in the beginning.
Or you could have simply looked it up through google like I did.
>> Second, low voltage electrolytics don't "dry out"
>>because they use a solid dielectric.
>
>That may be true, but they do change value and open up due to the
>affects of humidity and age.
Age, yes, they do change to a greater or lesser degree. Usually
lesser. Humidity? I don't think so, Dave. Not when they are
hermetically sealed. And the only caps that aren't hermetically sealed
are air caps. Have you ever seen an "air electrolytic"? I haven't.
> What the process is that causes that to
>happen, I'm not sure of the exact name. But "drying out" is a common
>term that we use to describe it. If I am technically in error, then so
>be it.
Who is "we"?
>> Third, you went on to describe
>>your one single experience from which you inferred, "the problem is
>>almost always....". Well Dave, suppose you once burnt yourself with a
>>soldering iron. Using your logic, the reason other people have burns
>>"is almost always" because they burn themselves with soldering irons.
>>It doesn't work.
>
>Ok, I'll admit that using the term "almost always" was a bit strong.
It was a bit stronger than "a bit strong". You stated it as if it were
a rule of thumb.
>But you erred when you made the claim that it was a "single"
>experience.
"I once had an old 23....."
> I have had a few experiences with older radios where this
>has happened. I also had an old SBE were a noisy receiver was traced
>to a germanium transistor going bad. There is no hard fast rules, and
>troubleshooting a radio over the internet is an exercise in throwing
>darts.
>
>All I wanted to do was to provide other alternatives, based on my
>extensive experience in this business. The bottom line is to help
>other people who have problems. It's not supposed to be a chest
>pounding competition on who's had the most experience.
If you truly wanted to provide other alternatives then you probably
shouldn't have limited your "alternative" to the exclusion of any
others. Trouble-shooting by remote control may be difficult but it's
far from "throwing darts". You just have to refrain from jumping to
conclusions.
>>>Yet you told me to "shut my trap" as if I said something that was
>>>totally incapable of happening. Something which you have now softened
>>>to at least acknowledging that it can happen, even if you believe that
>>>it's not likely.
>>
>>
>>Ok, how do I say this in a language you can maybe understand.....
>>
>>You inferred from your single experience that the fault is most likely
>>a failed capacitor.
>
>Not single. And the symptoms were VERY similar.
A burn from a soldering iron is VERY similar to a burn from a
cigarette. Believe me, I've done both too many times.
>>Yet because I rejected your conclusion (or, if you
>>prefer, your 'declaration of high probability'), you assert that I am
>>claiming the fault is -not- a capacitor. That is a perfect example of
>>the logical fallacy 'Argumentum ad Ignorantiam' which, by definition,
>>is ignorance.
>
>Not at all. If I claim that there is a high probability of component
>"A" failing and you disagree. Then you are saying, in essence, that
>there is a very low probability that component "A" is the failure.
Wrong. Disagreement is -not- an assertion of the opposite. This is the
same fallacy you invoke in your political arguments: You label me as a
liberal because I do not agree with your admitted conservatism, which
is also wrong. For example, if you believe in God and I state that I'm
an athiest, that doesn't mean I hold satanic rituals in my basement
(or anywhere else).
If you want to assert yourself as a person who can think logically
then you need to understand logical fallacies enough not to make them.
>If you consider component "A" to have a low probability of failure it
>stands to reason that you place a higher probability on another
>component failing.
Again wrong, and for the same reason.
>Inductive reasoning.
Before you make any more logical leaps, let me send you a good book on
the subject. I'll pay shipping.
>>>>You have to locate the faulty component(s) AND you
>>>>must determine why it failed. A lot of caps do indeed go bad, but most
>>>>of the time it's a secondary fault. Finding the primary fault is what
>>>>seperates the professionals from the amateurs.
>>>
>>>But what you fail to consider, and my experience has shown me to be
>>>true, is that caps, especially electrolytic and tantalum, can and do
>>>fail for no other reason than age. They ARE the primary failure.
>>
>>
>>A bad cap CAN BE a primary failure. But more often they are only a
>>secondary failure. That is what MY experience has shown ME, and part
>>of that experience includes work in failure analysis at HP. So shall
>>we compare resumes, Dave?
>
>More chest thumping? What you've done at HP has no bearing on what
>I've seen in 30+ years of CB radio repair.
Really? Do capacitors behave differently in CB radios than they do in
spectrum analyzers and signal generators?
>>>Caps will change value with age, even if they don't completely open or
>>>short. Sometimes a change in value is all that it takes to make an
>>>audio amp into an oscillator.
>
>
>>>
>>>>No it's not. After replacing the cap it would have been a good idea to
>>>>push the audio level rail-to-rail to see if the amp locks up, change
>>>>freqs to see if there's any resonance that could swing the AC above
>>>>the cap's rating, heat-run the radio for a few hours, and then -maybe-
>>>>you could safely conclude that the cap was the primary failure.
>>>
>>>That's normal procedure for any of my repairs. I always exercise the
>>>radios for several days to make sure there are no hidden "gotcha's".
>>>Sometimes I get a surprise and another intermittent component shows
>>>its face. Usually though, what I replace is what went bad.
>>>
>>>There is nothing more strenuous in the audio chain of a radio, than
>>>the clipped and peaked operator yelling "auuuuuuuuudiooooo".
>
>
>>Maybe you should take a look at alignment procedures sometime. A
>>frequently used test involves driving the audio hard enough to slam it
>>rail-to-rail at every stage from mic to speaker, usually by feeding
>>the mic input with 500 mV or more; i.e, significantly more than you
>>can generate by screaming into the mic.
>
>While I can understand why you might want to do that, I've never seen
>it in any of the factory alignment procedures or in the SAM's
>photofacts.
Well, I just got done fixing the TRC-449 I got from Lance (thanks
again! Got the front-end fixed and receive is hot! FWIW, the second RF
amp TR-5 was the culprit as I suspected -- it's a 2SC1730 and it's
Vceo is just too low for the radio, a design flaw) and have the manual
right beside me. On page 15 is the transmitter alignment section, and
about the middle of the page is Step 4 where the procedure is to align
the transmit coils while shoving 500 mV into the mic jack. Yet Step 8
is to set the modulation for 98% with a 200 mV input. Now there is a
reason for driving the audio so hard in Step 4 and that is to make
sure the signal path can handle the stress, because if it can cope
with a 500 mV input then a mere 200 mV should be well within the safe
operating area.
Now you claim to have worked on CBs for 30 years, yet you have never
seen a procedure as I just described (on what was once a fairly common
and popular radio)?
>Often the only reason given in the procedures to increase the audio
>input (And I have a nice Tektronix generator) is to verify proper
>operation of the modulation limiter.
I can send a pair of reading glasses with that book if you want.
>> One of the purposes of the
>>test is to see if any stages lock up, a problem that is much more
>>common with high-gain OP-amps produced since the 70's.
>
>Most CB radios made in the 1960- 1980 time frame do not use Op-amps.
.....uh, yeah. That's perfectly consistent with what I said. What's
your point?
>Some Uniden radios used them more recently, but not as audio amps.
>They were used as AGC amps and T/R switching.
Most OP-amps in modern radios (CB and other) are integrated into
special-purpose chips. Some older radios had OP-amps made with
discrete components.
>I have never seen an audio stage on a CB radio "lock up". The Hy-Gain
>623 used to have an interestingly annoying "design flaw" where if the
>modulator was driven too hard, it would shut down for a few seconds.
>You would loose receive audio until it recovered.
That sounds like a lock-up to me.
>>Resonance is also an issue that needs to be appreciated. As you noted
>>before, components change value over time. So a circuit that was not
>>resonant at the time of manufacture may be resonant 20 years later.
>>It's not a common cause of failure but I have seen it a few times.
>
>You should see what happens when a 7812 regulator goes into
>oscillation due to bad/missing bypass caps.
You should see my curio cabinet of failed components. It even includes
a shorted resistor (47K, carbon comp), something I was always taught
was absolutely impossible. But I show it off to visitors who stop by
and don't believe me, have them measure it for themselves, and they
walk away with a paranoid glaze over their eyes.
Ken
"Frank Gilliland" <wīrenut@NOSPAMīcehouse.net> wrote in message
<BITCH SNIP>
Frank,I want to shove my main component deep into your love hole.
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>On Tue, 25 Jan 2005 13:22:02 -0500, Dave Hall <nojunk...@ptd.net>
>wrote in <5uvcv0haj7ml7kre6...@4ax.com>:
>
>>On Mon, 24 Jan 2005 15:53:20 -0800, Frank Gilliland
>><wīrenut@NOSPAMīcehouse.net> wrote:
>>1. Of course they don't blow up for no reason. The reason they blow is
>>due to several factors, including high resistance leakage.
>>
>>2. I HAVE seen them blow due to leakage. You can come up with all
>>sorts of ripple theories and other such to explain the exact reason
>>why it goes, but the primary event which set the chain into motion was
>>the cap going leaky.
>>
>>When you open up a tube rig or an amplifier and have to clean out
>>shards of aluminum foil and paper wadding, you don't really care about
>>the exact chain of events. You just know that the cap blew. When you
>>replace it (Once you determine that the rest of the supply is fine),
>>and the radio plays fine, it's a safe assumption that the cap was bad.
>
>
>Maybe that's happened to you but never on my bench, not unless the cap
>was too small for the load. Like I said before, I have never seen a
>cap that blew up on it's own accord, leaky or not (and I don't know
>how you can determine it was leakage that caused your cap to blow
>unless you have a time machine in your shop).
Shhhhh. That a "way back" machine, and it's only a prototype.
But you make a good point. But like Sherlock Holmes once said, when
all the other possibilities are excluded, what remains, however
unlikely must be the answer. The supply was otherwise in good shape,
the load was nominal. The only thing "bad" was the cap.
Oh, and I've seen equipment which did have leaky caps, and they were
hot to the touch and some have had the rubber plug in the bottom
partially ejected. They didn't fail violently in an explosion, but
they did out gas from excessive leakage.
>>>>Leaky doesn't mean "dead short". A cap could have just enough
>>>>resistance to heat it up and cause it to out gas or explode, but not
>>>>cause other components to fail.
>>>
>>>
>>>Never seen it happen. It's easy to understand why it doesn't happen
>>>when you realize that heat requires power; i.e, watts.
>>
>>800V X .25 amps = 200 watts. More than enough to heat up a leaky cap.
>
>
>Heat it? Yes. Will it pass that much current without fusing
>internally? Doubtful. If it -could- pass that much current would that
>be enough power to make it explode?
The evidence that I've seen would tend to answer that question as a
yes.
> Hardly, since a lot of that power
>is shunted to ground and not dissipated in the cap.
A dead short is a shunt to ground. When the cap is simply leaking, the
cap becomes a resistor, and heats up like one when too much current
runs through it. It would only take a leakage resistance of around 4K
Ohms at 800 V to make it happen.
> It will just vent
>until it's empty or shorted.
That depends on the design of the cap, and how much current is being
dissipated. Explain it anyway you want. I'm telling you what I've
actually seen.
>>>Begin with the power supply. It is designed to supply a limited amount
>>>of power, usually a little more than for the intended use.
>>
>>And if the "intended use" is intermittent, (Like only when
>>transmitting) the supply is usually sitting there idle, supplying
>>little to no current.
>>
>>>Next, your
>>>cap leaks. This creates a shunt in parallel with the intended load,
>>>and therefore draws more power from the power supply.
>>
>>Not necessarily. If the supply is capably of supplying .5 amps at 800V
>>when operating, but a leaky cap only draws .2 amps while on standby,
>>that's more than enough to heat the cap without overtaxing the supply.
>
>
>And you are forgetting that the bigger the load the bigger the cap,
>and therefore the more power it would take for a leak to heat that cap
>to the point of exploding.
Caps are rated by three things, capacitance, voltage rating and
temperature. In most sweep tube amps and tube CB radios, I rarely see
more than about 40 uF in filter caps. Voltage rating depends on the
circuit. Most amps use 450V caps in series.
>>> In fact, your
>>>leaky cap draws so much -additional- power that it evaporates the
>>>contents faster than the gasses can be vented by the vent hole -AND-
>>>builds up enough pressure to literally explode the metal case.
>>
>>That's exactly what happens.
>
>
>Uh-huh.
>
>
>>> I don't
>>>think so, Dave -- that's just too much power to not expect an earlier
>>>failure in the power supply.
>>
>>Not at all, considering the scenario I just presented. I can't count
>>the number of D&A Mavericks and Thunderbolt 305's and such that I've
>>replaced blown caps in, and not had any other fault.
>
>
>Did you ever think that maybe the cap wasn't large enough for the
>application?
Perhaps. Then I guess the manufacturer has some 'splainin to do.
>>>And like I said before, if the supply isn't fused then the failure is
>>>probably going to be a shorted transformer or choke, blown rectifier,
>>>or some other device that is less robust than the capacitor.
>>
>>Sometimes, if the cap hangs on long enough and the resistance drops
>>further, the rectifiers blow too. But that's more rare.
>
>
>It -is- rare because the caps usually fuse internally before anything
>else goes.
When you say "fuse" to you mean open or short?
>
>>>You might also realize that electrolytics blow up when subjected to a
>>>reverse current -not- because of any heat buildup or dissipation, but
>>>because the dielectric undergoes electrolysis and one of the products
>>>is a gas.
>>
>>Yes, I am fully aware of that. When I was a young pup on my first full
>>time job, I worked 2nd shift and the techs sometimes got bored when
>>their work was done, so we used to make "capacitor cannons". We'd
>>solder a set of leads to a radial lead cap (Usually 10 uF at 50V)
>>place it in a hunk of PVC pipe, hook it up to a power supply
>>backwards, aim it at our "enemy" and crank up the voltage. It made
>>quite a projectile. I used to play with the voltage sometimes and I
>>would often crank it up slowly just to see where the point of
>>avalanche current draw first started.
>>
>>
>>But I digress......
>>
>>Answer me this, how does a functioning DC power supply reverse current
>>under typical operating conditions?
>
>
>If the cap is subjected to reverse current then the operating
>conditions are not typical, are they? But if you're asking how a low
>voltage cap can be subjected to reverse current then there are a
>myriad of possibilities, the most obvious and most common is that the
>power supply was connected backwards.
Are we now talking about solid state CB radios? That's a whole
different set of circumstances from the tube amps and radios we were
using as examples.
> Also, electrolytics are not
>always fixed at ground potential, such as when they are used as a DC
>block in an audio line. If the offset fails, the cap pops.
Absolutely! No argument there. But that's not what we were talking
about before.
> There are
>also instances where an abnormal AC component swings the normally
>positive side of the cap negative. The cap voltage can get reversed by
>rectification when a pull-up resistor fails. Etc, etc. Geez, Dave,
>you've been doing this for 30 years and you didn't know this stuff?
Not as applied to the tube amps we were talking about. If you switch
the conditions of the discussion, then the outcomes will change as
well.
>>> This takes much less power than brute-force heating of a
>>>cap. If a low-voltage electrolytic blows up, the reason is almost
>>>always due to reverse current. High-voltage electrolytics don't blow
>>>up nearly as often because they are rarely subjected to reverse
>>>current.
>>
>>Which makes that an unlikely reason for cap failure.
>
>
>It means high-voltage electrolytics are less likely to fail than
>low-voltage electrolytics. Nothing more.
In my experience the opposite happens. You see, while a 5K leakage
resistance is more than enough to draw 200 watts from an 800 V supply
and heat up the cap until explosion, that same 5K Ohm leakage across
20V (A typical unregulated voltage in a CB power supply) will only net
about .08 watts. Hardly enough to heat the cap to explosion. So in
this case, if a cap blows, it's normally the result of either too high
a voltage (exceeds the cap's rating) or reverse voltage.
>>>>Also, if you recall how the HV power supplies were designed in a
>>>>typical sweep tube "CB" grade amplifier, the plate supply voltage was
>>>>usually designed at about 800V give or take. The filter caps were
>>>>usually (2) 40 uF, 450V caps wired in series across the DC output.
>>>>Sometimes there will be (2) 100K or 1M ohm resistors in parallel with
>>>>the filter caps as well. What happens is that one of the two caps
>>>>becomes leaky. It starts to draw current, and it raises the voltage
>>>>potential across the other cap, until it exceeds the voltage rating
>>>>and it's usually THAT cap which explodes.
>>>
>>>
>>>You're reaching, Dave.
>>
>>Not at all. This is what happened on at least 2 other amps that I
>>worked on. One cap had exploded, and the other one was leaky (As
>>measured with a simple VOM). I concluded, based on the evidence,
>
>
>.....LOL!
>
>
>> that
>>the leaky cap had caused the other cap to blow. The first time this
>>happened to me, I replaced the "exploded" cap only, and when I fired
>>up the amp, a short time later, the new cap exploded as well. Only
>>then did I learn to check the remaining caps for leakage. Once the
>>leaky caps were replaced, there was no further problem.
>
>
>I think you are making that up, Dave.
Really? Based on what? Disbelief?
>>>If the voltage across the 'other' cap is
>>>exceeded what usually happens is the dielectric will puncture (at the
>>>dielectric's puncture voltage) and cause a brief short as the cap
>>>discharges. If the first discharge doesn't pop the fuse, or fuse the
>>>cap internally, it will continue to charge and discharge just like a
>>>relaxation oscillator, shunting (not dissipating) huge amounts of
>>>power until something else fails, which usually happens in less than a
>>>couple seconds. That's why you sometimes hear a short 'whine' before
>>>the radio goes dead and smoke billows from the transformer.
>>
>>That "whine" is usually from a 12V to HV multi vibrator which is
>>forced to stop from excessive current draw. I've seen that on a couple
>>of Lafayette HA-250's.
>
>
>I have heard that "whine" from equipment that had no vibrator,
>mechanical -or- solid-state. It comes from the cap. If you don't
>believe me, do the following experiment: Take a good cap, put it in
>series with a resistor of, say, 100 to 1k ohms, then slowly push the
>voltage up above the rated voltage. I would suggest using a cap of
>only 100-150 volts because some of them won't puncture until they get
>up around 400-500 volts. Anyway, at some point you hear the capacitor
>begin to whine, and if the whine doesn't stop right away the resistor
>will commence smoking (unless your power supply blows first). Try it.
>It's fun!
That is something I've never heard. I've never heard a cap "whine". I
have heard them "sizzle" right before they out gas.
>>>>> Usually it's the
>>>>>fuse that goes first, and if there's no fuse then the transformer
>>>>>overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
>>>>>blow when subjected to too much AC, and even a few volts of ripple can
>>>>>be enough to pop a can on a tube rig.
>>>>
>>>>It's the filter cap's job to smooth out ripple. Of course they will be
>>>>subjected to it. If what you say were true, then there would be caps
>>>>popping like popcorn. Ripple is not A.C., as the polarity never
>>>>reverses. it only "pulses" to some degree. The charge-discharge action
>>>>of the filter caps is what smooths these pulses out.
>>>
>>>
>>>Ripple is the AC component of a DC current.
>>
>>Not exactly true. It is the pulsating D.C. current component. There is
>>no A.C. after a bridge rectifier.
>
>
>Like I said, call it whatever you like.
I like being accurate.
>>> But call it whatever
>>>tickles your fancy because the fact is that if a filter cap is doing
>>>it's job of smoothing the ripple then it's not being subjected to very
>>>much ripple!
>>
>>Exactly my point. What you've described is a classic chicken and egg
>>scenario.
>
>
>The rooster came first.
That much is true.
>
>
>> The cap can't be subjected to too much ripple if it's doing
>>it's job. If there is ripple, then the cap is usually bad.
>
>
>Not necessarily. A heavy load will increase the ripple.
Yes it will, but it won't kill the cap. Caps are designed to handle
ripple. That's their job.
>>> If there -is- much ripple then the cap isn't doing it's
>>>job, the reason being either the cap is weak or the load is too big.
>>
>>Right! But if the voltage is normal, then you can assume the current
>>draw is not excessive.
>
>
>Again, not necessarily. The DC component may stay within norms but the
>AC component (ripple) will increase under load. That's just a fact of
>life. Haven't you ever studied the basics of power supplies?
I could write a course on power supplies.
An unregulated power supply will decrease voltage as load increases.
If the supply is at normal expected voltage, then the load is not
excessive. A regulated supply is a different issue. The regulated
output may remain at normal, while the unregulated "reserve" drops,
along with a proportional increase in ripple.
The reason ripple increases is due to the R/C time constant of the
circuit. When the load increased, the apparent resistance across the
cap drops and the discharge rate increases to the point that it looses
the capacity to hold the supply up in between ripple pulses.
>>>Whatever the cause, the cap sees that ripple as AC current which
>>>causes heat dissipation in the cap due mostly to hysteresis (high
>>>hysteresis being a characteristic of electrolytic caps).
>>
>>Now it's my turn. I've never seen a cap blow from a supply being
>>dragged down to the point of excessive ripple. Over voltage yes,
>>reverse voltage yes, leaky cap yes. Too much ripple, no.
>
>
>Yet you claim that you can definitively determine if a completely
>destroyed capacitor blew up because it was leaking. So how do you do
>that? Do you reconstruct the capacitor like the FAA reconstructs
>planes? Because that's what -they- need to do in order to determine
>the cause of failure when there's no other obvious cause. Maybe they
>should just hire you and your crystal ball, huh?
If a cap explodes due to an excessive load, then the evidence of the
failure which cause that excessive load should be evident. If not then
you can't conclude that the cap died due to ripple.
You are right, in a sense, that you can't always tell what DID cause
the cap to blow, but you can determine what DID'NT cause it to fail.
It then becomes a simple process of elimination.
It's like forensic science. We reconstruct an incident based on the
evidence in the aftermath.
>>>You missed capacitor lab, didn't you? Excessive load will draw down
>>>the DC voltage but -increase- the ripple.
>>
>>I've never seen a cap blow from ripple. Never.
>
>
>You probably have but didn't know it. You can create and measure a
>controlled failure with a couple good chokes for AC isolation, a
>couple large motor-run capacitors for DC isolation, and an audio amp
>to supply the AC. If you need a schematic just let me know.
I don't understand the relationship of what you are trying to
simulate. I'm talking about filter caps in a power supply now. I
should be able to "simulate" this by simply placing a large load
across the power supply enough to drop down the voltage and increase
the ripple. If the rectifier or the transformer doesn't give up the
ghost first,, then the cap should blow according to your theory.
>>>>>>But no lower either. They certainly fail more often than resistors.
>>>>>
>>>>>
>>>>>Not when you realize that a light bulb is nothing more than a
>>>>>resistor
>>>>
>>>>A light bulb is a resistor as a consequence of it's intended use. A
>>>>part designed as a resistor rarely fails, unless too much current is
>>>>drawn through it. In which case, the failure is quite obvious.
>>>
>>>
>>>A light bulb is most certainly designed as a resistor. If it didn't
>>>have resistance it wouldn't work.
>>
>>Yes resistance is necessary for it's function. But it's not the
>>function of a light bulb to act as a resistor in a circuit. Light
>>bulbs are far too fragile for that use.
>
>
>Wanna bet? I use them all the time for dummy loads while messing with
>my generator experiments.
A GE #47 bulb is close to 50 Ohms at 4 watts of power. The impedance
changes, of course, with more or less power.
That said, they're still a fragile resistor.
> They also make great regulators because of
>their non-linear characteristics.
But they can't handle much current.
> And for the same reason, miniature
>light bulbs make pretty decent audio compressors.
I prefer Op-Amps.
> So you still say
>that light bulbs aren't resistors? Hogwash. If it's useful as a
>resistor then it's a resistor.
I never said they weren't resistors, I said that they aren't practical
resistors to substitute in most circuits. A light bulb's resistance
isn't constant and changes with voltage. Not something you normally
want.
>
>>> A resistor necessarily dissipates
>>>power, and in a light bulb that power is dissipated as light (and
>>>heat). The only real question here is why household light bulbs fail
>>>regularly while automotive light bulbs last for a much, much longer
>>>time.
>>
>>Higher voltage, and a smaller ,thinner and more shock prone filament.
>
>
>Tungsten, one of the strongest metals on the periodic chart, is "shock
>prone"? And what difference does the voltage make? It's the current
>that causes the filament to heat up. Which is true regardless of it's
>thickness, too.
Current is a function of voltage divided by resistance. A higher
voltage bulb can make the same current with a higher resistance. They
usually control this with the length and thickness of the filament.
Shorter, thicker filaments are more durable. That's also why a 25 watt
bulb normally lasts longer than a 100 watt bulb.
>>>You said, "I'm assuming it's an old 23 channel classic.
>>
>>Yes, I was more or less asking for confirmation.
>
>
>You don't have a web browser?
I'm lazy, so sue me.
>>> With old
>>>radios like that, the problem is almost always related to caps drying
>>>out," which demonstrated your ignorance about several things. First,
>>>the radio is a more modern 40 channel PLL model, contemporary with the
>>>TRC-453 and PC-122.
>>
>>Ok, you could have just said that in the beginning.
>
>
>Or you could have simply looked it up through google like I did.
Yea, I guess I could 've.
>
>>> Second, low voltage electrolytics don't "dry out"
>>>because they use a solid dielectric.
>>
>>That may be true, but they do change value and open up due to the
>>affects of humidity and age.
>
>
>Age, yes, they do change to a greater or lesser degree. Usually
>lesser. Humidity? I don't think so, Dave. Not when they are
>hermetically sealed. And the only caps that aren't hermetically sealed
>are air caps. Have you ever seen an "air electrolytic"? I haven't.
They are also affected by long term exposure to heat above their heat
rating.
>> What the process is that causes that to
>>happen, I'm not sure of the exact name. But "drying out" is a common
>>term that we use to describe it. If I am technically in error, then so
>>be it.
>
>
>Who is "we"?
Other techs and engineers that I know in the industry.
I wasn't doing that. I was only suggesting something else to check
out.
>>>Yet because I rejected your conclusion (or, if you
>>>prefer, your 'declaration of high probability'), you assert that I am
>>>claiming the fault is -not- a capacitor. That is a perfect example of
>>>the logical fallacy 'Argumentum ad Ignorantiam' which, by definition,
>>>is ignorance.
>>
>>Not at all. If I claim that there is a high probability of component
>>"A" failing and you disagree. Then you are saying, in essence, that
>>there is a very low probability that component "A" is the failure.
>
>
>Wrong. Disagreement is -not- an assertion of the opposite.
It is in this case.
>This is the
>same fallacy you invoke in your political arguments: You label me as a
>liberal because I do not agree with your admitted conservatism, which
>is also wrong. For example, if you believe in God and I state that I'm
>an athiest, that doesn't mean I hold satanic rituals in my basement
>(or anywhere else).
>
>If you want to assert yourself as a person who can think logically
>then you need to understand logical fallacies enough not to make them.
I know what I need to know about logical fallacies. You have committed
your share of them, coincidentally when you debate politics.
>>If you consider component "A" to have a low probability of failure it
>>stands to reason that you place a higher probability on another
>>component failing.
>
>
>Again wrong, and for the same reason.
Then what are you saying?
>
>>Inductive reasoning.
>
>
>Before you make any more logical leaps, let me send you a good book on
>the subject. I'll pay shipping.
I don't need any more books.
>>
>>More chest thumping? What you've done at HP has no bearing on what
>>I've seen in 30+ years of CB radio repair.
>
>
>Really? Do capacitors behave differently in CB radios than they do in
>spectrum analyzers and signal generators?
Sure. They're exposed to different conditions.
>>>Maybe you should take a look at alignment procedures sometime. A
>>>frequently used test involves driving the audio hard enough to slam it
>>>rail-to-rail at every stage from mic to speaker, usually by feeding
>>>the mic input with 500 mV or more; i.e, significantly more than you
>>>can generate by screaming into the mic.
>>
>>While I can understand why you might want to do that, I've never seen
>>it in any of the factory alignment procedures or in the SAM's
>>photofacts.
>Well, I just got done fixing the TRC-449 I got from Lance (thanks
>again! Got the front-end fixed and receive is hot! FWIW, the second RF
>amp TR-5 was the culprit as I suspected -- it's a 2SC1730 and it's
>Vceo is just too low for the radio, a design flaw) and have the manual
>right beside me.
Actually TR5 is the 1st RF front end. it's the very first active
device in the chain. TR4 is an AGC clamp. The second device is the
mixer and it's a 3SK458 FET.
> On page 15 is the transmitter alignment section, and
>about the middle of the page is Step 4 where the procedure is to align
>the transmit coils while shoving 500 mV into the mic jack. Yet Step 8
>is to set the modulation for 98% with a 200 mV input. Now there is a
>reason for driving the audio so hard in Step 4 and that is to make
>sure the signal path can handle the stress, because if it can cope
>with a 500 mV input then a mere 200 mV should be well within the safe
>operating area.
Hmm.... My test procedure for that rig used 5mV to set 50% modulation,
with VR6 and 200 mV to set the 100% limiter threshold with VR7.
The reason for the increase in audio drive is to test that the
modulation limiter (and mike AGC in this particular radio) is doing
its job. Think about it, if the modulation limiter is working and
doing its job, then you can increase the mike level all you want
(until you fry something) and the AMC will clamp it to the level that
makes 100% modulation. The only parts in the audio chain that would be
stressed are the passive components before the AMC audio shunt
regulator, which is TR18.
There is very little you could tell me about that radio (or more
specifically that chassis) that I don't already know. That UPD-858 PLL
chassis is my all time favorite. But be careful with the UPC-1156
audio amp chip. They're discontinued and no longer generally
available. That tantalum capacitor that I mentioned before, that had a
habit of shorting, is C179 on this rig, and it's a 2.2uF.
Also, depending on how old that rig is, there was a factory mod that
needs to be performed. On the oldest versions of that chassis, the RF
final transistor had a habit of blowing, especially when a lot of SSB
activity occurred. The problem was that the bias was being applied too
strongly and the transistor would avalanche. The mod involved removing
D66 and D67, and changing the values of R205 and R206 to 10 Ohms.
Using the bias pots VR15 (driver) and VR16 (final) set the bias
voltage for 40ma (driver) and 15ma (final). Newer versions of the
radio already had the mod done at the factory.
That rig also mods very well, and will operate on most of the 10 meter
band, (if you want to convert it) without the need to change any
crystals. The PLL is capable of running (with the proper VCO
adjustment) up to close to 30 Mhz. The PLL programmable divider is
controlled by a 3 digit BCD "word", and capable of valid combinations
ranging from 000 to 399. If you want to broadband the transmitter, you
can do so quite simply by cutting the center tap circuit trace on both
sides of L37 (a double can) and moving the trace to the empty outside
legs on each respective side of the can. After realignment, the rig
will now hold power for about a 2 Mhz wide band segment.
Clarifier range can be extended by jumping out R116. The pot can be re
centered again by changing the value of R117.
I also found the transmit audio to be a bit mushy and compressed. The
main reason for that is an audio bandpass filter installed between TR
20 and TR22. Removal and jumping out of L11, L10, and removing C's
71,72,73,74,& 75, yields a much better sounding audio fidelity. Or you
can custom tailor the parts to suit your individual taste.
>Now you claim to have worked on CBs for 30 years, yet you have never
>seen a procedure as I just described (on what was once a fairly common
>and popular radio)?
I have seen similar procedures, but they were not used in the way you
interpreted it. I told you why they increase the level, and that's to
verify proper AMC action.
>>Often the only reason given in the procedures to increase the audio
>>input (And I have a nice Tektronix generator) is to verify proper
>>operation of the modulation limiter.
>I can send a pair of reading glasses with that book if you want.
Your failure to properly identify the true reason why the audio level
is increased, is not suggestive of my poor eyesight.
>>> One of the purposes of the
>>>test is to see if any stages lock up, a problem that is much more
>>>common with high-gain OP-amps produced since the 70's.
>>
>>Most CB radios made in the 1960- 1980 time frame do not use Op-amps.
>
>
>.....uh, yeah. That's perfectly consistent with what I said. What's
>your point?
That audio lockups, like what you've described, would be rare on most
CB radios.
>>Some Uniden radios used them more recently, but not as audio amps.
>>They were used as AGC amps and T/R switching.
>Most OP-amps in modern radios (CB and other) are integrated into
>special-purpose chips. Some older radios had OP-amps made with
>discrete components.
There are very few CB radios that use "special purpose chips" other
than the PLL. Most parts are standard off-the-shelf components. It's a
blessing for techs. Most CB radios could be fixed with Radio Shack
parts. At one time, Radio Shack actually carried RF final amps
(2SC-1307). It's a shame that they have scaled back their parts so
much.
>>I have never seen an audio stage on a CB radio "lock up". The Hy-Gain
>>623 used to have an interestingly annoying "design flaw" where if the
>>modulator was driven too hard, it would shut down for a few seconds.
>>You would loose receive audio until it recovered.
>
>
>That sounds like a lock-up to me.
It was a design flaw (or "feature" depending on your perspective).
>>>Resonance is also an issue that needs to be appreciated. As you noted
>>>before, components change value over time. So a circuit that was not
>>>resonant at the time of manufacture may be resonant 20 years later.
>>>It's not a common cause of failure but I have seen it a few times.
>>
>>You should see what happens when a 7812 regulator goes into
>>oscillation due to bad/missing bypass caps.
>
>
>You should see my curio cabinet of failed components. It even includes
>a shorted resistor (47K, carbon comp), something I was always taught
>was absolutely impossible.
Nothing is impossible in electronics. I had an open resistor once,
which had no signs of burning or overheating.
> But I show it off to visitors who stop by
>and don't believe me, have them measure it for themselves, and they
>walk away with a paranoid glaze over their eyes.
Oh, I believe you. I think I remember once seeing a resistor which
became a near dead short. It really messed up the bias of the circuit
it was installed in.
Dave
"Sandbagger"
I AmnotGeorgeBush
> I believe in being accurate
Nothing is stopping you except your inner struggle. Off you go now,
(with your accuracy, of course) explaining your criteria for claiming
dxing makes on a federal criminal. Of course, the stifling comes when
you are presented with the fact that a mandated 70 MPH speed limit is
also a "federally enacted law", the only criteria you cite in claiming
the infraction of such, maes one a "federal criminal". Such ludicrous
statements illustrates you have a great deal to learn concerning the
law, judicial and fiducial matters.
(who is we)
>Other techs
mmmph,,,,not only have you hallucinated that other hams were agreeing
with you, now you are hallucinating "other techs" are fondling over
yourself. (sigh),,ok,,go ahead and name a single one these techs and
hams only you can see.
Frank Gilliland
wrote in <7omfv0p8kn99bi7on...@4ax.com>:
Sherlock Holmes was not a real person, and that poorly-quoted cliche
has one very important point you completely missed: "when all the
other possibilities are excluded". Care to apply that concept to the
debate between creationism and evolution? Or can you see that there
are "other possibilities" that may not be obvious but they do exist?
> The supply was otherwise in good shape,
>the load was nominal. The only thing "bad" was the cap.
>
>Oh, and I've seen equipment which did have leaky caps, and they were
>hot to the touch and some have had the rubber plug in the bottom
>partially ejected. They didn't fail violently in an explosion, but
>they did out gas from excessive leakage.
I've seen the same thing several times myself.
>>>>>Leaky doesn't mean "dead short". A cap could have just enough
>>>>>resistance to heat it up and cause it to out gas or explode, but not
>>>>>cause other components to fail.
>>>>
>>>>
>>>>Never seen it happen. It's easy to understand why it doesn't happen
>>>>when you realize that heat requires power; i.e, watts.
>>>
>>>800V X .25 amps = 200 watts. More than enough to heat up a leaky cap.
>>
>>
>>Heat it? Yes. Will it pass that much current without fusing
>>internally? Doubtful. If it -could- pass that much current would that
>>be enough power to make it explode?
>
>The evidence that I've seen would tend to answer that question as a
>yes.
If you knew how much power it takes to cause an explosion you might
change your mind.
Capacitors are rated by more than those three things: Variation in
capacitance with frequency and temperature; hysteresis; inductance &
complex impedance; etc.... and AC resistance, also dependent upon
temperature and frequency, and which is critical in power circuits due
to the usually dismal power dissipation capabilities of capacitors.
> In most sweep tube amps and tube CB radios, I rarely see
>more than about 40 uF in filter caps. Voltage rating depends on the
>circuit. Most amps use 450V caps in series.
I have a cap handy that's rated 40 uF @ 440 VAC. It must weigh at
least five pounds and measures 10" x 4" x 2" (want a pic?). It's big
because it's designed to handle lots of AC. I also have another
capacitor handy that's rated 40 uF @ 450 VDC. It's barely 2" long and
1" diameter. It can't handle much AC at all. I have seen the caps used
in sweep tube amps and they aren't designed to handle much AC at all.
Load them down with some ripple and **poof**.
>>>> In fact, your
>>>>leaky cap draws so much -additional- power that it evaporates the
>>>>contents faster than the gasses can be vented by the vent hole -AND-
>>>>builds up enough pressure to literally explode the metal case.
>>>
>>>That's exactly what happens.
>>
>>
>>Uh-huh.
>>
>>
>>>> I don't
>>>>think so, Dave -- that's just too much power to not expect an earlier
>>>>failure in the power supply.
>>>
>>>Not at all, considering the scenario I just presented. I can't count
>>>the number of D&A Mavericks and Thunderbolt 305's and such that I've
>>>replaced blown caps in, and not had any other fault.
>>
>>
>>Did you ever think that maybe the cap wasn't large enough for the
>>application?
>
>Perhaps. Then I guess the manufacturer has some 'splainin to do.
That happens more times than I would like. GE and Motorola usually
have well-engineered products; Alvin Amps and Back-Room Communications
usually don't.
>>>>And like I said before, if the supply isn't fused then the failure is
>>>>probably going to be a shorted transformer or choke, blown rectifier,
>>>>or some other device that is less robust than the capacitor.
>>>
>>>Sometimes, if the cap hangs on long enough and the resistance drops
>>>further, the rectifiers blow too. But that's more rare.
>>
>>
>>It -is- rare because the caps usually fuse internally before anything
>>else goes.
>
>When you say "fuse" to you mean open or short?
Open. The wire fuses; i.e, melts. Internal wires in electrolytics are
usually quite thin, and the connections to the plates aren't always
secure either.
You must remain focused, Dave. We have been discussing several issues
regarding capacitor failure. In the present topic I had stated that
electrolytics fail most often due to excessive AC (or ripple) -or-
reverse voltage, at which point you queried as to how a power supply
can reverse it's current. I then explained that failure due to voltage
reversal is rare with high-voltage caps (in tube rigs) but went on to
explain how the failure is common in low-voltage caps (SS circuits).
Are you up to speed now?
>> There are
>>also instances where an abnormal AC component swings the normally
>>positive side of the cap negative. The cap voltage can get reversed by
>>rectification when a pull-up resistor fails. Etc, etc. Geez, Dave,
>>you've been doing this for 30 years and you didn't know this stuff?
>
>Not as applied to the tube amps we were talking about. If you switch
>the conditions of the discussion, then the outcomes will change as
>well.
Then why did you bring up the subject of tube amps when the original
topic was about a solid-state CB? If you don't want to discuss it then
don't bring it up.
>>>> This takes much less power than brute-force heating of a
>>>>cap. If a low-voltage electrolytic blows up, the reason is almost
>>>>always due to reverse current. High-voltage electrolytics don't blow
>>>>up nearly as often because they are rarely subjected to reverse
>>>>current.
>>>
>>>Which makes that an unlikely reason for cap failure.
>>
>>
>>It means high-voltage electrolytics are less likely to fail than
>>low-voltage electrolytics. Nothing more.
>
>In my experience the opposite happens.
Perhaps that's because you work more with tubes while I work mostly
with solid-state equipment.
> You see, while a 5K leakage
>resistance is more than enough to draw 200 watts from an 800 V supply
>and heat up the cap until explosion, that same 5K Ohm leakage across
>20V (A typical unregulated voltage in a CB power supply) will only net
>about .08 watts. Hardly enough to heat the cap to explosion. So in
>this case, if a cap blows, it's normally the result of either too high
>a voltage (exceeds the cap's rating) or reverse voltage.
Caps (in general) usually short or fuse (or short -then- fuse) when
the working voltage is exceeded. Low-voltage electrolytics blow up
usually because of reverse voltage.
>>>>>Also, if you recall how the HV power supplies were designed in a
>>>>>typical sweep tube "CB" grade amplifier, the plate supply voltage was
>>>>>usually designed at about 800V give or take. The filter caps were
>>>>>usually (2) 40 uF, 450V caps wired in series across the DC output.
>>>>>Sometimes there will be (2) 100K or 1M ohm resistors in parallel with
>>>>>the filter caps as well. What happens is that one of the two caps
>>>>>becomes leaky. It starts to draw current, and it raises the voltage
>>>>>potential across the other cap, until it exceeds the voltage rating
>>>>>and it's usually THAT cap which explodes.
>>>>
>>>>
>>>>You're reaching, Dave.
>>>
>>>Not at all. This is what happened on at least 2 other amps that I
>>>worked on. One cap had exploded, and the other one was leaky (As
>>>measured with a simple VOM). I concluded, based on the evidence,
>>
>>
>>.....LOL!
>>
>>
>>> that
>>>the leaky cap had caused the other cap to blow. The first time this
>>>happened to me, I replaced the "exploded" cap only, and when I fired
>>>up the amp, a short time later, the new cap exploded as well. Only
>>>then did I learn to check the remaining caps for leakage. Once the
>>>leaky caps were replaced, there was no further problem.
>>
>>
>>I think you are making that up, Dave.
>
>Really? Based on what? Disbelief?
Based on the fact that I've never experienced what you described, and
don't see how it's even possible from what I -have- experienced. I
have never seen a high-voltage cap blow up because it was leaking, and
I have rarely seen the same blow up from excessive DC voltage. I
-have- seen them blow up from excessive AC voltage (ripple). Is it
possible that the cap which blew up was subjected to too much ripple
caused by leakage in the other cap?
>>>>If the voltage across the 'other' cap is
>>>>exceeded what usually happens is the dielectric will puncture (at the
>>>>dielectric's puncture voltage) and cause a brief short as the cap
>>>>discharges. If the first discharge doesn't pop the fuse, or fuse the
>>>>cap internally, it will continue to charge and discharge just like a
>>>>relaxation oscillator, shunting (not dissipating) huge amounts of
>>>>power until something else fails, which usually happens in less than a
>>>>couple seconds. That's why you sometimes hear a short 'whine' before
>>>>the radio goes dead and smoke billows from the transformer.
>>>
>>>That "whine" is usually from a 12V to HV multi vibrator which is
>>>forced to stop from excessive current draw. I've seen that on a couple
>>>of Lafayette HA-250's.
>>
>>
>>I have heard that "whine" from equipment that had no vibrator,
>>mechanical -or- solid-state. It comes from the cap. If you don't
>>believe me, do the following experiment: Take a good cap, put it in
>>series with a resistor of, say, 100 to 1k ohms, then slowly push the
>>voltage up above the rated voltage. I would suggest using a cap of
>>only 100-150 volts because some of them won't puncture until they get
>>up around 400-500 volts. Anyway, at some point you hear the capacitor
>>begin to whine, and if the whine doesn't stop right away the resistor
>>will commence smoking (unless your power supply blows first). Try it.
>>It's fun!
>
>That is something I've never heard. I've never heard a cap "whine". I
>have heard them "sizzle" right before they out gas.
Sizzle, whine, fizz, scream..... it doesn't always sound the same.
>>>>>> Usually it's the
>>>>>>fuse that goes first, and if there's no fuse then the transformer
>>>>>>overheats and shorts. Caps don't blow because of DC leakage. Caps -do-
>>>>>>blow when subjected to too much AC, and even a few volts of ripple can
>>>>>>be enough to pop a can on a tube rig.
>>>>>
>>>>>It's the filter cap's job to smooth out ripple. Of course they will be
>>>>>subjected to it. If what you say were true, then there would be caps
>>>>>popping like popcorn. Ripple is not A.C., as the polarity never
>>>>>reverses. it only "pulses" to some degree. The charge-discharge action
>>>>>of the filter caps is what smooths these pulses out.
>>>>
>>>>
>>>>Ripple is the AC component of a DC current.
>>>
>>>Not exactly true. It is the pulsating D.C. current component. There is
>>>no A.C. after a bridge rectifier.
>>
>>
>>Like I said, call it whatever you like.
>
>I like being accurate.
Ok then. Any voltage has a DC component and an AC component. Pure AC
has no DC component, and pure DC has no AC component. The term
"ripple" is literally defined as the AC component of a DC current,
usually from a power supply, and usually a harmonic of the input
power. The term "ripple" is old and has been replaced by "pard", which
includes all AC components (or fluctuations) including hum, noise and
spikes.
Is that accurate enough for you?
>>>> But call it whatever
>>>>tickles your fancy because the fact is that if a filter cap is doing
>>>>it's job of smoothing the ripple then it's not being subjected to very
>>>>much ripple!
>>>
>>>Exactly my point. What you've described is a classic chicken and egg
>>>scenario.
>>
>>
>>The rooster came first.
>
>That much is true.
>
>>
>>
>>> The cap can't be subjected to too much ripple if it's doing
>>>it's job. If there is ripple, then the cap is usually bad.
>>
>>
>>Not necessarily. A heavy load will increase the ripple.
>
>Yes it will, but it won't kill the cap. Caps are designed to handle
>ripple. That's their job.
SOME caps are designed to handle ripple. Power supply filter caps are
designed to REDUCE ripple; i.e, shut the AC component to ground. When
the load becomes large then so does the ripple, and if the cap has a
high AC resistance then it will get hot very quickly.
I have a book on power supply design that I could send with that logic
book.....
>>>> If there -is- much ripple then the cap isn't doing it's
>>>>job, the reason being either the cap is weak or the load is too big.
>>>
>>>Right! But if the voltage is normal, then you can assume the current
>>>draw is not excessive.
>>
>>
>>Again, not necessarily. The DC component may stay within norms but the
>>AC component (ripple) will increase under load. That's just a fact of
>>life. Haven't you ever studied the basics of power supplies?
>
>I could write a course on power supplies.
Just about everyone who has ever used a soldering iron thinks they can
write a course on power supplies.
>An unregulated power supply will decrease voltage as load increases.
>If the supply is at normal expected voltage, then the load is not
>excessive. A regulated supply is a different issue. The regulated
>output may remain at normal, while the unregulated "reserve" drops,
>along with a proportional increase in ripple.
>
>The reason ripple increases is due to the R/C time constant of the
>circuit. When the load increased, the apparent resistance across the
>cap drops and the discharge rate increases to the point that it looses
>the capacity to hold the supply up in between ripple pulses.
Nice textbook explanation. But the increased ripple can be caused by a
host of factors, not the least of which (and to the suprise of many,
the most common) is the AC resistance of the output capacitor, -not-
the R/C time constant of the cap with respect to the source and load
impedances (which, BTW, are in parallel). A different scenario happens
with high-voltage supplies because regulation is usually achieved with
a swinging (saturating) choke, which permits more ripple to pass under
increasing load in order to maintain the DC component.
So when is your book coming out?
>>>>Whatever the cause, the cap sees that ripple as AC current which
>>>>causes heat dissipation in the cap due mostly to hysteresis (high
>>>>hysteresis being a characteristic of electrolytic caps).
>>>
>>>Now it's my turn. I've never seen a cap blow from a supply being
>>>dragged down to the point of excessive ripple. Over voltage yes,
>>>reverse voltage yes, leaky cap yes. Too much ripple, no.
>>
>>
>>Yet you claim that you can definitively determine if a completely
>>destroyed capacitor blew up because it was leaking. So how do you do
>>that? Do you reconstruct the capacitor like the FAA reconstructs
>>planes? Because that's what -they- need to do in order to determine
>>the cause of failure when there's no other obvious cause. Maybe they
>>should just hire you and your crystal ball, huh?
>
>If a cap explodes due to an excessive load, then the evidence of the
>failure which cause that excessive load should be evident. If not then
>you can't conclude that the cap died due to ripple.
The cause of failure is not always evident in the equipment. Like when
someone brings in an amp (audio, RF, whatever) that they overloaded,
overdrove, hooked up wrong or whatever. The dope who blew up the amp
usually isn't going to admit his mistake, he just wants to get it
fixed and get the heck out of there so he can put the incident behind
him. Happens all the time. But contrary to your self-admitted
cynicism, some people ARE honest, and that's one way I can determine
what's causing a specific type of failure. So there -is- evidence of
the cause of the failure even though it may not be evident from an
examination of the equipment.
>You are right, in a sense, that you can't always tell what DID cause
>the cap to blow, but you can determine what DID'NT cause it to fail.
>It then becomes a simple process of elimination.
The possibilities are limitless so any such conclusion would be based
only on probability.
>It's like forensic science. We reconstruct an incident based on the
>evidence in the aftermath.
Dave, before you use the word "forensic" again, do yourself a favor
and look it up in the dictionary.
>>>>You missed capacitor lab, didn't you? Excessive load will draw down
>>>>the DC voltage but -increase- the ripple.
>>>
>>>I've never seen a cap blow from ripple. Never.
>>
>>
>>You probably have but didn't know it. You can create and measure a
>>controlled failure with a couple good chokes for AC isolation, a
>>couple large motor-run capacitors for DC isolation, and an audio amp
>>to supply the AC. If you need a schematic just let me know.
>
>I don't understand the relationship of what you are trying to
>simulate. I'm talking about filter caps in a power supply now. I
>should be able to "simulate" this by simply placing a large load
>across the power supply enough to drop down the voltage and increase
>the ripple. If the rectifier or the transformer doesn't give up the
>ghost first,, then the cap should blow according to your theory.
The objective is to cause the cap to fail under controlled conditions,
and hopefully -without- destroying anything else. But if you want to
do it your way then feel free to do so.
>>>>>>>But no lower either. They certainly fail more often than resistors.
>>>>>>
>>>>>>
>>>>>>Not when you realize that a light bulb is nothing more than a
>>>>>>resistor
>>>>>
>>>>>A light bulb is a resistor as a consequence of it's intended use. A
>>>>>part designed as a resistor rarely fails, unless too much current is
>>>>>drawn through it. In which case, the failure is quite obvious.
>>>>
>>>>
>>>>A light bulb is most certainly designed as a resistor. If it didn't
>>>>have resistance it wouldn't work.
>>>
>>>Yes resistance is necessary for it's function. But it's not the
>>>function of a light bulb to act as a resistor in a circuit. Light
>>>bulbs are far too fragile for that use.
>>
>>
>>Wanna bet? I use them all the time for dummy loads while messing with
>>my generator experiments.
>
>A GE #47 bulb is close to 50 Ohms at 4 watts of power. The impedance
>changes, of course, with more or less power.
>
>That said, they're still a fragile resistor.
That's probably why I use a bank of 100-250 watt bulbs because even my
smallest generator (1kW) would probably nuke a 47 in short order.
>> They also make great regulators because of
>>their non-linear characteristics.
>
> But they can't handle much current.
Tell that to my old Heathkit audio generator. And zeners can't handle
much current either, relatively speaking. Ever hear of a neat little
device called a 'power transistor'?
>> And for the same reason, miniature
>>light bulbs make pretty decent audio compressors.
>
>I prefer Op-Amps.
Yuk. Even a fast OP-amp like the OP-27 makes for a noisy compressor.
The sub-mini incandescents are clean to at least 5kHz. If you want
something better try discrete log amps or the mod-limit-demod method.
Or just go buy a DBX.
>> So you still say
>>that light bulbs aren't resistors? Hogwash. If it's useful as a
>>resistor then it's a resistor.
>
>I never said they weren't resistors, I said that they aren't practical
>resistors to substitute in most circuits. A light bulb's resistance
>isn't constant and changes with voltage. Not something you normally
>want.
A light bulb's resistance changes with temperature, but who said a
resistor -must- be linear? The textbooks?
>>>> A resistor necessarily dissipates
>>>>power, and in a light bulb that power is dissipated as light (and
>>>>heat). The only real question here is why household light bulbs fail
>>>>regularly while automotive light bulbs last for a much, much longer
>>>>time.
>>>
>>>Higher voltage, and a smaller ,thinner and more shock prone filament.
>>
>>
>>Tungsten, one of the strongest metals on the periodic chart, is "shock
>>prone"? And what difference does the voltage make? It's the current
>>that causes the filament to heat up. Which is true regardless of it's
>>thickness, too.
>
>Current is a function of voltage divided by resistance. A higher
>voltage bulb can make the same current with a higher resistance. They
>usually control this with the length and thickness of the filament.
>Shorter, thicker filaments are more durable. That's also why a 25 watt
>bulb normally lasts longer than a 100 watt bulb.
Not in my house. I replace 40 watt bulbs just as often as 60 watt and
100 watt bulbs. And again, the it's the temperature of the filament
that determines it's brightness, which is a direct consequence of its
current. IOW, the current density through a cross-section of filament
is the same regardless of it's thickness or length. And for the sake
of discussion, UV bulbs work at a higher temperature, and therefore
have a -higher- current density.
Care to give it another stab?
>>>>You said, "I'm assuming it's an old 23 channel classic.
>>>
>>>Yes, I was more or less asking for confirmation.
>>
>>
>>You don't have a web browser?
>
>I'm lazy, so sue me.
>
>
>>>> With old
>>>>radios like that, the problem is almost always related to caps drying
>>>>out," which demonstrated your ignorance about several things. First,
>>>>the radio is a more modern 40 channel PLL model, contemporary with the
>>>>TRC-453 and PC-122.
>>>
>>>Ok, you could have just said that in the beginning.
>>
>>
>>Or you could have simply looked it up through google like I did.
>
>Yea, I guess I could 've.
>
>>
>>>> Second, low voltage electrolytics don't "dry out"
>>>>because they use a solid dielectric.
>>>
>>>That may be true, but they do change value and open up due to the
>>>affects of humidity and age.
>>
>>
>>Age, yes, they do change to a greater or lesser degree. Usually
>>lesser. Humidity? I don't think so, Dave. Not when they are
>>hermetically sealed. And the only caps that aren't hermetically sealed
>>are air caps. Have you ever seen an "air electrolytic"? I haven't.
>
>They are also affected by long term exposure to heat above their heat
>rating.
Duh. Even to long-term exposure to heat -below- their heat rating.
>>> What the process is that causes that to
>>>happen, I'm not sure of the exact name. But "drying out" is a common
>>>term that we use to describe it. If I am technically in error, then so
>>>be it.
>>
>>
>>Who is "we"?
>
>Other techs and engineers that I know in the industry.
Well, I'm an engineer, you know me, and I haven't ever used the term
"drying out" to describe a capacitor failure. Maybe "dielectrically
challenged"..... no, I haven't used that either.
"...almost always..."
>>>>Yet because I rejected your conclusion (or, if you
>>>>prefer, your 'declaration of high probability'), you assert that I am
>>>>claiming the fault is -not- a capacitor. That is a perfect example of
>>>>the logical fallacy 'Argumentum ad Ignorantiam' which, by definition,
>>>>is ignorance.
>>>
>>>Not at all. If I claim that there is a high probability of component
>>>"A" failing and you disagree. Then you are saying, in essence, that
>>>there is a very low probability that component "A" is the failure.
>>
>>
>>Wrong. Disagreement is -not- an assertion of the opposite.
>
>It is in this case.
Well, I'm pretty sure about this, Dave. After all, I was the first
person to ever ace the logic class (Phil 201) in college, which is
probably why they brought me in a year later to teach it for a quarter
while the doctor was out.
>>This is the
>>same fallacy you invoke in your political arguments: You label me as a
>>liberal because I do not agree with your admitted conservatism, which
>>is also wrong. For example, if you believe in God and I state that I'm
>>an athiest, that doesn't mean I hold satanic rituals in my basement
>>(or anywhere else).
>>
>>If you want to assert yourself as a person who can think logically
>>then you need to understand logical fallacies enough not to make them.
>
>I know what I need to know about logical fallacies. You have committed
>your share of them, coincidentally when you debate politics.
Feel free to point them out.
>>>If you consider component "A" to have a low probability of failure it
>>>stands to reason that you place a higher probability on another
>>>component failing.
>>
>>
>>Again wrong, and for the same reason.
>
>Then what are you saying?
You remind me of the sitcom boyfriends when they get dumped by their
girlfriends who scream something like, "I never want to see you again
for the rest of my life!", and the boyfriend responds with, "So what
are you saying?"
>>
>>>Inductive reasoning.
>>
>>
>>Before you make any more logical leaps, let me send you a good book on
>>the subject. I'll pay shipping.
>
>I don't need any more books.
Yes, you do. Actually, this book is an older (but better) version of
the Copi book I used for the logic class. I highly recommend it
because the scope is a little broader than most logic textbooks. Copi
addresses both logic and language as distinct but integrated concepts.
>>>More chest thumping? What you've done at HP has no bearing on what
>>>I've seen in 30+ years of CB radio repair.
>>
>>
>>Really? Do capacitors behave differently in CB radios than they do in
>>spectrum analyzers and signal generators?
>
>Sure. They're exposed to different conditions.
Such as.....?
>>>>Maybe you should take a look at alignment procedures sometime. A
>>>>frequently used test involves driving the audio hard enough to slam it
>>>>rail-to-rail at every stage from mic to speaker, usually by feeding
>>>>the mic input with 500 mV or more; i.e, significantly more than you
>>>>can generate by screaming into the mic.
>>>
>>>While I can understand why you might want to do that, I've never seen
>>>it in any of the factory alignment procedures or in the SAM's
>>>photofacts.
>
>
>>Well, I just got done fixing the TRC-449 I got from Lance (thanks
>>again! Got the front-end fixed and receive is hot! FWIW, the second RF
>>amp TR-5 was the culprit as I suspected -- it's a 2SC1730 and it's
>>Vceo is just too low for the radio, a design flaw) and have the manual
>>right beside me.
>
>Actually TR5 is the 1st RF front end.
So it is. Close enough.
> it's the very first active
>device in the chain. TR4 is an AGC clamp. The second device is the
>mixer and it's a 3SK458 FET.
>
>> On page 15 is the transmitter alignment section, and
>>about the middle of the page is Step 4 where the procedure is to align
>>the transmit coils while shoving 500 mV into the mic jack. Yet Step 8
>>is to set the modulation for 98% with a 200 mV input. Now there is a
>>reason for driving the audio so hard in Step 4 and that is to make
>>sure the signal path can handle the stress, because if it can cope
>>with a 500 mV input then a mere 200 mV should be well within the safe
>>operating area.
>
>Hmm.... My test procedure for that rig used 5mV to set 50% modulation,
>with VR6
Step 7, and it's 0.5 mV.
> and 200 mV to set the 100% limiter threshold with VR7.
Step 8, set VR7 for 95 to 98% mod.
>The reason for the increase in audio drive is to test that the
>modulation limiter (and mike AGC in this particular radio) is doing
>its job. Think about it, if the modulation limiter is working and
>doing its job, then you can increase the mike level all you want
>(until you fry something) and the AMC will clamp it to the level that
>makes 100% modulation. The only parts in the audio chain that would be
>stressed are the passive components before the AMC audio shunt
>regulator, which is TR18.
>
>There is very little you could tell me about that radio (or more
>specifically that chassis) that I don't already know.
I just did. Twice in this post and once in the previous post.
Thanks anyway, but I already used too much time pulling out the mods
and restoring it to stock condition -- except for the faceplate. Got a
spare?
>>Now you claim to have worked on CBs for 30 years, yet you have never
>>seen a procedure as I just described (on what was once a fairly common
>>and popular radio)?
>
>I have seen similar procedures, but they were not used in the way you
>interpreted it. I told you why they increase the level, and that's to
>verify proper AMC action.
I suppose that's why the manual says to overdrive the audio -before-
setting the AMC, huh?
>>>Often the only reason given in the procedures to increase the audio
>>>input (And I have a nice Tektronix generator) is to verify proper
>>>operation of the modulation limiter.
>
>
>>I can send a pair of reading glasses with that book if you want.
>
>Your failure to properly identify the true reason why the audio level
>is increased, is not suggestive of my poor eyesight.
You need specs because you can't read what's in front of your face.
That's what happens when you get old.
>>>> One of the purposes of the
>>>>test is to see if any stages lock up, a problem that is much more
>>>>common with high-gain OP-amps produced since the 70's.
>>>
>>>Most CB radios made in the 1960- 1980 time frame do not use Op-amps.
>>
>>
>>.....uh, yeah. That's perfectly consistent with what I said. What's
>>your point?
>
>That audio lockups, like what you've described, would be rare on most
>CB radios.
Like I said, you need to get some optical assistance.
>>>Some Uniden radios used them more recently, but not as audio amps.
>>>They were used as AGC amps and T/R switching.
>
>>Most OP-amps in modern radios (CB and other) are integrated into
>>special-purpose chips. Some older radios had OP-amps made with
>>discrete components.
>
>There are very few CB radios that use "special purpose chips" other
>than the PLL. Most parts are standard off-the-shelf components.
I suppose that's true if you consider the chips to be off-the-shelf
components.
> It's a
>blessing for techs. Most CB radios could be fixed with Radio Shack
>parts. At one time, Radio Shack actually carried RF final amps
>(2SC-1307). It's a shame that they have scaled back their parts so
>much.
You have finally said something to which I agree without reservation.
I'll even go one step further and say that Radio Shack sucks.
>>>I have never seen an audio stage on a CB radio "lock up". The Hy-Gain
>>>623 used to have an interestingly annoying "design flaw" where if the
>>>modulator was driven too hard, it would shut down for a few seconds.
>>>You would loose receive audio until it recovered.
>>
>>
>>That sounds like a lock-up to me.
>
>It was a design flaw (or "feature" depending on your perspective).
It was intentional?
>>>>Resonance is also an issue that needs to be appreciated. As you noted
>>>>before, components change value over time. So a circuit that was not
>>>>resonant at the time of manufacture may be resonant 20 years later.
>>>>It's not a common cause of failure but I have seen it a few times.
>>>
>>>You should see what happens when a 7812 regulator goes into
>>>oscillation due to bad/missing bypass caps.
>>
>>
>>You should see my curio cabinet of failed components. It even includes
>>a shorted resistor (47K, carbon comp), something I was always taught
>>was absolutely impossible.
>
>Nothing is impossible in electronics. I had an open resistor once,
>which had no signs of burning or overheating.
That's common. A little stress from shock or vibration is all it takes
to make some of them crack. That's why I always mount replacements
slightly above the board.
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>>>Maybe that's happened to you but never on my bench, not unless the cap
>>>was too small for the load. Like I said before, I have never seen a
>>>cap that blew up on it's own accord, leaky or not (and I don't know
>>>how you can determine it was leakage that caused your cap to blow
>>>unless you have a time machine in your shop).
>>
>>Shhhhh. That a "way back" machine, and it's only a prototype.
>>
>>But you make a good point. But like Sherlock Holmes once said, when
>>all the other possibilities are excluded, what remains, however
>>unlikely must be the answer.
>
>
>Sherlock Holmes was not a real person
No, but Sir Arthur Conan Doyle was, and he wrote it.
>, and that poorly-quoted cliche
>has one very important point you completely missed: "when all the
>other possibilities are excluded". Care to apply that concept to the
>debate between creationism and evolution?
Why? That's an entirely different subject with limitless and
unverifiable possibilities. Unlike the subject we're currently
discussing. There are a finite amount of potential causes of cap
failure. Most of which can be reproduced.
> Or can you see that there
>are "other possibilities" that may not be obvious but they do exist?
Not in this particular case.
>>>Heat it? Yes. Will it pass that much current without fusing
>>>internally? Doubtful. If it -could- pass that much current would that
>>>be enough power to make it explode?
>>
>>The evidence that I've seen would tend to answer that question as a
>>yes.
>
>
>If you knew how much power it takes to cause an explosion you might
>change your mind.
I know how much, I've witnessed it both first hand and in the
aftermath.
>> In most sweep tube amps and tube CB radios, I rarely see
>>more than about 40 uF in filter caps. Voltage rating depends on the
>>circuit. Most amps use 450V caps in series.
>
>
>I have a cap handy that's rated 40 uF @ 440 VAC. It must weigh at
>least five pounds and measures 10" x 4" x 2" (want a pic?). It's big
>because it's designed to handle lots of AC. I also have another
>capacitor handy that's rated 40 uF @ 450 VDC. It's barely 2" long and
>1" diameter. It can't handle much AC at all. I have seen the caps used
>in sweep tube amps and they aren't designed to handle much AC at all.
>Load them down with some ripple and **poof**.
But ripple is NOT A.C. It's pulsating D.C. At no point in the cycle
does the voltage polarity reverse (alternate).
>>>Did you ever think that maybe the cap wasn't large enough for the
>>>application?
>>
>>Perhaps. Then I guess the manufacturer has some 'splainin to do.
>
>
>That happens more times than I would like. GE and Motorola usually
>have well-engineered products; Alvin Amps and Back-Room Communications
>usually don't.
I usually like at least a 20% more capacity in a part than what the
circuit operates at. I'm a bit uncomfortable with a supply running at
850V with two 450V caps in series as filters. But It was very hard
back then to get caps with much higher ratings (that would still
physically fit).
>>> Also, electrolytics are not
>>>always fixed at ground potential, such as when they are used as a DC
>>>block in an audio line. If the offset fails, the cap pops.
>>
>>Absolutely! No argument there. But that's not what we were talking
>>about before.
>
>
>You must remain focused, Dave. We have been discussing several issues
>regarding capacitor failure. In the present topic I had stated that
>electrolytics fail most often due to excessive AC (or ripple) -or-
>reverse voltage, at which point you queried as to how a power supply
>can reverse it's current. I then explained that failure due to voltage
>reversal is rare with high-voltage caps (in tube rigs) but went on to
>explain how the failure is common in low-voltage caps (SS circuits).
But you keep changing the criteria to make you point. What might
happen in a D.C. block in an audio amp would not happen in a power
supply filter.
>>> There are
>>>also instances where an abnormal AC component swings the normally
>>>positive side of the cap negative. The cap voltage can get reversed by
>>>rectification when a pull-up resistor fails. Etc, etc. Geez, Dave,
>>>you've been doing this for 30 years and you didn't know this stuff?
>>
>>Not as applied to the tube amps we were talking about. If you switch
>>the conditions of the discussion, then the outcomes will change as
>>well.
>
>
>Then why did you bring up the subject of tube amps when the original
>topic was about a solid-state CB? If you don't want to discuss it then
>don't bring it up.
There will be no abnormal A.C. component beyond the rectifiers of a
power supply Frank, unless the rectifiers fail.
>
>Perhaps that's because you work more with tubes while I work mostly
>with solid-state equipment.
I thought you just said you were one of those "experts" who restored
old tube radios? Surely you've worked on as many tube radios as I have
considering that I haven't touched anything with tubes in 10 years.
>Based on the fact that I've never experienced what you described, and
>don't see how it's even possible from what I -have- experienced. I
>have never seen a high-voltage cap blow up because it was leaking, and
>I have rarely seen the same blow up from excessive DC voltage. I
>-have- seen them blow up from excessive AC voltage (ripple). Is it
>possible that the cap which blew up was subjected to too much ripple
>caused by leakage in the other cap?
How? The ripple would still not be that great, and it would neither
exceed the voltage rating of th cap, nor reverse the polarity of it.
>>>>Not exactly true. It is the pulsating D.C. current component. There is
>>>>no A.C. after a bridge rectifier.
>>>
>>>
>>>Like I said, call it whatever you like.
>>
>>I like being accurate.
>
>Ok then. Any voltage has a DC component and an AC component.
>Pure AC has no DC component, and pure DC has no AC component.
Those two statements are contradictory.
>The term
>"ripple" is literally defined as the AC component of a DC current,
>usually from a power supply, and usually a harmonic of the input
>power.
No, ripple is the result of pulsating D.C. post the rectifier stage.
The frequency of the ripple depends on the type of rectifier and the
line frequency of the A.C. In the U.S., a half wave ripple frequency
is 60 hz. A full wave would be 120 hz.
>The term "ripple" is old and has been replaced by "pard", which
>includes all AC components (or fluctuations) including hum, noise and
>spikes.
That's because newer switching supplies create more noise, hash, and
spikes, then a simple linear supply would. Hence the need for a new
term, which encompasses a broader definition.
>Is that accurate enough for you?
I guess it'll do.
>>>> The cap can't be subjected to too much ripple if it's doing
>>>>it's job. If there is ripple, then the cap is usually bad.
>>>
>>>
>>>Not necessarily. A heavy load will increase the ripple.
>>
>>Yes it will, but it won't kill the cap. Caps are designed to handle
>>ripple. That's their job.
>
>
>SOME caps are designed to handle ripple. Power supply filter caps are
>designed to REDUCE ripple; i.e, shut the AC component to ground.
No, they do not shunt A.C. to ground. The capacitor stores current
during the ripple peak, and then discharges it as the ripple falls.
The rate of charge/discharge (which is hopefully slower than 60 or 120
hz) is what "smooths" the ripple. When the load increases, the R/C
time constant of the circuit changes and the cap is forced to
discharge faster, which results in less filtering of the ripple and it
increases.
> When
>the load becomes large then so does the ripple, and if the cap has a
>high AC resistance then it will get hot very quickly.
How? There is no A.C.! Pulsating D.C. does not switch polarity.
>>>Again, not necessarily. The DC component may stay within norms but the
>>>AC component (ripple) will increase under load. That's just a fact of
>>>life. Haven't you ever studied the basics of power supplies?
>>
>>I could write a course on power supplies.
>
>
>Just about everyone who has ever used a soldering iron thinks they can
>write a course on power supplies.
Well considering that they were the very first thing that I studied
and built in tech school, it's pretty much basic stuff.
>>An unregulated power supply will decrease voltage as load increases.
>>If the supply is at normal expected voltage, then the load is not
>>excessive. A regulated supply is a different issue. The regulated
>>output may remain at normal, while the unregulated "reserve" drops,
>>along with a proportional increase in ripple.
>>
>>The reason ripple increases is due to the R/C time constant of the
>>circuit. When the load increased, the apparent resistance across the
>>cap drops and the discharge rate increases to the point that it looses
>>the capacity to hold the supply up in between ripple pulses.
>
>
>Nice textbook explanation.
The truth usually is.
> But the increased ripple can be caused by a
>host of factors, not the least of which (and to the suprise of many,
>the most common) is the AC resistance of the output capacitor, -not-
>the R/C time constant of the cap with respect to the source and load
>impedances (which, BTW, are in parallel).
Of course the R/C is in parallel. The cap is across the load.
>A different scenario happens
>with high-voltage supplies because regulation is usually achieved with
>a swinging (saturating) choke, which permits more ripple to pass under
>increasing load in order to maintain the DC component.
That's a very poor regulator. The choke/cap is really just a low pass
filter, that designed to attenuate 60hz ripple.
>So when is your book coming out?
Since power supplies are so basic, and there is no new information
that I could tell that isn't already widely known, my effort would be
superfluous.
>>If a cap explodes due to an excessive load, then the evidence of the
>>failure which cause that excessive load should be evident. If not then
>>you can't conclude that the cap died due to ripple.
>
>
>The cause of failure is not always evident in the equipment.
No, it's not. But in a simple application of a HV power supply in a
tube amp, it's easy to eliminate practically every potential cause.
> Like when
>someone brings in an amp (audio, RF, whatever) that they overloaded,
>overdrove, hooked up wrong or whatever. The dope who blew up the amp
>usually isn't going to admit his mistake, he just wants to get it
>fixed and get the heck out of there so he can put the incident behind
>him.
Yes, but a good tech can usually tell what caused the problem.
> Happens all the time. But contrary to your self-admitted
>cynicism, some people ARE honest
Shame most of them aren't in politics.
>, and that's one way I can determine
>what's causing a specific type of failure. So there -is- evidence of
>the cause of the failure even though it may not be evident from an
>examination of the equipment.
>
>
>>You are right, in a sense, that you can't always tell what DID cause
>>the cap to blow, but you can determine what DID'NT cause it to fail.
>>It then becomes a simple process of elimination.
>
>
>The possibilities are limitless so any such conclusion would be based
>only on probability.
The possibilities are not limitless. There are a finite set of
conditions which a particular circuit can find itself in. Is it
probable that a 12V supply was exposed to 500V? Is it probable that a
rectified output of a supply reversed polarity? Absent of other
component failures, how many possibilities really exist?
>>It's like forensic science. We reconstruct an incident based on the
>>evidence in the aftermath.
>
>
>Dave, before you use the word "forensic" again, do yourself a favor
>and look it up in the dictionary.
Have you never heard of forensic criminalists, who gather evidence to
reconstruct a crime to prove innocence or guilt? Watch CSI sometime to
get familiar with them. Or, if you prefer a less dramatic example, the
Discovery channel has a show called "forensic files" which documents
actual criminal forensic cases. Court TV has similar shows.
The example I used was similar in method.
>>>>I've never seen a cap blow from ripple. Never.
>>>
>>>
>>>You probably have but didn't know it. You can create and measure a
>>>controlled failure with a couple good chokes for AC isolation, a
>>>couple large motor-run capacitors for DC isolation, and an audio amp
>>>to supply the AC. If you need a schematic just let me know.
>>
>>I don't understand the relationship of what you are trying to
>>simulate. I'm talking about filter caps in a power supply now. I
>>should be able to "simulate" this by simply placing a large load
>>across the power supply enough to drop down the voltage and increase
>>the ripple. If the rectifier or the transformer doesn't give up the
>>ghost first,, then the cap should blow according to your theory.
>
>
>The objective is to cause the cap to fail under controlled conditions,
>and hopefully -without- destroying anything else. But if you want to
>do it your way then feel free to do so.
I have a battery charger which is unfiltered (the battery acts as the
filter cap), which may serve the purpose nicely.
>
>>> They also make great regulators because of
>>>their non-linear characteristics.
>>
>> But they can't handle much current.
>
>
>Tell that to my old Heathkit audio generator. And zeners can't handle
>much current either, relatively speaking. Ever hear of a neat little
>device called a 'power transistor'?
Why use a light bulb to do a zener's job?
>>> So you still say
>>>that light bulbs aren't resistors? Hogwash. If it's useful as a
>>>resistor then it's a resistor.
>>
>>I never said they weren't resistors, I said that they aren't practical
>>resistors to substitute in most circuits. A light bulb's resistance
>>isn't constant and changes with voltage. Not something you normally
>>want.
>
>
>A light bulb's resistance changes with temperature, but who said a
>resistor -must- be linear? The textbooks?
The change in temperature is in direct proportion to the applied
voltage.
> And what difference does the voltage make? It's the current
>>>that causes the filament to heat up. Which is true regardless of it's
>>>thickness, too.
>>
>>Current is a function of voltage divided by resistance. A higher
>>voltage bulb can make the same current with a higher resistance. They
>>usually control this with the length and thickness of the filament.
>>Shorter, thicker filaments are more durable. That's also why a 25 watt
>>bulb normally lasts longer than a 100 watt bulb.
>
>
>Not in my house. I replace 40 watt bulbs just as often as 60 watt and
>100 watt bulbs.
Not me. I get almost twice the life from a 40 watt bulb than a 100
watt bulb.
>And again, the it's the temperature of the filament
>that determines it's brightness, which is a direct consequence of its
>current.
Right, and the resistance needs to be lowered for a brighter, higher
wattage bulb, given the same voltage. The temperature is higher, and
the MTBF is shorter.
> IOW, the current density through a cross-section of filament
>is the same regardless of it's thickness or length.
A 12V bulb will have a thicker filament than a 120V bulb for the same
given wattage. That's because the amount of current in a 12V bulb is
10 times higher than the same wattage bulb in a 120 V configuration.
>>>Age, yes, they do change to a greater or lesser degree. Usually
>>>lesser. Humidity? I don't think so, Dave. Not when they are
>>>hermetically sealed. And the only caps that aren't hermetically sealed
>>>are air caps. Have you ever seen an "air electrolytic"? I haven't.
>>
>>They are also affected by long term exposure to heat above their heat
>>rating.
>
>
>Duh. Even to long-term exposure to heat -below- their heat rating.
The time is much shorter when the heat is higher.
>>Other techs and engineers that I know in the industry.
>
>
>Well, I'm an engineer, you know me, and I haven't ever used the term
>"drying out" to describe a capacitor failure. Maybe "dielectrically
>challenged"..... no, I haven't used that either.
You are but one man.
>>>>>Yet because I rejected your conclusion (or, if you
>>>>>prefer, your 'declaration of high probability'), you assert that I am
>>>>>claiming the fault is -not- a capacitor. That is a perfect example of
>>>>>the logical fallacy 'Argumentum ad Ignorantiam' which, by definition,
>>>>>is ignorance.
>>>>
>>>>Not at all. If I claim that there is a high probability of component
>>>>"A" failing and you disagree. Then you are saying, in essence, that
>>>>there is a very low probability that component "A" is the failure.
>>>
>>>
>>>Wrong. Disagreement is -not- an assertion of the opposite.
>>
>>It is in this case.
>
>
>Well, I'm pretty sure about this, Dave. After all, I was the first
>person to ever ace the logic class (Phil 201) in college, which is
>probably why they brought me in a year later to teach it for a quarter
>while the doctor was out.
We have only your word on that.
>>>>If you consider component "A" to have a low probability of failure it
>>>>stands to reason that you place a higher probability on another
>>>>component failing.
>>>
>>>
>>>Again wrong, and for the same reason.
>>
>>Then what are you saying?
>
>
>You remind me of the sitcom boyfriends when they get dumped by their
>girlfriends who scream something like, "I never want to see you again
>for the rest of my life!", and the boyfriend responds with, "So what
>are you saying?"
Regardless of the other possibilities, if you disagree that a
particular component has a high probability of failure, then what you
are saying, in essence, is that you feel that it has a much lower
probability of failure Therefore, by extension, it can be assumed that
you then consider other components have a higher probability of
failure.
>>>>More chest thumping? What you've done at HP has no bearing on what
>>>>I've seen in 30+ years of CB radio repair.
>>>
>>>
>>>Really? Do capacitors behave differently in CB radios than they do in
>>>spectrum analyzers and signal generators?
>>
>>Sure. They're exposed to different conditions.
>
>
>Such as.....?
Application. Does the circuit supply large currents or small ones?
Does it work at high voltage or low? Is the ambient temperature high,
or low? Do they filter, block, or bypass?
>>Actually TR5 is the 1st RF front end.
>
>
>So it is. Close enough.
>
>
>> it's the very first active
>>device in the chain. TR4 is an AGC clamp. The second device is the
>>mixer and it's a 3SK458 FET.
>>
>>> On page 15 is the transmitter alignment section, and
>>>about the middle of the page is Step 4 where the procedure is to align
>>>the transmit coils while shoving 500 mV into the mic jack. Yet Step 8
>>>is to set the modulation for 98% with a 200 mV input. Now there is a
>>>reason for driving the audio so hard in Step 4 and that is to make
>>>sure the signal path can handle the stress, because if it can cope
>>>with a 500 mV input then a mere 200 mV should be well within the safe
>>>operating area.
>>
>>Hmm.... My test procedure for that rig used 5mV to set 50% modulation,
>>with VR6
>
>Step 7, and it's 0.5 mV.
It's 5mV in my book.
>> and 200 mV to set the 100% limiter threshold with VR7.
>
>
>Step 8, set VR7 for 95 to 98% mod.
Right. But there are no numbered steps.
>>The reason for the increase in audio drive is to test that the
>>modulation limiter (and mike AGC in this particular radio) is doing
>>its job. Think about it, if the modulation limiter is working and
>>doing its job, then you can increase the mike level all you want
>>(until you fry something) and the AMC will clamp it to the level that
>>makes 100% modulation. The only parts in the audio chain that would be
>>stressed are the passive components before the AMC audio shunt
>>regulator, which is TR18.
>>
>>There is very little you could tell me about that radio (or more
>>specifically that chassis) that I don't already know.
>
>
>I just did. Twice in this post and once in the previous post.
What you "told" me wasn't of any real info about the radio itself.
>
>Thanks anyway, but I already used too much time pulling out the mods
>and restoring it to stock condition -- except for the faceplate. Got a
>spare?
No I don't. Most of my spare faceplates belong to the bonepile radios,
most of which are crappy little AM mobiles.
>>>Now you claim to have worked on CBs for 30 years, yet you have never
>>>seen a procedure as I just described (on what was once a fairly common
>>>and popular radio)?
>>
>>I have seen similar procedures, but they were not used in the way you
>>interpreted it. I told you why they increase the level, and that's to
>>verify proper AMC action.
>
>
>I suppose that's why the manual says to overdrive the audio -before-
>setting the AMC, huh?
My manual says or implies nothing of the sort. It instructs you to set
VR6 with 5 mV for 50% modulation, and then set the generator to 200mV
and set 98% with VR7. It doesn't tell you to overdrive anything.
Nowhere does it call for raising it to 500mV as you initially claimed.
>>>>Often the only reason given in the procedures to increase the audio
>>>>input (And I have a nice Tektronix generator) is to verify proper
>>>>operation of the modulation limiter.
>>
>>
>>>I can send a pair of reading glasses with that book if you want.
>>
>>Your failure to properly identify the true reason why the audio level
>>is increased, is not suggestive of my poor eyesight.
>
>
>You need specs because you can't read what's in front of your face.
>That's what happens when you get old.
Maybe you are the one "getting old", as I see nothing of the sort.
>> It's a
>>blessing for techs. Most CB radios could be fixed with Radio Shack
>>parts. At one time, Radio Shack actually carried RF final amps
>>(2SC-1307). It's a shame that they have scaled back their parts so
>>much.
>
>
>You have finally said something to which I agree without reservation.
>I'll even go one step further and say that Radio Shack sucks.
Once upon a time it was better, as was Lafayette Radio. Then there
were placed like Almo electronics and other regional parts houses.....
>>>>I have never seen an audio stage on a CB radio "lock up". The Hy-Gain
>>>>623 used to have an interestingly annoying "design flaw" where if the
>>>>modulator was driven too hard, it would shut down for a few seconds.
>>>>You would loose receive audio until it recovered.
>>>
>>>
>>>That sounds like a lock-up to me.
>>
>>It was a design flaw (or "feature" depending on your perspective).
>
>
>It was intentional?
I assume so, since everyone I've ever seen has done it.
Dave
"Sandbagger"
I AmnotGeorgeBush
Quick search reveals 4ax.com appearing in Hall's headers and posts in
addition to Doug,s also. Is fact, when that search is complete, it's
very interesting to retrace the threads and note just where 4ax.com
entered this group and who was involved in EVERY conversation relating
to the forges and gay postings whereas 4ax.com is to be seen. I love
usenet,,it;s extremely consistent! : )
dug88
you do give dismal new meaning
let me explain
if he took a ride north
he could not cope with 73 of 86 industires
crying for workers
gold emeralds diamonds oil sands.
gay reference
here is a hint
about 3 to 9 million are needed here in edmmonton in less than ten years
little dea of YOUR potential
the potential of this area is known
crying for techs
time to go take a swim and sauna
so, what is important to you
your BIG nuts
ha ha ha
become a man of the world
when you are ready to sell the farm
i am likely to look to buy the farm
other notices pending
"moparholic" <mopar...@hotmail-nospam.com> wrote in message
news:41f08819$1...@news.usenetzone.com...
>
> "Steveo" <mopar...@hotmail-nospam.com> wrote in message
> news:20050120232903.592$G...@newsreader.com...
>> itoldyouiamnotiamnotgeorge <itoldyouiamno...@yahoo.comSpam>
>> wrote:
>>> Frank you really are suckinfg twistys dick hard these days,
>>>
>> There's no call for that. You can disagree with someone without using
>> gay references.
>
> Steveo's right. when you make gay references it makes Steveo feel
> unwelcome here. Just because Steveo is homosexual doesn't mean that he
> doesn't have as much right as you to post here.
>
> Plus you bring up memories of Steveo's father leaving the home when Steveo
> was very young for another man.We don't need to remind him of this over
> and over again.
Frank Gilliland
weekend, so what the hell.....
On Fri, 28 Jan 2005 10:45:15 -0500, Dave Hall <nojunk...@ptd.net>
wrote in <bahkv05kjt6rpkidv...@4ax.com>:
>On Thu, 27 Jan 2005 08:23:13 -0800, Frank Gilliland
><wīrenut@NOSPAMīcehouse.net> wrote:
>
>>>>Maybe that's happened to you but never on my bench, not unless the cap
>>>>was too small for the load. Like I said before, I have never seen a
>>>>cap that blew up on it's own accord, leaky or not (and I don't know
>>>>how you can determine it was leakage that caused your cap to blow
>>>>unless you have a time machine in your shop).
>>>
>>>Shhhhh. That a "way back" machine, and it's only a prototype.
>>>
>>>But you make a good point. But like Sherlock Holmes once said, when
>>>all the other possibilities are excluded, what remains, however
>>>unlikely must be the answer.
>>
>>
>>Sherlock Holmes was not a real person
>
>No, but Sir Arthur Conan Doyle was, and he wrote it.
Sir Arthur Conan Doyle wrote fiction.
>>, and that poorly-quoted cliche
>>has one very important point you completely missed: "when all the
>>other possibilities are excluded". Care to apply that concept to the
>>debate between creationism and evolution?
>
>Why? That's an entirely different subject with limitless and
>unverifiable possibilities. Unlike the subject we're currently
>discussing. There are a finite amount of potential causes of cap
>failure. Most of which can be reproduced.
If a cap can fail due to an act of God then it can't be verified. Your
argument is therefore invalid unless you are an athiest.
>> Or can you see that there
>>are "other possibilities" that may not be obvious but they do exist?
>
>Not in this particular case.
That's a conclusion based on your own ignorance.
<snip>
>>> In most sweep tube amps and tube CB radios, I rarely see
>>>more than about 40 uF in filter caps. Voltage rating depends on the
>>>circuit. Most amps use 450V caps in series.
>>
>>
>>I have a cap handy that's rated 40 uF @ 440 VAC. It must weigh at
>>least five pounds and measures 10" x 4" x 2" (want a pic?). It's big
>>because it's designed to handle lots of AC. I also have another
>>capacitor handy that's rated 40 uF @ 450 VDC. It's barely 2" long and
>>1" diameter. It can't handle much AC at all. I have seen the caps used
>>in sweep tube amps and they aren't designed to handle much AC at all.
>>Load them down with some ripple and **poof**.
>
>But ripple is NOT A.C. It's pulsating D.C. At no point in the cycle
>does the voltage polarity reverse (alternate).
Despite your archaic definition of AC, the point is still valid; Those
large AC caps ("motor run" caps, to be specific) are frequently used
by audiophiles as DC power supply caps in SE tube amps because they
have a very low AC resistance, much lower than electrolytic caps.
Power supply caps are selected for their AC resistance, the capacity
in uF being only one of -several- factors figuring into the equation.
<snip>
>>>> Also, electrolytics are not
>>>>always fixed at ground potential, such as when they are used as a DC
>>>>block in an audio line. If the offset fails, the cap pops.
>>>
>>>Absolutely! No argument there. But that's not what we were talking
>>>about before.
>>
>>
>>You must remain focused, Dave. We have been discussing several issues
>>regarding capacitor failure. In the present topic I had stated that
>>electrolytics fail most often due to excessive AC (or ripple) -or-
>>reverse voltage, at which point you queried as to how a power supply
>>can reverse it's current. I then explained that failure due to voltage
>>reversal is rare with high-voltage caps (in tube rigs) but went on to
>>explain how the failure is common in low-voltage caps (SS circuits).
>
>But you keep changing the criteria to make you point. What might
>happen in a D.C. block in an audio amp would not happen in a power
>supply filter.
I see you have finally caught up.
>>>> There are
>>>>also instances where an abnormal AC component swings the normally
>>>>positive side of the cap negative. The cap voltage can get reversed by
>>>>rectification when a pull-up resistor fails. Etc, etc. Geez, Dave,
>>>>you've been doing this for 30 years and you didn't know this stuff?
>>>
>>>Not as applied to the tube amps we were talking about. If you switch
>>>the conditions of the discussion, then the outcomes will change as
>>>well.
>>
>>
>>Then why did you bring up the subject of tube amps when the original
>>topic was about a solid-state CB? If you don't want to discuss it then
>>don't bring it up.
>
>There will be no abnormal A.C. component beyond the rectifiers of a
>power supply Frank, unless the rectifiers fail.
A highly inductive load can drive the caps backwards via flyback. I've
never seen it happen myself but I know it's possible.
>>Perhaps that's because you work more with tubes while I work mostly
>>with solid-state equipment.
>
>I thought you just said you were one of those "experts" who restored
>old tube radios? Surely you've worked on as many tube radios as I have
>considering that I haven't touched anything with tubes in 10 years.
Maybe I have done more work on tubes than you. Maybe not. Maybe my
bench has seen more receivers while your's has seen more sweep-tube
amps. Maybe not. I merely suggested a possibility.
>>Based on the fact that I've never experienced what you described, and
>>don't see how it's even possible from what I -have- experienced. I
>>have never seen a high-voltage cap blow up because it was leaking, and
>>I have rarely seen the same blow up from excessive DC voltage. I
>>-have- seen them blow up from excessive AC voltage (ripple). Is it
>>possible that the cap which blew up was subjected to too much ripple
>>caused by leakage in the other cap?
>
>How? The ripple would still not be that great, and it would neither
>exceed the voltage rating of th cap, nor reverse the polarity of it.
Well, if the AC resistance of the leaky cap dropped because of the
leak, then a greater AC voltage would be developed across the good
cap. This in turn would cause a greater AC current in the good cap,
and therefore more heating. So maybe the cap blew because of the extra
heat in combination with the higher DC stresses. I can see how -that-
could happen.
>>>>>Not exactly true. It is the pulsating D.C. current component. There is
>>>>>no A.C. after a bridge rectifier.
>>>>
>>>>
>>>>Like I said, call it whatever you like.
>>>
>>>I like being accurate.
>>
>
>>Ok then. Any voltage has a DC component and an AC component.
>
>>Pure AC has no DC component, and pure DC has no AC component.
>
>Those two statements are contradictory.
Not at all. Your definition that AC must be a zero-crossing current
was repealed along with prohibition. The problem was that there were
getting to be too many terms for variations of direct current, like
'varying direct current' (vdc), 'pulsating direct current' (pdc),
'interrupted direct current'.....etc. So the concept has since been
simplified to say that any current has an AC component and a DC
component.
>>The term
>>"ripple" is literally defined as the AC component of a DC current,
>>usually from a power supply, and usually a harmonic of the input
>>power.
>
>No, ripple is the result of pulsating D.C. post the rectifier stage.
>The frequency of the ripple depends on the type of rectifier and the
>line frequency of the A.C. In the U.S., a half wave ripple frequency
>is 60 hz. A full wave would be 120 hz.
That's what I said. If the input power is 60 Hz then 120 Hz would be a
harmonic, right?
>>The term "ripple" is old and has been replaced by "pard", which
>>includes all AC components (or fluctuations) including hum, noise and
>>spikes.
>
>That's because newer switching supplies create more noise, hash, and
>spikes, then a simple linear supply would. Hence the need for a new
>term, which encompasses a broader definition.
If you say so.
>>Is that accurate enough for you?
>
>I guess it'll do.
>
>
>>>>> The cap can't be subjected to too much ripple if it's doing
>>>>>it's job. If there is ripple, then the cap is usually bad.
>>>>
>>>>
>>>>Not necessarily. A heavy load will increase the ripple.
>>>
>>>Yes it will, but it won't kill the cap. Caps are designed to handle
>>>ripple. That's their job.
>>
>>
>>SOME caps are designed to handle ripple. Power supply filter caps are
>>designed to REDUCE ripple; i.e, shut the AC component to ground.
>
>No, they do not shunt A.C. to ground. The capacitor stores current
>during the ripple peak, and then discharges it as the ripple falls.
>The rate of charge/discharge (which is hopefully slower than 60 or 120
>hz) is what "smooths" the ripple. When the load increases, the R/C
>time constant of the circuit changes and the cap is forced to
>discharge faster, which results in less filtering of the ripple and it
>increases.
A capacitor blocks DC and passes AC; at least that's the simplified
version of it's function. So if you have a cap connected between a DC
source and ground then the DC is blocked while any AC is passed to
ground. How well the capacitor passes AC to ground depends upon it's
AC resistance. What you describe above is essentially the same thing,
but it's overcomplicated and full of holes because you never learned
the basics.
>> When
>>the load becomes large then so does the ripple, and if the cap has a
>>high AC resistance then it will get hot very quickly.
>
>How? There is no A.C.! Pulsating D.C. does not switch polarity.
Update your lexicon and then read it again.
>>>>Again, not necessarily. The DC component may stay within norms but the
>>>>AC component (ripple) will increase under load. That's just a fact of
>>>>life. Haven't you ever studied the basics of power supplies?
>>>
>>>I could write a course on power supplies.
>>
>>
>>Just about everyone who has ever used a soldering iron thinks they can
>>write a course on power supplies.
>
>Well considering that they were the very first thing that I studied
>and built in tech school, it's pretty much basic stuff.
I hate to break this to you but your tech school did a piss-poor job
if the curriculum started with the study and construction of complex
circuits before covering the basics.
>>>An unregulated power supply will decrease voltage as load increases.
>>>If the supply is at normal expected voltage, then the load is not
>>>excessive. A regulated supply is a different issue. The regulated
>>>output may remain at normal, while the unregulated "reserve" drops,
>>>along with a proportional increase in ripple.
>>>
>>>The reason ripple increases is due to the R/C time constant of the
>>>circuit. When the load increased, the apparent resistance across the
>>>cap drops and the discharge rate increases to the point that it looses
>>>the capacity to hold the supply up in between ripple pulses.
>>
>>
>>Nice textbook explanation.
>
>The truth usually is.
>
>
>> But the increased ripple can be caused by a
>>host of factors, not the least of which (and to the suprise of many,
>>the most common) is the AC resistance of the output capacitor, -not-
>>the R/C time constant of the cap with respect to the source and load
>>impedances (which, BTW, are in parallel).
>
>Of course the R/C is in parallel. The cap is across the load.
The cap is in parallel with -both- the source and load impedances; and
while the load impedance may be linear, that's rarely the case for the
source impedance. Also, the capacitance of an electrolytic isn't a
constant; it's non-linear with respect to the voltage (a factor of
it's AC resistance, and another reason why audiophiles prefer motor
caps over electrolytics). Both these nonlinearities end up converting
your overly-simplistic R/C time constant analysis into an integral
function. The issue is further complicated because of the equivalent
series resistance of the capacitor (which, BTW, is a factor of it's AC
resistance, and is also non-linear).
The concepts involved here don't need to be nearly as complicated as
you make them out to be, yet you try to describe the complexities with
over-simplified theory. It's like you are trying to describe baseball
with dumbed-down Newtonian analysis. You still think you can write a
course on the subject? I don't think so.
>>A different scenario happens
>>with high-voltage supplies because regulation is usually achieved with
>>a swinging (saturating) choke, which permits more ripple to pass under
>>increasing load in order to maintain the DC component.
>
>That's a very poor regulator. The choke/cap is really just a low pass
>filter, that designed to attenuate 60hz ripple.
That's the static interpretation. But a swinging choke is a component
with dynamic (i.e, nonlinear) characteristics and makes a pretty good
regulator, which is why they have been used so extensively over the
past hundred years, and why they are still in production today.
>>So when is your book coming out?
>
>Since power supplies are so basic, and there is no new information
>that I could tell that isn't already widely known, my effort would be
>superfluous.
I think you might have meant 'redundant', but 'superfluous' is really
a much better choice.
>>>If a cap explodes due to an excessive load, then the evidence of the
>>>failure which cause that excessive load should be evident. If not then
>>>you can't conclude that the cap died due to ripple.
>>
>>
>>The cause of failure is not always evident in the equipment.
>
>No, it's not. But in a simple application of a HV power supply in a
>tube amp, it's easy to eliminate practically every potential cause.
It's not always easy.
<snip>
>>>You are right, in a sense, that you can't always tell what DID cause
>>>the cap to blow, but you can determine what DID'NT cause it to fail.
>>>It then becomes a simple process of elimination.
>>
>>
>>The possibilities are limitless so any such conclusion would be based
>>only on probability.
>
>The possibilities are not limitless. There are a finite set of
>conditions which a particular circuit can find itself in. Is it
>probable that a 12V supply was exposed to 500V?
Depends on the circuit.
> Is it probable that a
>rectified output of a supply reversed polarity?
I never said it was.
> Absent of other
>component failures, how many possibilities really exist?
Many, simply because the cause of a failure isn't limited to internal
influences. But even if you limit the scope to the equipment, the
failure of other components is not always obvious. And here's a
perfect example:
The first Peavey 1.3K I ever worked on came into the shop with a blown
bank of power transistors (not uncommon with Peavey amps) and a bad
OP-amp. The obvious conclusion is that the OP-amp locked up and drove
the bank into overload (the amp also has a triac clamp but it rarely
works -- it just nukes a trace on the board and lets the bank blow).
Anyway, that wasn't the case at all. I replaced damn near everything
in that amp and it would still lock up. To make an already long story
short... it turns out that Peavey, in their economical wisdom, used a
dual-diode as bias seperation for opposite banks of power transistors
without noting it in their literature, and one of those diodes had
shorted. The component appeared to be perfectly functional, showing a
forward voltage drop characteristic of a typical silicon diode, but it
was supposed to be -twice- that drop.
So there are -many- possibilities that may or may -not- be obvious.
>>>It's like forensic science. We reconstruct an incident based on the
>>>evidence in the aftermath.
>>
>>
>>Dave, before you use the word "forensic" again, do yourself a favor
>>and look it up in the dictionary.
>
>Have you never heard of forensic criminalists, who gather evidence to
>reconstruct a crime to prove innocence or guilt? Watch CSI sometime to
>get familiar with them. Or, if you prefer a less dramatic example, the
>Discovery channel has a show called "forensic files" which documents
>actual criminal forensic cases. Court TV has similar shows.
>
>The example I used was similar in method.
Gee, another excuse.
>>>>>I've never seen a cap blow from ripple. Never.
>>>>
>>>>
>>>>You probably have but didn't know it. You can create and measure a
>>>>controlled failure with a couple good chokes for AC isolation, a
>>>>couple large motor-run capacitors for DC isolation, and an audio amp
>>>>to supply the AC. If you need a schematic just let me know.
>>>
>>>I don't understand the relationship of what you are trying to
>>>simulate. I'm talking about filter caps in a power supply now. I
>>>should be able to "simulate" this by simply placing a large load
>>>across the power supply enough to drop down the voltage and increase
>>>the ripple. If the rectifier or the transformer doesn't give up the
>>>ghost first,, then the cap should blow according to your theory.
>>
>>
>>The objective is to cause the cap to fail under controlled conditions,
>>and hopefully -without- destroying anything else. But if you want to
>>do it your way then feel free to do so.
>
>I have a battery charger which is unfiltered (the battery acts as the
>filter cap), which may serve the purpose nicely.
Go for it! Don't forget to use several different caps, measuring both
AC and DC volts across the cap, as well as the time it takes until
failure. Post the results.
>>>> They also make great regulators because of
>>>>their non-linear characteristics.
>>>
>>> But they can't handle much current.
>>
>>
>>Tell that to my old Heathkit audio generator. And zeners can't handle
>>much current either, relatively speaking. Ever hear of a neat little
>>device called a 'power transistor'?
>
>Why use a light bulb to do a zener's job?
The point was that the characteristics of a low-power device can be
amplified. And yes, a light bulb is sometimes preferable to a zener
because zeners are noisy little bastards.
>>>> So you still say
>>>>that light bulbs aren't resistors? Hogwash. If it's useful as a
>>>>resistor then it's a resistor.
>>>
>>>I never said they weren't resistors, I said that they aren't practical
>>>resistors to substitute in most circuits. A light bulb's resistance
>>>isn't constant and changes with voltage. Not something you normally
>>>want.
>>
>>
>>A light bulb's resistance changes with temperature, but who said a
>>resistor -must- be linear? The textbooks?
>
>The change in temperature is in direct proportion to the applied
>voltage.
Wrong. The resistance is directly proportional to the temperature,
hence the term "temperature coefficient". A light bulb can even be
used as a thermistor for an electronic thermometer. IOW, you can feed
a bulb's filament with a fixed voltage and a constant current will
flow (after the temperature stablizes). Cool the filament and the
current will increase (light bulbs make pretty good wind-chill
detectors). Heat the filament and the current will decrease (good for
determining the temperature of a flame). All happen -irrespective- of
voltage.
Again, you need to go back and learn the basics.
<snip>
>> IOW, the current density through a cross-section of filament
>>is the same regardless of it's thickness or length.
>
>A 12V bulb will have a thicker filament than a 120V bulb for the same
>given wattage. That's because the amount of current in a 12V bulb is
>10 times higher than the same wattage bulb in a 120 V configuration.
True, but the cross-sectional current density must be the same in
order to reach the same temperature where it emits the same spectrum.
So why would a thicker filament last longer?
<snip>
>>>Other techs and engineers that I know in the industry.
>>
>>
>>Well, I'm an engineer, you know me, and I haven't ever used the term
>>"drying out" to describe a capacitor failure. Maybe "dielectrically
>>challenged"..... no, I haven't used that either.
>
>You are but one man.
Feel free to name some engineers who use the term "drying out" to
describe capacitor failure. Even better, try referencing it in some
authoritative literature. If it's a common term then it should be well
published, right? (Actually, it -is- a common term with engineers but
only in reference to batteries.)
<snip>
>>>>>If you consider component "A" to have a low probability of failure it
>>>>>stands to reason that you place a higher probability on another
>>>>>component failing.
>>>>
>>>>
>>>>Again wrong, and for the same reason.
>>>
>>>Then what are you saying?
>>
>>
>>You remind me of the sitcom boyfriends when they get dumped by their
>>girlfriends who scream something like, "I never want to see you again
>>for the rest of my life!", and the boyfriend responds with, "So what
>>are you saying?"
>
>Regardless of the other possibilities, if you disagree that a
>particular component has a high probability of failure, then what you
>are saying, in essence, is that you feel that it has a much lower
>probability of failure
You are constructing a syllogism from a conditional statement by
asserting the second premise, but you assume that second premise
-must- be the only other possibility. It is not. Using your own
subjective assessment of probability, the blank line in the statement
"the problem is _____________ caused by capacitors"
can be filled by any one of a long list of possible descriptors, only
a few of which are listed here:
A. always
B. almost always
C. often
D. sometimes
E. infrequently
F. almost never
G. never
The syllogism that you constructed is thus:
P1: Not B (my statement).
P2: If not B then F.
C: Therefore F.
Although the conclusion is perfectly valid, you ASSUMED that F is the
only possible alternative to B. The conclusion, therefore, is neither
true nor false; it can have no more truth than your assumed premise.
Just because I don't live at the North Pole doesn't mean I -must- live
at the South Pole. Or, for a better example, if you were to say that
the sky is white, and I said it is -not- white, that doesn't mean I'm
saying it's black. It could be blue, red, gray, etc. The colors of the
sky are limited to black and white only in your fantasy realm where,
coincidently, people must be either conservative or liberal.
> Therefore, by extension, it can be assumed that
>you then consider other components have a higher probability of
>failure.
Key word: "assumed". You assumed wrong, as is usually the case
whenever people ASSume things.
>>>>>More chest thumping? What you've done at HP has no bearing on what
>>>>>I've seen in 30+ years of CB radio repair.
>>>>
>>>>
>>>>Really? Do capacitors behave differently in CB radios than they do in
>>>>spectrum analyzers and signal generators?
>>>
>>>Sure. They're exposed to different conditions.
>>
>>
>>Such as.....?
>
>Application. Does the circuit supply large currents or small ones?
>Does it work at high voltage or low? Is the ambient temperature high,
>or low? Do they filter, block, or bypass?
All of the above in both applications. So how are they different?
<snip>
>>>Hmm.... My test procedure for that rig used 5mV to set 50% modulation,
>>>with VR6
>>
>>Step 7, and it's 0.5 mV.
>
>It's 5mV in my book.
I don't know what book you have (probably Sams) but mine is the
service manual provided by Radio Shaft. I can scan the page if you
want. If your book is in error that might explain your need to bypass
the LPF. Higher drive at mid-modulation will push the audio into
compression earlier, which would easily account for your audio
sounding "mushy and compressed", just as you described. My radio
sounds just fine without the modification, so it's very likely that
-your- book is in error. Not only that, I'm suprised that a person of
your self-proclaimed experience would not have suspected the error
earlier -- after all, your own description of the audio quality is
indicative of overmodulation. Yet you didn't bother to find a more
authoritative reference, or compare the calibration specification to
that of similar radios (I did, and 50% modulation at 0.5mV is common
for Uniden radios that come with the same mic). You could have also
compared the cal-spec the radio's rated mic sensitivity. But no, you
decided -not- to look for the primary cause of the problem, assumed
that the error was in the design and proceeded to chop the filter.
Dave, it's clear that in -both- electronics and politics you have a
real problem accepting any possibilities outside the realm of your own
prejudice.
>>> and 200 mV to set the 100% limiter threshold with VR7.
>>
>>
>>Step 8, set VR7 for 95 to 98% mod.
>
>Right. But there are no numbered steps.
There are in my book. And your's seems to be missing some steps.
>>>The reason for the increase in audio drive is to test that the
>>>modulation limiter (and mike AGC in this particular radio) is doing
>>>its job. Think about it, if the modulation limiter is working and
>>>doing its job, then you can increase the mike level all you want
>>>(until you fry something) and the AMC will clamp it to the level that
>>>makes 100% modulation. The only parts in the audio chain that would be
>>>stressed are the passive components before the AMC audio shunt
>>>regulator, which is TR18.
>>>
>>>There is very little you could tell me about that radio (or more
>>>specifically that chassis) that I don't already know.
>>
>>
>>I just did. Twice in this post and once in the previous post.
>
>What you "told" me wasn't of any real info about the radio itself.
I told you some -very- real info about the radio -- you just didn't
know how important that info is, or how it demonstrates that you
aren't as good with electronics as you think you are.
>>Thanks anyway, but I already used too much time pulling out the mods
>>and restoring it to stock condition -- except for the faceplate. Got a
>>spare?
>
>No I don't. Most of my spare faceplates belong to the bonepile radios,
>most of which are crappy little AM mobiles.
Heard that. Did you see the pic of my "bonepile" that I posted last
fall? I must have a couple dozen of those cheap, little Cobras buried
in the pile, not to mention a half-dozen or so mini-Midlands (which
weren't too awfully bad little radios, just not worth fixing). I
should go through the pile someday, make a list and post it for anyone
looking for parts.
<snip>
----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
Dave Hall
<wīrenut@NOSPAMīcehouse.net> wrote:
>>>Sherlock Holmes was not a real person
>>
>>No, but Sir Arthur Conan Doyle was, and he wrote it.
>Sir Arthur Conan Doyle wrote fiction.
But the principle is still sound.
>
>>>, and that poorly-quoted cliche
>>>has one very important point you completely missed: "when all the
>>>other possibilities are excluded". Care to apply that concept to the
>>>debate between creationism and evolution?
>>
>>Why? That's an entirely different subject with limitless and
>>unverifiable possibilities. Unlike the subject we're currently
>>discussing. There are a finite amount of potential causes of cap
>>failure. Most of which can be reproduced.
>
>
>If a cap can fail due to an act of God then it can't be verified. Your
>argument is therefore invalid unless you are an athiest.
What act of God kills caps? Lightning? A lightning strike can be
easily eliminated as it's fairly obvious when it occurs.
>>> Or can you see that there
>>>are "other possibilities" that may not be obvious but they do exist?
>>
>>Not in this particular case.
>
>
>That's a conclusion based on your own ignorance.
No, it's one based on probability and practicality.
>>But ripple is NOT A.C. It's pulsating D.C. At no point in the cycle
>>does the voltage polarity reverse (alternate).
>
>
>Despite your archaic definition of AC
It is THE definition of A.C.
>, the point is still valid; Those
>large AC caps ("motor run" caps, to be specific) are frequently used
>by audiophiles as DC power supply caps in SE tube amps because they
>have a very low AC resistance, much lower than electrolytic caps.
>Power supply caps are selected for their AC resistance, the capacity
>in uF being only one of -several- factors figuring into the equation.
Motor run caps are not the ones that blow up in amplifier power
supplies.
>>But you keep changing the criteria to make you point. What might
>>happen in a D.C. block in an audio amp would not happen in a power
>>supply filter.
>
>
>I see you have finally caught up.
I never needed to "catch up", but when we're discussing a particular
failure mode in one application, and you bring up a different
application as a justification for a differing point, that does
nothing but cloud the point.
>>>>> There are
>>>>>also instances where an abnormal AC component swings the normally
>>>>>positive side of the cap negative. The cap voltage can get reversed by
>>>>>rectification when a pull-up resistor fails. Etc, etc. Geez, Dave,
>>>>>you've been doing this for 30 years and you didn't know this stuff?
>>>>
>>>>Not as applied to the tube amps we were talking about. If you switch
>>>>the conditions of the discussion, then the outcomes will change as
>>>>well.
>>>
>>>
>>>Then why did you bring up the subject of tube amps when the original
>>>topic was about a solid-state CB? If you don't want to discuss it then
>>>don't bring it up.
>>
>>There will be no abnormal A.C. component beyond the rectifiers of a
>>power supply Frank, unless the rectifiers fail.
>
>
>A highly inductive load can drive the caps backwards via flyback. I've
>never seen it happen myself but I know it's possible.
Yes, that is certainly possible. But not in the equipment that I've
worked on. There is no inductive components past the power supply
filter caps, with the exception of a small 10 uH RF blocking choke.
Hardly enough to generate enough "kick back" to reverse polarity.
>
>>>Perhaps that's because you work more with tubes while I work mostly
>>>with solid-state equipment.
>>
>>I thought you just said you were one of those "experts" who restored
>>old tube radios? Surely you've worked on as many tube radios as I have
>>considering that I haven't touched anything with tubes in 10 years.
>Maybe I have done more work on tubes than you. Maybe not. Maybe my
>bench has seen more receivers while your's has seen more sweep-tube
>amps. Maybe not. I merely suggested a possibility.
I can accept that. My bench has probably seen the most 12V mobile
radios, with amplifiers a distant second. But the equipment which has
seen the most capacitor failures due to outgassing or explosion has
been the tube amps. In the solid state equipment, cap failures are
less obvious and usually not total. On older radios (25+ years) the
main cap failure mode is excessive deviation from original value. You
don't like the term "drying out", but whatever process is responsible
for this phenomenon is quite prevalent.
>>>Based on the fact that I've never experienced what you described, and
>>>don't see how it's even possible from what I -have- experienced. I
>>>have never seen a high-voltage cap blow up because it was leaking, and
>>>I have rarely seen the same blow up from excessive DC voltage. I
>>>-have- seen them blow up from excessive AC voltage (ripple). Is it
>>>possible that the cap which blew up was subjected to too much ripple
>>>caused by leakage in the other cap?
>>
>>How? The ripple would still not be that great, and it would neither
>>exceed the voltage rating of th cap, nor reverse the polarity of it.
>Well, if the AC resistance of the leaky cap dropped because of the
>leak, then a greater AC voltage would be developed across the good
>cap. This in turn would cause a greater AC current in the good cap,
>and therefore more heating. So maybe the cap blew because of the extra
>heat in combination with the higher DC stresses. I can see how -that-
>could happen.
That's exactly what I told you a few posts back. Only I didn't use the
term A.C.. A leaky cap decreases D.C. resistance, which decreases A.C.
resistance as well. Nice of you to finally get my point.
Wait, you're trying to play with my head aren't you? Trying to make me
change my tune? That only works when the "tune" is sufficiently
different.
>>>Ok then. Any voltage has a DC component and an AC component.
>>
>>>Pure AC has no DC component, and pure DC has no AC component.
>>
>>Those two statements are contradictory.
>
>
>Not at all. Your definition that AC must be a zero-crossing current
>was repealed along with prohibition. The problem was that there were
>getting to be too many terms for variations of direct current, like
>'varying direct current' (vdc), 'pulsating direct current' (pdc),
>'interrupted direct current'.....etc. So the concept has since been
>simplified to say that any current has an AC component and a DC
>component.
You stated: "Any voltage has a DC component and an AC component. "
Then you stated: "Pure AC has no DC component, and pure DC has no AC
component."
Yet you cannot see how these two statements contradict each other?
If you change the first statement to: "Certain voltages have a DC
component and an AC component. ", then the two statements would not
conflict.
For the record, I agree with your second statement.
>>>The term
>>>"ripple" is literally defined as the AC component of a DC current,
>>>usually from a power supply, and usually a harmonic of the input
>>>power.
>>
>>No, ripple is the result of pulsating D.C. post the rectifier stage.
>>The frequency of the ripple depends on the type of rectifier and the
>>line frequency of the A.C. In the U.S., a half wave ripple frequency
>>is 60 hz. A full wave would be 120 hz.
>
>
>That's what I said. If the input power is 60 Hz then 120 Hz would be a
>harmonic, right?
Technically, it's not a harmonic, it is the result of the full wave
rectification process. But there would be harmonics of 120 hz, but at
a substantially reduced level.
>>>The term "ripple" is old and has been replaced by "pard", which
>>>includes all AC components (or fluctuations) including hum, noise and
>>>spikes.
>>
>>That's because newer switching supplies create more noise, hash, and
>>spikes, then a simple linear supply would. Hence the need for a new
>>term, which encompasses a broader definition.
>
>
>If you say so.
I did.
>>>SOME caps are designed to handle ripple. Power supply filter caps are
>>>designed to REDUCE ripple; i.e, shut the AC component to ground.
>>
>>No, they do not shunt A.C. to ground. The capacitor stores current
>>during the ripple peak, and then discharges it as the ripple falls.
>>The rate of charge/discharge (which is hopefully slower than 60 or 120
>>hz) is what "smooths" the ripple. When the load increases, the R/C
>>time constant of the circuit changes and the cap is forced to
>>discharge faster, which results in less filtering of the ripple and it
>>increases.
>
>
>A capacitor blocks DC and passes AC; at least that's the simplified
>version of it's function.
Yes, that's the fundamental definition. But that does not describe
what is happening when a high capacity cap is used as a power supply
filter.
>So if you have a cap connected between a DC
>source and ground then the DC is blocked while any AC is passed to
>ground.
That depends on the reactance of the capacitor and the frequency of
the A.C.. In a power supply a large electrolytic cap acts much like a
battery, by providing a short duration storage for peak voltage in
order to "smooth" out the momentary loss of voltage between ripple
peaks. The higher the current draw of the supply, the more capacitance
you need to filter it effectively.
>>> When
>>>the load becomes large then so does the ripple, and if the cap has a
>>>high AC resistance then it will get hot very quickly.
>>
>>How? There is no A.C.! Pulsating D.C. does not switch polarity.
>Update your lexicon and then read it again.
There is still no A.C. If a voltage "swings" or "ripples" between +9V
and +10V, it is still keeping the cap charged in the proper polarity.
If the voltage swings negative, then you will have a problem, unless
the cap is a non-polarized electrolytic, like the type commonly found
in audio crossover circuits and other places.
>>>Just about everyone who has ever used a soldering iron thinks they can
>>>write a course on power supplies.
>>
>>Well considering that they were the very first thing that I studied
>>and built in tech school, it's pretty much basic stuff.
>
>
>I hate to break this to you but your tech school did a piss-poor job
>if the curriculum started with the study and construction of complex
>circuits before covering the basics.
That's HOW we studied the basics. How better to study Ohm's and
Kirchoff's law, and basic D.C. and A.C. fundamentals than to use a
basic building block, found in all electronic devices? What could be
more basic than a power supply?
My tech school was a "hands-on" style of teaching. They could have
bored everyone to tears by writing all sorts of math formulas on a
chalk board and demonstrating how things were supposed to work. But it
was much more rewarding and easier to relate to the circuits if we
could actually see it work in real life.
>>> But the increased ripple can be caused by a
>>>host of factors, not the least of which (and to the suprise of many,
>>>the most common) is the AC resistance of the output capacitor, -not-
>>>the R/C time constant of the cap with respect to the source and load
>>>impedances (which, BTW, are in parallel).
>>
>>Of course the R/C is in parallel. The cap is across the load.
>The cap is in parallel with -both- the source and load impedances; and
>while the load impedance may be linear, that's rarely the case for the
>source impedance.
But resistances and impedances in parallel tend to act like resistors
in parallel, and the results are similar. The source impedance, in the
case of a power supply is what is contained in the rectifiers. The
transformer is pretty much isolated at this point.
>Also, the capacitance of an electrolytic isn't a
>constant; it's non-linear with respect to the voltage (a factor of
>it's AC resistance, and another reason why audiophiles prefer motor
>caps over electrolytics).
Well of course! Look at any R/C time constance charge/discharge cycle
and it's not a straight line. As the cap reaches 70% of it's
capacity, the charge rate slows.
> Both these nonlinearities end up converting
>your overly-simplistic R/C time constant analysis into an integral
>function.
Call it what you will. My "overly simplistic" analysis covers the
conditions of what is actually happening.
> The issue is further complicated because of the equivalent
>series resistance of the capacitor (which, BTW, is a factor of it's AC
>resistance, and is also non-linear).
>
>The concepts involved here don't need to be nearly as complicated as
>you make them out to be, yet you try to describe the complexities with
>over-simplified theory.
That sounds like double talk. How can a simple concept be described in
overly simplistic complexities?
> It's like you are trying to describe baseball
>with dumbed-down Newtonian analysis. You still think you can write a
>course on the subject? I don't think so.
It's time to remind you of Occam's razor. I tend to use simple terms
when they fit. There's no need to be any more complicated than
necessary. My explanations were just that. I could have gone to the
trouble of mathematically explaining the whole deal, but that would be
far more effort than this thread deserves.
The complexities have been introduced by you in an effort to discredit
my explanations. While you can't deny them, you can inflate their
complexity and then accuse me of not explaining it thoroughly enough.
>>>A different scenario happens
>>>with high-voltage supplies because regulation is usually achieved with
>>>a swinging (saturating) choke, which permits more ripple to pass under
>>>increasing load in order to maintain the DC component.
>>
>>That's a very poor regulator. The choke/cap is really just a low pass
>>filter, that designed to attenuate 60hz ripple.
>That's the static interpretation.
But valid nonetheless.
>But a swinging choke is a component
>with dynamic (i.e, nonlinear) characteristics and makes a pretty good
>regulator, which is why they have been used so extensively over the
>past hundred years, and why they are still in production today.
Where? Certainly not in solid state equipment, where much better
components exist to do the job more effectively.
>>>So when is your book coming out?
>>
>>Since power supplies are so basic, and there is no new information
>>that I could tell that isn't already widely known, my effort would be
>>superfluous.
>
>
>I think you might have meant 'redundant', but 'superfluous' is really
>a much better choice.
That's why I chose it. Redundant would fit too.
>>>>If a cap explodes due to an excessive load, then the evidence of the
>>>>failure which cause that excessive load should be evident. If not then
>>>>you can't conclude that the cap died due to ripple.
>>>
>>>
>>>The cause of failure is not always evident in the equipment.
>>
>>No, it's not. But in a simple application of a HV power supply in a
>>tube amp, it's easy to eliminate practically every potential cause.
>
>
>It's not always easy.
No, but it is doable.
>
>
><snip>
>>>>You are right, in a sense, that you can't always tell what DID cause
>>>>the cap to blow, but you can determine what DID'NT cause it to fail.
>>>>It then becomes a simple process of elimination.
>>>
>>>
>>>The possibilities are limitless so any such conclusion would be based
>>>only on probability.
>>
>>The possibilities are not limitless. There are a finite set of
>>conditions which a particular circuit can find itself in. Is it
>>probable that a 12V supply was exposed to 500V?
>
>
>Depends on the circuit.
A solid state CB radio power supply.
>> Is it probable that a
>>rectified output of a supply reversed polarity?
>
>
>I never said it was.
>
>
>> Absent of other
>>component failures, how many possibilities really exist?
>
>
>Many, simply because the cause of a failure isn't limited to internal
>influences. But even if you limit the scope to the equipment, the
>failure of other components is not always obvious. And here's a
>perfect example:
>
>The first Peavey 1.3K I ever worked on came into the shop with a blown
>bank of power transistors (not uncommon with Peavey amps) and a bad
>OP-amp. The obvious conclusion is that the OP-amp locked up and drove
>the bank into overload (the amp also has a triac clamp but it rarely
>works -- it just nukes a trace on the board and lets the bank blow).
>Anyway, that wasn't the case at all. I replaced damn near everything
>in that amp and it would still lock up. To make an already long story
>short... it turns out that Peavey, in their economical wisdom, used a
>dual-diode as bias seperation for opposite banks of power transistors
>without noting it in their literature, and one of those diodes had
>shorted. The component appeared to be perfectly functional, showing a
>forward voltage drop characteristic of a typical silicon diode, but it
>was supposed to be -twice- that drop.
Ah, so instead of a 1.2V drop, it was only showing .6V? A change in
bias can certain cause a problem like that. But your example does not
negate my statement. You found that another component failed, which
resulted in a chain reaction failure. The cause of the failure was
found. But if you can't pin down the cause of failure to any other
component, you are left to consider that either the bad component
failed on its own, or it was subjected to a momentary stress beyond
its limits through some external event.
>So there are -many- possibilities that may or may -not- be obvious.
I respectfully disagree. There ARE other possibilities yes, but I
disagree that there are "many". A simple process of elimination will
whittle that number down to a few highly probable possibilities.
>>>>It's like forensic science. We reconstruct an incident based on the
>>>>evidence in the aftermath.
>>>
>>>
>>>Dave, before you use the word "forensic" again, do yourself a favor
>>>and look it up in the dictionary.
>>
>>Have you never heard of forensic criminalists, who gather evidence to
>>reconstruct a crime to prove innocence or guilt? Watch CSI sometime to
>>get familiar with them. Or, if you prefer a less dramatic example, the
>>Discovery channel has a show called "forensic files" which documents
>>actual criminal forensic cases. Court TV has similar shows.
>>
>>The example I used was similar in method.
>
>
>Gee, another excuse.
A perfectly valid one.
>>I have a battery charger which is unfiltered (the battery acts as the
>>filter cap), which may serve the purpose nicely.
>
>
>Go for it! Don't forget to use several different caps, measuring both
>AC and DC volts across the cap, as well as the time it takes until
>failure. Post the results.
When I get a chance, I'll set it up. I need to find some high current
resistors that can stand dissipating 20 amps or so.
>>Why use a light bulb to do a zener's job?
>
>
>The point was that the characteristics of a low-power device can be
>amplified. And yes, a light bulb is sometimes preferable to a zener
>because zeners are noisy little bastards.
That shouldn't matter in a power supply.
>>>A light bulb's resistance changes with temperature, but who said a
>>>resistor -must- be linear? The textbooks?
>>
>>The change in temperature is in direct proportion to the applied
>>voltage.
>
>
>Wrong.
Not wrong.
>The resistance is directly proportional to the temperature,
>hence the term "temperature coefficient". A light bulb can even be
>used as a thermistor for an electronic thermometer. IOW, you can feed
>a bulb's filament with a fixed voltage and a constant current will
>flow (after the temperature stablizes).
That's true of any fixed resistor. A constant current will flow at any
given fixed voltage.
> Cool the filament and the
>current will increase (light bulbs make pretty good wind-chill
>detectors). Heat the filament and the current will decrease (good for
>determining the temperature of a flame). All happen -irrespective- of
>voltage.
That's called an optical pyrometer, and I used to make them many moons
ago, when I worked at a company called Leeds and Northrup. A filament
will "glow" at a certain lightwave frequency at a given temperature.
You adjust the filament to "blend in" to the background flame or heat
source, and you can determine the temperature.
What you have are 3 factors (Voltage, current and resistance). When
you hold voltage constant, you can measure a change in temperature as
a change in current. But what if you increase the voltage? Will the
bulb not increase current and therefore temperature?
>>> IOW, the current density through a cross-section of filament
>>>is the same regardless of it's thickness or length.
>>
>>A 12V bulb will have a thicker filament than a 120V bulb for the same
>>given wattage. That's because the amount of current in a 12V bulb is
>>10 times higher than the same wattage bulb in a 120 V configuration.
>
>
>True, but the cross-sectional current density must be the same in
>order to reach the same temperature where it emits the same spectrum.
>So why would a thicker filament last longer?
It's physically more durable. Less prone to shock damage.
>>>>Other techs and engineers that I know in the industry.
>>>
>>>
>>>Well, I'm an engineer, you know me, and I haven't ever used the term
>>>"drying out" to describe a capacitor failure. Maybe "dielectrically
>>>challenged"..... no, I haven't used that either.
>>
>>You are but one man.
>
>
>Feel free to name some engineers who use the term "drying out" to
>describe capacitor failure.
Now you're taking lessons from twisty. I could put down a hundred
different names and what would that prove? You won't know them.
>Even better, try referencing it in some
>authoritative literature. If it's a common term then it should be well
>published, right? (Actually, it -is- a common term with engineers but
>only in reference to batteries.)
I entered the term "capacitors drying out" into google and came up
with 63 links to the term used in a similar context to what I've
promoted here. I didn't have the time to read them all, but be my
guest if you are so inclined.
I accept your apology.
>>
>>Regardless of the other possibilities, if you disagree that a
>>particular component has a high probability of failure, then what you
>>are saying, in essence, is that you feel that it has a much lower
>>probability of failure
>
>
>You are constructing a syllogism from a conditional statement by
>asserting the second premise, but you assume that second premise
>-must- be the only other possibility. It is not. Using your own
>subjective assessment of probability, the blank line in the statement
>
> "the problem is _____________ caused by capacitors"
>
>can be filled by any one of a long list of possible descriptors, only
>a few of which are listed here:
>
>A. always
>B. almost always
>C. often
>D. sometimes
>E. infrequently
>F. almost never
>G. never
>
>The syllogism that you constructed is thus:
>
>P1: Not B (my statement).
>P2: If not B then F.
>C: Therefore F.
>
>Although the conclusion is perfectly valid, you ASSUMED that F is the
>only possible alternative to B.
What you described is commonly called the Affirming a Disjunct logical
fallacy. Under the conditions you've described, this would be the
case. But it does not apply to my example. You stated that you do not
agree that a particular component has a high probability of failure,
so the converse to that is that you believe it has a lower probability
of failure. There are no other possibilities in this case. The
syllogism stands as valid.
> The conclusion, therefore, is neither
>true nor false; it can have no more truth than your assumed premise.
>Just because I don't live at the North Pole doesn't mean I -must- live
>at the South Pole. Or, for a better example, if you were to say that
>the sky is white, and I said it is -not- white, that doesn't mean I'm
>saying it's black. It could be blue, red, gray, etc. The colors of the
>sky are limited to black and white only in your fantasy realm where,
>coincidently, people must be either conservative or liberal.
Show me where I ever stated that people are either conservative or
liberal.
>> Therefore, by extension, it can be assumed that
>>you then consider other components have a higher probability of
>>failure.
>
>
>Key word: "assumed". You assumed wrong, as is usually the case
>whenever people ASSume things.
What other possibilities are there? Other components either have a:
A. Higher probability of failure.
B. Lower probability of failure.
C. The same probability of failure.
>
>
>>>>>>More chest thumping? What you've done at HP has no bearing on what
>>>>>>I've seen in 30+ years of CB radio repair.
>>>>>
>>>>>
>>>>>Really? Do capacitors behave differently in CB radios than they do in
>>>>>spectrum analyzers and signal generators?
>>>>
>>>>Sure. They're exposed to different conditions.
>>>
>>>
>>>Such as.....?
>>
>>Application. Does the circuit supply large currents or small ones?
>>Does it work at high voltage or low? Is the ambient temperature high,
>>or low? Do they filter, block, or bypass?
>
>
>All of the above in both applications. So how are they different?
Usage, Voltage, conditions, environment.
><snip>
>>>>Hmm.... My test procedure for that rig used 5mV to set 50% modulation,
>>>>with VR6
>>>
>>>Step 7, and it's 0.5 mV.
>>
>>It's 5mV in my book.
>
>
>I don't know what book you have (probably Sams) but mine is the
>service manual provided by Radio Shaft. I can scan the page if you
>want. If your book is in error that might explain your need to bypass
>the LPF.
.5 mV won't even tickle the modulation in most radios. Certainly not
enough to make 100%.
BTW, there is no "need" to bypass the LPF. It's just that the radio
sounds better IMHO that way. The LPF overly restricts the audio
bandwidth and kills fidelity. If you like sounding like you're talking
into a paper cup, then fine. I admit that I'm somewhat of an audio
purist. I like smooth "broadcast quality" audio. That's my personal
preference. YMMV.
> Higher drive at mid-modulation will push the audio into
>compression earlier, which would easily account for your audio
>sounding "mushy and compressed", just as you described.
Yes it would. But that's not the reason in my case. What was that you
tried to say before about acknowledging other possibilities?
> My radio
>sounds just fine without the modification, so it's very likely that
>-your- book is in error.
There is no error in my book. It correlates with other radio alignment
procedures.
Perhaps your book is in error. You assume that because it came from
Radio Crap, it's infallible.
My procedure is similar to what other radios also prescribe. After
all, there isn't much difference in the output that a given stock
microphone presents.
The fact that your radio "sounds fine" is a matter of some
subjectivity. I didn't say that the radios (Radio shack, Cobra and
President versions of this chassis) that I've worked on sounded "bad",
only that I felt that they sounded "better" with a modified LPF.
Again, a matter of personal preference.
> Not only that, I'm suprised that a person of
>your self-proclaimed experience would not have suspected the error
>earlier -- after all, your own description of the audio quality is
>indicative of overmodulation.
Not at all. 5 mV is hardly too much audio drive. Both the mike preamp
compression circuits and the modulation AMC were working properly. The
compressed sound was due to lousy audio bandwidth, coupled to a
microphone (A D-104) which didn't mate well with the audio bandpass
characteristic that this radio with the LPF presented.
> Yet you didn't bother to find a more
>authoritative reference, or compare the calibration specification to
>that of similar radios (I did, and 50% modulation at 0.5mV is common
>for Uniden radios that come with the same mic).
Funny, I took a quick look through a bunch of different test
procedures for various radios. Audio input ranged from 5 mV
(Considering the gain of that mike preamp circuit, that's not
surprising), to 10, 20, 50, 70 and 100 mV, to set the AMC. In NONE of
them was there ever a call for anything less than that.
>You could have also
>compared the cal-spec the radio's rated mic sensitivity. But no, you
>decided -not- to look for the primary cause of the problem, assumed
>that the error was in the design and proceeded to chop the filter.
Again, read my lips, there was NO "problem". This was a personal taste
modification made to improve the overall audio fidelity. I assumed
nothing, however you are doing more than enough assuming for both of
us.
>Dave, it's clear that in -both- electronics and politics you have a
>real problem accepting any possibilities outside the realm of your own
>prejudice.
No, what's clear is that you take a scenario, make many speculative
assumptions about the reasons for it, and then come to a conclusion
that's based on those erroneous assumptions. It's no wonder you are so
easily brainwashed by left-wing propaganda.
>>>> and 200 mV to set the 100% limiter threshold with VR7.
>>>
>>>
>>>Step 8, set VR7 for 95 to 98% mod.
>>
>>Right. But there are no numbered steps.
>
>
>There are in my book. And your's seems to be missing some steps.
Not at all. The necessary steps are there.
>>>>The reason for the increase in audio drive is to test that the
>>>>modulation limiter (and mike AGC in this particular radio) is doing
>>>>its job. Think about it, if the modulation limiter is working and
>>>>doing its job, then you can increase the mike level all you want
>>>>(until you fry something) and the AMC will clamp it to the level that
>>>>makes 100% modulation. The only parts in the audio chain that would be
>>>>stressed are the passive components before the AMC audio shunt
>>>>regulator, which is TR18.
>>>>
>>>>There is very little you could tell me about that radio (or more
>>>>specifically that chassis) that I don't already know.
>>>
>>>
>>>I just did. Twice in this post and once in the previous post.
>>
>>What you "told" me wasn't of any real info about the radio itself.
>I told you some -very- real info about the radio -- you just didn't
>know how important that info is, or how it demonstrates that you
>aren't as good with electronics as you think you are.
Like I said, there is nothing *valuable* you could tell me in a
hundred lifetimes, about that radio that I don't know, or haven't
experienced in one form or another. What you just recently learned, I
was doing 25 years ago. I've literally worked on over a hundred
versions of that radio. I used to sit there and play "what if" on my
own radio just for fun when I was bored and had nothing else to do.
I've designed controlled carrier modifications for it, FM receive and
transmit, 10 meter conversions, Digital readout frequency selection,
photocell LED and meter light dimmer circuits, wideband transmit and
receive, and a host of other things that I've long forgotten about.
Like I said, that chassis (in its various forms) has been one of my
all time favorites for the sheer ease at which the radio modified and
the overall capability of it.
>>>Thanks anyway, but I already used too much time pulling out the mods
>>>and restoring it to stock condition -- except for the faceplate. Got a
>>>spare?
>>
>>No I don't. Most of my spare faceplates belong to the bonepile radios,
>>most of which are crappy little AM mobiles.
>
>
>Heard that. Did you see the pic of my "bonepile" that I posted last
>fall? I must have a couple dozen of those cheap, little Cobras buried
>in the pile, not to mention a half-dozen or so mini-Midlands (which
>weren't too awfully bad little radios, just not worth fixing). I
>should go through the pile someday, make a list and post it for anyone
>looking for parts.
I saw it. Your "bonepile" is a bit larger than mine. I've been slowly
trashing the real junk, as I've pretty much "retired" from the CB-tech
scene. It's just not worth the time any more, as I've got far too much
on my plate these days. Being a parent takes up a lot of time.
Dave
"Sandbagger"
http://home.ptd.net/~n3cvj
Frank Gilliland
wrote in <vhte01hgdtsgmifpj...@4ax.com>:
>On Fri, 04 Feb 2005 13:34:26 -0800, Frank Gilliland
><wīrenut@NOSPAMīcehouse.net> wrote:
>
>>>>Sherlock Holmes was not a real person
>>>
>>>No, but Sir Arthur Conan Doyle was, and he wrote it.
>
>>Sir Arthur Conan Doyle wrote fiction.
>
>But the principle is still sound.
It is "sound" only so far as you are able to exclude all other
possibilities -- which is, quite literally, impossible. If it -was-
possible then we would have proof of the existence (or nonexistence)
of God. We don't.
Once again, Doyle wrote fiction.
>>>>, and that poorly-quoted cliche
>>>>has one very important point you completely missed: "when all the
>>>>other possibilities are excluded". Care to apply that concept to the
>>>>debate between creationism and evolution?
>>>
>>>Why? That's an entirely different subject with limitless and
>>>unverifiable possibilities. Unlike the subject we're currently
>>>discussing. There are a finite amount of potential causes of cap
>>>failure. Most of which can be reproduced.
>>
>>
>>If a cap can fail due to an act of God then it can't be verified. Your
>>argument is therefore invalid unless you are an athiest.
>
>What act of God kills caps? Lightning? A lightning strike can be
>easily eliminated as it's fairly obvious when it occurs.
Lightning is an act of nature. And why would you think that God is
limited to throwing lightning bolts?
>>>> Or can you see that there
>>>>are "other possibilities" that may not be obvious but they do exist?
>>>
>>>Not in this particular case.
>>
>>
>>That's a conclusion based on your own ignorance.
>
>No, it's one based on probability and practicality.
You have made two different arguments here. One is inductive and based
upon probability; the other is deductive where the conclusion is an
absolute (Doyle). Quit flip-flopping and make up your mind.
>>>But ripple is NOT A.C. It's pulsating D.C. At no point in the cycle
>>>does the voltage polarity reverse (alternate).
>>
>>
>>Despite your archaic definition of AC
>
>It is THE definition of A.C.
>
>
>>, the point is still valid; Those
>>large AC caps ("motor run" caps, to be specific) are frequently used
>>by audiophiles as DC power supply caps in SE tube amps because they
>>have a very low AC resistance, much lower than electrolytic caps.
>>Power supply caps are selected for their AC resistance, the capacity
>>in uF being only one of -several- factors figuring into the equation.
>
>Motor run caps are not the ones that blow up in amplifier power
>supplies.
Gee..... why not, Dave?
>>>But you keep changing the criteria to make you point. What might
>>>happen in a D.C. block in an audio amp would not happen in a power
>>>supply filter.
>>
>>
>>I see you have finally caught up.
>
>I never needed to "catch up", but when we're discussing a particular
>failure mode in one application, and you bring up a different
>application as a justification for a differing point, that does
>nothing but cloud the point.
Then why did you bring up the topic of power supply caps in sweep tube
amps? Just to "cloud the point"? Don't blame -me- because you confused
yourself with your own obfuscations.
>>>>>> There are
>>>>>>also instances where an abnormal AC component swings the normally
>>>>>>positive side of the cap negative. The cap voltage can get reversed by
>>>>>>rectification when a pull-up resistor fails. Etc, etc. Geez, Dave,
>>>>>>you've been doing this for 30 years and you didn't know this stuff?
>>>>>
>>>>>Not as applied to the tube amps we were talking about. If you switch
>>>>>the conditions of the discussion, then the outcomes will change as
>>>>>well.
>>>>
>>>>
>>>>Then why did you bring up the subject of tube amps when the original
>>>>topic was about a solid-state CB? If you don't want to discuss it then
>>>>don't bring it up.
>>>
>>>There will be no abnormal A.C. component beyond the rectifiers of a
>>>power supply Frank, unless the rectifiers fail.
>>
>>
>>A highly inductive load can drive the caps backwards via flyback. I've
>>never seen it happen myself but I know it's possible.
>
>Yes, that is certainly possible.
As you say, "apology accepted".
> But not in the equipment that I've
>worked on. There is no inductive components past the power supply
>filter caps, with the exception of a small 10 uH RF blocking choke.
>Hardly enough to generate enough "kick back" to reverse polarity.
>
>>
>>>>Perhaps that's because you work more with tubes while I work mostly
>>>>with solid-state equipment.
>>>
>>>I thought you just said you were one of those "experts" who restored
>>>old tube radios? Surely you've worked on as many tube radios as I have
>>>considering that I haven't touched anything with tubes in 10 years.
>
>
>>Maybe I have done more work on tubes than you. Maybe not. Maybe my
>>bench has seen more receivers while your's has seen more sweep-tube
>>amps. Maybe not. I merely suggested a possibility.
>
>I can accept that. My bench has probably seen the most 12V mobile
>radios, with amplifiers a distant second. But the equipment which has
>seen the most capacitor failures due to outgassing or explosion has
>been the tube amps.
Oddly enough, I have seen caps blow more often in solid-state
equipment.
> In the solid state equipment, cap failures are
>less obvious and usually not total. On older radios (25+ years) the
>main cap failure mode is excessive deviation from original value. You
>don't like the term "drying out", but whatever process is responsible
>for this phenomenon is quite prevalent.
Again in contrast to your stated experience, I have rarely seen caps
in solid-state circuits deviate significantly from their rated
tolerance (unless there was a total failure). On the contrary, I
-have- seen such deviation much more frequently in high-voltage power
supplies. You call it "drying-out", I call it "aging" -- it amounts to
the same thing.
>>>>Based on the fact that I've never experienced what you described, and
>>>>don't see how it's even possible from what I -have- experienced. I
>>>>have never seen a high-voltage cap blow up because it was leaking, and
>>>>I have rarely seen the same blow up from excessive DC voltage. I
>>>>-have- seen them blow up from excessive AC voltage (ripple). Is it
>>>>possible that the cap which blew up was subjected to too much ripple
>>>>caused by leakage in the other cap?
>>>
>>>How? The ripple would still not be that great, and it would neither
>>>exceed the voltage rating of th cap, nor reverse the polarity of it.
>
>
>>Well, if the AC resistance of the leaky cap dropped because of the
>>leak, then a greater AC voltage would be developed across the good
>>cap. This in turn would cause a greater AC current in the good cap,
>>and therefore more heating. So maybe the cap blew because of the extra
>>heat in combination with the higher DC stresses. I can see how -that-
>>could happen.
>
>That's exactly what I told you a few posts back. Only I didn't use the
>term A.C.. A leaky cap decreases D.C. resistance, which decreases A.C.
>resistance as well. Nice of you to finally get my point.
>
>Wait, you're trying to play with my head aren't you? Trying to make me
>change my tune? That only works when the "tune" is sufficiently
>different.
Hardly. The mechanisms behind our descriptions are different, and
your's neglected to address the heat generated by AC resistance in the
-good- cap. In fact, you denied AC resistance was even a factor. Yet
you just agreed above that it -is- a factor while making the excuse
that you "didn't use the term A.C." Another flip-flop.
>>>>Ok then. Any voltage has a DC component and an AC component.
>>>
>>>>Pure AC has no DC component, and pure DC has no AC component.
>>>
>>>Those two statements are contradictory.
>>
>>
>>Not at all. Your definition that AC must be a zero-crossing current
>>was repealed along with prohibition. The problem was that there were
>>getting to be too many terms for variations of direct current, like
>>'varying direct current' (vdc), 'pulsating direct current' (pdc),
>>'interrupted direct current'.....etc. So the concept has since been
>>simplified to say that any current has an AC component and a DC
>>component.
>
>You stated: "Any voltage has a DC component and an AC component. "
>
>Then you stated: "Pure AC has no DC component, and pure DC has no AC
>component."
>
>Yet you cannot see how these two statements contradict each other?
>
>If you change the first statement to: "Certain voltages have a DC
>component and an AC component. ", then the two statements would not
>conflict.
In several posts you have demonstrated that you have a problem
differentiating voltage and current. You should address that issue.
Now..... since you want to invoke semantics, I'll state this more
clearly for you: A pure DC current has an AC component with a value of
zero, and a pure AC current has a DC component with a value of zero.
Feel better?
>For the record, I agree with your second statement.
For the record, you agree with both statements. Whether you want to
believe it or not, the definition of AC has changed so that it
"encompasses a broader definition".
>>>>The term
>>>>"ripple" is literally defined as the AC component of a DC current,
>>>>usually from a power supply, and usually a harmonic of the input
>>>>power.
>>>
>>>No, ripple is the result of pulsating D.C. post the rectifier stage.
>>>The frequency of the ripple depends on the type of rectifier and the
>>>line frequency of the A.C. In the U.S., a half wave ripple frequency
>>>is 60 hz. A full wave would be 120 hz.
>>
>>
>>That's what I said. If the input power is 60 Hz then 120 Hz would be a
>>harmonic, right?
>
>Technically, it's not a harmonic, it is the result of the full wave
>rectification process. But there would be harmonics of 120 hz, but at
>a substantially reduced level.
Technically, it -is- a harmonic. The source of the fundamental
frequency was acted upon by a non-linear function (rectification). The
products of rectification are the harmonics of the fundamental. A
frequency-multiplier circuit works on the very same principle, with
the additional function of tuning the output to a specific harmonic.
Basics, Dave. Learn the basics.
>>>>The term "ripple" is old and has been replaced by "pard", which
>>>>includes all AC components (or fluctuations) including hum, noise and
>>>>spikes.
>>>
>>>That's because newer switching supplies create more noise, hash, and
>>>spikes, then a simple linear supply would. Hence the need for a new
>>>term, which encompasses a broader definition.
>>
>>
>>If you say so.
>
>I did.
Twice asserted. Again, as you say, "apology accepted".
>>>>SOME caps are designed to handle ripple. Power supply filter caps are
>>>>designed to REDUCE ripple; i.e, shut the AC component to ground.
>>>
>>>No, they do not shunt A.C. to ground. The capacitor stores current
>>>during the ripple peak, and then discharges it as the ripple falls.
>>>The rate of charge/discharge (which is hopefully slower than 60 or 120
>>>hz) is what "smooths" the ripple. When the load increases, the R/C
>>>time constant of the circuit changes and the cap is forced to
>>>discharge faster, which results in less filtering of the ripple and it
>>>increases.
>>
>>
>>A capacitor blocks DC and passes AC; at least that's the simplified
>>version of it's function.
>
>Yes, that's the fundamental definition. But that does not describe
>what is happening when a high capacity cap is used as a power supply
>filter.
Yes, it does.
>>So if you have a cap connected between a DC
>>source and ground then the DC is blocked while any AC is passed to
>>ground.
>
>That depends on the reactance of the capacitor and the frequency of
>the A.C.. In a power supply a large electrolytic cap acts much like a
>battery, by providing a short duration storage for peak voltage in
>order to "smooth" out the momentary loss of voltage between ripple
>peaks. The higher the current draw of the supply, the more capacitance
>you need to filter it effectively.
Dave, can't you see that we're saying the same thing? The difference
is that your explanation is sourced from an education lacking in the
fundamental concepts of electronics.
A capacitor -does- store a charge, and the amount of that charge -is-
a function of it's capacitance. But it's ability to -accept- and
-deliver- that charge leaves a gaping hole in your explanation. As I
said before, the value in capacitance is -only one- factor of its AC
resistance -- i.e, its ability to "filter" AC from a DC current.
And for a better perspective, dust off your book on basic electronics
and look up the page that covers simple LC filters -- low-pass and
high-pass. Notice that a high-pass filter has the cap in series with
the source, while the cap in the low-pass filter is an AC -shunt-. Now
you have already admitted that a power supply cap is nothing more than
a component of a filter, and I think it's safe to conclude that you
meant a low-pass filter (since the intended output is DC). Now unless
you intend to argue that the capacitor in an LC low-pass filter does
-not- shunt AC to ground, you have once again shown that you don't
have a firm grasp of the basics. And if you -do- intend to argue that
point then I suggest you first read the chapter (as well as all
previous chapters in the book).
>>>> When
>>>>the load becomes large then so does the ripple, and if the cap has a
>>>>high AC resistance then it will get hot very quickly.
>>>
>>>How? There is no A.C.! Pulsating D.C. does not switch polarity.
>
>>Update your lexicon and then read it again.
>
>There is still no A.C. If a voltage "swings" or "ripples" between +9V
>and +10V, it is still keeping the cap charged in the proper polarity.
>If the voltage swings negative, then you will have a problem, unless
>the cap is a non-polarized electrolytic, like the type commonly found
>in audio crossover circuits and other places.
You're stalling with semanitcs, but even -then- you are wrong. Say
your PS cap is charged to a DC voltage. When the power supply pulls
the ripple to it's peak then current is flowing -out- of the cap. When
the ripple bottoms out then current is flowing -into- the cap. You
have described the very same action at least twice. The problem is
that you didn't even know that AC stands for "Alternating CURRENT"!!!
The current is indeed alternating, even according to the obsolete
definition of AC.
LEARN THE BASICS, DAVE!!!
>>>>Just about everyone who has ever used a soldering iron thinks they can
>>>>write a course on power supplies.
>>>
>>>Well considering that they were the very first thing that I studied
>>>and built in tech school, it's pretty much basic stuff.
>>
>>
>>I hate to break this to you but your tech school did a piss-poor job
>>if the curriculum started with the study and construction of complex
>>circuits before covering the basics.
>
>That's HOW we studied the basics. How better to study Ohm's and
>Kirchoff's law, and basic D.C. and A.C. fundamentals than to use a
>basic building block, found in all electronic devices? What could be
>more basic than a power supply?
Oh, gee, I don't know..... How about:
A battery & a resistor;
Resistors in series and parallel;
An inductor in a DC and AC current;
Inductors in series and parallel;
An inductor with a saturable core;
A capacitor in a DC and AC current;
Capacitors in series and parallel;
A resistor and capacitor in series and parallel;
A resistor and inductor in series and parallel;
An inductor and capacitor in series and parallel;
A resistor, inductor & capacitor in various networks;
A current transformer;
A voltage transformer;
A power transformer;
Impedance transformation of a transformer;
A transformer with a resistive load;
A transformer with a reactive load.......
And I haven't even touched on semiconductors yet. Shall I continue
with this list?
>My tech school was a "hands-on" style of teaching. They could have
>bored everyone to tears by writing all sorts of math formulas on a
>chalk board and demonstrating how things were supposed to work. But it
>was much more rewarding and easier to relate to the circuits if we
>could actually see it work in real life.
If observation alone was sufficient for an education then anyone who
ever listened to the radio or watched a television would be an
engineer. That isn't the case. Those boring equations are technical
and mathematical descriptions of how electronic components and
circuits work in "real life". If you didn't learn these fundamental
concepts because electronic theory "bores you to tears" then you
probably shouldn't be discussing the subject with an EE who -did-
learn the basics (and in much more depth than they are covered in any
"tech school").
>>>> But the increased ripple can be caused by a
>>>>host of factors, not the least of which (and to the suprise of many,
>>>>the most common) is the AC resistance of the output capacitor, -not-
>>>>the R/C time constant of the cap with respect to the source and load
>>>>impedances (which, BTW, are in parallel).
>>>
>>>Of course the R/C is in parallel. The cap is across the load.
>
>>The cap is in parallel with -both- the source and load impedances; and
>>while the load impedance may be linear, that's rarely the case for the
>>source impedance.
>
>But resistances and impedances in parallel tend to act like resistors
>in parallel, and the results are similar. The source impedance, in the
>case of a power supply is what is contained in the rectifiers. The
>transformer is pretty much isolated at this point.
Wrong, wrong, and even -more- wrong. First, an impedance is not only
resistance but also -reactance-. Second, the power transformer is a
source impedance with an inductive reactance that changes with the
load. Third, the load can be (and frequently is) reactive to some
extent. Fourth, the caps are -both- a source and load impedance,
depending on which end you are looking at. Fifth, the rectifiers are
nonlinear components but they -do- have a linear region (conduction)
which passes impedances quite nicely, and in both directions......
Shall I continue with -this- list?
>>Also, the capacitance of an electrolytic isn't a
>>constant; it's non-linear with respect to the voltage (a factor of
>>it's AC resistance, and another reason why audiophiles prefer motor
>>caps over electrolytics).
>
>Well of course! Look at any R/C time constance charge/discharge cycle
>and it's not a straight line. As the cap reaches 70% of it's
>capacity, the charge rate slows.
Holy crap, Dave..... didn't they teach you -anything- in that "tech
school"? The RC curve is the relationship between voltage and time,
and assumes that both the resistance and capacitance are constants.
What I said (and what you are ignoring) is that the capacitance of an
electrolytic cap is -not- a constant. And because it is not a constant
the RC curve does -not- follow the same path that it would if the
capacitance -was- a constant. And BTW, the RC curve becomes a straight
line when the current is constant because the charge rate of the cap
is a fuction of current, not voltage. This is yet another example of
your inability to distinguish current and voltage.
>> Both these nonlinearities end up converting
>>your overly-simplistic R/C time constant analysis into an integral
>>function.
>
>Call it what you will. My "overly simplistic" analysis covers the
>conditions of what is actually happening.
It's clear that you have absolutely no clue as to what is actually
happening. Learn the basics, Dave.
>> The issue is further complicated because of the equivalent
>>series resistance of the capacitor (which, BTW, is a factor of it's AC
>>resistance, and is also non-linear).
>>
>>The concepts involved here don't need to be nearly as complicated as
>>you make them out to be, yet you try to describe the complexities with
>>over-simplified theory.
>
>That sounds like double talk. How can a simple concept be described in
>overly simplistic complexities?
>
>> It's like you are trying to describe baseball
>>with dumbed-down Newtonian analysis. You still think you can write a
>>course on the subject? I don't think so.
>
>It's time to remind you of Occam's razor. I tend to use simple terms
>when they fit. There's no need to be any more complicated than
>necessary. My explanations were just that. I could have gone to the
>trouble of mathematically explaining the whole deal, but that would be
>far more effort than this thread deserves.
>
>The complexities have been introduced by you in an effort to discredit
>my explanations. While you can't deny them, you can inflate their
>complexity and then accuse me of not explaining it thoroughly enough.
You are once again confused. In your attempt to demonstrate that you
know what you are talking about you have ultimately proven that your
understanding is poor at best. THAT was my point, and THAT is what is
evident from your technical explanations. Maybe you think you can leaf
through a few textbooks, speed-read a few concepts, paraphrase them
and try to pass them off as if they were integrated into your
education. Well, you can't. I've been in the field long enough to
recognize that scam and it doesn't work. It didn't work for Skippy, it
didn't work for Toll, it didn't work for Eitner and it won't work for
you. But please don't stop trying because I find it very entertaining.
>>>>A different scenario happens
>>>>with high-voltage supplies because regulation is usually achieved with
>>>>a swinging (saturating) choke, which permits more ripple to pass under
>>>>increasing load in order to maintain the DC component.
>>>
>>>That's a very poor regulator. The choke/cap is really just a low pass
>>>filter, that designed to attenuate 60hz ripple.
>
>>That's the static interpretation.
>
>But valid nonetheless.
>
>>But a swinging choke is a component
>>with dynamic (i.e, nonlinear) characteristics and makes a pretty good
>>regulator, which is why they have been used so extensively over the
>>past hundred years, and why they are still in production today.
>
>Where? Certainly not in solid state equipment, where much better
>components exist to do the job more effectively.
Not often in solid-state equipment, but they are still quite common in
high-voltage supplies, power coverters and line conditioners. In fact,
your microwave oven has a regulating transformer that works on the
same principle. Should I make a list for this subject, too?
>>>>So when is your book coming out?
>>>
>>>Since power supplies are so basic, and there is no new information
>>>that I could tell that isn't already widely known, my effort would be
>>>superfluous.
>>
>>
>>I think you might have meant 'redundant', but 'superfluous' is really
>>a much better choice.
>
>That's why I chose it. Redundant would fit too.
I'll give you another opportunity to look up those words.
>>>>>If a cap explodes due to an excessive load, then the evidence of the
>>>>>failure which cause that excessive load should be evident. If not then
>>>>>you can't conclude that the cap died due to ripple.
>>>>
>>>>
>>>>The cause of failure is not always evident in the equipment.
>>>
>>>No, it's not. But in a simple application of a HV power supply in a
>>>tube amp, it's easy to eliminate practically every potential cause.
>>
>>
>>It's not always easy.
>
>No, but it is doable.
Another flip-flop. Make up your mind -- is it easy or not?
>><snip>
>>>>>You are right, in a sense, that you can't always tell what DID cause
>>>>>the cap to blow, but you can determine what DID'NT cause it to fail.
>>>>>It then becomes a simple process of elimination.
>>>>
>>>>
>>>>The possibilities are limitless so any such conclusion would be based
>>>>only on probability.
>>>
>>>The possibilities are not limitless. There are a finite set of
>>>conditions which a particular circuit can find itself in. Is it
>>>probable that a 12V supply was exposed to 500V?
>>
>>
>>Depends on the circuit.
>
>A solid state CB radio power supply.
Isolation failure in the line transformer can easily put a couple
hundred volts on your 12 volt supply. Seen it a few times (twice with
Pyramid power supplies -- cheap transformers that are too small for
the rated output).
Actually, it was a design flaw. The diode was supposed to be thermally
connected to the heatsink, but was mounted such that the thermal link
was nothing more than a blob of heatsink grease. The grease dried out
and the redidue became a very good insulator. Regardless, my point was
that the failure is not always intrinsic to the circuit or device;
many times the problem is caused by outside influences, such as
improper operation or maintenance, poor design, manufacturing flaws,
etc. And many of those faults are not always evident; i.e, they don't
leave evidence of the cause of the failure. The fact is that there was
no evidence of the cause of the failure in the Peavey. I learned of
the cause after harassing Peavey tech support until they forked over
the info (I really think they try to keep faults like that a secret
because they make so much money on spare parts, but I suppose that's
just one aspect of modern business practices).
>>So there are -many- possibilities that may or may -not- be obvious.
>
>I respectfully disagree. There ARE other possibilities yes, but I
>disagree that there are "many". A simple process of elimination will
>whittle that number down to a few highly probable possibilities.
A shorted resistor is not "highly probable", nor is a half-shorted
double diode, a PCB trace with a hairline crack, an incorrect slug in
an RF or IF transformer, etc. I can make a list here, too.
>>>>>It's like forensic science. We reconstruct an incident based on the
>>>>>evidence in the aftermath.
>>>>
>>>>
>>>>Dave, before you use the word "forensic" again, do yourself a favor
>>>>and look it up in the dictionary.
>>>
>>>Have you never heard of forensic criminalists, who gather evidence to
>>>reconstruct a crime to prove innocence or guilt? Watch CSI sometime to
>>>get familiar with them. Or, if you prefer a less dramatic example, the
>>>Discovery channel has a show called "forensic files" which documents
>>>actual criminal forensic cases. Court TV has similar shows.
>>>
>>>The example I used was similar in method.
>>
>>
>>Gee, another excuse.
>
>A perfectly valid one.
You -still- haven't looked up the word. Shame.....
What you initially described is called 'failure analysis' (or 'crime
scene analysis', as per your example). The term "forensic science" is,
literally, the -presentation- of a scientific analysis.
>>>I have a battery charger which is unfiltered (the battery acts as the
>>>filter cap), which may serve the purpose nicely.
>>
>>
>>Go for it! Don't forget to use several different caps, measuring both
>>AC and DC volts across the cap, as well as the time it takes until
>>failure. Post the results.
>
>When I get a chance, I'll set it up. I need to find some high current
>resistors that can stand dissipating 20 amps or so.
What do you need resistors for? Just hook the puppies up! But if you
want a resistor anyway try an electric space heater or stove element.
>>>Why use a light bulb to do a zener's job?
>>
>>
>>The point was that the characteristics of a low-power device can be
>>amplified. And yes, a light bulb is sometimes preferable to a zener
>>because zeners are noisy little bastards.
>
>That shouldn't matter in a power supply.
Power supplies are not the only circuits that use regulators. And yes,
the noise of a zener is amplified by a power transistor along with
it's nonlinear characteristics, which can be bad if the supply is to
be used for low-noise applications.
>>>>A light bulb's resistance changes with temperature, but who said a
>>>>resistor -must- be linear? The textbooks?
>>>
>>>The change in temperature is in direct proportion to the applied
>>>voltage.
>>
>>
>>Wrong.
>
>Not wrong.
>
>>The resistance is directly proportional to the temperature,
>>hence the term "temperature coefficient". A light bulb can even be
>>used as a thermistor for an electronic thermometer. IOW, you can feed
>>a bulb's filament with a fixed voltage and a constant current will
>>flow (after the temperature stablizes).
>
>That's true of any fixed resistor. A constant current will flow at any
>given fixed voltage.
You are ignoring the first part of the paragraph -- the part about its
temperature coefficient.
>> Cool the filament and the
>>current will increase (light bulbs make pretty good wind-chill
>>detectors). Heat the filament and the current will decrease (good for
>>determining the temperature of a flame). All happen -irrespective- of
>>voltage.
>
>That's called an optical pyrometer, and I used to make them many moons
>ago, when I worked at a company called Leeds and Northrup. A filament
>will "glow" at a certain lightwave frequency at a given temperature.
>You adjust the filament to "blend in" to the background flame or heat
>source, and you can determine the temperature.
That's not what I described. I described a thermistor. Try reading it
again for content.
>What you have are 3 factors (Voltage, current and resistance). When
>you hold voltage constant, you can measure a change in temperature as
>a change in current.
Very good! That's all I was trying to say! Sheesh......
> But what if you increase the voltage? Will the
>bulb not increase current and therefore temperature?
Congratulations, you found the missing step!
>>>> IOW, the current density through a cross-section of filament
>>>>is the same regardless of it's thickness or length.
>>>
>>>A 12V bulb will have a thicker filament than a 120V bulb for the same
>>>given wattage. That's because the amount of current in a 12V bulb is
>>>10 times higher than the same wattage bulb in a 120 V configuration.
>>
>>
>>True, but the cross-sectional current density must be the same in
>>order to reach the same temperature where it emits the same spectrum.
>>So why would a thicker filament last longer?
>
>It's physically more durable. Less prone to shock damage.
I don't buy that simply because an auto bulb encounters much more
shock and vibration than most household lightbulbs. And UV bulbs have
a much thinner filament than a standard white bulb (to achieve a
higher temperature, and therefore a higher part of the spectrum), yet
they don't seem to burn out significantly faster. So your explanation
is still lacking any substance.
>>>>>Other techs and engineers that I know in the industry.
>>>>
>>>>
>>>>Well, I'm an engineer, you know me, and I haven't ever used the term
>>>>"drying out" to describe a capacitor failure. Maybe "dielectrically
>>>>challenged"..... no, I haven't used that either.
>>>
>>>You are but one man.
>>
>>
>>Feel free to name some engineers who use the term "drying out" to
>>describe capacitor failure.
>
>Now you're taking lessons from twisty. I could put down a hundred
>different names and what would that prove? You won't know them.
I know lots of people and I meet more all the time. How about just
three engineers -- you can email me their addys so I can verify their
statement (as you claim it) and their engineer status. Or is three too
many?
>>Even better, try referencing it in some
>>authoritative literature. If it's a common term then it should be well
>>published, right? (Actually, it -is- a common term with engineers but
>>only in reference to batteries.)
>
>I entered the term "capacitors drying out" into google and came up
>with 63 links to the term used in a similar context to what I've
>promoted here. I didn't have the time to read them all, but be my
>guest if you are so inclined.
>
>I accept your apology.
63 hits? That's all? I'm sure there are -plenty- more references from
people who, like yourself, mistakenly attribute the term "drying out"
to capacitors. Heck, I got over a million hits for "Santa Claus" but
that doesn't mean he exists. How about an -authoritative- reference,
Dave? You know the difference, don't you?
.....oh brother
> Under the conditions you've described, this would be the
>case. But it does not apply to my example. You stated that you do not
>agree that a particular component has a high probability of failure,
>so the converse to that is that you believe it has a lower probability
>of failure. There are no other possibilities in this case.
So A, C, D, E and G -can't- be used to fill in the blank?
> The
>syllogism stands as valid.
The syllogism itself is valid, as I already stated. The -truth- of the
-conclusion-, however, is undefined, which I stated in the rest of the
paragraph as follows:
>> The conclusion, therefore, is neither
>>true nor false; it can have no more truth than your assumed premise.
>>Just because I don't live at the North Pole doesn't mean I -must- live
>>at the South Pole. Or, for a better example, if you were to say that
>>the sky is white, and I said it is -not- white, that doesn't mean I'm
>>saying it's black. It could be blue, red, gray, etc. The colors of the
>>sky are limited to black and white only in your fantasy realm where,
>>coincidently, people must be either conservative or liberal.
>
>Show me where I ever stated that people are either conservative or
>liberal.
Focus, Dave. Your conclusion is the result of an assumed premise.
>>> Therefore, by extension, it can be assumed that
>>>you then consider other components have a higher probability of
>>>failure.
>>
>>
>>Key word: "assumed". You assumed wrong, as is usually the case
>>whenever people ASSume things.
>
>What other possibilities are there? Other components either have a:
>
>A. Higher probability of failure.
>B. Lower probability of failure.
>C. The same probability of failure.
Earlier you stated that -only- the converse could be true and that
"there are no other possibilities in this case". Yet now you add a
third possibility (in direct contradiction to your first statement
which asserts that there are no other possibilities), and it is along
the same lines as the list of possibilities that I previously listed.
Aside from your obvious flip-flop, re-read your possibility C, then
compare it to the following statements:
"And after being in the electronics business
professionally for 30+ years I can say, without any doubt whatsoever,
that capacitor failure rate (i.e, the capacitor being the -primary-
cause of the failure) is no higher than for any other component."
"Cap failure as the -primary- cause of
failure is no more common than a weak transistor, bad IC, blown
diode/zener, or bad RF/IF can, or even poor manufacturing standards."
"I never
said that caps don't go bad. I said they aren't the primary cause of
failure any more than any other component."
Now what part or parts of those statements did you not understand? Did
you read them while wearing the same dirty eyeglasses you wore when
reading my statement saying that "I am not a liberal"? Or are you just
being bullheaded?
>>>>>>>More chest thumping? What you've done at HP has no bearing on what
>>>>>>>I've seen in 30+ years of CB radio repair.
>>>>>>
>>>>>>
>>>>>>Really? Do capacitors behave differently in CB radios than they do in
>>>>>>spectrum analyzers and signal generators?
>>>>>
>>>>>Sure. They're exposed to different conditions.
>>>>
>>>>
>>>>Such as.....?
>>>
>>>Application. Does the circuit supply large currents or small ones?
>>>Does it work at high voltage or low? Is the ambient temperature high,
>>>or low? Do they filter, block, or bypass?
>>
>>
>>All of the above in both applications. So how are they different?
>
>Usage, Voltage, conditions, environment.
.....uh, can you be more vague?
>><snip>
>>>>>Hmm.... My test procedure for that rig used 5mV to set 50% modulation,
>>>>>with VR6
>>>>
>>>>Step 7, and it's 0.5 mV.
>>>
>>>It's 5mV in my book.
>>
>>
>>I don't know what book you have (probably Sams) but mine is the
>>service manual provided by Radio Shaft. I can scan the page if you
>>want. If your book is in error that might explain your need to bypass
>>the LPF.
>
>.5 mV won't even tickle the modulation in most radios. Certainly not
>enough to make 100%.
Geez Dave, can't you read? The spec was 0.5mV for -50%- modulation.
And yes, it will most definitely "tickle" the modulation because mine
is set just as described and it works dandy. BTW, your lack of both
experience and education is showing -- typical output of a dynamic
microphone just -begins- in the low microvolt range. We're talking
-60dB or less. The dynamic range of a human voice isn't that great,
barely 20dB, yet the dynamic range of CB AVC circuits (or AMC, AGC,
CVA, etc) can be anywhere from 40 to 60dB, which is such a wide range
for a number of reasons, not the least of which is the wide variety of
microphones, voices, and distances between the mics and the voices.
And 0.5mV is plenty for 50% mod, especially when the radio's mic
sensitivity is specified for 1.0mV minimum at 50% mod. I offered to
scan the page..... are you sure you don't want to see it? After all,
Uniden made these radios for radio shaft, so it was -Uniden- that
provided the information in the service manual. Not like Sams which
got their info from reverse engineering. So which manual do -you-
think is more likely to be accurate, Dave? Don't you think Occam's
Razor applies in this case? Or is this another case for Sherlock
Holmes?
>BTW, there is no "need" to bypass the LPF. It's just that the radio
>sounds better IMHO that way. The LPF overly restricts the audio
>bandwidth and kills fidelity. If you like sounding like you're talking
>into a paper cup, then fine. I admit that I'm somewhat of an audio
>purist. I like smooth "broadcast quality" audio. That's my personal
>preference. YMMV.
CB is not intended for high-fidelity. Passing the higher freqs widens
the bandwidth of the channel and makes it more likely to bleed into an
adjacent channel. That's one reason why CB mods are illegal.
Regardless, you can get improved sound with nothing more than a better
mic. The stock microphones are usually designed for 300-3000Hz and are
as flat as the Matterhorn. Plug an electret into the radio and the
response flattens out, making it sound much better even -with- the LP
filter. This should come as no suprise since 3000Hz is a pretty high
frequency when compared to the human voice. Only Kate Bush could sing
-that- high.
>> Higher drive at mid-modulation will push the audio into
>>compression earlier, which would easily account for your audio
>>sounding "mushy and compressed", just as you described.
>
>Yes it would. But that's not the reason in my case. What was that you
>tried to say before about acknowledging other possibilities?
Why should I? My audio sounds fine. I'm not the one with "mushy and
compressed" audio. I'm not the one who uses a book with an error in
the alignment procedures. And I'm not the one who felt the need to
bypass the LP filter. I don't see any reason to even consider any
other possibilities. If it works, don't fix it.
>> My radio
>>sounds just fine without the modification, so it's very likely that
>>-your- book is in error.
>
>There is no error in my book. It correlates with other radio alignment
>procedures.
>
>Perhaps your book is in error. You assume that because it came from
>Radio Crap, it's infallible.
Wrong. I never assumed any such thing. You mentioned the discrepancy
so I referenced it's other specs, as well as specs from other radios.
The only discrepancy is with -your- book. It stands alone in the
modulation procedure you described. The question is, given how your
own description of the audio is indicative of misalignment, and that I
have a more authoritative reference that disputes yours, why do -you-
assume that -your- book is correct?
>My procedure is similar to what other radios also prescribe. After
>all, there isn't much difference in the output that a given stock
>microphone presents.
Wrong. I have lots of mics that have widely different levels. And the
useful dynamic range of any given mic can be 30 to nearly 100dB,
depending on the type. That's why one mic will sound great on one
radio but crappy on another, and vice-versa.
>The fact that your radio "sounds fine" is a matter of some
>subjectivity. I didn't say that the radios (Radio shack, Cobra and
>President versions of this chassis) that I've worked on sounded "bad",
>only that I felt that they sounded "better" with a modified LPF.
>Again, a matter of personal preference.
You said they sounded "mushy and compressed". That isn't bad? Geez
Dave, how many times can you flip-flop in one post?
>> Not only that, I'm suprised that a person of
>>your self-proclaimed experience would not have suspected the error
>>earlier -- after all, your own description of the audio quality is
>>indicative of overmodulation.
>
>Not at all. 5 mV is hardly too much audio drive. Both the mike preamp
>compression circuits and the modulation AMC were working properly. The
>compressed sound was due to lousy audio bandwidth,
Or maybe it was due to a misalignment which caused a downward
expansion in a region where a flat response is intended? Naw, that
couldn't be it. You and your crystal ball must be right, Dave, it
couldn't be anything else.
> coupled to a
>microphone (A D-104) which didn't mate well with the audio bandpass
>characteristic that this radio with the LPF presented.
D-104's suck. They have no low-end and the high-end is raspy. The
output level of the unamplified models is too high for most radios,
and the amplified models are usually turned up too high by the user.
People like them because they are loud, but they are loud because of
their high and narrow spectral response -- not a mic I would consider
using for high-fidelity. Want a good mic? Try a Shure 522 -- a lot
cheaper than a D-104 and sounds a lot better.
>> Yet you didn't bother to find a more
>>authoritative reference, or compare the calibration specification to
>>that of similar radios (I did, and 50% modulation at 0.5mV is common
>>for Uniden radios that come with the same mic).
>
>Funny, I took a quick look through a bunch of different test
>procedures for various radios. Audio input ranged from 5 mV
>(Considering the gain of that mike preamp circuit, that's not
>surprising), to 10, 20, 50, 70 and 100 mV, to set the AMC. In NONE of
>them was there ever a call for anything less than that.
Like I said, I can scan the page if you don't believe me. You can even
ask Lance since I got the manual from him with the radio. BTW, have
you ever seen a factory service manual? You know, one that wasn't
published by Sams?
>>You could have also
>>compared the cal-spec the radio's rated mic sensitivity. But no, you
>>decided -not- to look for the primary cause of the problem, assumed
>>that the error was in the design and proceeded to chop the filter.
>
>Again, read my lips, there was NO "problem". This was a personal taste
>modification made to improve the overall audio fidelity. I assumed
>nothing, however you are doing more than enough assuming for both of
>us.
Let's put it this way: Have you heard the audio after aligning the
radio according to factory specs? I have, and it doesn't sound "mushy
and compressed" at all. And I've worked with professional audio
equipment for more than ten years at the station so I -do- know what
"mushy and compressed" sounds like. Gee, I wonder why -your- radios
sound that way but not mine.....
>>Dave, it's clear that in -both- electronics and politics you have a
>>real problem accepting any possibilities outside the realm of your own
>>prejudice.
>
>No, what's clear is that you take a scenario, make many speculative
>assumptions about the reasons for it, and then come to a conclusion
>that's based on those erroneous assumptions. It's no wonder you are so
>easily brainwashed by left-wing propaganda.
That's about the best example of projection I have ever read, Twisty's
rants included.
>>>>> and 200 mV to set the 100% limiter threshold with VR7.
>>>>
>>>>
>>>>Step 8, set VR7 for 95 to 98% mod.
>>>
>>>Right. But there are no numbered steps.
>>
>>
>>There are in my book. And your's seems to be missing some steps.
>
>Not at all. The necessary steps are there.
Oh heck, I'll scan the page anyway. I'll post the link later after I
dig the scanner out of the pile (everything's a mess here right now).
>>>>>The reason for the increase in audio drive is to test that the
>>>>>modulation limiter (and mike AGC in this particular radio) is doing
>>>>>its job. Think about it, if the modulation limiter is working and
>>>>>doing its job, then you can increase the mike level all you want
>>>>>(until you fry something) and the AMC will clamp it to the level that
>>>>>makes 100% modulation. The only parts in the audio chain that would be
>>>>>stressed are the passive components before the AMC audio shunt
>>>>>regulator, which is TR18.
>>>>>
>>>>>There is very little you could tell me about that radio (or more
>>>>>specifically that chassis) that I don't already know.
>>>>
>>>>
>>>>I just did. Twice in this post and once in the previous post.
>>>
>>>What you "told" me wasn't of any real info about the radio itself.
>
>
>>I told you some -very- real info about the radio -- you just didn't
>>know how important that info is, or how it demonstrates that you
>>aren't as good with electronics as you think you are.
>
>Like I said, there is nothing *valuable* you could tell me in a
>hundred lifetimes, about that radio that I don't know, or haven't
>experienced in one form or another. What you just recently learned, I
>was doing 25 years ago. I've literally worked on over a hundred
>versions of that radio. I used to sit there and play "what if" on my
>own radio just for fun when I was bored and had nothing else to do.
Wow! Over a hundred -versions- of that radio? I didn't even think
there were a hundred CB companies at the time the radio was made! So
you must really know your stuff, huh? But wait..... you don't even
have a factory manual for the radio. So everything you know about the
alignment of that radio is from where..... the Sams manual?
Hmmm.....
>I've designed controlled carrier modifications for it, FM receive and
>transmit, 10 meter conversions, Digital readout frequency selection,
>photocell LED and meter light dimmer circuits, wideband transmit and
>receive, and a host of other things that I've long forgotten about.
>Like I said, that chassis (in its various forms) has been one of my
>all time favorites for the sheer ease at which the radio modified and
>the overall capability of it.
I'll be impressed when you can design one from scratch. But that would
be superfluous, wouldn't it?
>>>>Thanks anyway, but I already used too much time pulling out the mods
>>>>and restoring it to stock condition -- except for the faceplate. Got a
>>>>spare?
>>>
>>>No I don't. Most of my spare faceplates belong to the bonepile radios,
>>>most of which are crappy little AM mobiles.
>>
>>
>>Heard that. Did you see the pic of my "bonepile" that I posted last
>>fall? I must have a couple dozen of those cheap, little Cobras buried
>>in the pile, not to mention a half-dozen or so mini-Midlands (which
>>weren't too awfully bad little radios, just not worth fixing). I
>>should go through the pile someday, make a list and post it for anyone
>>looking for parts.
>
>I saw it. Your "bonepile" is a bit larger than mine. I've been slowly
>trashing the real junk, as I've pretty much "retired" from the CB-tech
>scene. It's just not worth the time any more, as I've got far too much
>on my plate these days. Being a parent takes up a lot of time.
So does the newsgroup, apparently.