'Dead' Atari 1050
Geoff McHugh
sake, and also to see if I could rescue some of my 5.25" floppies
using Ape/Pro before the contents were lost forever.
Unfortunately, my 1050 seems to have died (along with one of my 130XEs!)
It worked fine a year or two ago when I last suffered from a case of
nostalgia and I made a Pro cable up but then got sidetracked and packed
it all away again.
When the 1050 is switched on, the power LED loghts, but the drive
mechanism does not respond at all, no drive LED and no familiar grinding
noise that I recall fondly.
I connected the Pro cable to the PC and 1050, ran up the demo of Pro'98,
and tried interrogating the drive. It correctly saw that it was a 1050
US Doubler enhanced drive, capable of Double Density, but reported
'Drive Error!' almost everywhere else.
I tried leaving the drive on for a while as I read somewhere here that
it may cure the problem, but I noticed the heatsink in the drive getting
very warm, and the PSU got very very hot. (the heat even began to spread
up the power cable from the back of the 1050) Eventually, the PSU gave
up (I first thought the fuse in the plug had blown, but I now think the
PSU is fried, as replacing the plug fuse didn't help.)
Using another, older 400/800 PSU the 1050 still responds to Pro'98, but
the drive mechanism is still dead. I tried removing all the socketed
ICs in the drive and cleaning the pins slightly in case there was some
corrosion, but this made no difference.
Is there anything that might be a common cause for this, or should I try
to find a second hand drive to replace it. I would like to be able to
'rescue' the contents of my disks.
Also, when I was trying Pro'98, I gave Ap'98 a whirl too, but it locks
up once the shareware delay finishes, so it doesn't seem to work on my
PC, has anyone else come across this? I'll post more details later, I
haven't really had much time to study it closely, what with having bits
of 1050's and 130XEs spread about the place for the last few days!!
Just while I'm on my mad nostalgia trip, does anyone remember James
Bastable and The Ark, or The City BBS in the UK back in the 80s? I
spent my first few online years there, keeping the phone company rich
and my parents poor :)
--
Geoff
Sent via Deja.com
http://www.deja.com/
Rick Cortese
news:941m56$lrm$1...@nnrp1.deja.com...
<snip>
> When the 1050 is switched on, the power LED loghts, but the drive
> mechanism does not respond at all, no drive LED and no familiar grinding
> noise that I recall fondly.
>
> I connected the Pro cable to the PC and 1050, ran up the demo of Pro'98,
> and tried interrogating the drive. It correctly saw that it was a 1050
> US Doubler enhanced drive, capable of Double Density, but reported
> 'Drive Error!' almost everywhere else.
=> Your drive's smarts are working and your 12V motor supply inside the 1050
isn't.
>
> I tried leaving the drive on for a while as I read somewhere here that
> it may cure the problem, but I noticed the heatsink in the drive getting
> very warm, and the PSU got very very hot. (the heat even began to spread
> up the power cable from the back of the 1050) Eventually, the PSU gave
> up (I first thought the fuse in the plug had blown, but I now think the
> PSU is fried, as replacing the plug fuse didn't help.)
1050s run hot, so hot they actually break solder contacts from thermal
cycling, but no way should they run that hot.
Sounds like it is either the voltage doubling circuit <in the area between
the 3 large caps at the back of the board> or a fried 7812 regulator. There
are two regulators on the heat sink, 7805 and 7812, I would think about
replacing the 7812 and flow new solder onto all the power diodes on the
board. It is possible that one of the diodes fried, they freeze on in a dead
short when this happens and that will cause everything to over heat.
That pretty much sums it up. You may want to test the diodes before you do
anything else. They are the cheapest and easiest thing to replace and you
may get lucky.
Thomas Richter
> news:941m56$lrm$1...@nnrp1.deja.com...
> <snip>
> Sounds like it is either the voltage doubling circuit <in the area between
> the 3 large caps at the back of the board> or a fried 7812 regulator. There
> are two regulators on the heat sink, 7805 and 7812, I would think about
> replacing the 7812 and flow new solder onto all the power diodes on the
> board. It is possible that one of the diodes fried, they freeze on in a dead
> short when this happens and that will cause everything to over heat.
> That pretty much sums it up. You may want to test the diodes before you do
> anything else. They are the cheapest and easiest thing to replace and you
> may get lucky.
Definitely a good advice. These diodes died here in one of my drives, they
can't take all the current for a long time and are IMHO under-dimensioned for
their purpose. I would especially look for the two diodes between the
three big capacitors in the voltage doubler circuit and would replace them
by 5A types. IIRC, the original drive came with 1A diodes here but they are
a bit too week.
So long,
Thomas
______________don't_cut_here,_it_could_damage_your_terminal____________________
_______ _____ _____
/ / / / / / / EMAIL: th...@einstein.math.tu-berlin.de
/ /____/ / / /____/ http://www.math.tu-berlin.de/~thor/thor/index.html
/ / / / / / \ PGP available on request, finger print:
/ / / /____/ / / 11 FC 46 B0 7F 42 43 AC 38 A4 78 9A 24 BC 77 BE
_______________________________________________________________________________
Rick Cortese
news:t690c0p...@corp.supernews.com...
> "Geoff McHugh" <mcpu...@forty-seven.demon.co.uk> wrote in message
> news:941m56$lrm$1...@nnrp1.deja.com...
> <snip>
> anything else. They are the cheapest and easiest thing to replace and you
> may get lucky.
Whoops!!! Forgot one other biggie. I have had one of the large caps on the
back of the board go out. Only once in several dozen drives I have repaired,
but that would produce the same symptoms.
If it is a cap, it will probably be kind of puffy. The fluid inside them
leaks out, but it looks like Atari put some glue on the bottoms of the caps
as part of their manufacturing process so you have to be able to tell the
difference. Go with puffy being bad if you see that.
Geoff McHugh
"Rick Cortese" <rico...@netmagic.net> wrote:
>
> Sounds like it is either the voltage doubling circuit <in the area
between
> the 3 large caps at the back of the board> or a fried 7812 regulator.
There
> are two regulators on the heat sink, 7805 and 7812, I would think
about
> replacing the 7812 and flow new solder onto all the power diodes on
the
> board. It is possible that one of the diodes fried, they freeze on in
a dead
> short when this happens and that will cause everything to over heat.
>
> That pretty much sums it up. You may want to test the diodes before
you do
> anything else. They are the cheapest and easiest thing to replace and
you
> may get lucky.
>
>
Thanks for the advice, I think you have hit the nail on the head...
For a start, I now remember leaving the drive in for repair (probably 10
or more years ago, so can't remember the exact details of the fault) and
the two diodes (CR15 & CR16) between the 3 big caps seem to have been
replaced (they aren't as 'neat' as the others. They are on part of the
board where the most heat seems to be generated, and CR15 looks slightly
melted around the edges. Also, the heatsink heats up quickest around
the 7812.
Is there any simple way to test the diodes, e.g. with a multimeter, say,
without having to replace them completely? You wouldn't happen to know
what diodes to ask for to replace them? Thomas recommended 5A types,
but I'd hate to buy the wrong sort of diode and fry the house along with
the 1050!! :) (I'm more of a software type, but I can handle a
soldering iron ok!)
I will try to replace the diodes and the 7812 and see how that works, I
have nothing to lose as things stand :)
Thanks again for your help, I wouldn't have had the first idea where to
start looking without it.
Rick Cortese
> In article <t690c0p...@corp.supernews.com>,
<Snip>
> Is there any simple way to test the diodes, e.g. with a multimeter, say,
> without having to replace them completely? You wouldn't happen to know
> what diodes to ask for to replace them? Thomas recommended 5A types,
> but I'd hate to buy the wrong sort of diode and fry the house along with
> the 1050!! :) (I'm more of a software type, but I can handle a
> soldering iron ok!)
Not sure exactly how they did the voltage doubling circuit with them, but
you should be able to measure resistance from one side of the diode to the
other. If it is zero, it is blown or the cap like I said in another post has
shorted out.
Of course if the cap has shorted, you will be lucky if anything else
survived. When they go bad, they have a tendancy to act like a dead short to
ground and burn everything else around them out. The weak power supplies
Atari used 31 VA, may have saved you.
As far as replacement diodes, just about anything around 3 amps should be
good. I even replaced the diodes in one drive with some 3 amp germanium
diodes from Radio Shack, seemed to work fine.
RHamiIton5
>Is there any simple way to test the diodes, e.g. with a multimeter, say,
>without having to replace them completely? You wouldn't happen to know
>what diodes to ask for to replace them?
When my CR17 went out and blew the fuse in the power brick, I detected the bad
diode by measuring resistance in each direction across them with a VOM; about
the same = bad. High - low probably OK. I replaced with 3 amp RS #276-1141;
still crankin' almost a year.
-Ron
Inform everyone that communications have completely broken down.
Geoff McHugh
rhami...@aol.comedy.com (RHamiIton5) wrote:
>
> When my CR17 went out and blew the fuse in the power brick, I detected
the bad
> diode by measuring resistance in each direction across them with a
VOM; about
> the same = bad. High - low probably OK. I replaced with 3 amp RS
#276-1141;
> still crankin' almost a year.
There's nothing like too much information at the end of the night to
confuse me :-)
(it's 1:30am and I should really have turned in a long time ago if I'm
going to make it to work tomorrow!!)
I checked all the diodes in the vicinity of the big caps with a
multimeter (in each direction), and they all seem to be in the high-low
category!
None of the caps look 'puffy' like Rick described earlier, does that
mean my first step should be replacing the 7812? Could it be the sole
cause of the problem? Is there a simple way to test if the 7812 is
fried before replacing it?
Thanks again, I'll get my head round this soon, honest ;-)
Rick Cortese
<snip>
> I checked all the diodes in the vicinity of the big caps with a
> multimeter (in each direction), and they all seem to be in the high-low
> category!
>
> None of the caps look 'puffy' like Rick described earlier, does that
> mean my first step should be replacing the 7812? Could it be the sole
> cause of the problem? Is there a simple way to test if the 7812 is
> fried before replacing it?
Good idea to check it right where it is. The center leg and the tab are the
ground. Let's say you are looking at the LM7812 printed on the front. Legs
will be:
[input] [ground] [output]
Input should be at least 14V and the output should be within .6V of 12V
Geoff McHugh
"Rick Cortese" <rico...@netmagic.net> wrote:
<snip>
>
> Input should be at least 14V and the output should be within .6V of
12V
>
I checked this last night, and unless I've screwed up somewhere (which
has to be considered as a possibility I suppose <g>) the input voltage
seems to be somewhere around 6V, and the output is a bit less than 1V!!
The heatsink is giving off a lot of heat around the 7812, so I guess it
must be fried...
I have to admit that I was a little pushed for time, so I'll check it
again later this evening, and if I get the same results, I'll be
swapping out the 7812 just as soon as I can get hold of a replacement.
Rick Cortese
> In article <t6ah0cf...@corp.supernews.com>,
> "Rick Cortese" <rico...@netmagic.net> wrote:
> <snip>
> >
> > Input should be at least 14V and the output should be within .6V of
> 12V
> >
>
> I checked this last night, and unless I've screwed up somewhere (which
> has to be considered as a possibility I suppose <g>) the input voltage
> seems to be somewhere around 6V, and the output is a bit less than 1V!!
> The heatsink is giving off a lot of heat around the 7812, so I guess it
> must be fried...
I almost want to say you are on your own now. This much heat means your 1050
may let out the magic smoke they put in those chips.<grin>
>
> I have to admit that I was a little pushed for time, so I'll check it
> again later this evening, and if I get the same results, I'll be
> swapping out the 7812 just as soon as I can get hold of a replacement.
But... I don't think so. I think the 7812 us getting hot because it is
literally working so hard.
I can only think of two reasons you would get this condition.
One is there is a dead short in something down stream of the 7812. Maybe the
spindle motor burned out. I can't recall off the top of my head, but they
may run the head positioning stepper on 12V too.
The other is the voltage doubler circuit went south and is either not
putting out the voltage it should. Sometimes when you pass to much AC to a
regulator they run pretty hot.
With everything unplugged, I would check the resistance between the
output<and output> of the 7812 and ground. I just did on one of my drives
and you don't get a good reading because the caps on the board are charged
by the DVM testing voltage, but you should see several hundred Ohms
resistance on both sides.
The reason for this IIRC is everything that uses 12V on the board is either
switched by a descrete transistor or by a transistor array => something like
a 2907<6?> 14 pin dip.
If you get low resistance on the output => transistor array is bad.
If you get low resistance on the input => bad cap in the voltage doubler
circuit or maybe shorted diode.
I really think the 7812 may be OK. It sounds like its thermal protection
circuitry shut it down.
Geoff McHugh
"Rick Cortese" <rico...@netmagic.net> wrote:
I almost want to throw the thing out and just stick with my PC, but all
those floppies are calling to me! :) Anyway, I always wanted to see
some of that magic smoke, so I'll keep trying until you give up on me as
a lost cause ;)
>
> With everything unplugged, I would check the resistance between the
> output<and output> of the 7812 and ground. I just did on one of my
drives
> and you don't get a good reading because the caps on the board are
charged
> by the DVM testing voltage, but you should see several hundred Ohms
> resistance on both sides.
>
Ok, I did that, and I don't think you are going to like the result... :)
With the power disconnected, it seems that the resistance between input
& ground is about 500ohms, and between output & ground seems to be zero!
I tried the same on the 7805 and got about 750ohms on the input and
350ohms on the output.
Now, does that sound as if it indicates a short somewhere or a busted
7812 or a bad transistor array?
I'm going for a lie down in a quiet corner of the room :)
Rick Cortese
Heck, this really makes me feel good. I'm kind of perverse that way! Closer
we get to having it cornered w/o bringing out a soldering iron, the better
in my book.
Armed with my most recent dose of enthusiasm after reading this, I dug out
my SAMS Computer Facts.
In no particular order. The voltage into the 7812 is listed at 17.8V.
TP14<test point 14> is the output and is listed at 12V of course.There's
nothing but low value caps to filter the output, so IMO they would have
exploded like a popcorn kernal if they had shorted.
Looking at the stepper motor, Jumper J15 pretty much completely disconnects
it. With the drive facing you, this in in the back right hand side. Unplug
the jumper and test resistance on the 7812 output to ground and if it is
still near zero, your stepper is fine.
The spindle motor is J1, 3rd from the back left. It should be plugged in
REVERSED. This has been the case on every 1050 I have ever seen. Sometimes
people see the other jumpers nicely plugged in a row and put J1 in so it is
aligned with the other jumpers, this is wrong, it should stick out like a
sore thumb.
Unplug J1 and test for resistance. If it is still low, then it is probably
the 7812. Looking at the output transistor for the motor, it is a TIP110 and
if it blew the drive would spin continuously.
There is other stuff in there that uses 12V, but most of it is pretty low
power and I can't imagine it being the problem w/o giving off a dull red
glow when you turned the drive on.
There are some other parts in the stepper circuit that could go out, but
they are feed with 12V through a 47 Ohm resistor, so even if they were fried
they wouldn't give zero Ohms.
Geoff McHugh
"Rick Cortese" <rico...@netmagic.net> wrote:
> Heck, this really makes me feel good. I'm kind of perverse that way!
Closer
> we get to having it cornered w/o bringing out a soldering iron, the
better
> in my book.
I can only agree with you there :)
> Looking at the stepper motor, Jumper J15 pretty much completely
disconnects
> it. With the drive facing you, this in in the back right hand side.
Unplug
> the jumper and test resistance on the 7812 output to ground and if it
is
> still near zero, your stepper is fine.
I checked this, and it was still zero resistance on the output to
ground, so that looks ok...
> The spindle motor is J1, 3rd from the back left. It should be plugged
in
> REVERSED. This has been the case on every 1050 I have ever seen.
Sometimes
> people see the other jumpers nicely plugged in a row and put J1 in so
it is
> aligned with the other jumpers, this is wrong, it should stick out
like a
> sore thumb.
>
> Unplug J1 and test for resistance. If it is still low, then it is
probably
> the 7812. Looking at the output transistor for the motor, it is a
TIP110 and
> if it blew the drive would spin continuously.
As you said, J1 was plugged in reversed. On removing it, though, the
resistance at the 7812 output to ground is still zero. I was very busy
this weekend though, and I onl managed to grab 10 or 15 minutes to check
this, but I am pretty sure that the resistance was zero no matter which
of the jumpers were plugged or unplugged.
Hopefully this is narrowing things down, you certainly understand what's
going on in the 1050 better than I do :)
Is this suggesting any other options to you, or is everything pointing
to the 7812?
Thanks again
Rick Cortese
> In article <t6h0k25...@corp.supernews.com>,
<snip>
> Is this suggesting any other options to you, or is everything pointing
> to the 7812?
That is the leading suspect. I am trying to think of a cheesy way of doing
it, heat sink prevents easy removal and all that.
You have a couple of choices, either try to save the 7812 or do less damage
to the PC board by cutting the legs off before removing it. I would probably
go quick and dirty: Use a diagonal snip to cut the output pin and check for
resistance on both sides of the cut. If the side attached to the 7812 still
reads low/no resistance, it is the 7812. If the side attached to the 7812
reads high, that narrows it down to maybe 4-5 parts on the PC board and you
can just solder the leg back together.
There's another thing you should try if you look at the 7812 . While the
7812 output is disconnected<one way or another> check the solder pad for the
input. See if you get the nominal 17V input when the power is applied. Long
as you are fixing stuff, might as well make sure the voltage doubling
circuit is working.
Here's hoping it is the 7812.
Geoff McHugh
"Rick Cortese" <rico...@netmagic.net> wrote:
<snip>
> You have a couple of choices, either try to save the 7812 or do less
damage
> to the PC board by cutting the legs off before removing it. I would
probably
> go quick and dirty: Use a diagonal snip to cut the output pin and
check for
> resistance on both sides of the cut. If the side attached to the 7812
still
> reads low/no resistance, it is the 7812. If the side attached to the
7812
> reads high, that narrows it down to maybe 4-5 parts on the PC board
and you
> can just solder the leg back together.
>
> There's another thing you should try if you look at the 7812 . While
the
> 7812 output is disconnected<one way or another> check the solder pad
for the
> input. See if you get the nominal 17V input when the power is applied.
Long
> as you are fixing stuff, might as well make sure the voltage doubling
> circuit is working.
>
> Here's hoping it is the 7812.
>
So much for hoping :)
I cut the output (right hand) lead of the 7812 and checked the
resistance on both sides of the cut. The 7812 side reads about 80 Ohms,
and the pcb side reads zero.
So, those 4-5 parts on the pcb you mentioned...? ;)
I also checked the input voltage on the 7812 with the output leg cut,
and it seems to be slightly above 20V.
I suppose this puts us back to square one with these other 4-5
components (I hope it is one of those!)
Once again, thanks for taking the time to help.
--
Rick Cortese
Well, that is good news. I mean the quick and dirty method of testing
probably saved 30 minutes or so and $1.50 for a new 7812.
> and the pcb side reads zero.
This absolutely bites.
>
> So, those 4-5 parts on the pcb you mentioned...? ;)
>
> I also checked the input voltage on the 7812 with the output leg cut,
> and it seems to be slightly above 20V.
Dynamite. I would expect it to go that high under no load conditions. i.e.
7812 had a leg cut.
>
> I suppose this puts us back to square one with these other 4-5
> components (I hope it is one of those!)
I'm looking at the schematic right now and asking myself "Why did they do
things like this?" Probably some engineering thing that only an engineer
would understand.
They have 4 low value caps on the output side of the 7812 that could have
shorted. These are C70, C43, C42, 47 micro Frarad and C21 4.7 micro Farad.
C70, C43, and C42 are just forward of the TIP110 transistor mounted on the
heat sink.
C21 is mounted near the front of the board next to a RCA CA3086 transistor
array.
As far as I can see, the only chips that have 12V being fed directly into it
w/o having a low value current limiting resistor <~47 Ohm> are the U22
MC1733CP, U23 NE592, U24 CA311, chips. Kind of skeptical it could be one of
the, I mean the CA311 is just a little 8 pin dip/op amp thing. Putting 30
watts through that if it failed would have made it glow like a xmas tree
bulb.
First thing is to check the little caps I just mentioned for puffs. Maybe
hold the PC board up to a bright light and look for signs of browning
fiberglass resin around the chips/caps I just mentioned.
For ref, the area around the diodes in the voltage doubling circuits is
likely to be a taffy brown and that is the only place you should see that
condition on the board if they were mounted directly on the board vs ~1/2"
above it.
It is kind of risky, you will burn your finger for sure, but another "If it
was my thumb" kind of thing. I would plug a good transformer back in after
fixing the snipped leg of the 7812 and check each of the chips. Just moisten
a finger so you don't completely brand yourself and lightly touch the chips.
If there is a hot one, that is your culprit.
Geoff McHugh
"Rick Cortese" <rico...@netmagic.net> wrote:
<snip>
> I'm looking at the schematic right now and asking myself "Why did they
I have this strange feeling we're backtracking right back to where we
started out :)
I fixed the 7812 and checked the 4 caps for puffiness - they look pretty
much as I imagine capacitors should, so I doubt if it's them. The chips
look ok and don't seem to even warm up when power is applied.
So I looked at the board, and the only place I can see any signs of heat
are round the two diodes nestled among the 3 big caps at the back (CR15
and CR16) With power applied, and a licked fingertip, the lead into
CR15 is hot enough (within seconds) to make an audible hiss as the
saliva (and then my fingertip - you were right about getting branded!!)
boiled away!
Now, does that sound like CR15 & 16 are the problem? They both give a
different resistance when measured in both directions - swapping +ve &
-ve probe - but boy were they hot!! My fingertip is still stinging ten
minutes later!! :) Or does it point to some other possible cause?
Rick Cortese
I would expect them to be hot since there appears to be a dead short on the
output circuitry of the 7812. The seemed to be working fine when you cut off
the output leg of the 7812 anyway.
I hope you feel comfortable with pulling a few parts to isolate the short. I
will try to really keep it simple. Start by pulling L6, the inductor right
by the jumpers/heat sink and measure the resistance from the both solder
pads and the output of the 7812 to ground.
If the 7812 output goes to ~6k or so and one side of the former 'home' of
the inductor pads <not a real reading but what I see while the caps and
stuff charge to the resistance testing voltage of my DVM> then it is the
TIP110.
If it doesn't, look for inductor L4 up by the front of the board, next to
the CA3086. Same basic procedure.
Since all the chips are running cool, I don't think it could be one of them.
If they were blown, they would have been as hot as the power diodes. I mean
a 30 watt soldering iron is pretty hefty. It's got to be a part that is
capable of sinking one hell of a lot of current w/o glowing red. That kind
of leads me to suspect the TIP agains, maybe a genuine short like a lead or
solder goober, or maybe a shorted cap.
Geoff McHugh
"Rick Cortese" <rico...@netmagic.net> wrote:
<snip>
Ok, I removed L6 (which appears to be a little blackened in the middle,
around the area away from the pcb) and checked the resistance from each
solder pad to the ground of the 7812 (centre pin) the resistance
appears to be zero from both pads, and in fact, the resistance between
the pads is zero, as is the resistance across the inductor (I was
measuring everything by this stage!)
Then I tried removing L4. Looking from the front of the board, the
resistance between the left hand solder pad and the centre pin of the
7812 is about 60 Ohms, and between the right hand solder pad and 7812
ground is zero. The resistance between the two solder pads of L4 is 500
Ohms, and the resistance across the inductor is zero (the inductors
looked a bit like resistors, but are probably not related at all!!)
Is this pointing to a fault somewhere to the back of the board, round
L6? What's next? start checking the caps?
(I didn't replace L6 before removing L4, does that make any difference?)
Rick Cortese
> In article <t6ulqlq...@corp.supernews.com>,
> "Rick Cortese" <rico...@netmagic.net> wrote:
>
> <snip>
> around the area away from the pcb)
I hope this is the key observation.
> and checked the resistance from each
> solder pad to the ground of the 7812 (centre pin) the resistance
> appears to be zero from both pads, and in fact, the resistance between
> the pads is zero, as is the resistance across the inductor (I was
> measuring everything by this stage!)
Inductors here are just for filtering noise. They should be zero resistance,
but good to know it isn't burned out.<yet>
This kind of has me puzzled, but then you probably guessed that by now. I'm
thinking pull the TIP110.
I am more following heat then anything else. L6 only supplies the TIP110 and
C43 is the only part between it and the TIP110. Those should be the only
parts that could short if my schematic is right, nothing else there.
Just that it should have been like cutting a trace. There should be normal
on one side and shorted condition on the other. Very strange this didn't
happen.
I would remove the RFI shielding over the drives processor/RAM/ROM just to
be on the safe side. Atari in their wisdom appear to have routed the 12 V
output from the 7812 all the way to the front of the board (L4) and back to
the transisor on the heat sink. Could be a mechanical short with the metal
shieldling.
>
> Then I tried removing L4. Looking from the front of the board, the
Recheck the L6 pads with L4 removed. It kind of goes
part part part<singular>
7812--|--L4-----|----L6---|--TIP110---- spindle motor
part part
Geoff McHugh
(Busy weekend out of the way)
> I would remove the RFI shielding over the drives processor/RAM/ROM
just to
> be on the safe side. Atari in their wisdom appear to have routed the
12 V
> output from the 7812 all the way to the front of the board (L4) and
back to
> the transisor on the heat sink. Could be a mechanical short with the
metal
> shieldling.
> >
> > Then I tried removing L4. Looking from the front of the board, the
>
> Recheck the L6 pads with L4 removed. It kind of goes
> part part part<singular>
> 7812--|--L4-----|----L6---|--TIP110---- spindle motor
> part part
I removed the RFI shielding, which didn't make any difference, and
checked around for any obvious signs of solder traces making unwanted
contact, but saw nothing glaringly obvious.
I checked the above route from the 7812 to the TIP110 as follows
(looking from the front of the drive)
7812 output --> L4 right hand solder pad 0 Ohms
L4 right hand solder pad --> L4 left hand solder pad ~60 Ohms
L4 left hand solder pad --> L6 right hand solder pad ~500 Ohms
L6 right hand solder pad --> L6 left hand solder pad 0 Ohms
L6 left hand solder pad --> TIP110 left hand leg ~60 Ohms
In the interests of measuring everything, in case I picked the wrong
path :) I also get
7812 output --> L4 left pad ~60 Ohms
L4 right pad --> either L6 pad 0 Ohms
either L6 pad --> TIP110 right hand leg ~60 Ohms
I think that covers all ways through the path above, given that both
pads of L6 appear to be shorted to ground.
I'm going to have a closer look at the pcb and see if I can't see
something obvious connecting to ground.
You'll be glad to hear that I'll have to give you a break from all this
for a while, I'm going to be working away from home for the next couple
of weeks, starting Wednesday (I'll be back for the weekend, but will
most likely be busy then too) and while I should have a net connection,
I'm not going to be carrying the 1050, multimeter and soldering iron
through an airport metal detector :)
I hope the above helps narrow this thing down a bit further...
Rick Cortese
Just a bit of info, since we aren't using the same DVMs, we will get
different resistances. My DVM always reads ~10x the resistance yours does.
I honestly thing you may have two problems here when I look at the results.
> > part part part<singular>
> > 7812--|--L4-----|----L6---|--TIP110---- spindle motor
> > part part
<snip>
> 7812 output --> L4 right hand solder pad 0 Ohms
same
> L4 right hand solder pad --> L4 left hand solder pad ~60 Ohms
Close to infinate with L4 removed ~same
> L4 left hand solder pad --> L6 right hand solder pad ~500 Ohms
Infinate ~same
> L6 right hand solder pad --> L6 left hand solder pad 0 Ohms
**************************
Infinate!
**************************
> L6 left hand solder pad --> TIP110 left hand leg ~60 Ohms
about the same.
>
> In the interests of measuring everything, in case I picked the wrong
> path :) I also get
>
> 7812 output --> L4 left pad ~60 Ohms
Near infinate
> L4 right pad --> either L6 pad 0 Ohms
Should be L6 Right to L4 Right 0 Ohms
L6 Left to L4 Right > 0 Ohms.
Double check me, but I think this elimnates everything between the 7812 and
L4.
That leaves us to C42, C21 as the leading suspects. There's a bunch of other
low value ones in there, but I don't believe they could have shorted w/o
exploding.
Before you do any more pulling, with L4 removed, you should power up the
drive again and check for 12V on the output of the 7812 and one of the L4
solder pads.
> either L6 pad --> TIP110 right hand leg ~60 Ohms
Remove and test in order
>
> I think that covers all ways through the path above, given that both
> pads of L6 appear to be shorted to ground.
No L6 is connected directly to the center leg of the TIP110. If you get 0
Ohms the left leg, your TIP110 has gone to electric lala land.
>
> I'm going to have a closer look at the pcb and see if I can't see
> something obvious connecting to ground.
>
> You'll be glad to hear that I'll have to give you a break from all this
> for a while, I'm going to be working away from home for the next couple
> of weeks, starting Wednesday (I'll be back for the weekend, but will
> most likely be busy then too) and while I should have a net connection,
> I'm not going to be carrying the 1050, multimeter and soldering iron
> through an airport metal detector :)
Have fun!
Geoff McHugh
of other
> low value ones in there, but I don't believe they could have shorted
w/o
> exploding.
>
> Before you do any more pulling, with L4 removed, you should power up
the
> drive again and check for 12V on the output of the 7812 and one of
the L4
> solder pads.
I did this, with both L4 & L6 removed, and got 12v on the output of the
7812 and the right pad of L4, I assume that is a good thing? :)
> > either L6 pad --> TIP110 right hand leg ~60 Ohms
>
> Remove and test in order
Before I make a catastrophic error, remove and test what in order, C42
& C21? I don't want to rip out more than I have to, or make things
worse! :) And should I put L6 & L4 back before doing anything else?
> > I think that covers all ways through the path above, given that both
> > pads of L6 appear to be shorted to ground.
>
> No L6 is connected directly to the center leg of the TIP110. If you
get 0
> Ohms the left leg, your TIP110 has gone to electric lala land.
I doubt I'll get a chance to remove anything until at least the
weekend, maybe even the following one...
Rick Cortese
> In article <t7bn6qp...@corp.supernews.com>,
> "Rick Cortese" <rico...@netmagic.net> wrote:
> <snip>
> >
> > That leaves us to C42, C21 as the leading suspects. There's a bunch
> of other
> > low value ones in there, but I don't believe they could have shorted
> w/o
> > exploding.
> >
> > Before you do any more pulling, with L4 removed, you should power up
> the
> > drive again and check for 12V on the output of the 7812 and one of
> the L4
> > solder pads.
>
> I did this, with both L4 & L6 removed, and got 12v on the output of the
> 7812 and the right pad of L4, I assume that is a good thing? :)
Absolutely! Means all those parts in between the 9VA jack and the L4 are
working. We knew a lot of them were working from snipping the output
terminal of the 7812 anyway, but this is a double positive and adds a
handful of parts into the "Test Good" side of the ledger.
>
> > > either L6 pad --> TIP110 right hand leg ~60 Ohms
> >
> > Remove and test in order
>
> Before I make a catastrophic error, remove and test what in order, C42
> & C21? I don't want to rip out more than I have to, or make things
> worse! :) And should I put L6 & L4 back before doing anything else?
I wouldn't, but this is where your own judgement comes into play. Remember I
am the guy who got your thumb branded.
C42 and C21 are really pedestrian parts, you could step on a transistor
radio and get 5-6 ones just like them. I really wouldn't worry about losing
them. If you are careful, you could just remove them, check the resistance
across their pins and see if it is anything other then a dead short. Really
if you had a capacitance meter instead of a DVM it would be better.
With your DVM you should see ~0 resistance when you first apply the probes
and it will steadily climb to infinite resistance.
The alternative procedure would be to jumper L4 for now and make sure it is
something between L4 and L6. You could just stick a piece of wire in where
L4 used to go, apply power and see if you still get 12V.
I think you can see what we are doing here. We are kind of going along the
12V circuit in inchworm fashion. Removing L4 and L6 just isolated the 12V
into sections so we could check each section individually and narrow the
suspect parts down.
So if you put L4 back in <or jumper it> and test for 12V, it is just making
doubly sure the fault<one anyway> is between L4 and L6. Completely up to you
as the results are valuble for isolating the problem.
Don't touch the diodes again if you don't get 12V on both sides of L4
though.<grin>
>
> > > I think that covers all ways through the path above, given that both
> > > pads of L6 appear to be shorted to ground.
> >
> > No L6 is connected directly to the center leg of the TIP110. If you
> get 0
> > Ohms the left leg, your TIP110 has gone to electric lala land.
>
> I doubt I'll get a chance to remove anything until at least the
> weekend, maybe even the following one...
No hurry, I think you have it cornered now.