Actual amp modding question (Electar Tube 30)

[Bruce Morgen]

I've an Electar Tube 30 for
several years -- long ago I
retubed with JJs and put in
a JBL MI-10 speaker, a model
that sounds great in other
amps I own. This amp lacks
clarity and chime -- because
of that, I now only use it
for bass (at low volume
levels -- I know it's open-
back and not really a bass
amp).

IIRC, JJ's EL84s are unusual
compared to most others --
given that the bias is
fixed, could that be the
problem? The circuit is
pretty odd (lots of diodes),
so being a non-expert I'm
hesitating to to play with
it without some savvy advice
from experienced modders --
so, any guidance would be
much appreciated, up to and
including, "Shitcan the dumb
thing and keep the speaker
-- that circuit is a total
misdesign!"

Here's the diagram:

<http://www.freeinfosociety.com/media/pdf/3346.pdf>

[WB]

On 12/3/2011 2:22 PM, Bruce Morgen wrote:

> <http://www.freeinfosociety.com/media/pdf/3346.pdf>

I love that site.

[RS]

On Sat, 03 Dec 2011 15:22:38 -0500, Bruce Morgen <edi...@juno.com>
wrote:

>[Electar Tube 30]
>... This amp lacks
>clarity and chime -

>IIRC, JJ's EL84s are unusual
>compared to most others --
>given that the bias is
>fixed, could that be the
>problem? The circuit is
>pretty odd (lots of diodes),

...

>
>Here's the diagram:
>
><http://www.freeinfosociety.com/media/pdf/3346.pdf>

Hi Bruce,

Starting from the input: The two 1n4148 diodes from the input to
ground are there to snub spikes. They'll conduct at around 0.65v, so
most low output guitars won't break over, but there's always the
possibility that peaks of higher output pickups will get clipped.
Either do without, or maybe put more diode pairs in series.

R5, the cathode resistor of the first stage, is way too low. This will
bias the tube in a slightly higher gain region but at the expense of
headroom, since the plate voltage will not have as much swing
available on the negative side (the tube is conducting more than it
should, so the voltage on the plate will be lower). Increase that to
1500 ohms or so, and bypass with a 10uf cap. Similar to the cathode of
V2a.

If you don't like bright top-end, lift one side of C5. That's
effectively a hard-wired equivalent of Fender's bright switch.

R7 is an error--it can't be 2K. See what that value actually is.

The V1b circuit is odd. with R12 left unconnected, it's a unity gain
phase splitter, but if they're getting switchable gain by connecting
the other side of R12 to ground, that would generate a loud pop.
Without knowing exactly what's in the LDR module, part of this is just
guesswork. The strange thing is that the two back to back diodes are
diagrammed as zeners, and there's no call for them. They're back to
back, so the forward conduction voltage (0.65v) of the opposite diode
will come into play before the zener gets anywhere near the reverse
breakover voltage.

Also unusual to see those directly in front of the plate without any
resistance in between. The short version: If you're not using
overdrive, lift the connection to the diodes (or one side of R101).
See if that signal cleans up.

The tone stack is configured to sound more Marshally than Fendery. If
that matters, I'll advise on changes. You could easily adapt that.

The bottom side of the volume pot should go to ground. Another error
on the diagram.

The diagram looks like R21 and R22 are in parallel, but there's no
need for that. No big deal. Leave it.

R6 is wrong. That must be 120 ohms (?), and it should be decent
wattage.

The grid stopper (parasitic suppression) resistors R27 and R28 are
higher than normal, but that's not a bad thing.

A word on bias methods: "Fixed bias" doesn't really mean that it's
stationary. Ironically, that's the method that can easily be made
adjustable. It just implies that bias is set by applying a negative
voltage to the control grids.

Your amp is not fixed bias though. It's cathode biased. That
effectively self-regulates, so you don't really need to adjust it. And
it means that it will adapt to tube changes.

Having said that, there's one occasional glitch: If one of the output
tubes starts conducting more, it will hog most of the available
current. And the converse: If one of the tubes conducts way less (or
is unplugged), the remaining tube can run -way- hot. The solution is
to use separate cathode resistors for each tube. In that case, they
need to be bypassed well with high value caps.

If you choose to use separate resistors there, you can easily measure
individual output tube current (voltage across the resistor, use ohm's
law).

Otherwise, you can try hooking 1 ohm or 10 ohm resistors in between
each cathode and the cathode resistor R6 (remember to check R6).
Then calculate current by measuring voltage across the resistor, etc.

Misc: Check power supply resistors R15 and R19 to see if they appear
charred. There are no B+ voltages specified, but you may want to
measure to get an idea.

That's all for first glance.

[Bruce Morgen]

RS <R...@nospam.com> wrote:

>On Sat, 03 Dec 2011 15:22:38 -0500, Bruce Morgen <edi...@juno.com>
>wrote:
>
>>[Electar Tube 30]
>>... This amp lacks
>>clarity and chime -
>
>>IIRC, JJ's EL84s are unusual
>>compared to most others --
>>given that the bias is
>>fixed, could that be the
>>problem? The circuit is
>>pretty odd (lots of diodes),
>...
>>
>>Here's the diagram:
>>
>><http://www.freeinfosociety.com/media/pdf/3346.pdf>
>
>Hi Bruce,
>
>Starting from the input: The two 1n4148 diodes from the input to
>ground are there to snub spikes. They'll conduct at around 0.65v, so
>most low output guitars won't break over, but there's always the
>possibility that peaks of higher output pickups will get clipped.
>Either do without, or maybe put more diode pairs in series.

OK, without makes sense
even though I usually
use low-output pickups.

>
>R5, the cathode resistor of the first stage, is way too low. This will
>bias the tube in a slightly higher gain region but at the expense of
>headroom, since the plate voltage will not have as much swing
>available on the negative side (the tube is conducting more than it
>should, so the voltage on the plate will be lower). Increase that to
>1500 ohms or so, and bypass with a 10uf cap. Similar to the cathode of
>V2a.
>

I suspect that's much of
the problem -- good catch!

>If you don't like bright top-end, lift one side of C5. That's
>effectively a hard-wired equivalent of Fender's bright switch.

As of now, it's not bright
enough -- but correcting
and bypassing R5 might
well fix that.

>
>R7 is an error--it can't be 2K. See what that value actually is.

OK, will do.

>
>The V1b circuit is odd. with R12 left unconnected, it's a unity gain
>phase splitter, but if they're getting switchable gain by connecting
>the other side of R12 to ground, that would generate a loud pop.
>Without knowing exactly what's in the LDR module, part of this is just
>guesswork. The strange thing is that the two back to back diodes are
>diagrammed as zeners, and there's no call for them. They're back to
>back, so the forward conduction voltage (0.65v) of the opposite diode
>will come into play before the zener gets anywhere near the reverse
>breakover voltage.

Good to know somebody else
is stumped by that.

>
>Also unusual to see those directly in front of the plate without any
>resistance in between. The short version: If you're not using
>overdrive, lift the connection to the diodes (or one side of R101).
>See if that signal cleans up.

OK, will do -- the overdrive
is useless to me, sounds
like crap.

>
>The tone stack is configured to sound more Marshally than Fendery. If
>that matters, I'll advise on changes. You could easily adapt that.

It matters -- I'm not a
Marshall guy.

>
>The bottom side of the volume pot should go to ground. Another error
>on the diagram.
>

Yeah, I thought wiring it
as a variable resistor
rather than a voltage
dividers was very odd.

>The diagram looks like R21 and R22 are in parallel, but there's no
>need for that. No big deal. Leave it.
>
>R6 is wrong. That must be 120 ohms (?), and it should be decent
>wattage.

OK.

>
>The grid stopper (parasitic suppression) resistors R27 and R28 are
>higher than normal, but that's not a bad thing.
>
>A word on bias methods: "Fixed bias" doesn't really mean that it's
>stationary. Ironically, that's the method that can easily be made
>adjustable. It just implies that bias is set by applying a negative
>voltage to the control grids.
>
>Your amp is not fixed bias though. It's cathode biased. That
>effectively self-regulates, so you don't really need to adjust it. And
>it means that it will adapt to tube changes.
>
>Having said that, there's one occasional glitch: If one of the output
>tubes starts conducting more, it will hog most of the available
>current. And the converse: If one of the tubes conducts way less (or
>is unplugged), the remaining tube can run -way- hot. The solution is
>to use separate cathode resistors for each tube. In that case, they
>need to be bypassed well with high value caps.

By fixed I just meant there
was no way to adjust it
manually -- wrong term.

>
>If you choose to use separate resistors there, you can easily measure
>individual output tube current (voltage across the resistor, use ohm's
>law).
>
>Otherwise, you can try hooking 1 ohm or 10 ohm resistors in between
>each cathode and the cathode resistor R6 (remember to check R6).
>Then calculate current by measuring voltage across the resistor, etc.
>
>Misc: Check power supply resistors R15 and R19 to see if they appear
>charred. There are no B+ voltages specified, but you may want to
>measure to get an idea.
>
>That's all for first glance.

Very helpful -- thanks!

[RS]

On Sat, 03 Dec 2011 23:21:58 -0500, Bruce Morgen <edi...@juno.com>

wrote:

>RS <R...@nospam.com> wrote:
>
>>On Sat, 03 Dec 2011 15:22:38 -0500, Bruce Morgen <edi...@juno.com>
>>wrote:
>>
>>>[Electar Tube 30]
>>>... This amp lacks
>>>clarity and chime -

>>>Here's the diagram:
>>>
>>><http://www.freeinfosociety.com/media/pdf/3346.pdf>
>>

>>Starting from the input: The two 1n4148 diodes from the input to
>>ground are there to snub spikes. They'll conduct at around 0.65v, so
>>most low output guitars won't break over, but there's always the
>>possibility that peaks of higher output pickups will get clipped.
>>Either do without, or maybe put more diode pairs in series.
>
>OK, without makes sense
>even though I usually
>use low-output pickups.

Very few amp designs use those protection diodes. Tubes are pretty
rugged.

>>R5, the cathode resistor of the first stage, is way too low. This will
>>bias the tube in a slightly higher gain region but at the expense of
>>headroom, since the plate voltage will not have as much swing
>>available on the negative side (the tube is conducting more than it
>>should, so the voltage on the plate will be lower). Increase that to
>>1500 ohms or so, and bypass with a 10uf cap. Similar to the cathode of
>>V2a.
>>
>I suspect that's much of
>the problem -- good catch!

It may be part of it, but if there's a profound lack of clarity, I'd
suspect a secondary problem.

>>If you don't like bright top-end, lift one side of C5. That's
>>effectively a hard-wired equivalent of Fender's bright switch.
>
>As of now, it's not bright
>enough -- but correcting
>and bypassing R5 might
>well fix that.

No big deal. Its corner frequency will be higher than the Fender
bright switch--pretty much 'air zone' band. If that bothers you later,
you know what to do.

>The strange thing is that the two back to back diodes are
>>diagrammed as zeners, and there's no call for them. They're back to
>>back, so the forward conduction voltage (0.65v) of the opposite diode
>>will come into play before the zener gets anywhere near the reverse
>>breakover voltage.
>
>Good to know somebody else
>is stumped by that.

I've seen lots of nonsensical stuff propagate in tube amp designs
until a competent engineer comes in and reviews schematics. A couple
possibilities here:

It could be that they just thought zeners in forward mode sound good.
I've never tried them for overdrives, though I've tried everything
else, from Schottky barriers to various colors of LEDs, to old
seleniums. (BTW, I like IN400x rectifiers)

The odd thing about that particular circuit is that they specified the
zener voltage, so it would look like they didn't intend them to be
used as they are shown (in which case the zener voltage doesn't
matter). If they did want to hard-clip at +/- 25v (not a nice sound),
the correct way to hook them would have been in series.

Or maybe it's just drawn wrong. Also likely, given the other mistakes.

>>Also unusual to see those directly in front of the plate without any
>>resistance in between. The short version: If you're not using
>>overdrive, lift the connection to the diodes (or one side of R101).
>>See if that signal cleans up.
>
>OK, will do -- the overdrive
>is useless to me, sounds
>like crap.

If you have no use for the overdrive, the best way to deal might be to
simply bypass that whole section. Pull up the right side lead of C7
and hook it to pin 2 of V2a, with an additional ground-ref resistor to
ground. And then bypass R9 with a low value resistor in parallel with
a cap (again, like V2a).

Or if you don't need the gain, you could bypass V2a as well by hooking
the tone stack directly to the plate of V1b. Or alternately, by
hooking R8 to the grid of V2a rather than V1b.

>>The tone stack is configured to sound more Marshally than Fendery. If
>>that matters, I'll advise on changes. You could easily adapt that.
>
>It matters -- I'm not a
>Marshall guy.

Let's go over that after the initial problems are solved. But for now,
find a good quality 250pf cap, a .047uf, and a 100k resistor if you
want to effectively duplicate Fender's response and plate loading.
C12 and the Bass control value are not crucial.

>>R6 is wrong. That must be 120 ohms (?), and it should be decent
>>wattage.
>
>OK.

That resistor value could have a big impact. So could a faulty output
tube, abnormal power supply voltages, or a couple other things we
didn't cover.

>>The grid stopper (parasitic suppression) resistors R27 and R28 are
>>higher than normal, but that's not a bad thing.
>>
>>A word on bias methods: "Fixed bias" doesn't really mean that it's
>>stationary. Ironically, that's the method that can easily be made
>>adjustable. It just implies that bias is set by applying a negative
>>voltage to the control grids.

>By fixed I just meant there
>was no way to adjust it
>manually -- wrong term.

The terms are misleading--the opposite of what you might think. There
are a couple downsides to cathode bias, but you can get away with a
lot when you use it. The auto-bias characteristic is a time-saver.

>>That's all for first glance.
>
>Very helpful -- thanks!

Sure thing, Bruce.

RS
(Actual Expert...haha!)

[jh]

#


...Hi Bruce,

in addition to what RS already wrote:

there's nothing "unusual" about the JJEL84. It works and biases like any
other EL84.

regards

Jochen

[Stephen Cowell]

On 12/3/2011 9:58 PM, RS wrote:
> On Sat, 03 Dec 2011 15:22:38 -0500, Bruce Morgen<edi...@juno.com>
> wrote:
>
>> [Electar Tube 30]
>> ... This amp lacks
>> clarity and chime -
>
>> IIRC, JJ's EL84s are unusual
>> compared to most others --
>> given that the bias is
>> fixed, could that be the
>> problem? The circuit is
>> pretty odd (lots of diodes),
> ...

We can't reach through the wires and
swap tubes for you, Bruce. If you
don't own a spare pair of cheap Sovtek
EL84's, then get one. A Peavey blister-pak
goes great in the gig bag.

>>
>> Here's the diagram:
>>
>> <http://www.freeinfosociety.com/media/pdf/3346.pdf>

The courtesy of a diagram is most welcome.
Thank you, WarBaby.

> Starting from the input: The two 1n4148 diodes from the input to
> ground are there to snub spikes. They'll conduct at around 0.65v, so
> most low output guitars won't break over, but there's always the
> possibility that peaks of higher output pickups will get clipped.
> Either do without, or maybe put more diode pairs in series.

OK, now my take: the diodes are there to set a known
spike region to protect the crappy LDR's and clipping
zeners coming later. Just the start of a crappy tone design, see
below.

> R5, the cathode resistor of the first stage, is way too low. This will
> bias the tube in a slightly higher gain region but at the expense of
> headroom, since the plate voltage will not have as much swing
> available on the negative side (the tube is conducting more than it
> should, so the voltage on the plate will be lower).

The tube is working very hard (look at the load also, 2K!) and
also generating even harmonics (one side different from other).
There is sparkle lost in this process... any time V1A
doesn't bypass the K resistor you're going to get a more Marshally
tone. Sparkle is hard to do right, that's where the nasty
parasitics live.

> Increase that to
> 1500 ohms or so, and bypass with a 10uf cap. Similar to the cathode of
> V2a.

This will totally fuck up the rest of the circuit design, believe me.
This is not a platform to start messing with (unless you rip it apart,
not economical).

> If you don't like bright top-end, lift one side of C5. That's
> effectively a hard-wired equivalent of Fender's bright switch.

It's meant to compensate for the tone suck of the 1M Gain control.
At full volume it has no effect.

> R7 is an error--it can't be 2K. See what that value actually is.

The moment you allow high-impedance voltages to start flying around
you start popping anything made of sand. This amp uses 2K loads
(note 1K on dirty channel) for a reason.

> The V1b circuit is odd. with R12 left unconnected, it's a unity gain
> phase splitter, but if they're getting switchable gain by connecting
> the other side of R12 to ground, that would generate a loud pop.

The zeners are clippers. R12 is always grounded in my print.

> Without knowing exactly what's in the LDR module, part of this is just
> guesswork. The strange thing is that the two back to back diodes are
> diagrammed as zeners, and there's no call for them. They're back to
> back, so the forward conduction voltage (0.65v) of the opposite diode
> will come into play before the zener gets anywhere near the reverse
> breakover voltage.

The resistor R12 evens out the effect... note the 1K plate load for
this side (the dirty channel). It's called asymmetrical clipping...
more even harmonics.

You can understand the LDR circuit better if you visualize that
there are pass units (the phototransistor) across the top side
and switch units (the LED/light source) across the bottom. What
looks like device pins are actually jumpers between devices.
I found a circuit diagram for one (Isocom 90980) that shows
this layout. Jumpering the LED voltages like they do shows
that one is off while the other is on (kathode jumped to anode).
Pin 1 is bottom left in the diagram, by C5. It's a 16-pin DIP.

http://www.farnell.com/datasheets/90980.pdf

>
> Also unusual to see those directly in front of the plate without any
> resistance in between.

They are protected by the low-impedance load and the clipping diodes
back at the beginning. This is all crappy, BTW.

> The short version: If you're not using
> overdrive, lift the connection to the diodes (or one side of R101).
> See if that signal cleans up.

It will probably sound crappier. That's your basic problem here...
most anything you do will make it sound crappier, and it's not an
amp that will reward use of premium tubes. A basic waste of thermionic
emission.

> The tone stack is configured to sound more Marshally than Fendery. If
> that matters, I'll advise on changes. You could easily adapt that.

Unequivocally not worth it. These things go well on CL, as in gone.

> The bottom side of the volume pot should go to ground. Another error
> on the diagram.

That would not be wise if the intent (as drawn) is as a series
resistor. The argument could be made that the bottom of the
pot should be connected to the wiper, another matter entirely.

>
> The diagram looks like R21 and R22 are in parallel, but there's no
> need for that. No big deal. Leave it.

In order to balance the output they probably used R21 to trim the
68K kathode resistor down some to make up for the extra 1.8K/1M
self-bias network (what's 2.1M || 68K? 65.867K.. +1.8 = 67.667K).
Really stupid considering all the attention to generating even
harmonics done earlier in the circuit (with no inductor to sweeten
the sauce). Note the difference (a factor of 100) between the two
phase coupling caps. I'd be very surprised if this was really
the case.

> R6 is wrong. That must be 120 ohms (?), and it should be decent
> wattage.

RS is correct, this is an error. I'd like to see a 120K 1/4 watt here.

> The grid stopper (parasitic suppression) resistors R27 and R28 are
> higher than normal, but that's not a bad thing.

Say goodbye to sparkle and tone. Resistors are not tone's
friend... any time a resistor appears in a circuit the opportunity
to have the signal response spectrum skewed appears. The obvious
intent was to inhibit highs and make parasitics a non-issue.
This kind of amp is easy and cheap to make.

(snip bias lecture)

The LDR's are there for channel switching... there are four of
them because two are used for each (dirty/clean) channel, one
for passing the signal and one for shorting it. This is crappy
because you'll never get a decent clean while there's a dirty
signal being generated right next door... this stuff runs
around the power supply and ground circuit, not to mention
that LDR's suck mightily as switches, they are very temperature
dependent and do drift noticeably. The LDR choice brought the
low impedance circuit choice... which brings the crappy tone
choice. Tubes are high-impedance things, mostly.

Bruce, enjoy it for what it is or sell it and get something
that sounds better. Or if you want to mod, get a kit or
something else... this amp is so crazy non-standard that
you can't touch one part without the rest of the stack
of cards falling. Get a Tiny Terror like I did and be done.
__
Steve
.

[Bruce Morgen]

Thanks, Steve -- much food
for thought there, including
your closing advice, which I
anticipated might be the best
"mod" available. I like the
cab (which isn't too small for
the speaker like my otherwise
*far* superior Pignose G40V's)
and physical layout of the
amp, but at this point
experimenting with its oddball
circuitry seems like a pretty
dicey use of my time.

[boardjunkie1]

Whatta shitbag design. Looks horrible on paper...I can't imagine what
it sounds like....

I'd gut it and start over.

[Bruce Morgen]

It's pretty danged poor
-- but you ain't heard
"horrible" until you
press that OD button,
simply gawdawful.

>
>I'd gut it and start over.

...or put in a cheap
Asian speaker and sell
it for a few $$.

[Bruno Puntz Jones]

On Sun, 4 Dec 2011 15:49:43 -0800 (PST), boardjunkie1
<boardj...@gmail.com> wrote:

>Whatta shitbag design. Looks horrible on paper...I can't imagine what
>it sounds like....

D3 & D4 are completely un-needed. Is R6 REALLY 120K ?

The whole 'drive' circuit is laughable too. Far too much for
what little it does. Someone mentioned D1, D2, D12, D13
as being there to protect the LDR mess. I wouldn't doubt it.

Attach sep resistors to pin 9 of V3, V4. 220 to 1k, 5wt.

Notice no NFB. Yet a AC30 this isn't. Also, R36 is only 1/2
watt, yet attached to it is R38, a 5 watt. R37 is unneeded.

R27 and R28 are too high, IMHO. What does this retail for?

>I'd gut it and start over.

I doubt if there is enough iron there worth bothering with.

Being named a 'ELECTAR TUBE 30' do they imply it's 30 watts ?

Maybe like the old stereo TOTAL PEAK RMS ratings of old...

I expected more from EPI.



JJTj




-------------------------------------------------------------
Don't go away MAD, here's a SASE for you to fill out telling
me how tasty my dust is, and an order blank for re-fills....

..ok....NOW you can go......
-------------------------------------------------------------

[Bruno Puntz Jones]

I've been getting many emails about this..

(2 many of them with postage due...HA HA HA..)

..and, I'm amazed about how many read this NG,
yet sit in the background. Goes 2 show ya never can tell..

Brian said he had one of these and gutted it, The iron seems
to be good enough to dabble with. It's now some FMIC clone
thing of a boob. He's happy. He uses it as a 3rd amp in waiting.

dogbreath666 (love the name) seems to like his as stock, adores
the drive crap, it's his practice amp. Still has stock tubes..

Mic has tweaked his. He's sending me a schematic...

Michelle has studied the schematic, and thinks it's the nightmare
some team of people who all had certain designs they jerked off to,
and wanted them in ANY product. There IS alot of design ideas from
old fart days (.com) products. She also said the stock iron is better
then ya'd think. In 40+ years of this biz, she is one of only (now) 8
women who know far too much about this shit. We're trading between
our collection of schematics/etc. Her husband sells new Corvettes...

JJTj




Let me ride on the Wall Of Death one more time
Let me ride on the Wall Of Death one more time
You can waste your time on the other rides
This is the nearest to being alive
Oh let me take my chances on the Wall Of Death

You can go with the crazy people in the Crooked House
You can fly away on the Rocket or spin in the Mouse
The Tunnel Of Love might amuse you
Noah's Ark might confuse you
But let me take my chances on the Wall Of Death

On the Wall Of Death all the world is far from me
On the Wall Of Death it's the nearest to being free

Well you're going nowhere when you ride on the carousel
And maybe you're strong but what's the good of ringing a bell
The switchback will make you crazy. Beware of the bearded lady
Oh let me take my chances on the Wall Of Death

Let me ride on the Wall Of Death one more time
Oh let me ride on the Wall Of Death one more time
You can waste your time on the other rides
This is the nearest to being alive
Oh let me take my chances on the Wall Of Death
Let me take my chances on the Wall Of Death
Oh let me take my chances on the Wall Of Death

[Bruce Morgen]

Thanks, Jay -- please share
what these folks have done
with theirs and I'll
eventually decide whether
to mod, gut/rebuild, or
just sell the thing. No
hurry here -- plenty of
other projects are ahead in
the queue. :-)

[RS]

On Sun, 04 Dec 2011 11:54:59 -0600, Stephen Cowell
<sco...@no.austin.spam.rr.com> wrote:

>On 12/3/2011 9:58 PM, RS wrote:

>> On Sat, 03 Dec 2011 15:22:38 -0500, Bruce Morgen<edi...@juno.com>:

>>> [Electar Tube 30]
>>> ... This amp lacks clarity and chime -

>> Starting from the input: The two 1n4148 diodes from the input to
>> ground are there to snub spikes. They'll conduct at around 0.65v, so
>> most low output guitars won't break over, but there's always the
>> possibility that peaks of higher output pickups will get clipped.
>> Either do without, or maybe put more diode pairs in series.

>OK, now my take: the diodes are there to set a known
>spike region to protect the crappy LDR's and clipping
>zeners coming later. Just the start of a crappy tone design, see
>below.

Steve, I really hope that this is not a continuation of your declared
vendetta on behalf of your friend Valve. In the past I've given you
credit for knowing more about electronics than he does, but that
doesn't explain some of what you've posted below--I know you
understand this stuff better than that. Still, in the event that
they're honest mistakes and not just another attempt to snipe,
I'll be nice. Maybe we can stick to scientific fact and keep the
thread informative.

First, LDR's: Everyone has seen these in Fender trem circuits from the
60's on. In Fender circuits, they are hooked directly inline with the
very high-signal 2nd preamp stage. So there's no reason to regard
LDR's as particularly fragile. Fender doesn't put diodes on their
inputs.

The zeners: As shown they will both forward bias, so the voltage
across them at any given time would be only 0.65 volts. There's no way
that will exceed either the voltage or wattage spec. That leaves
current: A lowly 1n4148 is rated for way more current than the tube
stage can source. They're safe.

On the chart, there is also a 1K resistor in parallel with the whole
thing which would bring those numbers down even further, but I'm
going to ignore that as yet another drafting error.

I've advised Bruce to bypass the ugly zener clipping circuit anyway,
so this could all be irrelevant. The distortion is raspy and there's
no reason to use it.

>> R5, the cathode resistor of the first stage, is way too low. This will
>> bias the tube in a slightly higher gain region but at the expense of
>> headroom, since the plate voltage will not have as much swing
>> available on the negative side (the tube is conducting more than it
>> should, so the voltage on the plate will be lower).
>
>The tube is working very hard (look at the load also, 2K!) and
>also generating even harmonics (one side different from other).
>There is sparkle lost in this process... any time V1A
>doesn't bypass the K resistor you're going to get a more Marshally
>tone. Sparkle is hard to do right, that's where the nasty
>parasitics live.

There are some very bad drafting mistakes on this schematic. That 2K
also has to be another one. I've already commented, but for the sake
of technical explanation:

Consider the full volume conditions--volume control full up. The
output impedance of the preceding tube stage is its plate resistance
(62k) in parallel with the plate load (100k). That's around 40k. So
even at full volume, that would be a 20x loss. The gain of a 12ax7
stage is only 60 when biased correctly (this one is not), so gain
would be under 3!??

Even worse, what happens when the 1M volume control is turned down
just a fraction: At just 90%, there's now 100k + the 40K in series
with the signal path. That's a 70x signal hit! IOW a very sharp
dropoff at the slightest rotation.

Conclusion: Gotta be a drafting error.

>> Increase that to
>> 1500 ohms or so, and bypass with a 10uf cap. Similar to the cathode of
>> V2a.
>
>This will totally fuck up the rest of the circuit design, believe me.

A rash pronouncement, and believe me, it's wrong. There is no reason
that the amp -can't- be transformed into a decent-sounding low-wattage
Fender-like amp. Changing the bad biasing on the input stage is a
very logical first step toward Bruce's goal (Fender, not Marshall).

>This is not a platform to start messing with (unless you rip it apart,
>not economical).

There's no $ investment aside from a few resistors and caps. And if
Jay's comments about the transformers are correct, this would be a
good return on time spent. I'd do it in a second.

>> If you don't like bright top-end, lift one side of C5. That's
>> effectively a hard-wired equivalent of Fender's bright switch.
>
>It's meant to compensate for the tone suck of the 1M Gain control.

It would radically overcompensate for 'tone suck'. As I mentioned,
it will boost the top end like a bright switch, but with a corner
frequency an octave higher. Not necessarily icepick, so not
imperative--hence my comment.

Miller capacitance of a 12ax7 is around 150pf, which does become a
factor for higher inline resistances, but consider that the 3.3M
resistor in Fender reverb amps is bypassed by only 10pf.

>> R7 is an error--it can't be 2K. See what that value actually is.
>
>The moment you allow high-impedance voltages to start flying around
>you start popping anything made of sand.

Not sure what you mean by 'high-impedance voltages.' If a given
voltage is sourced by a high impedance, it's less likely to cause
harm. See Van de Graf generator.

> This amp uses 2K loads
>(note 1K on dirty channel) for a reason.

Already commented on schematic typos above.

>> The V1b circuit is odd. with R12 left unconnected, it's a unity gain
>> phase splitter, but if they're getting switchable gain by connecting
>> the other side of R12 to ground, that would generate a loud pop.
>
>The zeners are clippers. R12 is always grounded in my print.

It is? The schematic at the link shows R12 going to the LDR module.
R101 above is 12K and it's grounded. Is that what you were thinking
of?

>> Without knowing exactly what's in the LDR module, part of this is just
>> guesswork. The strange thing is that the two back to back diodes are
>> diagrammed as zeners, and there's no call for them. They're back to
>> back, so the forward conduction voltage (0.65v) of the opposite diode
>> will come into play before the zener gets anywhere near the reverse
>> breakover voltage.
>
>The resistor R12 evens out the effect...

See comments about R12 above. It comes off the cathode. I'd have
to see a correct drawing to tell how much effect R101 actually has.
As drawn, it would overwhelm the effects of the diodes.

> note the 1K plate load for
>this side (the dirty channel). It's called asymmetrical clipping...
>more even harmonics.

Same comments about gain and R7 relate to R14. This whole circuit
is probably drawn incorrectly, as the zeners are not going to
work as zeners, etc.

>You can understand the LDR circuit better if you visualize that
>there are pass units (the phototransistor) across the top side
>and switch units (the LED/light source) across the bottom. What
>looks like device pins are actually jumpers between devices.
>I found a circuit diagram for one (Isocom 90980) that shows
>this layout. Jumpering the LED voltages like they do shows
>that one is off while the other is on (kathode jumped to anode).
>Pin 1 is bottom left in the diagram, by C5. It's a 16-pin DIP.
>
>http://www.farnell.com/datasheets/90980.pdf

The Epi circuit's chip says 4xLDR. That diagram is for an optoisolator
based on phototransistors. It's a very different animal from an LDR.

>> Also unusual to see those directly in front of the plate without any
>> resistance in between.
>
>They are protected by the low-impedance load and the clipping diodes
>back at the beginning. This is all crappy, BTW.

Well, I've already commented on the dissipation of the zeners.
There's no need to protect them. And yes, it's a bad idea anyway,
so best bridged out.

>> The short version: If you're not using
>> overdrive, lift the connection to the diodes (or one side of R101).
>> See if that signal cleans up.
>
>It will probably sound crappier. That's your basic problem here...
>most anything you do will make it sound crappier, and it's not an
>amp that will reward use of premium tubes. A basic waste of thermionic
>emission.

Disagree, but already addressed. Crappy drawing, and no doubt a bad
circuit. Low resale value. Supposedly decent transformers. All adds
up to: good mod platform.

>> The tone stack is configured to sound more Marshally than Fendery. If
>> that matters, I'll advise on changes. You could easily adapt that.
>
>Unequivocally not worth it. These things go well on CL, as in gone.

Again, addressed. I've seen them for about $200 at GC. But all
depends on whether Bruce wants to solder or not.

>> The bottom side of the volume pot should go to ground. Another error
>> on the diagram.
>
>That would not be wise if the intent (as drawn) is as a series
>resistor. The argument could be made that the bottom of the
>pot should be connected to the wiper, another matter entirely.

??? You certainly have to know that this is not correct. A volume
control works as a divider. Lift ground...no divider. The only
effect of the series resistance would be a few db, working against
R20.

>> The diagram looks like R21 and R22 are in parallel, but there's no
>> need for that. No big deal. Leave it.
>
>In order to balance the output they probably used R21 to trim the
>68K kathode resistor down some to make up for the extra 1.8K/1M
>self-bias network (what's 2.1M || 68K? 65.867K.. +1.8 = 67.667K).

It would make sense if those were not 5% tolerance components.
That's within that 5%. I doubt there was any strategy there.

>Really stupid considering all the attention to generating even
>harmonics done earlier in the circuit (with no inductor to sweeten
>the sauce). Note the difference (a factor of 100) between the two
>phase coupling caps. I'd be very surprised if this was really
>the case.

I'm going with "Yet another drafting error."

>> R6 is wrong. That must be 120 ohms (?), and it should be decent
>> wattage.
>
>RS is correct, this is an error. I'd like to see a 120K 1/4 watt here.

Finally agreed on something?

>> The grid stopper (parasitic suppression) resistors R27 and R28 are
>> higher than normal, but that's not a bad thing.
>
>Say goodbye to sparkle and tone. Resistors are not tone's
>friend... any time a resistor appears in a circuit the opportunity
>to have the signal response spectrum skewed appears.

Nope. There's no significant Miller Effect in a pentode. The screen
grid isolates the plate, and it's only the grid to plate capacitance
that gets amplified by the tube. I don't use 47k's there, but there's
no drastic tone loss. There should be far more concern about
the rolloff from the 68k input resistors, or maybe Fender's use
of inline 220k resistors as their channel mixing net. In both
of those cases, Miller capacitance is indeed a factor.

I'm getting near my typing limit, so I'll refer you to Randall
Aiken's site: http://www.aikenamps.com/InputRes.htm
Randy is a very competent tube amp designer. That doc covers grid
stopper resistors. The information is correct.

The other interesting side-benefit of higher resistor values in that
spot is that they help to suppress grid blocking.

The main concern is actually the rating for "max DC resistance
from grid to ground", but in this case, it's in spec.

>(snip bias lecture)

OK then. Done for now.

[RS]

On Sun, 04 Dec 2011 15:33:34 -0500, Bruce Morgen <edi...@juno.com>
wrote:

>Thanks, Steve -- much food
>for thought there, including
>your closing advice, which I
>anticipated might be the best
>"mod" available. I like the
>cab (which isn't too small for
>the speaker like my otherwise
>*far* superior Pignose G40V's)
>and physical layout of the
>amp, but at this point
>experimenting with its oddball
>circuitry seems like a pretty
>dicey use of my time.

Hi Bruce,

I've already commented on the technical side. If you bypass the
overdrive circuit as I recommended, the circuit becomes pretty direct.
I think it all depends on whether you want to solder or not. The best
way to learn tubes is to try your hand at circuitry. I had an amp
chassis on my bench from the time I was 14--prototyped thousands of
circuit variations until I knew what to expect from each component
change. That's how I learned. I figured that's what you were after.
You'd probably get good very fast.

I just started another project (a book, more math than music this
time), so I've been a bit burned out on typing the past few days.
But if you decide to modify the amp, I'll check in and make sure
you're on track. Can't guarantee that you won't get a solder burn or
two, but it will sound good.

[Bruce Morgen]

VERY educational -- getting
close to the critical mass
of info needed for this
decision. Perfectly willing
to solder, do it frequently
on guitar innards -- don't
relish disassembly/reasembly
(and testing after each mod)
tedium of such projects, but
I remember enjoying my old
Gibby Explorer 1x10 combo
with EL84s and a JBL very
much, so this one might be
worth running the gauntlet.

You know, sometimes these
arguments actually serve to
clarify the technical issues
because they make folks
attempt to make their point
with different (and often
more effective) wording.
Trying to reach somebody
you're confident is wrong
can wind up being a de facto
inspiration even if it comes
out of exasperation. :-)

[RS]

On Sun, 04 Dec 2011 20:55:38 -0500, Bruno Puntz Jones <Somewhere in
yer dreams.com> wrote:

>
>On Sun, 4 Dec 2011 15:49:43 -0800 (PST), boardjunkie1
><boardj...@gmail.com> wrote:
>
>>Whatta shitbag design. Looks horrible on paper...I can't imagine what
>>it sounds like....
>
>D3 & D4 are completely un-needed.

Those are spike snubbers. Errr... optional.

> Is R6 REALLY 120K ?

Nope.

>The whole 'drive' circuit is laughable too. Far too much for
>what little it does.

Yep.

> Someone mentioned D1, D2, D12, D13
>as being there to protect the LDR mess. I wouldn't doubt it.

D1 and D2 are also spike snubbers, but not many amps have used those.
Fender uses LDRs...no suppressors. Tons of other amps have solid state
circuits. Again, best put more diodes in series to make sure hot
pickups don't break over, but other than that, no harm done.

>Attach sep resistors to pin 9 of V3, V4. 220 to 1k, 5wt.

There ya go. Bypassed by large caps.

>Notice no NFB. Yet a AC30 this isn't.

No harm though. NFB can actually -cause- oscillation if it's not done
right.

>Also, R36 is only 1/2
>watt, yet attached to it is R38, a 5 watt.

With a lower value resistance you don't need as much wattage. It's E
squared / R, or I squared * R. (That's a letter "I"). Assuming
same current through both:

(I squared) would have to exceed 0.10 to put R38 over rated
dissipation. At that current, (I squared) = 0.10, the 0.5 ohm would
be within rated dissipation. With a normal short, R38 would probably
burn up first.

> R37 is unneeded.

Again no harm. Voltage drainage path if you're working on the amp.

>R27 and R28 are too high, IMHO.

Addressed in another post. Again, no harm. Possibly good when the amp
is driven very hard--not for anything relating to dissipation; for an
effect known as 'grid blocking' (see Norm Crowhurst).

>>I'd gut it and start over.
>
>I doubt if there is enough iron there worth bothering with.

That's a significant factor, but in another post you said the iron was
good.

>Being named a 'ELECTAR TUBE 30' do they imply it's 30 watts ?

Ahem...

>Maybe like the old stereo TOTAL PEAK RMS ratings of old...

I think the best one was "Peak Music Power". Usually about 1 zillion
imaginary watts.

>I expected more from EPI.

I'd have to see it drawn correctly, but the zeners are not a good
indication even if they're hooked up right.

I've worked with a lot of draftsmen. Sometimes drawings are done by
the design engineer, sometimes it's a hired draftsman who usually
doesn't know electronics (else they'd be an engineer), sometimes one
of the techs gets lassoed. In the two latter cases, there will be
errors.

I once worked for a keyboard manufacturer who bought another company
and the design crew spent the afternoon reviewing the hilariously
sillly schematics we inherited. A couple got pinned to the door. One
of those had stuff like B+ hooked to ground. There was a signal line
that went around the outside of the whole page, then came back and
connected to where it started (a schematic ground loop!). The Electar
chart is bad though.

[RS]

On Tue, 06 Dec 2011 23:25:10 -0500, Bruce Morgen <edi...@juno.com>
wrote:

>VERY educational -- getting
>close to the critical mass
>of info needed for this
>decision. Perfectly willing
>to solder, do it frequently
>on guitar innards -- don't
>relish disassembly/reasembly
>(and testing after each mod)
>tedium of such projects, but
>I remember enjoying my old
>Gibby Explorer 1x10 combo
>with EL84s and a JBL very
>much, so this one might be
>worth running the gauntlet.

Hi Bruce. I think one of the main questions is whether you'd like to
do this for fun. It's the best educational experience. And rewarding
once you get it sounding good (and it will). But it takes some
patience. You may have to trace through the component layout for some
sections, given the poor schematic.

You probably won't need to reassemble anything in order to test it.
Just leave it open on the bench. And remember not to touch the guitar
strings and the power supply :-) Or even let a dangling string end
get into the amp (I believe that's one of the few things I've never
done!)

Do you have a meter? A scope?

>You know, sometimes these
>arguments actually serve to
>clarify the technical issues
>because they make folks
>attempt to make their point
>with different (and often
>more effective) wording.
>Trying to reach somebody
>you're confident is wrong
>can wind up being a de facto
>inspiration even if it comes
>out of exasperation. :-)

In the early days of aga, there were some great tech discussions, and
despite some divergent opinions and design philosophies, they usually
stayed on a courteous practical level. Hopefully this one won't run
off into the weeds.

In any event, you can probably tell that I've done this kind of thing
many times (since I was a kid, actually) and I've worked for some of
the major instrument and FX manufacturers. If you stick with it, I'll
do my best from here.

It might be good to find out the real values of the mystery
components, but that could also be done on the fly as you're modding a
given section. Still, I'm curious about that. I bet there's an
accurate schematic somewhere that's not going to have 2k resistors
loading tube plates.

[It's that Guy again...]

On Tue, 06 Dec 2011 23:36:12 -0500, RS <R...@sorrynospam.com> wrote:

>>Maybe like the old stereo TOTAL PEAK RMS ratings of old...
>
>I think the best one was "Peak Music Power". Usually about 1 zillion
>imaginary watts.

I remember selling Pioneer receivers that were, ahem, 300 watts
"..total peak RMS music power.." yet in real terms, were only
75 true per channel. I think, IMMSMW, they or someone made
(advertised) a 1000 watt receiver under those same terms.

BOY HOWDY, kids in dorms wanted them so bad, but they burnt up
speakers for lunch, so of course we had to sell them those home
Cerwin-Vega spkrs, that would not die but sounded like shit.

One day, we decided to hold a "..trade in anything.." sale, where
we would take anything in trade and give ya double it's cost off
the LIST price of anything in stock. Cases of beer, about 50
pizza's (which we gave to a homeless shelter..the pizza, not the beer)
weed, one guy offered his wife..we turned her down.. The sale
was a overnight sale, and about 2am, we were all pretty drunk,
so we cranked up the Pioneer monster into a pair of Altec 5's.

The cones caught fire, we dragged it onto main street, and doused
it out with beer. Months later, some Gov office made all these brands
use real terms in power advertising. AH, the old Atlantis Sound days.


JJTj



I know you know what you know
but you should know by now that you're not me
Talk about a month of Sundays
Toffee nosed wet weekend as far as I can see

Hey diddle diddle .. the cat and the fiddle
Piggy in the middle
Doo-a-poo-poo

Bible punching heavyweight evangelistic boxing kangaroo
Orangutang and anaconda
Donald Duck and Mickey Mouse...even Pluto too

Hey diddle diddle .. the cat and the fiddle
Piggy in the middle
Doo-a-poo-poo

(Stig is dead, honestly)

One man's civilization is another man's jungle, yeah
They say revolution's in the air
I'm dancing in my underwear...'cause I don't care

Hey diddle diddle .. the cat and the fiddle
Piggy in the middle
Doo-a-poo-poo....Doo-a-poo-poo

Walky talky man says hello, hello, hello
with his ballerina boots
you can tell he's always on his toes
Hanging on a Christmas tree
Screaming like a bogey man getting up my nose

Hey diddle diddle .. the cat and the fiddle
Piggy in the middle
Doo-a-poo-poo...a-poo-poo-poo

This little piggy went to market
This little piggy stayed home
This little piggy had roast beef
and this little piggy had none
This little piggy went.....Woooo!
All the way home

This little piggy went to market
This little piggy stayed home
This little piggy had roast beef
and this little piggy had none
(little piggy had..)
This little piggy went
Woooo! Wooo . . .
All the way home

[Bruce Morgen]

RS <R...@sorrynospam.com> wrote:

>On Tue, 06 Dec 2011 23:25:10 -0500, Bruce Morgen <edi...@juno.com>
>wrote:
>
>>VERY educational -- getting
>>close to the critical mass
>>of info needed for this
>>decision. Perfectly willing
>>to solder, do it frequently
>>on guitar innards -- don't
>>relish disassembly/reasembly
>>(and testing after each mod)
>>tedium of such projects, but
>>I remember enjoying my old
>>Gibby Explorer 1x10 combo
>>with EL84s and a JBL very
>>much, so this one might be
>>worth running the gauntlet.
>
>Hi Bruce. I think one of the main questions is whether you'd like to
>do this for fun. It's the best educational experience. And rewarding
>once you get it sounding good (and it will).

Your confidence certainly
help. Just to be clear,
I have a great deal of
experience with tube
electronics -- built my
share of Heath, Eico, and
Knight kits back in the
day and managed a repair
ship with 5-8 techs
through most of the 1970s.
Mods are a different story
from my P.O.V. -- they
require understanding
circuit dynamics in much
more detail than repairs,
and I never learned enough
about active component
behavior to rescue a
working-but-deficient amp.

I did do a trivial mod or
two to my beloved little
Piggy. I added a Fender-
style series resistor to
the input and put in a DC
filament supply for the
first-hole preamp tube in
an attempt to minimize
hum. The latter really
didn't do very much at all
-- but I left it in since
it did no harm either.

>But it takes some
>patience. You may have to trace through the component layout for some
>sections, given the poor schematic.

That I can do. Given
enough light and reading
glasses, I can even read
resistor color codes! :-)

>
>You probably won't need to reassemble anything in order to test it.
>Just leave it open on the bench. And remember not to touch the guitar
>strings and the power supply :-) Or even let a dangling string end
>get into the amp (I believe that's one of the few things I've never
>done!)

I know the safety drills
well, having supervised
techs who were fond of
shortcuts.

>
>Do you have a meter?

Yup.

>A scope?

Nope.

>
>>You know, sometimes these
>>arguments actually serve to
>>clarify the technical issues
>>because they make folks
>>attempt to make their point
>>with different (and often
>>more effective) wording.
>>Trying to reach somebody
>>you're confident is wrong
>>can wind up being a de facto
>>inspiration even if it comes
>>out of exasperation. :-)
>
>In the early days of aga, there were some great tech discussions, and
>despite some divergent opinions and design philosophies, they usually
>stayed on a courteous practical level. Hopefully this one
>won't run off into the weeds.

Sure would be nice if
that's the way it goes.

>
>In any event, you can probably tell that I've done this kind of thing
>many times (since I was a kid, actually) and I've worked for some of
>the major instrument and FX manufacturers. If you stick with it, I'll
>do my best from here.

I appreciate the support.
I've got the basic physical
and troubleshooting skills,
but I'm nobody's circuit
designer, especially when
active components are
involved.

>
>It might be good to find out the real values of the mystery
>components, but that could also be done on the fly as
>you're modding a given section.

Yup, color codes and/or
actual ohmmeter readings
are your friends. :-)

>Still, I'm curious about that. I bet there's an
>accurate schematic somewhere that's not going to have 2k
>resistors loading tube plates.

Probably worth some further
Googling -- it'll be some
weeks before I attempt this
and I'll see what I can
find in the meantime.

Thanks for everything!

[Bruno Puntz Jones]

On Wed, 07 Dec 2011 02:20:16 -0500, RS <R...@sorrynospam.com> wrote:

>In the early days of aga, there were some great tech discussions, and
>despite some divergent opinions and design philosophies, they usually
>stayed on a courteous practical level. Hopefully this one won't run
>off into the weeds.

I know I've been guilty of shiet, but yes, I came here for the amp
content. I've saved decades of all posts, and have learned big time.

>In any event, you can probably tell that I've done this kind of thing
>many times (since I was a kid, actually) and I've worked for some of
>the major instrument and FX manufacturers. If you stick with it, I'll
>do my best from here.

I got into it young too, about 14-15. Back then it was easy to enter
kayfabe. What I did, was if I had interest in a product, I would
call them. UK calls back then were $$$$$. If someone from the US
called the UK about a amp, you were listened to. They'd send ya
all the catalogs/schematics/whatever ya asked for. US companies too.

I was a Univox dealer at 17. 'MF' music was thought to mean as
'mother fucker' music, but in reality, it was 'mother fiananced', as
my family backed up my 'college degree'. I was the GMT east coast
(for lack of a better badge, no one else cared) rep at 19. Being one
got me to meet Zappa, the Who,Genesis and tons of others.

Working in Marlboro MA for 'VOX/Pr1mo' got me to meet U2.

Today, ya don't have the chance. Yet Ian Hunter was playing
nearby last month. He asked me to show up at one show, but
as it happens, names get mis-placed. I still had it, I talked my
way in. Ya are kayfabe or you are not, I guess..

>It might be good to find out the real values of the mystery
>components, but that could also be done on the fly as you're modding a
>given section. Still, I'm curious about that. I bet there's an
>accurate schematic somewhere that's not going to have 2k resistors
>loading tube plates.

One hopes so, as the 129k EK84 cath resistor could not fly.

I talked to a friend who works for Gibson/Epi, and he said it's
the schematic they have. He's digging deeper. Seems they
can't stock enough of them, and yes, they claim it at 30 watts.

He laughed when he told me. Seems t's a joke down in Tenn.


JJTj







God Bless us, because we is funky...
Making our own rules, no matter the price..
and you best believe...we NEVER think twice...

[RS]

On Wed, 07 Dec 2011 17:04:34 -0500, Bruno Puntz Jones <Somewhere in
yer dreams.com> wrote:

>On Wed, 07 Dec 2011 02:20:16 -0500, RS <R...@sorrynospam.com> wrote:
>
>>In the early days of aga, there were some great tech discussions, and
>>despite some divergent opinions and design philosophies, they usually
>>stayed on a courteous practical level. Hopefully this one won't run
>>off into the weeds.
>
>I know I've been guilty of shiet, but yes, I came here for the amp
>content. I've saved decades of all posts, and have learned big time.

Someone once argued that the off-topic and flaming crap that started
about 15 years ago didn't matter cause all the tech topics were
already covered in depth by the first generation.

Back then, it did seem like there were discussions on just about
anything and everything to do with tube amps, and often at very deep
technical level. But most of the great stuff is not archived by Google
(It was "DejaNews" back then, bought by Google later and presumably
purged of old content).

>>In any event, you can probably tell that I've done this kind of thing
>>many times (since I was a kid, actually) and I've worked for some of
>>the major instrument and FX manufacturers. If you stick with it, I'll
>>do my best from here.

>I was a Univox dealer at 17. 'MF' music was thought to mean as
>'mother fucker' music, but in reality, it was 'mother fiananced', as
>my family backed up my 'college degree'. I was the GMT east coast
>(for lack of a better badge, no one else cared) rep at 19. Being one
>got me to meet Zappa, the Who,Genesis and tons of others.
>
>Working in Marlboro MA for 'VOX/Pr1mo' got me to meet U2.

I was still in school then, but started designing instruments for
commercial manufacturers when I was pretty young. Some top name bands
were using them (mostly keyboards and synths). Ran custom design shops
in a few locations in big cities for a while, and there you meet tons
of good musicians and bands. But I was never real star-struck. Just
into the technology and what could be done next. Some of the guys were
cool and fun to hang out with, but that's all that registered at the
time.

>>It might be good to find out the real values of the mystery
>>components, but that could also be done on the fly as you're modding a
>>given section. Still, I'm curious about that. I bet there's an
>>accurate schematic somewhere that's not going to have 2k resistors
>>loading tube plates.
>
>One hopes so, as the 129k EK84 cath resistor could not fly.

You'd think they would have had an engineer check the schematic. A lot
of those are sold by Music-Yo, and they have the same ugly chart.
Gotta wonder how they layed out the circuit board.

I once worked for a top studio FX manufacturers whose president and
main designer would do eccentric stuff (he had an image to uphold)
like going straight to PC layout. (Won't be mentioned here, cause you
know who it is). He'd actually hand a PC layout to a draftsman and
tell him to draw the schematic.

Plenty weird, but it would appear that they did something like that
here. How else would the schematic be so far from workable reality.

>I talked to a friend who works for Gibson/Epi, and he said it's
>the schematic they have. He's digging deeper. Seems they
>can't stock enough of them, and yes, they claim it at 30 watts.
>
>He laughed when he told me. Seems t's a joke down in Tenn.

I guess you could name the amp anything that you want. Electar 300.
If it's got good transformers (you did say that, right?) it could
sound pretty loud. I've been around where UTC used to operate, and
I've bought a bunch of old unused surplus UTC transformers. Put 'em in
a guitar amp and the thing really lights up. Almost too much
sometimes.

[RS]

On Wed, 07 Dec 2011 11:37:28 -0500, Bruce Morgen <edi...@juno.com>
wrote:

> RS Wrote:
>>Hi Bruce. I think one of the main questions is whether you'd like to
>>do this for fun. It's the best educational experience. And rewarding
>>once you get it sounding good (and it will).
>
>Your confidence certainly
>help.

Approach anythingj with a defeatist "Can't do it" attitude and you're
already beaten. No point in false confidence of course, but this is
very doable.

I did a quick search today to try to find a better schematic. I
didn't, but I found tons of comments about people buying that amp for
cheap ($50-$100) and modding it. Most did use the typical Fender
first stage, BTW. Some took out the whole board and started from
scratch. But of course that's not what I was suggesting. The changes
can be done incrementally so you can localize and problems if they
occur.

> Just to be clear,
>I have a great deal of
>experience with tube
>electronics -- built my
>share of Heath, Eico, and
>Knight kits back in the
>day and managed a repair
>ship with 5-8 techs
>through most of the 1970s.

I figured the former, knew the latter. You'd have a good aptitude for
design if you wanted to do that.

>Mods are a different story
>from my P.O.V. -- they
>require understanding
>circuit dynamics in much
>more detail than repairs,

True enough, but it just takes logic and determination. You'll get
that if you want to.

>>But it takes some
>>patience. You may have to trace through the component layout for some
>>sections, given the poor schematic.
>
>That I can do. Given
>enough light and reading
>glasses, I can even read
>resistor color codes! :-)

I was once hired on a consulting project to compact a given circuit
board for a demoable prototype, and I had to do it before the NAMM
show--a few weeks. Had to chop the existing board, cutting traces
with an XActo, essentially folding the board in half (luckily only two
layer), then I worked fanatically for days, soldering tons of digital
traces to ribbon connectors to link the two halves. My eyes are good,
but at the end of that they felt like they were jittering. Weirdest
thing. For a couple days, I thought my eyes were permanently screwed
up.

Also had to design and perf-board the rest of the circuits, but it was
a piece of cake after the first part.

>>You probably won't need to reassemble anything in order to test it.
>>Just leave it open on the bench. And remember not to touch the guitar
>>strings and the power supply :-) Or even let a dangling string end
>>get into the amp (I believe that's one of the few things I've never
>>done!)

>I know the safety drills
>well, having supervised
>techs who were fond of
>shortcuts.

For the sake of any possible reputation I may have, I won't tell you
some of the stupid things that I've done. You deal with the stuff,
the volts will eventually find you. The best rule: Make sure there's
someone in the room with you who knows what to do.

>>Do you have a meter?
>
>Yup.
>
>>A scope?
>
>Nope.

Meter will probably do for now. Scope later if you want to pursue it
further.

>>In the early days of aga, there were some great tech discussions, and
>>despite some divergent opinions and design philosophies, they usually
>>stayed on a courteous practical level. Hopefully this one
>>won't run off into the weeds.
>
>Sure would be nice if
>that's the way it goes.
>>
>>In any event, you can probably tell that I've done this kind of thing
>>many times (since I was a kid, actually) and I've worked for some of
>>the major instrument and FX manufacturers. If you stick with it, I'll
>>do my best from here.
>
>I appreciate the support.
>I've got the basic physical
>and troubleshooting skills,
>but I'm nobody's circuit
>designer, especially when
>active components are
>involved.

You'll do fine. Make sure you want to do it first. From what I've
seen, the resale value of that amp won't be very high, but maybe you
want to check that as well. Still, it's tough to find a proto
platform with good iron for cheap. I don't know if you'd be able to
find a kit or parts for what you'd get for the amp.

>it'll be some weeks before
> I attempt this
>and I'll see what I can
>find in the meantime.
>
>Thanks for everything!

Sure thing. I'll be here.

[Bruce Morgen]

The most impressive iron I
ever personally encountered
was in a huge McIntosh tube
power amp I sold to a Long
Island club called "My
Father's Place" when I was
selling audio for Arrow
Electronics in the early
1970s -- iirc the amp put
out an honest 600 RMS watts
and could do that all night
with no problem (it had two
big cooling fans). The
(eight!) output tubes were
TV horizontal output tubes
with plate caps up top
because the B+ was up too
high for any available
audio output tube to handle
due to arcing risk.

The shop I managed later
fixed one for Jefferson
Airplane -- it wasn't for
house sound, Jack Casady
used it in his bass rig! I
later took a cue from him
and used an old Heathkit
preamp and Williamson power
amp (with real USA Tung-Sol
5881s) for bass in my mid-
70s acoustic trio -- our
bassist (and lead singer)
strung his old P Bass with
black nylon wrapped Fender
strings, and with that amp
and a 15" Utah you'd swear
he played an upright bass!
Heathkit was another outfit
that used excellent iron in
their hi-fi stuff -- that
output transformer was at
least twice the weight of
what Leo used in the Tweed
Bassman and the filter choke
(the amp also powered the
preamp via an octal socket
and cable) was nearly as big.

[Bruno Puntz Jones]

On Wed, 07 Dec 2011 18:13:04 -0500, RS <R...@nospam.com> wrote:

>Someone once argued that the off-topic and flaming crap that started
>about 15 years ago didn't matter cause all the tech topics were
>already covered in depth by the first generation.

I just always hoped someone would see something new, I would
send schematics to anyone, etc. Then the shit started.

>Back then, it did seem like there were discussions on just about
>anything and everything to do with tube amps, and often at very deep
>technical level.

Which may/may not of scared some folks off. In my early years, I had
*NO* (repeat:* NO*) tech know how. I'd guess now, most have little
if any. Today people attach gear to musicians. I never did. Really.

Best I got back then, was being able to draft 1:1 dual sided PCBs
with templates. Blew people's mind for some reason. P'O/cake..

I remember the day I got hooked. A local band was playing around
the corner from me. LOUD. I walked up, crashed the party, and saw
this 8-10 Traynor making more sound then I ever heard. The Band's
name was 'Proof'. Long gone. I was 12 or so. Got religion..

Gear also was cheaper then, and alot of folks traded gear for amp
repairs. One guy, had a 100 wt Marshall (late 60's) he attached white
fur letters of his name. Paul something. EVERY week he wanted me to
tweak his amp. I butchered it. I glued parts in, made it more of a
VOX/Sunn type of beast he loved, every week, for over 5 yrs he'd
hand me this amp...and paid me decent $ to do it. Don't know where
he is today, but may God Bless and keep him always. A good player.

>.. But most of the great stuff is not archived by Google

>(It was "DejaNews" back then, bought by Google later and presumably
>purged of old content).

I remember DejaNews. Wondered what happened to them.

>I was still in school then, but started designing instruments for
>commercial manufacturers when I was pretty young. Some top name bands
>were using them (mostly keyboards and synths). Ran custom design shops
>in a few locations in big cities for a while, and there you meet tons
>of good musicians and bands.

See, you got lucky. You were smart enough to learn the craft, and
make it worked. In my days, there were FEW open jobs for tube amp
designers. I was a well paid troubleshooter on demand..

>...But I was never real star-struck.

Looking back in my life, I too couldn't of cared less who the band
was. Really. I never heard of the band Genesis till one MR came
into the music store I worked at to rent a bass amp. He was playing
the 'Orph' in Boston that night, and my boss "Vito" asked me if I'd
go with it. Sure. I've told the tale before, won't again (and yes,
folks, the world is still turning..yuk yuk). I hated (still do) U2
yet still fixed da amp. Other bands were just too cool to not be
amazed at. But it was always the amps, and maybe that was
what got me in the door. There for a reason. Kayfabe..

>...Just

>into the technology and what could be done next. Some of the guys were
>cool and fun to hang out with, but that's all that registered at the
>time.

90% of the time, it was all work for me. The time off work was just
not my red wagon. I noticed the 'guys' treated ya with more respect
if ya were into the gear..kayfabe..not just some rube wanting a A/G.

>You'd think they would have had an engineer check the schematic. A lot
>of those are sold by Music-Yo, and they have the same ugly chart.
>Gotta wonder how they layed out the circuit board.

At PV, before the schematic reached the PCB folks, who REALY !!
know their shit, they caught many errors, from pins to part values.

>I once worked for a top studio FX manufacturers whose president and
>main designer would do eccentric stuff (he had an image to uphold)
>like going straight to PC layout. (Won't be mentioned here, cause you
>know who it is). He'd actually hand a PC layout to a draftsman and
>tell him to draw the schematic.

HP. He'd walk into the building. He'd drive his DeLorean up the
side, produce some scrap of paper, want a p/t in hours. With
HP,at least, he KNOWS his shit tech wise, so it wasn't like RZ
of Mighty Mity who made up dreams he hoped would be cool.

BTW..Those old Mighty Mite P/Us were wound at the Camarillo
CA mental hospital Larry 'Wild Man' Fischer had stayed at...

I don't know the time frames.. RIP Wild Man...

>Plenty weird, but it would appear that they did something like that
>here. How else would the schematic be so far from workable reality.

Simple. Today they import boxes. If enough people bitch, they
correct it. If not, who flocking cares. Later it's damage control.

>I guess you could name the amp anything that you want. Electar 300.

Yet face it, call ya amp a '30' it implies 30 watts.

>If it's got good transformers (you did say that, right?)

So I've been told...

>..... it could

>sound pretty loud. I've been around where UTC used to operate, and
>I've bought a bunch of old unused surplus UTC transformers. Put 'em in
>a guitar amp and the thing really lights up. Almost too much
>sometimes.

Those tranz were overkill. Like Hammond's of today's. The old
Marshall's were not the same as the newer Drake equipped ones.

Hiwatts with Partridge tranz were over kill but made the Hiwatt sound.

JJTj



I have always thought in the back of my mind

"Cheese and Onions".

I have always thought that the world was unkind

"Cheese and Onions".

Do I have to spell it out ?

"C-H-E-E-S-E A-N-D O-N-I-O-N-S"

oh ho.

Man or machine (Man or machine)
Keep yourself clean (Keep yourself clean)
Or be a has-been (Ah-ah)
Like a Dinosaur, oh oh-oh.


Man of advise (Man of advise)
For ev'rything nice (Ev'rything nice)
You'd better think twice (Ah-ah)
At least once more, oh oh.

[Bruno Puntz Jones]

On Wed, 07 Dec 2011 19:40:08 -0500, Bruce Morgen <edi...@juno.com>
wrote:

>The most impressive iron I
>ever personally encountered
>was in a huge McIntosh tube
>power amp I sold to a Long
>Island club called "My
>Father's Place" when I was
>selling audio for Arrow
>Electronics in the early
>1970s -- iirc the amp put
>out an honest 600 RMS watts
>and could do that all night
>with no problem (it had two
>big cooling fans). The
>(eight!) output tubes were
>TV horizontal output tubes
>with plate caps up top
>because the B+ was up too
>high for any available
>audio output tube to handle
>due to arcing risk.

McIntosh always made great iron. Even the SS
state stuff had nasty overkill iron. The Dead toured
with those SS rack amps. they could not die. At the
Watkin's Glen moment, they were about 20 miles away
from B/NY where McIntosh was. I toured the place.

And going there got me backstage at the nightmare, where
it was warm and dry, and everyone had weed and beer.

>The shop I managed later
>fixed one for Jefferson
>Airplane -- it wasn't for
>house sound, Jack Casady
>used it in his bass rig! I
>later took a cue from him
>and used an old Heathkit
>preamp and Williamson power
>amp (with real USA Tung-Sol
>5881s) for bass in my mid-
>70s acoustic trio

The JA and Dead were around the Alembic folks, who made
those kool stackable cabs powered by them dar McIntosh
SS rack mount amps. JBLs and EV's. In Keene NH, they
did a show with LA/HT/JS (great package job there) where
the local radio station (WAAF) bused us to the show n back.

Standing behind 2 Twins with Altecs, JK shut them off as I
mentioned the spkrs. With a guitar around his neck, sweating
big time, he mentioned how he found the combo of the two.

Later the 'Bagend' folks tried to do the same stackable
Alembic thing, yet the times, they were a changing.

>Heathkit was another outfit
>that used excellent iron in
>their hi-fi stuff -- that
>output transformer was at
>least twice the weight of
>what Leo used in the Tweed
>Bassman and the filter choke
>(the amp also powered the
>preamp via an octal socket
>and cable) was nearly as big.

They were smart. They KNEW it had to work in the den of the
guy smoking a pipe, s/iron in hand. Good iron hides mistakes.

Anyone ever play the SS Heathkit combo? Really, it was amazing.

Face it, it was alot more fun back then...


JJTj




Charva, I loved you
I loved you through and through
I loved you since in grammar school
When we were sniffing glue
I loved you purty baby doll
And I don't know what in the world to do about it
Boppa-bah-boppa-choo-wah

Charva, my darling
The only love I had
I hope you will forgive me, dear
For punching out your dad
I loved you, I loved you
And I don't know what in the world I'm gonna do about it
Ooohh-oh-oooh
Lahm-buh-buhm-buhm

I remember, remember the junior prom
And I remember the time I broke your father's arm
And I remember, remember all the love we shared
Every place and everywhere

Oh Charva, Charva
I love you more and more
I swear it ain't because your father owns a liquor store
Charva, my baby
I love you and I don't know what to do about it
Oh-oh Charva

Ooooohh oh-oh Charva
Ooooohh, come back my little darling Charva

I love you so much, honey
Come back to me, Charva
Please Charva, please come back to me
I miss you so much

[RS]

On Wed, 07 Dec 2011 19:40:08 -0500, Bruce Morgen <edi...@juno.com>
wrote:

>The most impressive iron I
>ever personally encountered
>was in a huge McIntosh tube
>power amp I sold to a Long
>Island club called "My
>Father's Place" when I was
>selling audio for Arrow
>Electronics in the early
>1970s --

I know of "My Father's Place", famous club! And of course I've dealt
with Arrow for sourcing. Didn't know you worked for them.

>iirc the amp put
>out an honest 600 RMS watts
>and could do that all night
>with no problem (it had two
>big cooling fans).

No doubt rugged. The Mc's were terrific amps. There are actually some
good design ideas in there for guitar amps, like their output tube
driver circuits.

> The
>(eight!) output tubes were
>TV horizontal output tubes
>with plate caps up top
>because the B+ was up too
>high for any available
>audio output tube to handle
>due to arcing risk.

Yeah, those caps are dangerous though. A lot of the high-end audio
guys use big transmitter tubes like 211's and 845's:
http://www.antiqueradios.com/gallery/d/127031-2/845+tube.jpg

Beautiful looking tubes, and no plate cap. But they do run at very
high voltages. A friend of mine ended up wearing a sling for a week
cause he touched the envelope of one of his 211's and it evidently
arc'ed over into his hand. Paralyzed his arm for a few days.

Some cutting amps for old vinyl pressing plants also used huge
rack-mount 211 and 845 rigs to drive the cutting heads for the initial
master. Fairchild and others made those.

>The shop I managed later
>fixed one for Jefferson
>Airplane -- it wasn't for
>house sound, Jack Casady
>used it in his bass rig! I
>later took a cue from him
>and used an old Heathkit
>preamp and Williamson power
>amp (with real USA Tung-Sol
>5881s) for bass in my mid-
>70s acoustic trio -- our
>bassist (and lead singer)
>strung his old P Bass with
>black nylon wrapped Fender
>strings, and with that amp
>and a 15" Utah you'd swear
>he played an upright bass!

Those strings do sound very acoustic-like. Not a great idea for
guitar, but I could see how they may have been perfect for jazz bass.
That was probably a great sounding rig.

>Heathkit was another outfit
>that used excellent iron in
>their hi-fi stuff -- that
>output transformer was at
>least twice the weight of
>what Leo used in the Tweed
>Bassman and the filter choke
>(the amp also powered the
>preamp via an octal socket
>and cable) was nearly as big.

And those were regarded as 'consumer' kits back then. Those days will
never be back. Most of Heath's circuits were designed pretty well too.

[RS]

On Wed, 07 Dec 2011 20:27:03 -0500, Bruno Puntz Jones <Somewhere in
yer dreams.com> wrote:

>On Wed, 07 Dec 2011 18:13:04 -0500, RS <R...@nospam.com> wrote:
>
>>Someone once argued that the off-topic and flaming crap that started
>>about 15 years ago didn't matter cause all the tech topics were
>>already covered in depth by the first generation.
>
>I just always hoped someone would see something new, I would
>send schematics to anyone, etc. Then the shit started.

Well, realistically, you know the main root of the problem. This was
a great tech group about 15 years ago. Then...

>>Back then, it did seem like there were discussions on just about
>>anything and everything to do with tube amps, and often at very deep
>>technical level.
>
>Which may/may not of scared some folks off.

No, actually the engineers were generally good at replying on the
level of the poster's skill. It's the discussions between engineers
that took off into some real interesting side-tracks. You can always
learn from someone else who has also spent their life doing this
stuff.

> In my early years, I had
>*NO* (repeat:* NO*) tech know how. I'd guess now, most have little
>if any. Today people attach gear to musicians. I never did. Really.

I was lucky to have grown up with a couple engineers, and they worked
in and around the music industry. So I don't even remember learning
some of the tech stuff.

>Best I got back then, was being able to draft 1:1 dual sided PCBs
>with templates. Blew people's mind for some reason. P'O/cake..

Those are usually done 2:1 and then photoreduced. Easier to get good
resolution between the two sides. But I've also done a few 1:1. Now,
of course, it's easier to use software.

>Gear also was cheaper then, and alot of folks traded gear for amp
>repairs.

A lot of my early guitar playing clients had girlfriends with more
money. :-) Kidding... but I used to let them slide if I knew they
were dedicated and poor.

>One guy, had a 100 wt Marshall (late 60's) he attached white
>fur letters of his name. Paul something.

Classy!

>>.. But most of the great stuff is not archived by Google
>>(It was "DejaNews" back then, bought by Google later and presumably
>>purged of old content).
>
>I remember DejaNews. Wondered what happened to them.

I think everything before 1995 is gone. I watched it disappear slowly
after the Google takeover.

>>I was still in school then, but started designing instruments for
>>commercial manufacturers when I was pretty young. Some top name bands
>>were using them (mostly keyboards and synths). Ran custom design shops
>>in a few locations in big cities for a while, and there you meet tons
>>of good musicians and bands.
>
>See, you got lucky. You were smart enough to learn the craft, and
>make it worked. In my days, there were FEW open jobs for tube amp
>designers. I was a well paid troubleshooter on demand..

I've worked a lot with tubes, but many of the better paying gigs were
regular analog or digital, and now I mostly do software. Made a
living teaching music for a while too. Loved it, but as you've
noticed, it's not as lucrative as engineering.