Converting a tube pa to a guitar amp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

ethervalve

Well-known member
Joined
Nov 19, 2008
Messages
211
Location
Montreal
Hi all,

After a few of you here gave me some great advice on tweaking my Garnet amp I got thinking about an old, fairly useless tube PA that I have and the idea of gutting it and building a decent guitar amp with as many of its parts as possible.

I was thinking that the Vox AC 10 circuit might make sense because of some similarities with the PA:
-Both use a pair of EL84s for output tubes
-Both use an EZ81 as a rectifier
-Both use an EF86 as a preamp tube
-Both use a 12AX7 as a PI
The only tube difference is the Vox calls for a ECF82 whereas the PA has a 12AU7.

As far as I can tell, provided the existing transformers are usable, the priciest missing part for this conversion would be the 8H 120mA choke in the Vox schematic.

The PA amp's output transformer has 4, 8, 16 ohm taps which would make it a pretty versatile studio amp. (It also has 25v and 70v CV taps which I wouldn't be using)

The PA amp's power transformer has a centre-tapped B+ winding (I get 365VDC at the cathode of the EZ81 with all other tubes removed--the Vox schematic shows 320VDC at the same spot under load).
The filament windings aren't centre-tapped (they don't seem to be in the AC10 schematic either)

back panel
https://drive.google.com/file/d/0B3eNXo7qrc4yZHF5RTRKbEdvMEE/view?usp=sharing

chassis guts
https://drive.google.com/file/d/0B3eNXo7qrc4yTGFRRi03VTFLZkU/view?usp=sharing

pa amp top view
https://drive.google.com/file/d/0B3eNXo7qrc4yaGMwclFTZy1Mb1E/view?usp=sharing

Vox AC10 schematic
https://drive.google.com/file/d/0B3eNXo7qrc4yOVZMZXVxdWpoVDg/view?usp=sharing

Does this seem like a sane first tube amp project? 
 
Did you try just putting a guitar into the mike input?

That often works well, with maybe some tonestack tweaks.
 
Yes I did.  When I first got it years ago I tried and I found it a little too thin sounding.

Unfortunately it was also the victim of my very first attempt at soldering (I tried replacing the small electrolytics and somehow silenced it).  It's been on a shelf for years.

I'm also interested in converting it to a 'proper' guitar amp circuit (if it makes sense) for the learning experience. Maybe if it goes well, it'd be a good platform for playing around with the Duncan Tone Stack calculator
 
check the output transformer for frequency response,

vocals occupy a very narrow slice of the audio spectrum so this is where PA builders know they can skimp on a hi-fidelity OPT,

EL84 is probably one of the best tubes available for guitar amps,

that pentode input tube might need special care in order to get it to behave over time,

tricks to do this are dropping the screen voltage and mounting the tube socket  on rubber grommets,

go for a 6X4 rect tube, cheap, and orig equip on Vox amps,

plug your new amp into a 2-12 cab, celestion bass spk teamed up with an old jensen 12 and you will not want to stop playing,
 
ethervalve said:
The only tube difference is the Vox calls for a ECF82 whereas the PA has a 12AU7.
The pentode half of the ECF82 is used for tremolo. Do you want it? Anyway, you have one extra tube socket available, so you may decide to install the tremolo there.
As far as I can tell, provided the existing transformers are usable, the priciest missing part for this conversion would be the 8H 120mA choke in the Vox schematic.
The choke in the AC10 carries all the current including that for the end stage. Many guitar amps use a lower current choke that carries only the low currents that go to the preceding stages. That reduces the bulk and cost of the choke.
The PA amp's output transformer has 4, 8, 16 ohm taps which would make it a pretty versatile studio amp. (It also has 25v and 70v CV taps which I wouldn't be using)
This arrangement has a drawback in the sense that there is a high-impedance winding  (about 500 ohm) that adds significant stray capacitance and lowers significantly the self-resonance frequency of the transformer. That is not generally a problem in itself but it results in a noticeable sonic signature that may or may not be to your taste. As CJ mentioned, PA amps are designed with the vocal range in mind, which typically orients the design towards a lower limit of 50 Hz. Considering the lowest guitar note of interest is about 60 Hz (that's for a B-tuned 7-string, you could get a lot of mileage. Anyway, don't forget that your guitar cabinet is probaly limited to 70-80 Hz. [/quote]
The PA amp's power transformer has a centre-tapped B+ winding (I get 365VDC at the cathode of the EZ81 with all other tubes removed--the Vox schematic shows 320VDC at the same spot under load). [/quote] This is absolutely fine.
The filament windings aren't centre-tapped (they don't seem to be in the AC10 schematic either)
  A virtual center-tap, such as provided by two 100R resistors connected to ground is as efficient as a real center-tap. Anyway, this will never be a so-called "high gain" amp, so the need for extra care regarding heatre-induced hum is not mandatory. The AC10 just grounds one leg of the heater's voltage.
Does this seem like a sane first tube amp project?
If I were you, I would try to retain as much as possible of the original design; remember that it is an industrial product, on which designers and engineers have spent hours in designing and debugging.
First thing is drawing a clear and accurate schemo of the current state.
Make a proper assessment of the actual performance with a good guitar and a good cabinet. You mentioned sounding "thin"; I would suggest you check the grid resistors on the inputs. They may be too low. You need at least 470k for guitar (1 Meg is better).
Then we can help you in accordance with your wishes.
 
Thanks for the great info guys! I think it's really sage advice to get the thing working properly stock and try to trace the schematic as a starting point. 
If the output transformer ends up having too poor a frequency response, do any of you have any experience with the Hammond 1750Y http://www.hammondmfg.com/pdf/EDB1750Y.pdf? It's supposed to be their AC15 replacement. (Hammond stuff tends to be a bit easier to find/cheaper here in Canada).
 
CJ said:
that pentode input tube might need special care in order to get it to behave over time,

tricks to do this are dropping the screen voltage and mounting the tube socket  on rubber grommets,

I have a humongo thread elesewhere on making an ef86 mini build.  But this screen grab gets to the heart of it and worked very well even at the og vox ac15 preamp bias voltages.
Sadly, the amp was too unstable for other reasons (too close to the sun) and sits on the shelf of rotational revists, i'll get back to it eventually :)
 

Attachments

  • New Picture.jpg
    New Picture.jpg
    77.3 KB · Views: 34
an old, fairly useless tube PA

What model?

I assume you've already auditioned it and found it lacking?  My main guitar amp for one of the rock bands I play in is 35W tube PA - 100% stock circuit.  I did jumper the channels to get one input clean and one high gain.

It can be deceptive when played at lower volume - sounds maybe too muffled compared to typical boutiques and classics.  But when you turn it wide open the sound gets correct and perfect.  I wouldn't change a thing on it.
 
andyfromdenver said:
I have a humongo thread elesewhere on making an ef86 mini build.  But this screen grab gets to the heart of it and worked very well even at the og vox ac15 preamp bias voltages.
Sadly, the amp was too unstable for other reasons (too close to the sun) and sits on the shelf of rotational revists, i'll get back to it eventually :)

Oh yeah that's a great idea for the EF86 mounting Andy.  I can't find the thread you speak of, is it on another forum?

lassoharp said:
an old, fairly useless tube PA

What model?

I assume you've already auditioned it and found it lacking?  My main guitar amp for one of the rock bands I play in is 35W tube PA - 100% stock circuit.  I did jumper the channels to get one input clean and one high gain.

It's a obscure Canadian thing called a Keil (Kitchener Electronic Industries Limited) A17. I picked it up because my first ever amp was a little Vox-esque model by them called the 77T.  I still have it but it's on the blink (next project).  I also have an interesting germanium transistor reverb box by the same company.

I'm still working on drawing the schematic (It's my first time trying this; slow going). I'm going to try to post it soon.

As an aside, what do you guys think of Jack Darr's Electric Guitar Amplifier Handbook http://www.diyguitarist.com/Misc/J_Darr.htm? Someone just recommended it to me.  I like his troubleshooting procedure.
 
> what do you guys think of Jack Darr's Electric Guitar Amplifier Handbook

Just read it. Has to be done.

 
That's great to hear it gets your OK PRR--so much charlatanism* in guitar amp tech circles.
I already found a leaky coupling capacitor in my Bassman using his technique.

*exhibit A: https://web.archive.org/web/20100603073326/http://kendrick-amplifiers.com/Merchant2/merchant.mvc?Screen=PROD&Store_Code=KOS&Product_Code=KEN0103-GWSA&Category_Code=
"Canarywood has uncommon clarity. Take a vintage amp design and play a closely voiced interval (such as a C and B, 1/2 step apart - like in a C major 7th). Notice how the notes compete for space and blur each other. Do this same test with the Canarywood and the notes will be clear, focused and discernable(sic)."
"Maybe I shouldn't have told you about those "out of tune" sub-harmonics found in other classic amps. Ignorance is bliss. If you haven't noticed them before, you most certainly will now. The GWSA tone is pure and "in tune" at any volume - thanks to a superior design in the power supply and highest quality American-made filter capacitors."
 
I made a great guitar amp out of an old Stromberg Carlson AU-58.  It's actually my go-to amp for recording bass lately.  I ran the guitar right in to the mic input, removing the switch, and ran the speaker from the 8 ohm output tap.  No other modifications required.
 

Attachments

  • image.jpg
    image.jpg
    761 KB · Views: 31
> it gets your OK

Yeah, "OK". Not great, but unique, and therefore must be absorbed.

> so much charlatanism* in guitar amp tech

Ha! No, Jack Darr was not a charlatan.

There used to be shops that fixed radios. Many of these men learned their chops and slick-tricks from Darr's writings in magazines and books. Long-running column in R-E, and also wrote books for mail-order schools.

517c-ClVNXL._AA160_.jpg
51I4OKUqYwL._AA160_.jpg
51u0-Gq4fNL._AA160_.jpg


Between AVC loops and garage door openers, he did the Guitar Amp book. On one hand, a g-amp is a big radio with no tuner, or a PA amp, and he knew that stuff cold. OTOH a g-amp is a music creation tool, and that's outside his field.

He touches on most of the extras that guitar amps have, but not all the ways those frills can be implemented.

He has no brand/style bias because he's just a fixer, not a player. (Some bias toward makers who were generous with information.)

And of particular note: this is probably the ONLY book covering guitar amps from the period when many of our traditional designs were fresh. All others (that I know) were written LONG-LONG after the golden age. Much has been forgotten (or never learned). A lot of B.S. has accumulated.

So while it nominally covers the topic, it plods along in a workmanlike mechanical way, basically telling the radio-shop man how guitar-amps work so he can fix them also.

And then there is SAMS publishing. Sams' re-drawn schematics have errors. Sams is not as bad as TAB became, but both were just pushing paper to market, not really about quality books.

Still, the Darr book is one you have to read.
 
That's great information PRR; Very interesting to know his background. I read the book and found it immensely helpful. Funny how he's deferential to musicians ("The player's ears are sensitive to the sound of an instrument in a way the serviceman can never be") and has that self-effacing 'don't ask me how I found this out' tone throughout.

I decided it would be a wiser to fix my Keil 77T guitar amp before I tackle the PA (I suspect I'm going to find a leaky coupling cap).
I redrew a sketched schematic that Bill Gill of Electroglide amps kindly shared with me.

Just wondering if any of you could help me understand a few things about this amp..

- what is the purpose of the string of .005uf and 1MOhm resistors going to ground for channel 1? I noticed a similar circuit in the early Vox AC15.
- Bill put the tone stack cap values in quotes because his examples all measured .1uf or higher. The values for the tone stack don't resemble any examples I've found (Duncan, etc.). Any thoughts?
- even when it was working well, the amp was a tad dark (treble usually on 10; bass on zero) and mushy at high volumes. Would it make sense to lower the value of the coupling caps going to the EL84 grids from .05uf to .022uf?

update:
I got it working again (it was a coupling cap). I currently have the channel 2 disconnected and the tone stack bypassed and it sounds very good.  My desoldering pump gave me some sass and threw its contents into my coffee cup,  perhaps a sign to take a break?
 

Attachments

  • Keil-77t schematic.pdf
    137.9 KB · Views: 18
ethervalve said:
- what is the purpose of the string of .005uf and 1MOhm resistors going to ground for channel 1? I noticed a similar circuit in the early Vox AC15.
It's a high-order high-pass filter that eliminates the VLF frequency introduced by the tremolo.
- Bill put the tone stack cap values in quotes because his examples all measured .1uf or higher. The values for the tone stack don't resemble any examples I've found (Duncan, etc.). Any thoughts?
I would say the one that's marked as 180M could well be 180pF, but the rest doesn't make much sense.
- even when it was working well, the amp was a tad dark (treble usually on 10; bass on zero) and mushy at high volumes. Would it make sense to lower the value of the coupling caps going to the EL84 grids from .05uf to .022uf?
It may cut a little bass, which would be beneficial to the recovery time, but it wouldn't change much the overall tone.
I believe the "tone stack" needs some attention.
It's not what is usually referred as a tone stack, which seems to be generally describing the Fender/Marshall arrangement of 2 or 3
pots in series. Here, it's a James (often called improperly "passive Baxendall" - an oxymoron), which is a parallel arrangement. It is preferred in HiFi because the neutral response can be made acceptably flat, when the series tone-stack cannot be made flat. In neutral position it has a noticeable midrange dip that has become the signature "Fender clean" sound. I have tried a number of guitar amps with a James EQ ( Ampeg, Dynacord) and have never been pleased with it.
 
Thanks for that info.  I'm going to try to read up on these James tone controls.
Now, I've bypassed the tone control and hooked up both channels. And I've put the .022uf coupling caps in the power section. Both channels sound as good they ever had in my memory. In particular channel 2, which was probably too dark almost any pickup, sounds great with my Jaguar. The first channel is a maybe a bit too bright with single coils until I roll off a bit on the guitar, but it's still a very useful sound.
I wonder if some sort of treble and bass switch arrangement might be my best bet since this would give me access to this raw no-tone-control sound.
 
I see you switched gears.
I am fairly excited for your original idea of gutting that PA and building a vox ac10.  I've been curious as to how that amp sounds irl.  Wondering if there's a reason Vox never re-released it (aside from it's closeness to the ac15).

On your other amp work, did you fully recap it and get the voltages and tubes sorted to the original schematic?  Or are you tweaking as you work on it? 

I can pm you the shock mount thread

*edit P.S. , oh yeah I really liked that super8 video on your band page :)*


 
OK here is what I would do with your PA.

1)        Sell the GZ34 on ebay and it will finance your project,  at least pay for your choke.  Use cheap EZ81 instead or use SS diodes.

2)        Strip out the tremolo circuit to free up the 12AX7's.

3)        Copy the Mullard 10W output stage and strip out the feedback on the input to the EL84's.

4)        Use one of the 12AX7's as a cathode follower to drive your tone stack, use Vox, Fender or Marshall (JR is right about James)

5)        Check out Valve Wizard as to why you should do this http://www.valvewizard.co.uk/dccf.html

6)        Maybe use one channel with no tone stack a la Brian May.

7)        Renew/update all your electrolytic caps.

8)          Read about sag in the power supply and decide if your guitar style would benefit from it.

9)          Try with and without feedback (Vox) and decide which you like best.

10)        Find out the reasons for every change before you make it, that way you will know why you are doing it, don't just take our word for it!

11)        Beware you are embarking on an addictive pastime!

Have fun!
DaveP
 
Before you do all that, assuming it works,  just plug a guitar into the mic input and see how it sounds. You might like it.
 

Latest posts

Back
Top