Here's a strange guitar amp...

I was at work and found this way back on a shelf-Some student must have made it with scrap parts. I drew the schematic here

This is my first time trying out OrCAD. Sorry about the bad schematic layout :? still learning. This uses two input transformers (probably because that's what was around?) And feeding two 6X4 rectifiers. OT is a Triad S27A. (8 audio watts).

I hooked it up and found that the top of the audio wave will clip very quickly and that the bottom of the wave stays good for much longer. Is this biasing? Any thoughts?

Joel

thats an odd little preamp section. with a 470k plate resistor and a 13k resistor on the cathode you are going to get some pretty good asymmectrical clipping. The other half of the 12AX7 does give abit more gain and probably adds some more pretty sevre clipping. The preamp could be made quite useable with very little work. a few resistor a cap changes. The power amp seems to be somewhat in order nothing jumping out like on the preamp. But I am not to familure with EL84s. You could just give the preamp 100k plate resistors and 1.5k resistors on the cathode and see how it sounds. They most likely will not be the best values but it will work alot better. Without knowing some voltages its really kinda hard to say anything else.

adam

Thanks! I'll get some voltages on it tomorrow if I can get the time...I wish I understood more about what those resistors actually "do" in the circuit. That 470K resostor will limit the current going to the plate and therefore decrease gain on that stage? Is that correct? Then, what is the function of the 13K resistor on the cathode?


I know these must be very basic things-so thanks for your patience!

Joel

the 13k resistor on the cathode sets the grid bias and the 25uF cap parrallel to it add in some feedback over the whole audio spectrum. at lower values this cap can act as a low cut filter.

Grounded_Cathode_Amplifier.pdf

There is a good article on the grounded cathode amplifier goes through most everything you will need. But it is aimed towards the audiophile crowd and not the guitarist.

http://www.aikenamps.com
and
http://members.aol.com/sbench101/

Those both has a good assortment of information also. Look in the tech section on the aikenamps pages, it is more geared towards guitarist, but still alot of good info. Steve Benchs page is just full of good all around usefull info. Has info on doing loadlines for power output stages on pentodes. so you can figure out if the EL84 is happy or not.

adam

Well,
After doing some experements, I'm more confused than ever! AC coming out of the two PTs is 240VAC. The DC voltage after the rectifiers is 280VDC. After going thru the Caps in the power supply, it drops to 96VDC(!?!).

Changing those two resistors didn't seem to help the clipping on the top of the wave. The clipping occurs at the plate of the #1 pin of the AX7.

We tried to reduce the 2W 560K resistor on the power caps to a 55K. That raised voltages after the caps but now the signal oscelates (right word?) The AC signal literally goes up and down seemingly following a ripple of the ground? This is a great learning thing, and I may be kicking a dead horse by trying to make this work right, but I'm happy to be learning from it. Any thoughts?

Thanks.
Joel

Dumb question - how old does it look? Is it possible that components have started going bad or drifting in value?

Hard to say,
Tubes look pretty new. Iron looks old. The caps are in one of those all-in-one cans that are usually older. I'd guess assembled within 10 years max. Components may be older...

Joel

Here is how i would start, the power supply is changed up just to be a standard guitar amp style power supply, nothing fancy, but does the job, and then made those few changes on the preamp. Or once you get the power supply done and get some more decent voltages comeing around you can refer to those links and do the math for the preamp. I am not completely sure what is going on with the power supply design on it. kinda funky. Just make sure any caps on the plates will be able to handle the higher voltages. same with plate resistors. while the bias on the preamp most likely will not be perfect it should be alot better with the new voltages and you can take some measurements and work form there. or read those articles and redesign the preamp completely. Although I find it to be alot of fun to take a mess like this and make it good and useable.

adam

Cool,
Just to be sure-what is the B+ line coming up out of the middle of the caps in the power supply?

Joel

> the top of the audio wave will clip very quickly

Since we can't know if the "top" on the output is in/out of phase with what is happening at the tube plate, this can mean different things.

However I spotted something that WON'T work:



The output tube wants to work into about a 5K impedance. But the screen grid network is a 560Ω resistor and cap to ground. If it is really wired that way, the 560Ω is taking all the power.

If the 560Ω really goes to the other side of the primary, fine.

If not, you have two choices. Move it to the other end of the primary (pentode operation) or remove the 10uFd cap (triode operation).

However that cap may already be dead. It gets most of the ~350V plate supply, but is marked 150V.

The power supply is pretty silly. There is no good reason to use two 6X4 here. I hope the two HV windings are connected out of phase, so it works full-wave. The 60+60uFd caps are really too big for what is in effect a Fender Champ, though I suppose it won't hurt (and may be necessary if the power supply runs half-wave). The 3.16K in the power supply does.... what?

You'd probably also like two 100Ω resistors from the 12AX7 heater wires to ground.

Unless you have papers that prove Eric Clapton played Brighton Beach on this amp, rip the silly thing apart and do it right. Between possible transcription errors on your part, and possible slap-dash original design, I can't see much merit in "restoring it".

Thanks PRR,

The 3.16K in the power supply does.... what?

Click to expand...

That may be my error. When I found this, the wire going from the 3.16K back to the cap was not connected anymore (pulled off or something). I connected it where I "thought it should go" It definately went to one of those four caps in the PS and that seemed like the one it wanted to go to.

the "top" on the output is in/out of phase with what is happening at the tube plate, this can mean different things.

Click to expand...

I'm taking the measurements in the circuit with a scope IE: directly after the plate, etc. -not sure if that will help you clear that confusion up.

There is no good reason to use two 6X4 here

Click to expand...

Could you clarify this? Is it because another tube would work better, or because one 6x4 would be better? I have a feeling that this amp was purely a "build what you can with what is in these drawars" type of thing, and there aren't any 5AR4's around (I checked ). The PT's only have 6.3 v heaters on them too.

I feel silly running you guys around on this bad design, but it is really a great learning tool to try and understand the problems with this other than just build something that I know works.

I'll have to check on that 560R resistor to see if it was my error, but I think that's how it is.

Unless you have papers that prove Eric Clapton played Brighton Beach on this amp

Click to expand...

Nah, just something from some guy named "Heindricks". Ever heard of him?:green:

Joel :green:

check out www.angela.com and look in their how to section for the Super Single Ended 6V6, it is a two rectifier amp and goes into what two rectifiers gets you over one. And I aggree with you on getting this amp to work over scrapping it and starting over. I have one of those mystery amps myself, someone built it out of the amp from an old organ extension cab. Its a real mess, I have not even been able to trace most of the circuit yet, there is alot of stuff going on in that small case.

If you refer back to Steve Benchs webpage and check out his threory section on Loadlines. The first part of that will show you how to design a power stage for triodes and then pentodes, read that. Then go over to tdsl.ducanamps.com and do a search for tube data sheets, find one for the EL84, you will have many to choose from. The info there should give you enough knowledge to redesign the power amp section and get it running nice and happy.

The B+ comeing up in the middle should just be connected to the other line Labled B+, You will notice that at the plate resistor of the second stage of the 12AX7. Once you get a good power supply going and 100k plate resistors (although you can use other values, 100k is a good place to start with) on both 12AX7s you can use what you learned in The Grounded Cathode Amplifier pdf and In the Common Cathode article at aiken amps to make the preamp good and stable. The oscilations you hear are because the amplifier stage is unstable, there are a few things that can cause this, but those two articles should get you a nice stable amp.

adam

> Super Single Ended 6V6, it is a two rectifier amp and goes into what two rectifiers gets you over one.

"If one 5Y3 is enough for the power supply, let's try two of 'em parallel like some of the old hi-fi amps used."

Doubling the power tube (and halving the load) doubles the power output. A modest 3dB increase in volume, with possibly more distress in the speaker.

Unless the rectifier is woefully inadequate, doubling-up on it gives only 10%-20% more power, a barely audible 1dB difference. And it is not impossible that the extra heater sucks-down a skimpy transformer more than the reduced rectifier loss gains.

> theory section on Loadlines.... how to design a power stage for triodes and then pentodes

Yes, good theory. But in fact you won't go wrong using the tube-manual values as a starting point, and then applying the "Fender Goose" (jack the plate voltage WAY up until the tubes don't quite fry). For this tube (about any power pentode), 5K is a good load but anything from 2K to 10K will work almost as loud. The book also gives plate current values (35mA-50mA). 6BQ5 is not rated over 300V design-center, but if you actually get 350V you will be OK for rock-n-roll. You can go higher, with some risk of short tube life, but a Champ-size amp is not about Rated Power. 5 funky watts beats 8 "good" watts. Low-buzz may be a bonus for some uses; then it makes sense to use a "wasteful" power supply with extra filtering for a little less test-power and a lot less buzz.

I'm inclined to NOT study this thing too hard: it may be a Bad Example. But whatever works for you.

For a quick and known-good amp, compare what you got in the 12AX7 against any classic Fender input. The Angela SSEGAP is also a fine plan. Wire the 6BQ5 the same way as a Champ or SSEGAP, but adjust the cathode resistor to get an operating current like the one in the 6BQ5 spec-sheet (it isn't quite the same as 6V6/6AQ5, has a lot more gain). Check all the connections: don't assume the original builders knew what they were doing.

For big power and low buzz in a deluxe non-portable amp, use the power supply choke. For low-buck or super-portable use, replace the choke with a 1K 10W resistor.

As seen in the SSEGAP, you really want three stages of power supply filter. The Plate-feed is the biggest load and does not have to be super-clean. Screen feed should be cleaner. And the preamp stages probably should not be fed from the same point as the screen grid: two cascaded stages are stable, three stages can motorboat. Here it is less likely because the screen grid throws only small signal into its power rail, but the first 12AX7 stage needs much cleaner power than the screen needs so it is well to add another filter stage.

Damn PPR you are good. I love makeing these kind of things work and work well without changing to much, lost causes are fun. But you convinced me, if it is a nice chasis rip it apart and make a new amp otherwise salavage it for parts.

Although I would still love to see you make that thing work well.

adam

I'm just gonna try a few more things, and then rip it apart and use the parts...

Joel