Hi,

I recently recapped (all of them) a silverface (70's) fender champ amp and biased it. It was working fine and sounded good. After a month of daily use (2-3 hrs/day), I began to hear the very occasional short crackles when I left the amp on and wasn't playing anything. I played with the amp knobs. Changing volume produced a crackle. I turned the amp off and then after that no sound whatsoever.

I've swapped all the tubes. All tubes glow and have the proper voltages. The speaker has been swapped out as well and tests fine.

Any ideas where to look for the problem? I know the output transformer for these amps is underrated. Is this a good place to start?

Thanks,
Den

When you said you rebiased it, do you mean you adjusted the cathode resistor to some value other than 470 Ohm?

The cathode resistor should be at least a 5 watt.

Is that an AA764 chassis?

I would check the plate, screen and cathode voltages on the 6V6 to start.

I just re-read you post and you said the voltages are normal. What is the cathode voltage? It's possible the output transformer is bad, but if the plate and screen voltages are correct it's unlikely the OPT.

If the primary of the OPT were open (the plate voltage would have to be very low if this were the case), the screen of the 6V6 would draw current like crazy and probably be destroyed.

Check all of your resistors. Sometimes resistors can develop cracks (down to hairline in size sometimes) that can cause the resistor to become noisy as a connection is made and then broken or sometimes the resistor can fail outright. Heating and cooling cycles only exacerbate this.

And, as it was making noise when you tweaked it, verify the pot itself and related solder connections. I do agree that cathode resistor/solder connection may be culprit.
In situations like this, consider "what has changed since it last worked reliably?"

could be as simple as a solder joint to the speaker. start from the speaker continuity check. then front end input continuity check

Couldn't cost you more than $5 and an hour or so to replace all the carbon resistors as well. And blast out the controls and tube sockets and jacks with deoxit, and let it dry overnight after doing so. As stated take a good look at all connections, you can resweat solder connections if there is any doubt.

Good luck.

Thanks for all the hints, info and advice!

I powered up the amp. Checking the volume pot, I found nothing unusual. Then using a wooden stick to probe around the power tube (6v6gta) for bad solder joints, I found the problem. I had added screen and grid-stopper resistors (5 watt cement) . The solder joint at pin 4 was the problem.

I also went through the rest of the amp with the stick.

btw, is the coax between the volume pot and pin7 on the 12ax7a really that sensitive to placement/wire dress?

Den

You should not use wire-wound resistors for grid or screen stoppers. They are inductive, and can actually make a parasitic oscillation problem worse. Here's a case where old-fashioned carbon composition resistors are better, and that's what you should use. You don't need power resistors there -- there is very little power in grid and screen circuits. 1/2 watt resistors will be fine.

-David

Note well the good advice about stopper resistors. I like to use 1 watt on screen stoppers only because of the better voltage safety margin.

Still recommend you replace the rest of the carbon resistors, and follow the rest of the suggestions before you let it out the door lest the damn thing be brought to your door a couple of months from now.

Properly terminated shielded cable dress should not be too critical, but ground only as the original wire was, sometimes at one end only, sometimes both. Best follow original layout exactly. unless you think you can do better than Leo.

Oh, one more thing, you could change over to a 3 prong grounded plug and lose the death capacitor, but that's another thread that you should research on your own.

You can simply use metal film resistors in place of the carbon comp resistors.

shouldnt be any dress issue with shielded coax cable shield grounded on one end

Thanks again.

@Cdoose

Yes. I changed the 470 ohm cathode resisitor to a 2 watt millitary spec'd pot to set the bias but will likely change it back once the amp settles down and I try some more experiments with it (like checking the wave form with a software oscilloscope).

Yes this is the AA764 chassis.

I have some other possible concerns with the OT. After checking on the web, I found some material stating that its (possibly) ok for the screen voltage to be a bit over the plate voltage, but reading more closely, "ok' is around a 1-2 volt difference. My plate reads 357, screen 367, and cathode 19 without the screen and grid-stopper resistors, with the new bias resistor pot around 1.2k ohms. This seems high. A screen resistor of 3.3k brings it down to about 358 volts, but I was wondering if I should try a higher values so that I am getting something around 355? Is the 10v difference between plate and screen voltage without the screen resistor an indication of something wrong or deteriorating with the OT?

I ws thinking that with such a large difference, and if there's nothing wrong with the OT, I could try increasing the value of the Ik resistor on the B+ (the one after the rectifier, before the 10K resistor). Is this an acceptable way to bring down the screen voltage as well?
(I have no theoretical background and not a lot of experience).

@codefox
I have cleaned what I can with deoxit and will check/replace all the resistors shortly. 3 prong plug was replaced when I got it but I did remove the death cap and reordered the wiring to have the power come to the fuse before the switch.

@dberman51
Is metal film instead of carbon comp for the screen and grid-stopper resistors ok?

--
re: coax cable.

The coax has the shield grounded at one end, at the volume pot. There is no exposed shielding at the other. Small movements (sliding the wire a little towards the pot or the tube) while still "sandwiched" between the two capacitors seems to result in a big change in hum. In my vibro champ, doing the same thing doesn't have such a large effect on hum. I swapped in a new piece of RG174 but the same degree of change in hum. There is a spot where I can position the wire where the hum is very low but I thought maybe something wasn't right.

I have to go and order some more parts now. I broke off one of the tube socket lugs with a solder sucker

It is ok for screen grid voltage to be a little higher than plate voltage provided the tube is nort run above any rating.

If it came that way from the manufacturer and ran fine all these years why go changing things?

Besides you've altered the tone of the amp by those changes and now the whole guitar amp community will shun you

Most tube failures occur from an excess of screen voltage and not plate voltage. Fender was known for pushing 6V6's past their rating of 250V plate and screen and beyond. Most old stock 6V6's can take it just fine. Make sure your bias is set correctly for the plate voltage and you should be okay. If the carbon composition resistors test fine and none of them are intermittent then just leave them alone for originality's sake.

N.B. Most 6V6's manufactured in last 20 years or so will not do well in old Fenders, while scrouges from olt TV's etc... do fine. And, yes, the screen voltage is really very important. You can dumb it down a notch to tame the beast. And those tube specs... Always presented as plate or whatever to cathode. Cathode is not necessisarily ground, so look again. Assume nothing.

By 20 years ago you're referring to the late production Phillips-ECG tubes as well, correct? You're right about the voltages being in respect to the cathode but I still maintain that Fender did push the ratings of the 6V6.

No dout about it. Fender tested the limits in many designs, and later production 6V6's will red plate most of the time, and pulls from old TV's or car radios will plug along for 1000 hours of playing time before crapping out. Go figure. Am now reading Tom Wheeler's book about the Fender Amps.

I think metal film resistors are OK for stoppers but am not totally sure. I would use carbon film if you're not comfortable with carbon comp for some reason.

Now I know what that horrible noise coming from Heavy Metal bands is -- the poor 6V6s screaming in pain.

-David

Metal film resistors should be ok. I forget where but I read in a post that someone had measured the metal film resistors up to several hundred MHz without any inductive effects.

I would suggest setting your cathode bias pot to 470 Ohms, then measure the B+(red wire of the OT), the plate, screen and cathode voltages of the 6V6 and posting the results here. I can then tell you if there are any problems. The primary of the OT should be about 300 Ohms.

I would also suggest to use a fixed 5 W cathode resistor instead of the 2 W pot. The resistor will dissipate about 1 W, so with a 2 W pot your just asking for reliability problems.

You wrote that you have the cathode bias resistance set to 1.2 kOhms and the voltage across it is 19 V. If this is correct the cathode current is only 16 mA where it should be about 40 mA. This is why I suggest getting the cathode resistance back to 470 Ohms where it belongs. You can use a 1 kOhm g1 stopper, but don't insert any more resistance in series with the g2 since it already has the 1 kOhm.

Below is a bad scan of the AA764, and I've added the calculated currents and plate dissipation in red.
The original design is fine for plate dissipation it's not being operated beyond the 6V6 design limits, no need to change anything. The 6V6 max plate dissipation is 12 W and the Champ is designed for about 11 W which is fine.

The Champ is a great little circuit with a minimal amount of parts, why change it?

Edit: One quick point, The max design DC plate and screen voltage of a 6V6GT is 315 V, so the AA764 Champ is running it a bit high at 330 V, but since the plate dissipation is within limits I wouldn't worry about it. I've seen some 6V6 amps running the plate voltage at 380 V and plate dissipation of 19 W without toasting the tube. I wouldn't recommend doing this, but the Mercury Magnetics mod runs the Champion 600 at these levels.