5F6A Bassman total custom build - super low output

[tomgilmartin]

My OT has three outputs for heaters, all 6VAC nominal. One feeds the rectifier through a pair of 0.3ohm resistors to achieve 5VAC. The other two are tied together to produce 12VAC. Then this is split with the 500ohm humdinger to produce two 6VAC outputs for the two power tube filaments, and the 12VAC is used for the three little tubes. Not sure it's really "humdinging" at this point. All I can really do by adjusting it is balance the voltage on the power tubes filaments. Unless I have insufficient heat in the tubes, this should not affect the ability to output a signal as far as I can tell. Nothing is connected to pin 9 on the small tubes. Voltages all measure good and the tube filaments are glowing nicely. I can post a schematic later.

Like I said, this thing is stupidly elaborate. I'm obsessive.

Lots of things were done with the idea of longevity. Like the cermet potentiometers. Pointless as far as tone is concerned, but should last many, many years. I want this thing to be inherited some day.

One thing I wonder about is the coupling caps and polarity. Between the PI and power section is a pair of Mojotone caps which have a polarity on them, but really the two sides are at similar voltages. I put positive toward the power tubes as that seems to be slightly higher voltage than output of PI.

[Chuck H]

Unless I have insufficient heat in the tubes, this should not affect the ability to output a signal as far as I can tell.

That's easy enough to rule out. Ohms law. Novices use it all the time. Shouldn't be any stretch for a guy with your creds. Even easier, just measure the filament voltage to see if it's grossly low. That would be an indicator of excessive current draw.

You really should scope the thing to see if it's oscillating.

[tomgilmartin]

Sorry Humdinger is 500ohm, not 500K.

[tomgilmartin]

Yeah, except with the crazy tight wiring it's VERY difficult to get to the V1-V2-V3 terminals 4-5 with a probe. With the tubes out it's easy and they all measure 13.2VAC from 4 to 5.
I think I have a short somewhere, like pin-to-pin. The pins on the small tube sockets are SOOO close together. I checked and checked and checked....

Also, thanks to everyone responding, even the snarky guys! ;-)

Also, when I say it's louder without the soak resistor, I mean like it becomes a whisper. Before, I had to put my ear right up to the speaker to hear anything. But it was there...

[Chuck H]

Get a scope if you can and see if there's an ultrasonic frequency present when the amp is trying to conduct signal. I can see in your photo's that you have some leads sharing proximity over long lengths that are typically, for stability purposes, kept short and far away from each other.

[Justin Thomas]

Any possibility of a miswire on/around the speaker jacks? Is there an Extension Speaker jack that you plugged into by mistake? I think there are other issues likely, too, but maybe... yes, a schem & a scope would be handy.

Justin

[tomgilmartin]

Any possibility of a miswire on/around the speaker jacks? Is there an Extension Speaker jack that you plugged into by mistake? I think there are other issues likely, too, but maybe... yes, a schem & a scope would be handy.

Justin

I didn't do it exactly like the Bassman with RCA jacks. There is only one speaker output and it's a 1/4" isolated switchcraft. The speaker sounds incredible fed from my Studio amp (Alesis RA-100)

[Chuck H]

Short of having a scope on hand you could try disconnecting the feedback/presence lead FROM THE OUTPUT JACK END. You have it bundled with your preamp leads now. That lead has the full output of the amp on it before it get's to the feedback resistor and shouldn't be anywhere near preamp leads because it can couple via capacitance and EMF with earlier like phase circuits and cause positive feedback and oscillation. Lifting that lead will eliminate your negative feedback circuit and presence control, but if it changes your problem it may offer a clue.

[tomgilmartin]

Get a scope if you can and see if there's an ultrasonic frequency present when the amp is trying to conduct signal. I can see in your photo's that you have some leads sharing proximity over long lengths that are typically, for stability purposes, kept short and far away from each other.

The amp has three sections: Preamp, Controls, Power

All the leads tied together are low-power signals, and each signal is twisted with its corresponding ground.
Because these are only pre-amplified signals, I wasn't worried about coupling between them.

8 pairs of wires are tied together going between controls and preamp.

• Input from guitar to preamp V1 (2 pairs)
• Output from preamp to volume controls (2 pairs)
• Volume controls to CF input V2 (2 pairs)
• CF V2 output to tone stack. (1 pair)
• Boost switch wires (1 pair)

Separately, there are 2 more pairs. I would consider these also low-power.

• Tone stack output to PI V3 (1 pair)
• Feedback from OT (1 pair)

Kept as far away as possible in the tight confines:

• Filament wires to preamp tubes (1 pair)
• Output from PI V3 to Power tubes (1 pair)
• Power tubes output to OT (1 pair)
• Preamp DC power from filter caps

[tomgilmartin]

Short of having a scope on hand you could try disconnecting the feedback/presence lead FROM THE OUTPUT JACK END. You have it bundled with your preamp leads now. That lead has the full output of the amp on it before it get's to the feedback resistor and shouldn't be anywhere near preamp leads because it can couple via capacitance and EMF with earlier like phase circuits and cause positive feedback and oscillation. Lifting that lead will eliminate your negative feedback circuit and presence control, but if it changes your problem it may offer a clue.

Thanks I will try it.

Would you agree that the feedback is at low voltage being on the secondary of the OT, and the feedback resistor limits current at the presence pot, so this is not a power signal, even though it's connected to the high-power output to the speaker? Therefore I was not too concerned with coupling into other signals. Maybe I have gotten it wrong...

[Chuck H]

You've clearly trumped my rough analysis with your engineering experience. You're good to go then Sorry I doubted you.

Oh, wait!?! The amp still doesn't work, right?

Your feedback lead isn't a low power lead any more than the output transformer secondary prior to it's connection to the feedback dependent circuits. Signal amplifier grid leads are typically kept short and isolated from any like phase leads because they are very sensitive. Long preamp grid and plate leads act like send and receive antennas. Guitar amplifier grounding scheme can even be critical because the sensitivity is such that even the resistance of lead wire can be enough to detrimentally couple signals. Your post above indicates that are voluntarily ignoring these long developed design principles in favor of your own ideals. I actually expected this. You must be a hell of an engineer to possess such confidence. I still believe that even if you do find a circuit fault causing your low power issue the amp will still suffer from stability problems because of the above mentioned issues. You took some trouble to explain why I was wrong and why my suggestion can't be the problem because you feel you've already seen to certain criteria in the design and refuse to fault your assessment by simply unsoldering one connection. So good luck with that.

[tomgilmartin]

You've clearly trumped my rough analysis with your engineering experience. You're good to go then Sorry I doubted you.

Oh, wait!?! The amp still doesn't work, right?

Your feedback lead isn't a low power lead any more than the output transformer secondary prior to it's connection to the feedback dependent circuits. Signal amplifier grid leads are typically kept short and isolated from any like phase leads because they are very sensitive. Long preamp grid and plate leads act like send and receive antennas. Guitar amplifier grounding scheme can even be critical because the sensitivity is such that even the resistance of lead wire can be enough to detrimentally couple signals. Your post above indicates that are voluntarily ignoring these long developed design principles in favor of your own ideals. I actually expected this. You must be a hell of an engineer to possess such confidence. I still believe that even if you do find a circuit fault causing your low power issue the amp will still suffer from stability problems because of the above mentioned issues. You took some trouble to explain why I was wrong and why my suggestion can't be the problem because you feel you've already seen to certain criteria in the design and refuse to fault your assessment by simply unsoldering one connection. So good luck with that.

Chuck, thank you for your response. I never meant to dismiss your rough analysis. Your experience building lots of amps definitely trumps mine, which is one, and that one doesn't work. I wasn't explaining that you were wrong, just trying to explain my thought process, and offering you an opportunity to correct it, which you did. I'm trying to learn here, I admit I'm not a radio engineer, I am a power and controls engineer. Tube circuits are out of my true field of expertise, and my experience only gives me the fundamentals to understand voltage and current. I have zero experience in tube amps, and less than that with RF antennas.

[Chuck H]

I agree that the NFB lead is a low voltage lead. And even though current is mitigated by the FB resistor at the termination of that branch of lead, the other branch of the lead is the actual amplifier output. This is at the same voltage as that lead, of course. Low voltage, but high current. The NFB lead can still act like a broadcast antenna. Below I've attached an actual 5f6a layout. Notice the location and length of the NFB lead (highlighted red). In fact, notice how much care has been given to keeping leads short and routing preamp and late output stage associated leads literally shoved away from each other. The signal path starts at one end and leads are connected in succession as they occur in the circuit and everything moves physically from one end to the other. Even builders that go with the stock format for this amp have been known to get a little sideways and end up with instability. There are other ways to layout an amp that don't follow all the known practices, but you can easily go sideways getting out of the box (pun intended). Just calling the amplifier "high gain" and saying that layout, grounding scheme and lead proximity "can be critical" is almost an understatement.

EDIT: In support of my position I'l call some later Fender designs to the stand. Some CBS era Fenders didn't follow the same careful lead dress practices of their earlier couterparts even though the circuits were very similar. On these later designs, that were only a little cavalier in lead dress, it was necessary to add HF bleeder capacitors for stability. Sometimes these caps can be snipped out. But sometimes they can't before some careful rework of the lead routing has happened. These are some lower gain circuits than your build. That's how critical it can be.

[tomgilmartin]

I agree that the NFB lead is a low voltage lead. And even though current is mitigated by the FB resistor at the termination of that branch of lead, the other branch of the lead is the actual amplifier output. This is at the same voltage as that lead, of course. Low voltage, but high current. The NFB lead can still act like a broadcast antenna. Below I've attached an actual 5f6a layout. Notice the location and length of the NFB lead (highlighted red). In fact, notice how much care has been given to keeping leads short and routing preamp and late output stage associated leads literally shoved away from each other. The signal path starts at one end and leads are connected in succession as they occur in the circuit and everything moves physically from one end to the other. Even builders that go with the stock format for this amp have been known to get a little sideways and end up with instability. There are other ways to layout an amp that don't follow all the known practices, but you can easily go sideways getting out of the box (pun intended). Just calling the amplifier "high gain" and saying that layout, grounding scheme and lead proximity "can be critical" is almost an understatement.

EDIT: In support of my position I'l call some later Fender designs to the stand. Some CBS era Fenders didn't follow the same careful lead dress practices of their earlier couterparts even though the circuits were very similar. On these later designs, that were only a little cavalier in lead dress, it was necessary to add HF bleeder capacitors for stability. Sometimes these caps can be snipped out. But sometimes they can't before some careful rework of the lead routing has happened. These are some lower gain circuits than your build. That's how critical it can be.

Incredibly good and interesting info here to ponder. I am in contemplation mode. The build isn't just about the sound, it's also about the adventure and expanding my knowledge base. I'm at work, so can't respond for a while. :-)

[tomgilmartin]

Would a humdinger pot be 500k? Didn't we just say on a different thread that said his measured 0-30k that even THAT was grossly out of whack?

Justin

Checked Rob's site - specs a 100-250R linear pot.

It's 500ohms, typo. Couldn't get a 250ohm in the type I wanted.