Built a 5f4 Tweed Super but have low distorted volume!

[Alan0354]

I'll source an adequate cap for that tomorrow when I can get to the store. So that's what motorboating is! I think it can also be a constant motorboat sound too which means your caps are suspect to fail. This only comes when it's cranked and I strum a heavy chord and the motorboating comes in as the note decays and then fizzles out after a second. Thanks for that tip off!

All too common when you crank it up. I am more sure it's motorboating. The reason is the signal get progressively larger going down the chain. The signal from either the power amp or PI feed through the power supply, through the 10K resistor back to the two 100K plate resistor. This cause a positive feedback. The higher the volume, the more it feed back.

I usually do not suspect bad components as this is a new build. So I would not suspect the cap is bad yet until it is proven.

8uF is too small. People normally use 20uF. This gets much worst in high gain amps with 3 or more stages. This is a common problem.

If you want to be safe before spending money, and if you have a 20K resistor handy, you can try changing the 10K to 20K, you might stop the motor boating. but now, you'll drop more voltage on the plates of the two preamp tubes, but the amp should still work. If that solve the problem or change the character of the problem, a bigger cap definitely help.

[hamilcaster]

I just woke up and realized there's an error in the layout that I was using:


The 1k5 2w resistors off of pin 5 on the 6l6's should have a wire running to the 220k resistors in the bias supply, instead the wire connects to pin 5, effectively leaving the 1k5 resistors out of the circuit since pin 6 doesn't have a tube connection!! That has to be significant but probably not sound related. I'm surprised there was such a blatant error on an official layout like that.

EDIT: apparently the company that made that layout intentionally wired it up like that when they found that the 1k5 wasn't needed for proper function. They left them there anyway just in case you wanted to hook them up and use them...

[g1]

Well, the company seems to have a pretty loose definition of "needed for proper function", there's probably lots of other parts they could eliminate then .
Connect those wires to pin 6 instead of pin 5 so you actually have grid stoppers. Fender added them to later models to serve a function. Just because the amp works without them doesn't mean they serve no purpose.

[hamilcaster]

Well, the company seems to have a pretty loose definition of "needed for proper function", there's probably lots of other parts they could eliminate then .
Connect those wires to pin 6 instead of pin 5 so you actually have grid stoppers. Fender added them to later models to serve a function. Just because the amp works without them doesn't mean they serve no purpose.

No kidding. I'll fix that and then I think I'll add a pair of 1k 2-3w screen resistors to the power tubes also. 470ohm is common but I might increase it to 1k on suggestion of others. Today I'll try seeing if the preamp filter cap is the culprit for my farty overdriven tone.

Should I be able to sit in the same room and crank the amp wide open and not have to cover my ears? I was hoping the 24-watt amp would be a bit louder with two 10's is all. It's loud but not painfully so.

[g1]

I find my friends Deluxe to be painfully loud, he does not , depends on the ears.
Best check would be to compare to another similar model. But get it working right first, maybe you will solve the volume with the other fault.

[eschertron]

Should I be able to sit in the same room and crank the amp wide open and not have to cover my ears? I was hoping the 24-watt amp would be a bit louder with two 10's is all. It's loud but not painfully so.

Regarding the loudness, I have a couple questions:
If you're playing through ch1, do you have the vol for ch2 turned down? From the schematic, it looks like there will be some attenuation if you don't have ch2 vol up.
Have you tried a 12AX7 in the first position? A 12AY may not be so bad, but a 12AT7 (if that's what is there now) will dull the pain a little.
Have you determined that the speakers are in phase? It can be a real bugger to scour the amp and have the issue not in the chassis proper.

[hamilcaster]

Regarding the loudness, I have a couple questions:
If you're playing through ch1, do you have the vol for ch2 turned down? From the schematic, it looks like there will be some attenuation if you don't have ch2 vol up.
Have you tried a 12AX7 in the first position? A 12AY may not be so bad, but a 12AT7 (if that's what is there now) will dull the pain a little.
Have you determined that the speakers are in phase? It can be a real bugger to scour the amp and have the issue not in the chassis proper.

I was thinking I should get a 12ay7 to replace the AT in there now. I think the AT is just the wrong tube. I haven't messed with volume interaction yet but I will certainly try that tonight. I'll also take a look at speaker polarity. Thanks!

[hamilcaster]

Ok, turns out I had the speakers out of phase. Once in phase, the blatty and quacky sound went away. I don't so much hear any motorboating so I'll hold off on upping that preamp filter cap unless you think I should. I moved the bias wires over to the correct side of the 1k5 resistors so those are now in the circuit and I added a 750ohm screen resistor to each power tube. The amp seems mostly there. However, two things I've noticed:

1) I read the AC volts across the speaker terminals while cranking the amp and playing heavy chords. It read on average about 7vac. I squared that to get 49 and divided by 4 (the ohms of the two speakers in parallel), which gives 12.25 watts, correct? That seems sorta low for an amp said to be in the 20-30W range. Perhaps the donor OT isn't correctly suited for 4ohms?

2) My bias voltage keeps running away (in either direction). Imagine trying to balance a ball on your head. I can't ever seem to adjust the bias pot in a way to make the tube bias stay still indefinitely. I can get it to gradually stop at a desired mA but it doesn't stay there long and either starts to climb or fall. Something is amiss. Here's my bias schematic that I'm using:


So these two things are plaguing me at the moment. The amp sounds pretty good though the overdriven tone is a little farty in the bass still but that could be the alnico speakers distorting. I truly appreciate everyone sticking with me so far. This is wonderful progress!

EDIT: also, I apparently had a 12ax7 in V1 the whole time. I will still get a 12AY7 sometime soon though.

SECOND EDIT: I swapped power tubes to see if that was the issue and that gave slightly different results regarding the bias behavior and wattage. I managed about 8vac off the speaker terminals, which equates to about 16 watts now, still not as high as I'd thought. THe bias still climbs or falls but doesn't stay in one spot very long. I'll bias it at say 34mA, play heavy then stop and see that the mA jump way down to about 20mA and then veeeery slowly creep up and up past the original 35mA and beyond. Quiet strange.

[Enzo]

2) My bias voltage keeps running away (in either direction). Imagine trying to balance a ball on your head. I can't ever seem to adjust the bias pot in a way to make the tube bias stay still indefinitely. I can get it to gradually stop at a desired mA but it doesn't stay there long and either starts to climb or fall. Something is amiss. Here's my bias schematic that I'm using:

WHAT exactly is running away? Bias is the grid voltage that sets the tube current. Is the bias voltage running away or is the tube current running away?

If the current runs away, look at the voltage, is it ALSO running away?

Set a context. First measure AC voltage in your mains - your house wiring. Watch it for a while. is it stable? Or is it moving around? If it is stable, move on. If it moves around, consider that means the voltages in the amp also move around. And I see +415 as the B+ voltage in the schematic, call it 420. That meas with 120v mains, for every volt that changes, that B+ will change 3.5 volts. A 6 volt change (say 117v to 123v) in mains would result in a big 21v change in the B+. And that would surely affect tube current. Worth checking, no point in trying to get the amp past that.

But if that is not much of an influence here, then monitor the bias voltage. The voltage at the grid, not the voltage on the bias probe. Schematic says -40v. I don;t recall what you were setting it to, but whatever the voltage, does it stay stable or does it travel up or down? We need to know if the grid voltage is changing, and the tube current follows merrily along, or if the bias voltage is stable, but the tube current is ramping up.

[Alan0354]

I am puzzled, I could clearly hear the beating sound in the recording. That was definitely not caused by the out of phase of speakers. Listen more careful to see whether you still have the beating sound.

Best to record another clip so people can listen to it.

[hamilcaster]

I'll definitely record another clip after I figure out this bias problem.

Good call to work forward from the mains. So my wall is steady at 122vac.

With my jerry-rigged bias supply circuit, upon power on I derive -70vdc at the input of the bias pot. Over the course of 5 minutes, that voltage slowly climbed to about -78dc. Here's the kicker: I have my bias pot turned all the way negative, tubes are biased cold, I have one meter with a bias socket plugged into one of the tubes while I watch it, and another measuring the negative voltage. As soon as I try to measure the output of the bias pot that goes to the board, the reading at this point suddenly starts to drop down while the power tube starts to pass more and more current until running away. I pull the probe off and the negative voltage goes back down and the tube shuts down again. This can be repeated over and over. Does putting my voltmeter on the output of the bias pot make the bias voltage more positive somehow?

To be clear, I derive a steady negative voltage from the rectifier tap PERIOD. The negative voltage that the tube sees after the bias circuit does NOT stay steady!

I'm also going to throw in yet another pair of power tubes just to see.

[hamilcaster]

New power tubes yield same result. I am not thrilled with my bias circuit. No one has said that it is inferior but to start from scratch, how would one get -41dc from the rectifier pin like I showed? I'm curious about which resisitors in the original circuit would I adjust. I don't think that circuit and its values was designed for -41dc but I have no idea how I'd alter it. Is it a voltage divider? I can put together a tube amp from scratch but apparently I don't know WHY it works!

I'm doing some sleuthing. I'll find a way to get the correct voltage! Looking at other fender schematics that may have gone that route.
Found it! I'll be back with the new circuit!

[frus]

why don't you just lift the bias circuit from Jim

http://www.drtube.com/schematics/marshall/78331-2.gif

it's pretty similar already, only yours is probably "touchier", and secod, you should never trust the wiper actually making the contact. if it fails (or the pot is scratchy), you would end up with no bias voltage on grids

[hamilcaster]

why don't you just lift the bias circuit from Jim

http://www.drtube.com/schematics/marshall/78331-2.gif

it's pretty similar already, only yours is probably "touchier", and secod, you should never trust the wiper actually making the contact. if it fails (or the pot is scratchy), you would end up with no bias voltage on grids

Thank you for that schematic. It is very similar to the one I have. You think it'll be a little less "touchy" than my current one? I'll have to get new parts at the store tomorrow so it'll be a day or two before I can report back with the new circuit. Hang tight everyone!

Do you think this bias business is the reason for the low power output? I know it's the crucial component for any output at all.

Question: how do you all feel about getting a small 12v transformer and reverse wiring it to say, the heater wires to get something less than 120 on the other side: I'm thinking 6vac from the heaters into a backwards 12v transformer would yield 60vac? Then you just rectify and filter a little?

[g1]

Your meter is loading down the high impedance grid circuit when you are measuring right at the grid. Depending on your meter impedance, you will get more or less of this effect. Monitor the bias with the bias probe, without the meter on the grid and see if you can get a stable reading.
Bias takes a few minutes with NO signal to stabilize. So if you check, strum, check, you will always get some movement.
You should be able to get proper bias derived from HV winding via resistor, no need for another transformer.
So figure out if you really have a bias problem, then move on to the lack of power.