Tremolo Issue on 6G3

[martin manning]

Also check the amplitude of the sine wave at the left side of the intensity pot. It's peak voltage should be approximately the same as the bias voltage. If you don't have a scope, measure the AC voltage there and multiply by 1.41.

[bal704]

I measured across the intensity pot and here's what I got:

-31V DC on the RH side (the side going to the AC tap marked -26V on the schematic)

The LH side of the intensity pot varies from about -23 V DC to -37 V DC. This is the side going towards the Tremolo Pedal jack on the schematic. It basically seems to be oscillating on the LH side.

These are DC voltages, not AC.

[martin manning]

Yes that's right DC on the right, AC (LFO) signal on the left. Can you measure the AC on the left? Your meter probably isn't giving you a good peak-peak value measuring DC and it's shifted negative because of the bias voltage on the other end. While you're at it get the DC voltage at both ends of the oscillator's 220k plate load resistor (with the oscillator off). That will tell you if the operating point is centered.

[bal704]

There's not much AC on either side. I measure 37 mV on the LH side, about 160 mV on the RH side.

[martin manning]

Looks like your meter won't measure AC at such a low frequency, or the trem wasn't on(?).

How about the DCV on either side of the plate load resistor with the trem off?

[jjman]

You could remove only the preamp tube and report if the thump stops. Reinstall it and then remove only the PI tube and check again. Could also reinstall then ground the grid on the "other" half of the trem tube, which is the driver for the PI. Doing all of these individually will help verify/rule out an injection point of the thump.

[bal704]

I pulled each tube. Pulling any 1 of the tubes stops the thump. Only all 3 get the thump.

[martin manning]

No trimmer, just the 22k pot on the schematic. Sounds like I may need to add one.

22k pot? The intensity pot is 250k, right?

[bal704]

Yeah, typo. I checked the intensity pot, and it is 250k. I was intending to say install a trimmer after the 333 V AC input, instead of using a 100k fixed, install a trimmer. I haven't done it yet, but I can if needed.

[bal704]

I've made some progress:

1) I confirmed the intensity pot is 250k
2) resoldered connections and joints
3) unsoldered B+ from Tremolo circuit and hooked it up with alligator clip.

When tremolo hooked to B+, I get the thump. When hooked to B+1 (node after 1k, 2w resistor), I get the thump but much quieter. When hooked to B+2, I get a very very faint thump and the tremolo is actually usable. When hooked to B+3, I get less thump, but the tremolo doesn't seem as intense, but still usable. To me, the B+2 sounds best. Basically no thump and a nice sound to it.

So what is this telling me? Can I just hook the tremolo up to B+2, rather than B+ on the schematic, and go with it?

[martin manning]

Yes go with moving the LFO supply to the PI node if that works well enough. I'd also look at adding adjustable bias, but do not replace the 100k resistor with a pot- the current draw is too high, and I'd rather not have the high voltage on the pot. A much better circuit is shown below. If you use a 10k Fender-type bias pot you can hang the extra cap and resistor off the pot, going to a ground lug. Make the 75k resistor a 2W. Once you get the bias set correctly maybe you can revisit the LFO supply connection. What plate voltage do you have now, and do you have a way to measure idle current?

[bal704]

Plate voltage is 430 V. Not sure how to measure idle current.

I will say the amp sounds great, with and without the tremolo. I would prefer more grit, but I can only run this thing at about 9 o'clock without a run in with the wife.

[martin manning]

Plate voltage is 430 V. Not sure how to measure idle current.

I will say the amp sounds great, with and without the tremolo. I would prefer more grit, but I can only run this thing at about 9 o'clock without a run in with the wife.

The easiest and safest way to measure idle current is to ground the power tube cathodes through 1-ohm, 1% resistors so that the measured voltage on the cathodes in mV is numerically equal to the cathode current in mA.

An alternative is to measure the DC resistance of the OT primary on each side and use those resistances to calculate the current for each power tube. This is more of a PITA, and you are probing points that are at high voltage, but it is true plate current. To get it that way, measure the DC resistance across the OT primaries with power off (red to blue, and red to brown), and then get a measurement of the voltage drop across each side with the power on (Vred -Vblue and Vred - Vbrown). Then you can calculate plate current for each power tube as V/R.

With 430V on the plates, and 6V6's being rated at 12W max plate dissipation, you would want to idle at about 60-70% of max or (0.6 x 12W)/430V = 0.017A, 17 mA. It would be a good idea to find out where you are on this since you are running a pretty high plate voltage.

[jjman]

Moving the PS connection down the B+ rail reduces the trem's plate voltage and therefore the strength of the trem (and thump.) Since removing the preamp tube kills the thump, it is being injected into the preamp in error. Maybe there is a trem wire laying parallel to a preamp wire?