Thomas Vox V1032 troubleshooting help?

[MWaldorf]

I disconnected the external speaker jack from the chassis, leaving it only connected to ground through the PCB. It definitely changed the behavior of the amp: now it's much quieter, still with output with the vol at zero, dropping off to nothing, coming back around 6 and then oscillating above 8.

Next, I tried disconnecting the speaker jack and using the internal speaker connection. This had more output, again making sound with the volume at zero, but this time oscillating as soon as I raised the volume to 2.

So, still funny behavior, but different funny behavior!

[R.G.]

My reply from last night seems not to have made it to the thread.

I'm still pondering what that means. But we have highlighted one issue - if removing the external speaker jack from chassis makes it quieter, there was a grounding issue in play. Just maybe not THE grounding issue.

Here's a question I think is relevant - where is the chassis grounded? If you remove the input jacks from the chassis and the external speaker jacks from the chassis, does your meter still think that ground at the (-) of the on-pcb first filter cap is connected to chassis?

[MWaldorf]

With the external speaker jack disconnected and the three inputs unscrewed from the chassis, I'm reading 294 ohms from the PCB ground (the negative leg of C33) to the chassis.

[R.G.]

Hmmm. Here's a question - where is that 294 ohms? The PCB is self-contained as far as ground, including power, signal, and speaker return. Maybe on the power transistor heat sink? Reverb tank wiring?

That's off the top of my head. I'll go look at the schemos and pictures of the ones from the past and see if I can figure out what's grounded.

[MWaldorf]

Progress!!!

I'm not sure where the 294 ohms are coming from, but I do know that disconnecting the reverb pan breaks the connection from the chassis to ground entirely.

Then I had a closer look at the reverb transformer, and I noticed that the leg of one of the capacitors from the e-tuner (C37 I think) was touching the frame of the transformer. I put a piece of electrical tape in there as seen in the photo. It didn't change the 294 ohms from the PCB ground to the chassis, but the amp no longer falls into oscillation as the volume increases. So, the distortion issue still exists, with some crunchy output with the volume at zero, dropping off and then increasing, but now it keeps going the full sweep of the volume pot.

[R.G.]

Then I had a closer look at the reverb transformer, and I noticed that the leg of one of the capacitors from the e-tuner (C37 I think) was touching the frame of the transformer. I put a piece of electrical tape in there as seen in the photo. It didn't change the 294 ohms from the PCB ground to the chassis, but the amp no longer falls into oscillation as the volume increases.

Hot stuff!

I'm sitting here pondering about how no amount of theoretical thought about how it ought to work will ever trump an accidental wire connection.

Nice work, and hard to find.

[R.G.]

I haven't come up with a new theory about the cause, but I have thought of a new diagnostic.

The issue of the funny volume pot behavior could be caused by the preamp, the stuff after the preamp but before the power amp, or the power amp, or maybe their interactions. The trick is to separate these so we can find where else to dig.

One way to separate them is to break the chain of signals and apply external signals to test where the funny-ness leaks in.

One obvious way to do this is to remove the wire from the CW/hot/input pin of the volume control and connect a signal generator onto the CW pin, and the generator ground (a) on the cold/CCW pin of the volume pot and as a second test, (b) on the (-) pin of the main filter cap on the PCB. The volume pot behavior is now independent of the preamp, and this could get some interesting information.

[MWaldorf]

OK, I got the test done. Removing the CW pin of the Vol control and inserting the signal generator there creates a distortion free signal, both with the ground on the CCW pin and on the negative leg of C33. Also, in this configuration, the amp is silent when the volume is at zero!

Next I put the signal on input 1 and read the signal at the wire normally attached to the CW pin of the Vol and here there was distortion in the signal.

First, this is the signal as it is inserted at Vol CW.


This is the signal at the wire before the CW pin.


At this point I figured I'd check the voltages on the transistors in hopes that the previous fix on the cap leg ground short might have made things better. They haven't. Q5 reads 0.7C, 0.6B and 0.06E which is substantially low on the collector (should be in the 2.5-4V range). Q6 is wrong across the board at 12C, 5B and 4.5E where it should be 15-16C, 1.4-2.4B and 1-2E. Q11 reads 26.4C, 12.2B and 11.5E where it should read 27C, 4.5B and 3.8-3.9E

After that, I tried removing the e-tuner from the circuit, using JP8 and removing the e-tuner wires from the PCB to see if that had anything to do with the voltage problems, and it made no change.

[R.G.]

Catching up here.

That's a very useful set of info. I need to speculate on it for a bit. Back soon.

[R.G.]

With two days more coffee on board, I'm ready to proceed.

OK, I got the test done. Removing the CW pin of the Vol control and inserting the signal generator there creates a distortion free signal, both with the ground on the CCW pin and on the negative leg of C33. Also, in this configuration, the amp is silent when the volume is at zero!

The result of that test indicates that everything past the volume control is not involved in this mess. At least until the reverb and tremolo get tested, but those are possibly OK too, just untested. This test partitions the issue back into the preamp. Somewhere.

Next I put the signal on input 1 and read the signal at the wire normally attached to the CW pin of the Vol and here there was distortion in the signal.
First, this is the signal as it is inserted at Vol CW.
This is the signal at the wire before the CW pin.

There's definitely extra junk on the input side. Here's another question: is your scope set to AC coupling or DC coupling? If on DC, I can ignore this and go back to pondering why ground isn't ground. If it was set to AC, flip it to DC and see if the signal that ought to go to the CW lug of the volume pot has a DC offset. It shouldn't.

In fact, get out your DC meter and test for DC voltage on all three terminals of the treble and bass pots, with your meter's ground connection clipped to the main filter cap's negative lead. Maybe C5 is leaking.

If it were on my bench ( who am I kidding? If it was on my bench I couldn't get to it for a month! ) I'd then tack-wire a short across the MRB inductor to see if it's picking signals up from the air, and check the wiring to the footswitch for MRB enabling.

At this point I figured I'd check the voltages on the transistors in hopes that the previous fix on the cap leg ground short might have made things better. They haven't. Q5 reads 0.7C, 0.6B and 0.06E which is substantially low on the collector (should be in the 2.5-4V range). Q6 is wrong across the board at 12C, 5B and 4.5E where it should be 15-16C, 1.4-2.4B and 1-2E. Q11 reads 26.4C, 12.2B and 11.5E where it should read 27C, 4.5B and 3.8-3.9E

That's interesting, but since those parts seem to pass signal undistorted, pending testing the tremo and reverb, we ought to save that for another day. It may be normally-huge variation. Thomas designs were not rock solid as to normal conditions. Which is another thing that makes them a joy to work on.

On a side note, one of the early adopters of the JFET-input "Big Head" replacement boards found out that even though modern JFETs and PNP transistors as used in the preamps can replace the older Thomas parts one at a time, when you put in both new parts, the gain and bandwidth are so much better that you get either RF oscillation or distortion. Or both. We had to insert a new resistor to tame the loop gain down, and now it's much better. If we did the measurements, probably better than the Thomas design, but that's neither here nor there. Modern parts can be too good in some designs.

That's not your issue in this circuit, I don't think. Don't go there.

Yet.

[MWaldorf]

My scope is set to DC and has been all along. No offset that I see in the readings.

Here are my DC voltage readings with my DMM, with a 440Hz 50mV signal inserted at input 1:

Input 1: -4.2mV
Vol: 0mV all pins
Treb: 0mV all pins
Bass: 0mV all pins
Reverb: 66.4mV all pins
Speed: 0-400mV oscillating at CW pin, 0mV others
Intensity: 0mV all pins

A jumper across the MRB inductor made no difference to the behavior or the voltage readings.

Since grounding seems to be an issue, I took resistance measurements at all the inputs to the PCB, with C33 negative as the ground and all potentiometers are turned all the way up. Here's what I got. Maybe something jumps out as trouble?

W1:101k (with 1/4 plug inserted; otherwise reads 0)
W2:101k (with 1/4 plug inserted; otherwise reads 0)
W3:101k (with 1/4 plug inserted; otherwise reads 0)
W4:0
W5:0
W6:0.4
W7:6.6k
W8:6.6k
W9:7.2k
W10:6.6k
W11:6.6k
W12:7.2k
W13:0
W14:0
W15:0
W16:7.1k
W17:2.4k
W18:2.3k
W19:0
W20:183r
W21:148r
W22:3.6k
W23:3.6k
W24pen
W25:157k
W26:20.4k
W27:0
W28:0
W29:0
W30:1M
W31:0
W32:1.6k
W33:1.6k
W34:0
W35:470r
W36:484r
W37:484r
W38:953r
W39:967r
W40:4.1r
W41:0.1r
W42:0.8r
W43:0.1r
W44:0.8r
W45:10r

Also took measurements at some junctions in the preamp.
R1/R2/R3/C1: 33k
C1/C2/R4: 124k
C2/R5/C6: 6.4k
TP1/R9: 5.5K
R4/R5/R6/C3: 25k
C3/R8/C4: 100r
R8/C4/R55:3.2r
R10/C9: 7.2k
C9/R13: 100r
C5/C7/R10: 8.7k
C7/C8/L1: open (reads 1.8r with the foot switch activated)

The foot switch operates as expected on the MRB and reverb. On the tremolo, there's an odd "thrumming" (motorboatish?) sound when the tremolo is switched off which varies based on the speed and intensity settings. Goes away and tremolo operates as expected when the trem is switched on. Doesn't happen without the foot switch. Not sure if this could be related or is just another oddity.

Crazy about the big head JFET - who knew parts could be too good?

Am I the first guy building a V1032, or just the first non-EE to make the attempt?

[R.G.]

I'll do some digesting on the numbers.

No, you're not the first guy to build one - that was me - nor the first non-EE, not even the second of either of those. You're about number seven. That's one of the very puzzling things about this. The others just seemed to work once the wiring was right. But none of them included the E-tuner.

Hmm. The E-tuner raises one of the preamp's emitters off real ground. It just struck me that it's possible that the bug is hiding there. I'll go look.

[drewl]

Fwiw, got an odd late 60's solid state power amp in this week.
Possibly an AC Delco? Or at least a bunch of the parts are but anyway, it is very similar to a TV Vox power amp.

Two germanium PNP's, driver tranny, but built much better!
Better parts, layout, works like a charm.

[MWaldorf]

Gah! I got my hands on a proper schematic for V1032 from NCM and there are more differences from V1031 than just the e-tuner! By item, here are the new values, with the 1031 values in parens
C2: 10pf (18pf)
R15: 15K (27K)
R17: 100K (330K)
R19: 52R (82R)
C14: 470pf (180pf)

Also, there is a 680 cap to ground at the junction of C12/R18/Q4C!

There are two voltage readings that are difference at well: Q3C/Q4B: 15V (11.5V) and Q4E 1.3V (9V)

Are these differences enough to be causing my problems?

vox-berkley-ii-v1082-1032R1.pdf

[drewl]

No.