Thomas Vox V1032 troubleshooting help?

[drewl]

Which is why I wrote in large red ink "disconnecting E-tuner must ground emitter"
As a note to myself on my schematic!

[MWaldorf]

Made the changes and still have the distortion and odd volume behavior.

Here's a video of the e-tuner where I start with the volume at 0, bring it up to 10 and then back down. Note that there's sound when the vol is at 0, and crazy oscillation when the vol reaches about 8.
https://youtu.be/MQGjOvLWzLs

Next is a video where I roll the volume up and down, but this time plucking the low E on a guitar. Note the same vol travel behavior and the distortion on the guitar signal.
https://youtu.be/dUmx4vWhfeQ

You can't hear it well on the e-tuner video, but the volume drops between 0 and 2 and then starts getting louder after that.

[R.G.]

Yep, it's got an oscillation. The volume drop at 8 is likely to be where the internal screaming goes ultrasonic (ish) and eats all the available swing.

The second video is "private" and I can't see it.

What does Mr. Oscilloscope say about what's happening?

[MWaldorf]

R.G.,
Looks like I forgot to publish the second video. Should be available now. As for the oscilloscope, any particular points I should check?

[R.G.]

Hook one channel to the input jack and one to the top of the volume control. Then re-try that second video and see what you see.

The gain of these channels is small-ish, about 8. So with a 100mV input signal, you'll see something under a volt at the top of the volume control.

This is a little sneaky. The signal at the top of the volume control should be independent of the volume setting. If it's not, you're getting feedback from somewhere else to the preamp.

[MWaldorf]

I'm not sure what you mean by top (is this before the volume?) but here's a video of the oscilloscope reading at the volume wiper, starting from zero, up to full and back down. You can really hear how strong the signal is with the volume at zero, how it drops off, increases again, and then goes crazy!

[52 Bill]

Have you taken a resistance reading from the CCW terminal of the Volume pot to ground?

And I think that R.G. means the CW terminal when he sez "top" of the pot.

[R.G.]

Bill is right - "top" means the CW lug on the volume pot, where the signal is highest, and is not affected by the wiper position - or should not be.

And he's right about first checking ground on the CCW terminal of the pot. A good place to clip your ground lead for any measurements concerning voltages or continuity to ground is on the negative terminal of the biggest cap at the top right of the PCB. That's connected to the star ground for the board.

A signal with the wiper turned full CCW indicates that most likely the wiper is not connected to ground when the pot is turned down, and the first place to check there is whether the CCW end of the pot is grounded.

[R.G.]

Here's another question for you. Is your chassis grounded by the input jack bushings?

[MWaldorf]

OK, I'm a bit of a putz on the CW/CCW stuff, so I'll reference which leg is connected where on the replacement board. The leg connected to W8/VolHot reads 6.9K ohms to ground. The leg connected to W6/VolCold reads 0 ohms to ground. Do I have them backwards???

The input jack grounds read 0 ohms to the chassis. The negative leg of C33 also reads 0 ohms to ground.

Here's a video of the signal at W8:

[R.G.]

First a bit of un-putzing. The wiper of a pot may be mechanically moved so that it electrically connects to either of the other two terminals. For a rotary (as opposed to straight line) pot, the shaft can be turned either clockwise (CW) or counter-clockwise (CCW). It's almost universal to turn knobs clockwise to turn them "up". On the Cambridge/Berkeley, the knobs are labeled ) through 8, with 0 being counter clockwise and 8 being fully clockwise.

With that as background, the CW/clockwise terminal is the terminal that the wiper has the lowest electrical resistance to when the shaft is turned fully CW/clockwise. The CCW/counter-clockwise terminal is the terminal that has the lowest electrical resistance to the wiper when the shaft is turned fully CCW/counterclockwise.

It has become common to call the terminals 1, 2, and 3, with those translating to CW, wiper, and CCW. I refuse to learn the correspondence to 1, 2, and 3 because that's yet another layer of abstraction to put on top of this whole mess.

The mechanical/physical positions of the terminals on the pot body are commonly obviously CCW/wiper/CW, but sometimes are dramatically different on oddly-made pots. But what never changes is the relationship of mechanical motion (CW - CCW) to electrical reading. So I call the terminals CCW (being the terminal that the wiper is nearly zero resistance to when turned fully "down") regardless of where it is on the physical pot body), wiper (being the one that moves electrically), and CW (the one **electrically** closest to the wiper when the shaft is turned fully up/clockwise.

And I usually just call them hot, wiper and cold. "Hot" is CW, and it's where the signal goes into a normal volume control. The signal is "hottest" there. "Cold" is the terminal that's connected to ground on the typically-wired volume control.

I *think* I put all the CW/wiper/CCW pins in the same order on the boards. It was a design objective and I can't remember any exceptions. But in your case, yes, pad W8 is hot/CW, W7 is wiper, and W6 is cold/CCW/ground for the volume control.

And the scope trace helped. I think that your power amp wiring is set up in a way that causes the amp to oscillate when the signal is big enough. If I were sitting there working on this, I would temporarily disconnect the wires from W17 and W18, which disconnects the reverb driver transformer, and also W32 and W33, the connections to the output stage driver transformer. This eliminates the reverb and power outputs from causing oscillation. Then I'd repeat that test.

I think that reverb or the output stage is getting back into the volume control there and causing oscillation. Reverb is more likely, but let's start with both to see that the input preamp is working.

[R.G.]

First a bit of un-putzing. The wiper of a pot may be mechanically moved so that it electrically connects to either of the other two terminals. For a rotary (as opposed to straight line) pot, the shaft can be turned either clockwise (CW) or counter-clockwise (CCW). It's almost universal to turn knobs clockwise to turn them "up". On the Cambridge/Berkeley, the knobs are labeled ) through 8, with 0 being counter clockwise and 8 being fully clockwise.

With that as background, the CW/clockwise terminal is the terminal that the wiper has the lowest electrical resistance to when the shaft is turned fully CW/clockwise. The CCW/counter-clockwise terminal is the terminal that has the lowest electrical resistance to the wiper when the shaft is turned fully CCW/counterclockwise.

It has become common to call the terminals 1, 2, and 3, with those translating to CW, wiper, and CCW. I refuse to learn the correspondence to 1, 2, and 3 because that's yet another layer of abstraction to put on top of this whole mess.

The mechanical/physical positions of the terminals on the pot body are commonly obviously CCW/wiper/CW, but sometimes are dramatically different on oddly-made pots. But what never changes is the relationship of mechanical motion (CW - CCW) to electrical reading. So I call the terminals CCW (being the terminal that the wiper is nearly zero resistance to when turned fully "down") regardless of where it is on the physical pot body), wiper (being the one that moves electrically), and CW (the one **electrically** closest to the wiper when the shaft is turned fully up/clockwise.

And I usually just call them hot, wiper and cold. "Hot" is CW, and it's where the signal goes into a normal volume control. The signal is "hottest" there. "Cold" is the terminal that's connected to ground on the typically-wired volume control.

I *think* I put all the CW/wiper/CCW pins in the same order on the boards. It was a design objective and I can't remember any exceptions. But in your case, yes, pad W8 is hot/CW, W7 is wiper, and W6 is cold/CCW/ground for the volume control.

And the scope trace helped. I think that your power amp wiring is set up in a way that causes the amp to oscillate when the signal is big enough. If I were sitting there working on this, I would temporarily disconnect the wires from W17 and W18, which disconnects the reverb driver transformer, and also W32 and W33, the connections to the output stage driver transformer. This eliminates the reverb and power outputs from causing oscillation. Then I'd repeat that test.

I think that reverb or the output stage is getting back into the volume control there and causing oscillation. Reverb is more likely, but let's start with both to see that the input preamp is working.

[MWaldorf]

R.G. - thanks for the de-putzification on the CW/CCW definitions. It was the part about lower resistance between wiper and leg that I was missing.

So, I just ran three tests:

First I removed the T1 reverb driver transformer (W17/18). Same behavior as before.

Next, I removed the T2 output stage driver transformer (W32/33), still leaving W17/18 out also. Now an unmolested signal regardless of volume pot position!

Finally, I replaced the reverb driver wires (W17/18) leaving only W32/33 disconnected. This is an unmolested signal as well.

Here's an image of the signal (440Hz, 100mV). Looks like the issue is with T2 and/or the connection between T2 and the board? I've got the blue wire from T2 connected to W32 and the red wire to W33. I believe this is correct, or at least matches how my amp was wired when I got it with T2 red as the 27V connection.

[R.G.]

Ok. It was the output stage.

Now we're into the wonderland of getting the six wires of the driver transformer correct. I wrote six pages in the Vox Owner's Safety Net on the connections of the driver transformer. There are um... eight ways to wire the driver transformer, since each winding can be correct or incorrect. One of these eight is correct for an NPN output stage, and one is correct for a PNP output stage.

I wish I could tell you to just put <wire color A> here and <wire color B> there and have it come out right, but I have found two counter examples in factory wired Vox amps, so while most of them got wired the right way, some of them didn't, and the wires are therefore inconsistent.

The real story is this.
Transformers have polarity. It has to do with the direction of flux change of the magnetic field and Lenz' Law and ... well, they have polarity. In a transformer with two windings, if you put a signal on one winding, then the secondary will follow that signal for the wires one direction, and invert the signal if you invert the second two wires. The polarities are "follow" and "invert", or what the transformer craft calls either "start" or "dotted".

You can designate one wire of a winding as "dotted", and apply a signal between that wire and the other wire of the winding, considering the non-dotted wire as "ground" or common. For every other winding, one of the two wires will follow the dotted wire in polarity, and one will be inverted.

You can think of this as all the dotted ends of windings will go positive with respect to the other end of the same winding at the same time. And negative at the same time. So the dot tells you which wires of a winding are "in phase".

This is crucially important in the Thomas Vox amps because the driver transformer relies on this. The output stage is a stack of two same-sex output devices across the power supply, with the load in the middle. Each output device has one of the secondary windings connected between its base and emitter. The "dot" of one secondary is connected to one of the output devices' base, the non-dot of the other winding is connected to the other output device's base. When the transformer is driven from the third/primary winding, the polarity of the dots on the secondary windings turn either the positive-pulling transistor or the negative-pulling transistor on, but never both at the same time. The inverse connection of the driver secondaries and the dotted connections on them ensure the drive voltage waveform is out-of-phase for the two transistors, the the transistor's base-emitter diode determines which polarity of incoming signal they conduct on.

Yes, that's confusing. Took me a long time to sift that out.

For two NPNs, a positive-going signal on the primary dotted wire will pull the output one way. For two PNPs, a positive-going signal on the primary dotted wire pulls the output the other way. ==> You can invert the output signal at the speaker output by subbing in NPNs for PNPs.
Or
==> You can invert the output signal at the speakers by inverting BOTH secondaries at the same time.
Or
==> You can invert the output signal at the speakers by inverting ONLY the primary connections.

Your output amp is causing oscillations. I think, based on the info here, that it may be that it's inverting and the feedback is letting it oscillate. Maybe. There are a couple of dark-horse candidates.

So some questions first:
- are you using the original output transistors?
- did you disconnect the output transistors from the resistors and such on that tagboard out on the heat sink? I subsumed that stuff onto the new PCB.

[MWaldorf]

R.G.
The output transistors are modern replacements - MJ15016G which are PNP. There are no transistors or anything else for that matter on the heat sink (see photos).

I looked at the datasheet for this transistor and it says that leg 1 is the base, leg 2 is the emitter and the case is the collector. However, I don't see anything about orientation to determine which is leg 1 and which is leg 2.

The wires are connected to the PCB as follows:
W32: T2 Blue
W33: T2 Red
W34: black wire (Q9C)
W35: T2 Green
W36: orange wire (Q9E)
W37: yellow wire (Q8C)
W38: T2 Black
W39: white wire (Q8E)