bf vibrochamp circuit with low low output

[mort]

uh oh.. I think I just used a signal generator for the first time. This is scary. I'm over 40 and under 60. I'm not supposed to learn new things


So with no power on the amp, I clipped in a signal(freq x 10, 10v range) across teh OT's primary and observed a nice clear audible signal through the speaker. Did I just test and confirm that the OT is in fact good? Without any direction from anyone?? For real???

[audiopete]

but did you really learn anything

[eschertron]

This is scary. I'm over 40 and under 60. I'm not supposed to learn new things

So what new thing am I supposed to learn after 60? How to wear diapers? I learned that a long time ago... I think!

Good work with the signal injector! Try setting it for a value of about 100mv, and inject it into the preamp. take AC readings at each stage's grid to follow the amplification. If you can, use the advice above to send that signal to something you can see or hear, to know if it is distorting at points where it shouldn't.

[mort]

I think I learned that can rule out the OT as a problem(?)

And it's the beginning of a new method of circuit detection for me.

Now can I inject signal anywhere along the signal path[with amp power on] and not hurt the signal generator? As in work my way backwards from the OT and find where the signal begins to get lost? This is brand new territory for me so get ready for stupid questions

[eschertron]

Now can I inject signal anywhere along the signal path[with amp power on] and not hurt the signal generator?

Yes, with reservations!

Since you know that you're working with high voltages, you will take care not to expose you signal generator to any large DC voltage (or any potential high current source). Put the signal gen probe on the grid of the tube stage and it will be happy to provide a signal to that hi-Z point in the circuit.

[52 Bill]

Yes, with reservations!

Since you know that you're working with high voltages, you will take care not to expose you signal generator to any large DC voltage (or any potential high current source). Put the signal gen probe on the grid of the tube stage and it will be happy to provide a signal to that hi-Z point in the circuit.

Or until you are sure that you are injecting the signal into low voltage grids, use a large blocking cap to protect the signal generator.

[mort]

Or until you are sure that you are injecting the signal into low voltage grids, use a large blocking cap to protect the signal generator.

how large? Wire one in series with each signal lead?

[eschertron]

how large? Wire one in series with each signal lead?

The ground lead (black, if you have a black/red pair as I have on my sig gen) should be tied directly to ground. The cap? I'd guess normal interstage coupling cap size, 0.1uF or so. Big enough to allow the signal to wobble around without being filtered out...

[52 Bill]

how large? Wire one in series with each signal lead?

For signal tracing the capacitance value is not really that important, but the voltage rating is.

If you're going to be sticking this in a tube amp with 400-500 vdc potential, you will want to use a 600v cap. Maybe a 0.1uf.

One side of the output from the signal generator is grounded, so ground that to the chassis of the amp. Put the blocking cap in series with the hot lead of the generator and you're good to go.

Now if you should accidentally connect it to the wrong side of a coupling cap, the dc from the amp will not hit the output of the generator.

[g1]

Most of the software ones I looked at could not handle the voltages in a tube amp even with a 10x probe.

still, I recommend a scope that can handle the juice.

Here's an attenuator you can build, courtesy of JM Fahey:

[mort]

Ok so since it really needed the wiring cleaned up I went ahead and disconnected all wiring from the turret board, made sure there weren't any connections underneath it, and rewired the thing. I seem to have more output now but there's still a major problem with the signal distorting badly and not sounding full through the speaker. I'll list off some observations and things I tried.

I clipped in a .1v sine signal directly on the input and got clear audible signal through the speaker until the amplitude was 75% or so at which point the amp started distorting. When the signal is clear, the vibrato effect works like it should.

Since the turret board is laid out for a typical Fender chassis it made for some long wire leads in this chassis which is set up to be oriented the same way a Marshall chassis would be(head shell). I flipped over a few of the cathode caps to make for shorter wire runs and a neater layout. So if you notice any of them backwards, that is intentional and are wired correctly for their polarity.

Found that the 25k pot for intensity is not RA, but standard. Tried wiring it backwards, doesn't change the issue.

Read through various parts of the circuit with an ohm meter and did not get any wonky readings. Measured ohms through inputs with plugs in and out of each input and all readings appeared to be what they should be, so I don't suspect the problem to be there.. All resistors are at their spec value and caps are reading what they should. I even replaced the tone stack caps just to make extra damn sure.

I did find 3.8vdc on pin 8 of V1 where it should be closer to 1.5v. Any ideas why that would be? Also found some negative voltage in there that I don't quite understand but I'm assuming it has to do with the vibrato function.

V1
p1- 230.1
p2- 0
p3- 1.7
p6- 320.5
p7- 0
p8- 3.8

v2- 172
p2- [-.8]
p3- 1.8
p6- 366.1
p7- 172
p8- 175

v3
p3- 361.4
p4- 367.1
p5- fluctuates from -1.2 to +1.6


I'm also including a couple of photos of the chassis as it is after the rewire. Most of the wiring is visible with only a few leads running under teh board. The grounding scheme is a little unconventional but the power section is still grounded separately and the inputs, preamp section, and outputs are all grounded together. I've tried it a few different ways but none of it touches on the issue at hand.

[eschertron]

Check out everything around the NFB. V1b plate (320V) and cathode (3.8V) are abnormal, and shouldn't both be high! My hypothesis is that the 47 Ohm resistor on the NFB tail is much bigger. Maybe 4k7 or 47k, even. The feedback ratio will be wildly skewed if this turns out to be correct.

Note this is merely a hypothesis. If the resistors all check out, there is still something wonky with the voltages on that tube stage.

[mort]

that resistor reads 47.4 ohms. Now those v1b plate and cathode voltages are a little lower but still high. Right now the v1b plate voltage is bouncing around between 200-250v and the cathode voltage is fluctuating in the 2.1-2.5 range. There's 336vdc feeding the two 100k plate resistors for that tube. On 1 side I've got 196vdc going to plate v1a and the other 100k resistor has the fluctuating voltage on it feeding v1b plate. I tried swapping in a new 100k and symptom is still there.

What next? Should I drop in a pair of .01uF caps in place of the .02? (out of .02's at the moment).

[mort]

Just observed that the v1b voltages are directly affected by the intensity pot. If I set it to minimum intensity then the voltages go to normal. As the intensity increases, so do the voltages.

[mort]

Here's one more oddball observation before I let it go for the evening...


The pin 5 grid on the 6v6 has about 3vac on it. There's also up to -10vdc on that pin depending how high I turn the volume up. It goes to 0vdc with teh volume all the way down.

With the lead lifted off of the pin there is neither +vac nor -vdc present on either the pin or the lead.


And here I thought I was getting half way decent at troubleshooting. Seems like the closer I get the further away it becomes