New build AB763 DR: Channel two treble pot kills volume

[72Blazer]

Tried to figure out the issue myself but no luck.  Amp is quiet and everything works except this issue.  Anybody ever encountered something like this?  Stock except for for Sluckey's trem-o-nator mod-works great.  Brand new JJ tubes for this build.  Voltages fine on tube pins.  Replace treblel pot and even tried a new 47pf on the volume pot.

Vr
J.

[sluckey]

Almost got to be a wiring error. You got two identical circuits. Start comparing.

[72Blazer]

Will do Sluckey.  The problem has got to be right under my nose.

[sluckey]

What layout did you use? If you don't find it right away, then post some hi-rez pics.

[72Blazer]

I used rob robinettes layout (eyelet board) and his grounding scheme off his website.  If this helps, the treble pot kills volume when turned up 75% max or so.  I'm gonna hit the grounds again with solder.

[72Blazer]

Here you go

[thetragichero]

sounds like one end of the treble pot track is connected to ground when it shouldn't be

[Blind Lemon]

I thought the same thing, like a clipped tag end of a wire got in the pot and was shorting out the signal.


But in the the first post he said he replaced the pot.


BL

[sluckey]

Here you go

I was hoping for big hi-rez pics that show enough detail to be able to read resistor color codes, etc.

[thetragichero]

I thought the same thing, like a clipped tag end of a wire got in the pot and was shorting out the signal.


But in the the first post he said he replaced the pot.

Almost got to be a wiring error. You got two identical circuits. Start comparing.

[NewYorker]

Check under the eyelet board, where that Treble pot connects to the 250 pF cap.  Could something be making contact with the chassis there?

Ed

[72Blazer]

Checked under eyelet board as suggested above.  Nothing touching/grounding.  None of tags on the treble pot have coninuity with ground.  Here are some more pics and I hope these are better for viewing.

[72Blazer]

more

[zircontweezer]

Could you explain what happens with the treble pot, is it as soon as you turn the pot from nil to 1 or at the top of the range. Does volume affect the issue?
  Reason I ask is I've got a Deluxe Reverb clone that I built, works fine... But, when someone was blasting it at full volume (not something I can do unless I get some earplugs and a rehearsal venue) they turned the treble to 10 and there was a noticable volume drop of 10-20 percent and I thought well the tone stack would be mostly bypassed? at that point and some hi freq oscillation is creating the issue?. Put it in the to do list and it's still there. My amp is the single channel version from this site.

[Latole]

sounds like one end of the treble pot track is connected to ground when it shouldn't be

I agree.

[SILVERGUN]

Almost got to be a wiring error. You got two identical circuits. Start comparing.

I looked at your pics briefly and the first thing that stood out to me was this spot where a cap is in series with a cap. Just from memory I cant think of a spot, other than the trem circuit where this would exist.

Could you link to the specific layout from Rob's site? There is so much info there and I just don't have the time to sort through it.

My other thought is...throw the layout in the trash, take the schematic and check component by component wiring until you find it.
Nothing can hide from a good old highlighter.

[pdf64]

Could you explain what happens with the treble pot, is it as soon as you turn the pot from nil to 1 or at the top of the range. Does volume affect the issue?
  Reason I ask is I've got a Deluxe Reverb clone that I built, works fine... But, when someone was blasting it at full volume (not something I can do unless I get some earplugs and a rehearsal venue) they turned the treble to 10 and there was a noticable volume drop of 10-20 percent and I thought well the tone stack would be mostly bypassed? at that point and some hi freq oscillation is creating the issue?. .

I suspect your amp may be flipping into ultrasonic oscillation.
Typically due to lead dress / layout being less than ideal.

Do preamp / output valve cathode voltages change when the issue manifests?

[72Blazer]

I'll check that voltage when I get more time. 

When watching the noise floor of the amp with my O-Scope at the output-it goes absolutely crazy.

[pdf64]

Ah, right, you’re scoped up.
Can you lock on and get a oscillation waveform / frequency?

[72Blazer]

PDF thanks-I've not a clue how to do that.  I've got a Iwatsu SS5705 40Mhz and a Kikusui COS4020 20 Mhz.

Are these capable of doing what you asked?

[72Blazer]

After double checking all wiring and resistor values-again, I decided disconnect all wires of the circuit board on the tube side.  I disconnected all jumper wires from underneath the board and ran them top side.  PDFs comment about lead dress was the one of the driving reasons.  After I did that the issue of the trebel pot killing the volume is gone.  I did have lead dress that were too long and not run as efficiently as could have been.  This is simply all I did.  Perhaps re-soldering all those connections played a part as well.  This was driving me nuts and I thank all that took the time to try and help me resolve this issue.  Another lesson learned.

Vr
J.

[pdf64]

Well done!
Any functional scope should be capable of showing oscillation. Mess with the timebase to show frequencies 10kHz-100kHz.
Most guitar amps are pretty close to instability.

[HotBluePlates]

I looked at your pics briefly and the first thing that stood out to me was this spot where a cap is in series with a cap. Just from memory I cant think of a spot, other than the trem circuit where this would exist.

Fender Deluxe Reverb

V2 Pin 6 runs to two series caps (0.02µF and 500pF) where the second cap feeds the reverb driver stage.

[Esquirefreak]

Not aiming to crap on your work, but the last pic in your first post show some not so great solderjoints. Both between the 2+1 connector and the chassis, but also on the back of the bias pot. I'd double check those to be sure they're solid. Looks like more heat is needed in order to make the solder flow and stick.

/Max

[72Blazer]

Noted and thanks.