Vox NT15 various problems

[rockman627]

Not since the last time I mentioned it. But I think I see a burn mark near R24. I'll try to attach a photo.


..

[rockman627]

Is there any good reason for there to be 6.5k ohms between the solder pads on C14 & C17? Yes I tried cleaning the area with isopropyl and got a different reading.
Also, would taking measurements in thick mode or triode mode be of any use? In thick mode I get 1.45k ohms between the solder pads.

[The Dude]

Remove all of that glue. It looks like it's become conductive.

[rockman627]

What is that stuff? It's hard like plaster/ceramic. Is there something like lighter fluid or something to soften it?

Isopropyl alcohol worked fine.

[rockman627]

It seems to me conductivity on the board can't be the issue, or the resistance wouldn't change between thick and bright modes.

[g1]

The parallel resistance of the circuit would change when you change modes, so that would affect the measurement.
Conductive glue has become more commonplace recently, so it's a very good suspect.
Did the isopropyl do the trick?

[rockman627]

Yes, it didn't exactly soften but I was able to chip it off with a jeweler's screwdriver.

[rockman627]

I edited the schematic to show signal paths for bright and thick modes. I also drew the r24 paths for each mode. Did I do it correctly? It seems in thick mode r24 is grounded and is thus a cathode resistor in parallel with r23, but is going on in bright mode between r24 and V4p7?

Vox_nt15 bright b.pdf
Vox_nt15 thick b.pdf

[g1]

Yes, in Thick R24 is in parallel with R23. Same happens with R38 and R28 of V4A. Also, at the top half of switch, tone stack is lifted from ground and bypassed via C28.
In Bright, the connection between R24 and R38 would not be significant I believe, a bit of inter-stage feedback. At the bottom of the switch, tone stack gets grounded. At the top of the switch, the bright cap is inserted around the gain pot.

The stuff about the switch was just to show that any conductivity of the 'glue' would change with changing of switch settings, due to changing parallel resistance paths.
If the problem you are having can't be explained by any component fault (lifting components didn't help), it must be either the glue or the board itself. The glue is a known problem for this model that has come up on the forum before, so that was the best place to start.
You need to follow all traces connecting that line from R13 > C14 > C17 > R26 > R9 > C9, both sides of the board. Any thing the trace goes under should be lifted for inspection under. Look for any sign of arcing. That may include V3 or V4 sockets. Anywhere the trace comes in contact with the glue, remove the glue.

[rockman627]

Sorry for delay, I have an Ampeg SVT 4 pro to work on. But this morning I looked at the traces and lifted one side of r26, r9, and c9. After each lift, I measured between the c14 solder pads and still measured around 7k ohms. I did not see any signs of arcing. I also replaced r22 because it looked like it was touching a copper ground tape, which I peeled back. As the components were lifted, I measured between their respective solder pads and got OL each time. I removed the V3 tube socket. No improvement. After r13, all glue is removed. It's looking like the board is the problem to me. Can anything be done? If not, I have a spare Fender Champ eyelet board, and it may be time for a project.

[g1]

You didn't mention C17, you should check resistance with that lifted as well.
Then with all 4 of those components lifted, see if the voltage on C14 comes up.

[rockman627]

Both c14 and c17 were totally out of the circuit. I will give it a try but it seems to me if there is always about 7k ohms between the "top" of c17, c14, r26, r9, c9 and ground, then excess current will leak to ground and keep the voltage low.

[rockman627]

Ok, I am currently caught up. Should I reinstall C14, then lift R13, R26, R9, and C9? And by lift you mean disconnect one side? Is lifting one side preferable to the other side?

[g1]

Yes to all. It doesn't matter which side of each component is disconnected.