Loss of signal after turning up the volume

[hasserl]

I've just got a new build fired up and checking it all out. This is a two channel amp, one channel is working just fine, the other one is giving me a bit of a problem. This channel is basically an AB763 circuit, without Reverb or Vibe. The problem is the signal just cuts out after turning the volume up past about 6. It will play fine as long as I leave the volume down low. Tracing the signal with a scope I find everything is fine until I get to the 3rd stage. I have a good signal going into the grid of that stage, but coming out it is very asymetrical, at about 3 or so the top half of the wave starts flattening. As I increase the volume the top just gets flatter and flatter, while the bottom half gets very pointed. At about 6 the tops are flattened so far and the wave has widened enough that it is almost a straight line, while the bottom is very elongated and pointed sharp. At this point the whole thing just collapses and the tops form a straight line and all I have is the bottom in sharp points, and I get no sound at all (if I plug in the speakers and plug in a guitar). If I turn it off and wait for a bit then turn it back on it will work fine again, as long as I leave the volume set low. If I turn it up it does the same thing all over again.

I can bypass that stage and connect the output of the second stage to the channel mixing resistor and it works fine like that. I can crank it up and have no problems. The other channel works just fine also. I did change the tube in that slot, just in case I had a faulty tube, but that was not it.

I have a 100k plate load resistor on this stage, a 1k5 cathode resistor and 25uf 25 volt bypass cap. Any ideas? Why would the output be so asymetrical?

[Tom Phillips]

This sounds very similar to another recent post. The problem could be a missing part. Check your problem stage and verify that it has a grid leak resistor properly installed. This could be any high reistance (500k to 1 Meg typical) path from the tube input grid to ground. This can be served by a gain pot path or a separate resistor.

[hasserl]

Ahh, of course! That's undoubtedly it. I was doing the Vibe channel without the reverb or vibe. So I just connected the second stage to the third without adding a grid to ground resistor there. I guess the reverb and/or the vibe provide this in the AB763 circuit. I eliminated those but didn't think about the need for a grid leak. OK, I guess that's tomorrow's job.

Thanks Tom, I appreciate it.

[hasserl]

Just an update on this issue for anyone interested. I've only had very limited time to work on this project, what with Holy Week activities and other obligations. But in the limited time I had I first installed a 470k grid leak resistor. I installed it between the coupling cap and the grid. The amp went into horrible oscillations right away with the volume control turned to about 2. I pulled that resistor and installed a 220k, I could then turn up the volume a little more before it started oscillating and screeching horribly. I then pulled the 220k and to make this process a little easier to find a good value I clipped a resistor decade box into place and started checking various values. Each time I decreased the value I could increase the volume control a little more, also I noticed that there was a tone of gain anywhere with the volume turned down below the point that the oscillations took over. My little ol Telecaster sounded outrageous with much sustain and distortion. I kept decreasing the value of the resistor until I got to 3k3. I can now turn the volume pot up to 10 with no oscillations, and there is still plenty of gain, much more than I have goten from Fender amps before. But 3k3 is way smaller than I anticipated, I fully expected the 470k to 220k to be the range that would work.

I'm going to be away from this for a few days as I take my family on a short vacation. So I won't be able to work on this until I get back. I haven't had it on the scope while I was doing this, so I will do that next to see what is happening. I'll follow up this up later.