Amp Going Into Cut-Off?

[mwelch55]

I built a single ended amp with 5f2 preamp and 5881 output tube.  This amp sounds great in every way until it goes into "cut-off".  I assume that is what is happening.  It happens when I overdrive the front end with a boost pedal. For example when I overdrive it with a pedal, it sounds great and then the volume goes way down and sounds "trebley".  This goes on for a few seconds and then goes back to normal volume.  While this is happening, I am monitoring the bias current and it looks correct during the "cut-off" period.

How can I tell where the problem is?  I have tried swapping the output tube and the preamp tube and still has the same problem.  I have attached a current schematic of my amp.  What is the best way to troubleshoot this?

[shooter]

looks correct during the "cut-off" period.

when a tube "cuts off" it no longer conducts, no, or very little current, hence NO sound, or really crappy low volume sound.

This amp sounds great in every way until

welcome to SE amps
by nature, they WANT to operate ~~~~ Class A, by overdriving them beyond the limits you turn them into switching amps 

[pdf64]

Two approaches are to reduce the charging time constants (eg RC coupling values, cathode bypass cap value) and to reduce the magnitude of the signal at the affected grid, presumably that of the power tube (eg try a 220k linear vol control in place of the power tube grid leak).
If you've got a scope you could investigate the signal symmetry when overdriven along the signal chain; maybe the mark space ratio of the waveform at the power tube grid is such that the 'troughs' last longer than the 'peaks'. Then you can tweak the bias, plate resistor values etc to try even things up.

[sluckey]

Monitor the three B+ nodes. What's happening with them during "cutoff"?

[mwelch55]

Thank you for the replies.

I am surprised that a 5f2 preamp overdriven by a Sweet Honey OD could create a signal large enough to drive a 5881 into cutoff.  Is it possible that it is the preamp that is switching off instead of the power tubes?  Could there be a bad component somewhere?

Shooter:
     when a tube "cuts off" it no longer conducts, no, or very little current, hence NO sound, or really crappy low volume sound.

When the amp is in "cutoff", I still get 61 ma plate current on the power tube with a plate voltage of 378 VDC.  Would the plate current still flow if the power tube is in cutoff?

I am at work now, but I will check the B+ stages during cutoff.

[shooter]

I still get 61 ma plate current

remember this is now a dynamic thing, not an idle DC thing.  So for the few milliseconds one of the lobes is beating the PA tube she may be cutting off

Your DCV reading at the cathode is your "limit", once you exceed that you're bending if not "stopping" the tube.
without a scope it's hard to "see" the dynamics in action, so it's guesses and swaps

[PRR]

> the volume goes way down and sounds "trebley".  This goes on for a few seconds and then goes back to normal volume.

SECONDS?? A part-second is normal recovery for normal grid R-C values. I strongly suspect one of your grid resistors is NOT connected (or is >10X bigger than you think).

Try your DC voltmeter on each grid. (A DMM may cause a racket?) When you find a grid that goes WAY negative and stays there "a few seconds" you have found the troubled node.