Epiphone Valve Senior problem

[bigdrums925]

I'm working on a Epiphone Valve Senior guitar amp that works fine when the gain & master pots are very low. When it is slowly turned up by either the gain or
master volumes will cause the output to drop to near nothing with noticeable distortion. Then put it back on standby for a few seconds, take it off standby and
it comes back to life until the gain reaches a certain level. Then it drops off again. I've changed the tubes, checked the heater and plate voltages and verified most
all the values of the resistors, and checked most of the caps. I don't understand why it resets itself when I turn on and off the standby since the plate voltages
don't change with the drop off. I will try to attach a schematic with some voltages I measured added, but had trouble doing that last time. As always any suggestions are much appreciated!

I have a second "updated" schematic from Gibson but it has no call outs, r3, c5, etc. Its a better drawing and largely the same but no labels.

[Enzo]

My first guess is that at elevated gain settings, the amp is going into oscillation. It is oscillating at frequencies above audio.

[bigdrums925]

Enzo: I don't see anything I can detect as oscillation on the scope. Any recommendation on a proper test for that?

[MarkusBass]

I would also vote for oscillation. How exactly are you checking this? With a dummy load, signal generator and oscilloscope on the output?
Since you cannot see it, I would suggest detecting at which stage the signal dissapears. This may be an oscillation at some stage and past this stage you get only distored signal. I would check the amp stage by stage. In this way you will learn at least at which stage the problem appears.

Mark

[Enzo]

Yes^^^^

It doesn't HAVE to be oscillation, that is just a potential cause that fits the symptoms.

[bigdrums925]

Yes I am checking with a speaker load, signal generator and oscilloscope. I was reluctant to probe with the scope around the plate voltages so used a meter set to AC volts.
Are my concerns about using a scope on those high voltages warranted?
First checked that all plate voltages did not change with the signal output dropping. Then checked the heater voltages on each tube, also stayed constant when the output dropped. I put a few notes on the schematic one of which says when I touched the VOM probe to junction between the cathode pins on V3 it caused the same effect as turning up the gains. Can you tell me anything about the 4 resistor array at that location? What is that supposed to do? That was the only place where I could get a changed response.
As far as oscillation, I looked for any trace of such on the speaker output with the scope starting at the lowest output and raising gain until output dropped.
Thanks

[MarkusBass]

I put a few notes on the schematic one of which says when I touched the VOM probe to junction between the cathode pins on V3 it caused the same effect as turning up the gains.

I haven't noticed the note because the image is displayed either very small, or huge. And what happens if you touch the junction with a resistor that has the same resistance as your meter? Do DC voltages on R16 and R14 change in this case?

Can you tell me anything about the 4 resistor array at that location? What is that supposed to do? That was the only place where I could get a changed response.

This is so called "Long Tailed Pair Phase Inverter". Shortly speaking it provides two signals in opposite phases for tubes in the power amp. Much better explanation can be found here: The Valve Wizard -Long Tail Pair , or here: The Long-Tail Pair

As far as oscillation, I looked for any trace of such on the speaker output with the scope starting at the lowest output and raising gain until output dropped.

Looking at the second remark on the schematic, I suggest checking in what phase is the feedback signal (compared to the input signal of the power amp). To check it you need 2 channel scope. Do you have such a scope? Is it possible that the wires of the output transformer are exchanged and the feedback signal is positive instead of negative? This could cause similar problems as you have.

Mark

[g1]

Another way to check for correct NFB, disconnect R15. Gain should increase. If gain actually decreases with R15 disconnected, then the OT is phased wrong.

[MarkusBass]

Right. This: How the Long Tail Pair Phase Inverter Works - Telecaster Guitar Forum is another very good explanation of the Long Tail Pair circuit.

Mark

[nickb]

Yes I am checking with a speaker load, signal generator and oscilloscope. I was reluctant to probe with the scope around the plate voltages so used a meter set to AC volts.
Are my concerns about using a scope on those high voltages warranted?

Just to address this question, you can read <this thread> for more info about that.

[bigdrums925]

So I checked the feedback and it appears to be correct as the output goes up when the feedback is removed
Then I chased the signal backwards. AC voltage at the plate of V4 climbs steadily with the increase in volume or master volume pots to 50 VAC, then drops to 1 VAC when
the volume at the load drops (as would be expected of course) but pin 5 of V4 rises from 0 to 8 VAC, then jumps to 9 VAC during the same test.
DC voltage remains steady on both plate and screen grids at 255 and 249 respectively with light bulb current limiter in line of the supply AC
Heater voltage pin 2 or 7 referenced to ground starts at 3 VAC until standby switch is turned off, then stays steady at 2.3 volts before and after the drop.
Seems to me the problem is in the power tubes, but they are brand new and matched.
Thoughts anyone?

[Jazz P Bass]

My only suggestion is that the OP removes the 'lamp limiter'.

It is serving no purpose.

[J M Fahey]

My only suggestion is that the OP removes the 'lamp limiter'.

It is serving no purpose.

Which lamp limiter?

As of measured voltages, I'd like to know what V3 plate voltages become when amp becomes attenuated.
Post before and after.

[Jazz P Bass]

Post #11.

"DC voltage remains steady on both plate and screen grids at 255 and 249 respectively with light bulb current limiter in line of the supply AC"

Unless I am reading it incorrectly.

[bigdrums925]

I removed the lamp limiter, before attenuation pin 1, V3 is 289 VDC, 288 VDC after attenuation
Pin 6, V3 before is 299 VDC, and 298 VDC after attenuation
I am starting to think it may have something to do with high freq oscillation as Enzo said because the amp will attenuate sooner with a higher freq input.
Anybody suggest which components I should start with to curb such a problem?