Here's my tech's report with everything we've tried.
Ok brain trust, GO!It seems very well assembled, voltages are right on, runs very well, except for one thing:
At <>180 & 240 Hz, there is a weird 'static' or 'arcing' sound that comes into the audio. Kind of a 'zzzt' in the background. The owner noticed it just playing his guitar. We are running it on a sine wave generator, so we can hear any distortion more clearly.
It only happens between 160 and 250Hz, in several 'resonances,' or specific frequencies. It only happens at medium to higher output levels.
Looking at the waveforms on the output tube plates, we do not see anything unusual. The amplifier is not clipping, nor is there 'notch' distortion at the crossover. Looking at the cathodes of the output tubes, at high level, we see a low level <>667 Hz sine wave appear. Adding a larger decoupling cap removes that sine wave, but does not clear up the output distortion. We feel strongly this is an important clue.
We have tried replacing or substituting;
1) All the tubes, all at once. No change of symptom.
2) The speaker and case. (Twice over, with different cabinets, just to be very sure) I even used a portable speaker, set up far from the amplifier, in case it was mechanical vibration induced. I used a bare 'Coffee Can' EV SRO 12" as the test speaker.
3) The output transformer. It has a very nice premium grade USA made one, and we tried a 'regular' metal case OEM style output xfmr.
4) We have tried adding more filter capacity to the power supply B+ lines, reduces the hum a little, but no other change in symptom.
Remember this is a kit built amp, all parts are new.
Every part we have tested reads good.
Our next step is to bring in another 5E3 we have access to, and bridging over the circuits in various places. If you look at the schematic, we can lift the .1 uF coupling caps after the 12AX7 driver tube, and we'll have four wires to jumper; The two caps, the power supply, and the ground. We will use the preamp and driver of one amp, and the output stage of the other. The resulting live test should narrow down whether the problem is in the driver stage or the output stage. We can also reverse the test for verification.
We can do a similar tests in other places too, and narrow it down to four or five components, at which time we will probably shotgun them.
Assuming of course, that the troubleshooting process above yields good results.
We're looking for anyone who has invented this wheel. I'm hoping someone has seen this problem, and can advise us on how to resolve it.
I doubt it is a wrong-wiring problem, as at higher or lower frequencies, the performance is flawless.
Suggestions or questions welcome.
768]
