JP amp - parasitic oscillation

[gtomax]

Did a clone of a JP amp a long ways back and just put an NOS RFT in v1 to get a bit more gain from it as I've found those to be pretty hot tubes. With the boost switch engaged i'm now getting parasitic oscillation at the top (3pm-5pm) of the input vol range or if the presence is up. It's not lead dress, everything is Hiwatt neat and short. Proper shielded cables for inputs and switch, the lot and yet this squeal is there.

Has anyone run into PO with a JP circuit? A bit stumped as to where to troubleshoot/look. It is using a 100k presence pot (rather than the 22k of the original amp) but I feel like any PO is something wrong elsewhere.

Preamp is wired just like shakti's latest build http://www.marshallforum.com/attachment ... peg.67337/

[gtomax]

.

[gtomax]

Potentially one more clue - the gain is definitely high when the switch is engaged. When it's off (single triode mode) no issues and the amp seems not loud enough for a 100w and with the switch engaged there's a very loud hiss and it's definitely full power (plus gives the PO).

[gtomax]

Welp, dumb mistake the 1n cap on the presence pot was hooked to the wrong side of the .047 i'm shocked it worked well for so long. The increased gain must have just brought it to light. More work to do stay tuned but that seems to have killed the squeal.

Now I get a motorboat but only at FULL vol, boost on and full MV.

More troubleshooting to go

[mhuss]

If we wait long enough, you'll fix everything yourself.

Motorboating is almost always bad filter caps.

[gtomax]

I hoped you were right but back to where I began

The circuit is wired up like Shakti's so basically has v3 and v4 connected like sn 894 but the v1 and v2 are JP spec. Sounds great and everything works perfectly when the boost is disengaged.

When the boost is engaged I get PO squeal when input vol is 3pm -> maximum or when presence is 4pm -> maximum.

Any suggestions? I haven't yet tried swapping the 470k resistor to ground on v3 pin 2 to a more standard 1M and I've also got the resistor coming off pin 3 as 100k rather than 220k so maybe v3 is biased too hot?

[gtomax]

If we wait long enough, you'll fix everything yourself.

Motorboating is almost always bad filter caps.

Motorboating issue went away when I realized I had mis-wired something in the midst of trial & error troubleshooting. Whoops + red herring

[mhuss]

The only way to kill parasitic oscillation:
1) fix the grounding
2) fix the layout
3) add additional local supply filtering

"fix" involving some juju, incantation muttering, and sometime incense.

[gtomax]

excellent!

How damaging is PO to tubes - aka should I put some beaters in while troubleshooting layout? Or is it just an unpleasant listening experience during?

I doubt layout is the issue given it's Hiwatt spec but I'll chopstick some of the wires around. Grounding I think is legit but I'll double-check it to standard.

The other answer is just "don't turn it up that far" - I usually don't turn presence up to 10 and certainly if the boost is engaged and master is high I'm likely deaf by then. But, being a detail oriented individual, that's a crappy answer.

Mark - what about adding a grid resistors to v4? Something like a 10k?

[mhuss]

A grid resistor can certainly help, and it won't hurt anything.

I might use beaters for the output tubes while doing any prolonged testing.

[gtomax]

Very happy to report that everything is fixed. It actually had NOTHING to do with the preamp. After extensive chopsticking the presence wire from the power supply board to the OT selector was a clear problem. That "Led" me to a full audit of the power section and when I installed an adjustable bias in 2008 I must not have been paying proper attention there was a massive loop problem where a jumper connected two incorrect turrets. It's a minor miracle the amp didn't kill itself long ago with dead power tubes from all the oscillation.

It only took rebuilding the whole amp twice to find it as the jumper was hidden under the board but all's well that ends well. I've learned a lot throughout the process and that's what matters.

NO grid resistors or other bandaids to hide symptoms necessary.

Now off to play...

[mhuss]

Glad to hear you got it sorted.

[shakti]

Late to the party, but glad it got sorted.

Just a word about the layout posted above - I had forgotten to draw in a 100k jumper. See if you can spot it...

Joking aside, I tried both pre-phase inverter arrangements - both the «classic» early 70s cathode follower style, and this late 70s style and found I ever so slightly preferred the later style. This is how one of Page’s originals is wired in the few existing photos. But it doesn’t look stock and it’s slightly strange that it would feature an arrangement that wouldn’t show up again in Hiwatts until many years later. But any discussion on that is welcome by me. It’s such a subtle difference anyway. My amp is built on a converted 1973 Slave amp chassis and sounds phenomenal.

[gtomax]

@shakti - great timing actually and funny story! The oscillation came back which was baffling. How could it be gone and then come back when the amp hasn't moved or changed?

It struck me that the issue could be with the SWITCH and its cabling since it was long and coiled up and if the cable was crummy quality this could be a problem.

My build has a 1/4" cliff jack and turns out the footswitch cable I was using wasn't shielded. When I used a blank 1/4" jack inserted in the switch jack to put it in boost mode, NO oscillation. Using the unshielded cable footswitch, oscillation.

I have a much shorter footswitch, also unshielded about 2 feet long and it also causes oscillation though less than the long cable footswitch.

Shakti what switch do you use?

@Mark would shielded cable in the footswitch solve this?

I should've done it with a relay like the Hi-Tone amps

[mhuss]

Yes, the internal audio signal is going downthat wire, shielded definitely.

As I've said before, this was an odd design decision, and I've never liked it.

(We use an internal switch in our HITONE JPs so only the DC control voltage is going out to the foot switch.)