THE HORROR!

[dehughes]

Marshall JCM 2000 DSL 50, early version with the "crappy" PCB (which I've not had any trouble with...perhaps due to the 50w vs. 100w setup?). Anyway.

Tinker tinker....replaced most (soon to be all) coupling caps in the .0047 to .1uf range with Sozo NextGen caps (nice), installed choke in place of resistor (nicer), and just the other day swapped the OT with a ClassicTone 50w Marshall OT (drop-in, laydown version of their stand-up 50w 3.6k primary OT).

Turn the amp on...LOUDEST MOST AWFUL IMMEDIATE GOD PLEASE MAKE IT STOP SCREECH-HOWL NOISE CONCEIVABLE. *standby on* Wait for hearing to return. Good Lord that was horrible, and the volume was super low. Hmm.....destroyed the speakers, mis-wired the OT and just torched the thing?

No smoke, no heat....enter the Internet....looks like I just created a *positive* feedback loop vs. the circuit's required negative feedback loop. Hmm....never installed/replaced an OT in this type of circuit before (all were Class A Vox-y cut control vs. presence type dealys built from scratch). Maybe, just MAYBE, all I need to do is swap primary wires.

BINGO! Amp sounds fantastic. Lesson learned. Tone improved.

Next time? Buy an o-scope? Would be wise. Learn from my mistakes, all.

[martin manning]

Something to always be aware of if there is NFB. No scope is needed, though. The spec sheet for the transformer has the phasing of the primary and secondary indicated (dots at the end of the coils), or if it didn't you could test it with an AC voltage source.

[tubeswell]

Turn the amp on...LOUDEST MOST AWFUL IMMEDIATE GOD PLEASE MAKE IT STOP SCREECH-HOWL NOISE CONCEIVABLE. ....looks like I just created a *positive* feedback loop

Common mistake. Occasionally even the factory gets the wire colour codes the wrong way around. The only way to really find out in any situation is to hook it up and see.

[pompeiisneaks]

Yeah I agree with Tubeswell, I just hooked up a champ with the colors per the datasheet and it still howled, I swapped and it worked fine I just get ready for the squeal on first power up. On my recent TW Liverpool it was right, lucky me Its a 50/50 chance.

~Phil

[dehughes]

Interesting. Didn't know it was so common.

I know I've read of this before, but like all things...if not practiced, gets relegated to the corners of memory. Never again.

[pdf64]

Note that with a newly fitted OT, the feedback may be positive but conditions may not be be right for free running audible oscillation to start up.
So no howl doesn't mean it's all good.
I find it best to test the amp open loop, measure the power amp gain, and then close the feedback loop; negative feedback will reduce the gain, positive will increase it.

[roberto]

Another option can be to swap the wires between the PI and the PA.
Often the wires to the OT are cut at different lenghts, so swapping the PI's outputs is easier.

[pompeiisneaks]

Oh man! I'd never thought of that, that's a way simpler way to do it! Thanks

~Phil

[roberto]

You are welcome.
Yes, it's way easier and yes, it's not a tip that you read so often online.
I came out with this solution years ago just because I had already cutted the OT wires to the PA sockets, and a joint there was not a possible solution. That's just porn aesthetics.

[norburybrook]

I always leave my primaries long after making 4 amps on the trot with them reversed!!! Last two have been OK..once you've had the squeal of death you don't forget in a hurry. I usually put ear defenders on now when I first fire up a new build

marcus