Carvin BelAir troubles

[Zouto]

Hi, I got some troubles on an up to now trusty Carvin BelAir:
the Drive Channel suddenly became weak and farty, while the Clean Channel was ok, so I replaced all the (rather old) tubes with new 12AX7 WB Sovteks and EL84Ms for power amp, rebiased the amp, and it worked fine for some hours. On next rehearsal after switching ON, the Clean Channel was totally weak and the Drive Channel had become weak and hissing again. I checked and cleaned all the jacks (Input, FX, Footswitch),tube sockets and pcb connectors/wires, but it did not fix it. Also I tried different new tubes( JJs, EHs, GTs) for no bettering.
Any ideas what parts to check next, any similar experiences with those Carvins?
Thanks for help
Zouto

[Zouto]

Any idea.....

No Carvin players around????

[Gtr_tech]

Your problem is prob'ly not specific to that particular amp. First off, check the efx loop. Plug a cable from the send right back into the return jack. Problem gone now? Have the loop jacks deoxidized. Problem still there? Could be a coupling cap or plate load resistor on the way out. Carvin has a tendency to use 1/4w plate loads, so its certainly possible one is about to eat it.

[Zouto]

Thanks, I checked and cleaned all the jacks, bridged the FX circuit, no fix at all, plate resistors give about 220K readings, but I'll replace them anyway.Voltages also seem to be ok, so I suspect the relay/supply.
Zouto

[Gtr_tech]

Ok...you need to find out where yer losing signal then. A scope is best here, but a meter set to AC volts can get you there if you know what to look for.

Passive components can check good on thier own (with simple multimeter testing) and then fail under working voltage. Just FYI.....

[pacAir]

Put the amp on an Oscilloscope...

FWIW:

I had a Carvin Nomad (same circuit as the Belair, smaller 1x12 chassis) that had very weak output with no apparent physical problem. When I monitored the output into a bench load on a scope it became obvious why it was so weak... all of the output power was being expended on amplifying an ultrasonic (40kHz or so) oscillation!

One problem that has surfaced with various generations of this design is that the various preamp input/output signals (from the front-panel preamp board) are routed to the the rear-panel connector board via computer ribbon cable(s) right over the output tube sockets! If the wire dress of these ribbon cables isn't just so, the cable's relative position to the output board can cause these ultrasonic oscillations to occur.

It seems the factory is aware of this as the latest schematics show an "optional" snubbing capacitor (56pf or 62pF I believe) be placed in the output stage to kill such instability. In addition to installing this cap, I also rerouted the ribbon cable as far away from the output tubes as I could and tie-wrapped them to the Reverb cables to keep them from drifting around. These two simple changes stabilized my Nomad and it should no further propensity for unwanted oscillations.

I was partially successful just re-routing the ribbon cable but it was the snubber capacitor that sealed the deal. The amps now sounds fabulous!


Steve

[Zouto]

Thanks Steve,
I'll give it a try and some snubbers.Could you actually hear some of the oscillations? I mean, 40khz is within the standard dog range..
Those ribbon cables are a drag and don't belong into tube amps.
I had lots of troubles with it in different contemporary tube amps (Fender, Mesa, Marshall plus cheapos).
Zouto

[pacAir]

Belair/Nomad Oscillations

I certainly can't hear 40kHz! This is why ultrasonic oscillations can be so insidious! They sap most of the amp's power, they can cause an amp to overheat while it is just just sitting there (depending on the frequency, transformer & load), and can cause you to incorrectly adjust almost anything in an amp that CAN be adjusted (like bias). You don't know it's there except by observing the effects it might have on the amplifier.

A dual-trace analog oscilloscope (w/ decent 10X probes) is an invaluable tool when repairing amps and I never even turn an amp on without having a scope attached to the load box. I monitor across the load with one channel and plug the other channel into the Sends of Effects Loops or insert points to monitor other points in the circuit. Not only can you see oscillations, you can isolate which section of the amp is having the problem. You can spot "snippets" of instability (can also be inaudible), unequal clipping, gross waveform distortions, hum, bad rectifiers, dried-out filter caps, bad tubes, etc... things that might only be hinted at if you are restricting your "measurements" to a meter and your ears!

Steve

[Zouto]

Amp fixed:
I soldered a 100pF polyprop cap directly to V5 tube socket, cleaned and resoldered all tube sockets, replaced 2 brandnew but apparently crapoly JJ 12AX7s (V1&3) with brandnew TungSols and did the hasserl presence mod,
voila, sweet sounds again.
BTW: there was a cheesy ceramic 56pf snubber cap fitted, I removed it.