1974 Garnet stencil 10-PT - AC isolation mod

[cancon]

Hello all, I'm new to this forum. I've been directed to you good people by Enzo since the last place I posted about this didn't look too kindly on non-isolated gear repair. My pro-education argument about learning how to safely deal with these amps fell on deaf ears.

Here is the schematic:

There are two notable changes in circuit for the unit I'm working on;

There is now a 50va isolation transformer installed. The fuse and power switch are now wired in series with the live line AC and transformer primary (neutral to primary). Secondary connected to ground and rectifier. Chassis is earth grounded.

The drawing doesn't show it, but this amp had a capacitor (0.22uf/600v) installed between circuit ground and chassis (except for the input jacks and one leg of the OT secondary which are direct chassis connections). When I re-capped the power supply I opted to keep this cap in circuit (new cap installed). I hear no difference between this cap in circuit or bypassed. Was this cap a fail-safe against high voltage DC leaking to chassis? Should I pull it or leave it?


Thank you
adam

[Jazz P Bass]

The general consenus is to pull the capacitor.
The real problem is not the circuit, it's the component.
The correct capacitor for that location is rated to go 'open circuit' on a failure.
The poly caps, no matter what the voltage rating, are not designed to go O.C.
Here is a real good read on Line Filter Capacitors.
Link: Line-Filter

[cancon]

Thanks for your reply.
I just want to make sure we're talking about the same thing...I believe you are referring to those caps which are used between the mains and chassis in conjunction with polarity switches. The cap which I'm referring to is not connected to the mains at all - but instead, it is installed between the "common ground" rail and chassis/earth ground. In this position, in the event of a failure, you definitely want the cap to short out - otherwise the cathodes and power supply won't be ground referenced. I just wanted to make sure we're on the same page, but I think I just gave myself another reason to pull that cap as well

[52 Bill]

The schematic that you posted shows a 0.047uF cap paralleled with a 220K resistor, which is common for that style of amp. So is there a resistor paralleled with the 0.02 cap?

[cancon]

That cap/resistor is too far back in the circuit. You'll see that the drawing shows the input jacks "chassis grounded", while the rest of the circuit is just "common ground" (not chassis). Why would Gar Gillies specifically separate these two grounds. "Common ground" connects to "chassis ground" via the cap in question. The only things that directly couple to chassis are the input jacks, one tap of the OT secondary, and the new earth ground. Shouldn't the circuit just connect to chassis and do away with the .25 cap?

[Enzo]

But don't overlook the fact that originally this was a "hot chassis" amp. You have a common in the circuit, but you don't connect it to chassis because the circuit is wired to the mains.

The original connected the mains to one side of the power supply and power amp. But the input stages needed a input circuit grounded to chassis. Ultimately the two halves of the amp must be referenced together, but we need to do it without a direct connection - and thus a direct commection to the mains on the chassis.

SO they connect the chassis common and power common together with the cap. Signal runs through it OK, but not the direct mains current.

Once you isolate the power supply from the mains, then your cap can become a wire.