Isolation 1:1 for widowmaker

[Smokebreak]

I reattached the floating ground system, and connected the secondary leg to it, and installed a shiny new 270K/.05 from all that, and I've got sound...granted the most primitive amplified guitar sound you've ever heard, through a damaged speaker, and some hum, but I'll address that tomorrow. I've got a whopping 105VDC on my 50L6 plate
Thanks a bunch for the help Martin and Lou.

[martin manning]

Yes, you have to ground the bottom of the iso transformer secondary. Since you now have isolation, I believe you can ditch the 5n//270k and take all the grounds to the chassis. That resistor and cap are there to add some safety by limiting the current that could flow from the mains supply to the chassis. Use standard procedure as far as taking the first filter negative lead directly to the transformer secondary.

[JazzGuitarGimp]

Yes, ground the bottom lead of the PT secondary. Without it, you have no return path for the HT supply. As for the heater string, measure the AC voltage across each tubes' heater, as well as the 150R dropping resistor.

[martin manning]

I reattached the floating ground system, and connected the secondary leg to it, and installed a shiny new 270K/.05 from all that, and I've got sound...granted the most primitive amplified guitar sound you've ever heard, through a damaged speaker, and some hum, but I'll address that tomorrow. I've got a whopping 105VDC on my 50L6 plate ;)
Thanks a bunch for the help Martin and Lou.

Check the bottom of the cathode resistor on the 50L6, it'll be -40V or so to ground because of the 270k//50n. Definitely try shorting that out and see if you like it better. V1a plate will go up to ~50V instead of ~25V.

Re your question, in the original circuit the return isn't grounded, it goes to the top of the 270k//5n. You need a return somewhere, but we seem to be throwing the term "ground" around carelessly.

[Smokebreak]

Alright I think we're good here. Shorting the 270K//5n to ground kicked up the 50L6 plate from 105VDC to 145VDC, and quieted the hum considerably, as well as adding some life to the tone, probably mostly from the preamp voltages coming up too.

For the heater voltages, I've got:

35Z5 : 37VAC (spec 35V)
50L6 : 56VAC (spec 50V)
12AX7 : 12VAC (spec 12V)

So I'm within 10% of spec. The 150R resistor in the string dropped 25VAC.
Adding these all up is 130VAC, which is what I'm measuring across the secondaries.

Check my math:
The tubes SHOULD drop 35+50+12=97V, so I really need the resistor to drop 130-97=33V

Tubes draw 150mA and current is same in series circuit so R=E/I : 33/.150 = 220R

Should I really bother with a 220R resistor instead of the 150R(which by the way doing the same math has drifted to ~166R, I think..) ?

Speaker is rattling but the tone isn't half bad at this point. Amp is about as loud as a loud acoustic guitar.

[martin manning]

I'd measure the dropping resistor in the filament string and calculate the current. If it's 150mA then it's good. Any of the tubes might be off spec re voltage drop. The 50L6 plate didn't really come up, it just has the same reference as you are measuring from now. I'm not surprised that it sounds better the way you have it. How's the isolation transformer doing?

[JazzGuitarGimp]

I'm not surprised that it sounds better the way you have it.

Agreed, without the 50nF cap across the 270K resistor, the first stage was basically a concertina PI, with 270K in the plate, and 270K in the cathode. Unity gain, which the 50nF cap helped a bit - but even at 329Hz (open E, 1st string), that 50nF cap only looks like about 10K-ohms. At the open low E string, the Xc of the 50nF cap is about 40K-ohms.

[Smokebreak]

Iso transformer seems to be doing OK. I turned the amp on 20min ago and left the room. Just checked and it's gotten a little warm, but I keep my hand on it indefinitely. I ordered a Jensen MOD 8" speaker 4ohm for it, so I'll revisit in a week when I get it.
Only thing that's buggin me a little is that the first filter cap is 150V, and I've got 47u 160V caps in there.

[JazzGuitarGimp]

I noticed the filter cap for the preamp stage in the schematic. I think it won't be a problem as long as you keep the 35Z4 (ir whatever it is) in the circuit. If you eventuall replace that with a SS recto, you will likely see higher voltages than is safe across that cap until the amp warms up. That could be a problem.

[martin manning]

...without the 50nF cap across the 270K resistor, the first stage was basically a concertina PI, with 270K in the plate, and 270K in the cathode. Unity gain...

I don't think so, Lou. The input signal is ground referenced at the bottom of the 3k3 cathode resistor so the input stage has 30+ dB gain, and the 50n cap is dropping some of the low end.