SF Pro Reverb UL bias and voltage question

[gui_tarzan]

I put a 6500/4 ohm Heyboer non-ultra linear OT transformer in this to see what it did and it sounds great, but the OT is only rated at 45w so it can't stay there for long. I'm getting the following voltages:

Bias: -59v

V7 & V8
Pins 3: 538v
Pins 4: 440v
Pins 8 to grd via 1 ohm resistor: 20mv

I ran the pins 4 (with the 470R) from the second stage of B+ to replace the UL taps. So there is almost a hundred volt drop between pins 3 & 4 which I don't think would be unusual with this configuration. The BM Amps calculator shows that should produce 79w (I know, that's not accurate) with this configuration but that seems way out of line since the UL rating was only 70w. But again, this is with a pair of 6L6GCs so in reality it shouldn't put out more than 55w, but at a current of 20mv each where do I stand with this?

I'd like to get the initial B+ down to the 450v range to try to soften this up a little if that's even possible. This is the stock schematic so you can see what the voltages are from the factory.

[martin manning]

To reduce the B+ you'd have to resort to something like a MOSFET in the CT of the PT (see R.G.'s MOSFET Follies page).

The way you have it wired up now the screen current is being pulled through that 2k7 10W resistor, which may be more than it can handle. Adding another R-C branch from the first filter would be a better way to go, but you would want to make it a stacked arrangement to handle the high voltage. Something like a 3k3 or 4k2 10W resistor and a pair of 350V 47uF caps.

At 20mA you are biased pretty cold right now, somewhere around 0.02A*538V/30W=36% of max. About 30mA would probably be better.

I get about 60W on the load line with the voltages and primary impedance you listed, but it will be less than that in the real word.

Did Heyboer respond to your inquiry about a replacement UL OT?

[gui_tarzan]

No, and they usually respond fairly quickly so I sent another email last night.

[pdf64]

Bias: -59v
V7 & V8
Pins 3: 538v
Pins 4: 440v
Pins 8 to grd via 1 ohm resistor: 20mv
...
The BM Amps calculator shows that should produce 79w

I guess those are static voltages?
For the calculator to give accurate, theoretical max results, the voltages need to be measured at full load, ie max sine into rated load before clipping.
Expect plenty of B+ sag.

[Stevem]

I hope your stack of the two main filters are rated for 350 volts each with that 538 volts sitting on them?
My silverface UL PR is pulling the plate voltage down to 480 at idle with the outputs drawing 38ma each.

[gui_tarzan]

Yep, they're F&T 220/300v in series.

My customer said to give the Tung-Sol 6V6GTs a try so I still want to lower the first stage of the power rail closer to 450v (or lower) if possible.

[gui_tarzan]

Bias: -59v
V7 & V8
Pins 3: 538v
Pins 4: 440v
Pins 8 to grd via 1 ohm resistor: 20mv
...
The BM Amps calculator shows that should produce 79w

I guess those are static voltages?
For the calculator to give accurate, theoretical max results, the voltages need to be measured at full load, ie max sine into rated load before clipping.
Expect plenty of B+ sag.

Yes, they are idle voltages with all the tubes in place.

[gui_tarzan]

This is what I had in mind. Tell me honestly if this will work reliably or not.

[martin manning]

It'll smoke! Calculate the current and power through those resistors and you'll see. The schematic mark-up and the drawing don't match, BTW.

[gui_tarzan]

Thanks Martin, I figured as much even though the source I got it from said "no problem". I just didn't feel right about it, which is why I asked.

So what is your opinion of a MOSFET voltage reducer or "power scaler"?

[gui_tarzan]

I think I need to re-think this whole proposal, there's no way I'm going to be able to drop the B+ voltage enough to run 6V6s in this amp. It just wasn't designed to do that.

Back to the drawing board...

[gui_tarzan]

After careful consideration, I'm not going to mod this amp for 6V6s. It's way too much work and taking a real chance on it not being a reliable amp.

Thanks for the suggestions guys, but I think I'll put this one back to stock and move on.

[potatofarmer]

I posted in your other thread, but for anyone else interested in significantly reducing the power in one of these late Silverfaces with the center-tapped bridge rectifier, look at the "1/4 power" switch on the "red knob" Twin schematic. 6V6s not necessary.