Reverb Driver - why such high voltage?

[Buddha's Guitar Tech]

It is my understanding that a 12AT7 plate-cathode max voltage is spec'd around 250.

So how is it that, according to a Princeton schematic, we can get away with around 400?

The Princeton's reverb driver is a paralleled 12AT7 with 400 volts on the plates.
Since the two triodes are paralleled, does this change how each sees the available voltage?

How would the same layout with a 12AX7 at a lower voltage compare, i wonder.

[martin manning]

Parallel circuit elements see the same voltage just like always. In short, Fender just ignored the spec limit, and these amps are known to be hard on reverb driver tubes. The Twin Reverb schematic shows over 430V Va-k on the driver. For a 12AX7 version see the 6G16 Vibroverb, same high plate voltage, though. Later amps (Concert II, e.g.) had a 10k resistor in series with the reverb transformer secondary, which would lower the plate voltage some, but it's still ~400V.

[diagrammatiks]

the upper limit is actually the maximun average at the pin.

to figure out the maximun voltage before the dropping resistor you need to load line calculate the expected max dissipation of the tube.

[David Root]

This circuit is tough on vintage 12AT7s allright, never mind current production.

I have a '62 6G15 Reverb that I use a used '70s Sylvania 12AT7 in the driver position, ie nothing too expensive. Still sounds like '62 Prom nite!

I wonder if a 12BZ7 would do better, it's essentially two 12AX7s in parallel in the same envelope. Or maybe a 12BH7.

[Buddha's Guitar Tech]

i'm wondering more how lowering that voltage will change results.

If there was 250 on the plates, would it just be very weak?
(my gut says yes, but if anyone's experimented, why not ask.)

[martin manning]

There will be less voltage swing available, so less headroom before clipping, and I think that the clipping will be asymmetrical. Result: reduced and distorted reverb. The Concert II I mentioned (1980's) is the only Fender circuit I can find (with a quick look) that seems to be easing up on the driver's plate voltage, and I see it is biased hotter, probably to center the operating point a little better. The one-tube reverb circuit discussed here (TAG) also has a resistor (27-47k suggested) in series with the reverb driver transformer, and uses a single 12AX7 triode for the driver stage. I'd guess it runs mid-300's on the plate from a 440V supply

[diagrammatiks]

i'm wondering more how lowering that voltage will change results.

If there was 250 on the plates, would it just be very weak?
(my gut says yes, but if anyone's experimented, why not ask.)

again it depends on whether you are talking about before or after the anode resistor.

[Buddha's Guitar Tech]

i'm wondering more how lowering that voltage will change results.

If there was 250 on the plates, would it just be very weak?
(my gut says yes, but if anyone's experimented, why not ask.)

again it depends on whether you are talking about before or after the anode resistor.

I was under the impression that the reverb transformer primary was a kind of anode resistor.

[diagrammatiks]

right right.

at that point it's just about figuring out where the tube's biased on the load line.

you can drop the voltage but they run it as high as they do because a transformer coupled triode is already a really inefficient reverb driver so they need to drive it really hard for the reverb to work at all.

[Buddha's Guitar Tech]

"a transformer coupled triode is already a really inefficient reverb driver so they need to drive it really hard for the reverb to work at all."

That's exactly what i thought.
i wonder what other ways there are to do it.

[Firestorm]

Anybody have any thoughts on what voltage to run it push-pull instead of SE?

[diagrammatiks]

well if you were running it push pull you'd be sending a balanced signal to the reverb pan...

it's been done before in a couple of orange amps and sound cities before.

merlin has a good write up on the srpp on his website.

http://www.freewebs.com/valvewizard2/reverbdriver.html