HV Direct Drive tube amp.

[sasha]

Hello all,
I bought "TCJ Push-Pull calculator" software and simulated this circuit.
It looks very promising. It should be much better than Acoustat
at least it will have much more current for driving electrostatics.
And if bias is increased it will operate in class-A at least first few "Watts"
Now, there is a problem. Software doesn't calculate parts values
resistors, caps ...
And this is where I need your help, tube gurus.
I hope it is not too much work to do this.
Thanks in advance.
P.S.
output tube is 6HV5
gain stage, your choice

[Miles Prower]

WTF!

I'd say right now, you'd better rethink this. One of those tubes fails, and your output goes 2500V hot. That's sufficient to kill should you come in contact with the speaker connection. Furthermore, it's going to be extreamly difficult to reduce the DC offset to an acceptable minimum (and keep it there as the tubes age) with +/- 2500V rails.

Furthermore, check out the heater/cathode voltage rating: it's no more than 200V. The lower VT has that exceeded by better than an order of magnitude.

Nor is it a good design: worse than a Futterman, which isn't as good as a Circlotron.

This is a really bad idea.

[sasha]

Output of course will be coupled with capacitors, so if tube fail or dc voltage offset appears it is not concern. If You have heater transformer good enough to support voltage difference between heater and cathode
without arcing it is OK. Circlotron doesn't provide voltage gain
which is essential for this circuit. In that case as my simulation shows gain is 1.95 so I should have over 1KV drive to get max swing.
With original circuit I need only 6V for 4.5KV swing.
Regards.

[Geek]

Even if capacitor coupled, the charge of the cap at that voltage is enough joules to make you a vegetable in a wheelchair.... if you're lucky

I agree, this is a bad idea for permanent design, but maybe a good learning experience on the bench (mind furry 4-legged kids with no voltage sense in your home if you have 'em )

[sasha]

Hello,
Thanks for your concerns guys, but I think it is not really that much
dangerous simply because I have experienced 25KV discharge
from TVs CRT couple of times. And CRT can hold much more charge
than 50nF cap. It is very unpleasant get shocked by this voltage
but far from lethal or even dangerous. Only dangerous part is power supply because it will continue to send current not only pulse like
capacitor. So, stay away from PS and You are fine.
Maybe this fear of high voltage is preventing a lot of diyers to take on something like this.
Regards

[Miles Prower]

Maybe this fear of high voltage is preventing a lot of diyers to take on something like this.

"Fear" ain't got nuttin' to do with it. Big difference between fear and respect, and the prudence the latter demands. I am leery of any VT OTL for that very reason. It's OK for solid state. In that case, blow a final and you burn out your speeks. There isn't usually enough voltage to be deadly. Lose a final in a VT OTL:

Thanks for your concerns guys, but I think it is not really that much
dangerous simply because I have experienced 25KV discharge
from TVs CRT couple of times.

This comparison is not valid. The current requirements for a CRT are quite low, no more than 1.0mA, and so the CRT PS tends to be high impedance. There isn't enough current there to kill under most circumstances. The 2500V rails of this amp will have to source a lethal current, otherwise, it won't work at all.

And CRT can hold much more charge
than 50nF cap. It is very unpleasant get shocked by this voltage
but far from lethal or even dangerous. Only dangerous part is power supply because it will continue to send current not only pulse like
capacitor.

Wrong again! The nominal capacitance for a CRT is 365pF. At 25KV, that's: Q= CV.

(25E3)(365E-12)= 9.125uC

(2500)(50E-9)= 125uC

That's quite a bit more charge on that 50nF capacitor. In terms of energy stored: W= 0.5CV^2:

(0.5)(365E-12)(25E3^2)= 0.114 N * m

(0.5)(50E-9)(2500^2)= 0.156 N * m

You have more charge and more energy in that 50nF coupling capacitor. And you will be storing that much in case there is a failure of one of the finals.

Any way you look at it, this is not a safe design. It would be much better to use a coupling xfmr, so as to isolate that HV from any external connection.

If you insist on doing this, then I'd absolutely forget about trying to turn this into either a kit or production amp. This won't be something "cool" for you to share with anyone who isn't real experienced with HV circuitry.

[tubelab.com]

I believe that you are asking for some serious fireworks at the least, but this has been done before. I really like the design with 4 833A tubes per channel. I thought that I was crazy building an 833A guitar amp.

See this thread :

Tube amp specifically for driving ESLs

[sasha]

I don't know if you are familiar with Acoustat X commercially available
in mid 80's. This is exactly same amp I already have cloned.
It is up and running but since it is class-A amp doesn't have enough
current capabilities. All I wonted is to build push-pull capable of delivering more current and compare the two.
If You do the search for Acoustat on this forum You will find this link
http://www.twinstaticaudio.nl/twin-dual-mono/index.html and see
that Yes, commercial unit is available and probably approved to be sold. Beside all this I was looking for help finding parts values in
proposed schematic and not to argue safety.
Thanks.

[sasha]

Miles Prower,
I see You are good with math, maybe You can help me finding those values?

[gorgon53]

Hi
It wont work.
There is no way to get with those tubes (and the voltages you suggest) enough current to supply the heater transformers stray capacitance at the hf side of the spectrum. A few mA (or even tens) just arent enough. There are also corona effects that have to be taken into account.
wont work, forget it