Constant current source -3000V

[olvrkrg]

Cathode to heater voltage is max. 150volts for 4x150a, and I dont think its reasonable to heat swinging around 1500V at 2.6 amperes

 

[SY]

Re: Leakage

With the heater connected to the cathode , the transformer leakage capacitance needs to be driven 🙁
I'm also a 1KV virgin , even 700V scares the crap out of me . Mind you I've been lit up by the 1400V diaphragm supply a few times , not too bad , like a static shock but it brings home the dangers of what a large capacitance high current EHT supply can do

316a

Two interesting points here.

You can (can, not will!) survive a shock from a diaphragm or diaphragm supply because of the low current capabilities. The voltages for the power stage will come from sources that can run an electric chair. An unintentional contact with that is VERY likely to kill you.

I think you've nailed it that the transformer reactances are going to be important. And there's a design paradox: to reduce the stray reactances between this large transformer and surrounding components and grounds, you want to leave lots of space around it. But then you need long leads that are swinging high voltages, not a good thing.

 

[Jax]

Re: Re: Leakage

I think you've nailed it that the transformer reactances are going to be important. And there's a design paradox: to reduce the stray reactances between this large transformer and surrounding components and grounds, you want to leave lots of space around it. But then you need long leads that are swinging high voltages, not a good thing.

That is indeed true. The majority of the stray capacitance will be between the heater winding and the electrostatic shield in the transformer (it has to have a shield to avoid coupling signal back to the mains over the interwinding capacitance). I'm not sure which capacitance will be larger, this stray C or the ESL.

Anyway, the tapped choke idea seems to be the best way to go. Cascading several tube CCS will create the same problem with the heaters as with the SRPP.

But being openminded is a Good Thing 😀

 

[olvrkrg]

I have already thought about chokes. But I dont know who sells chokes, lets say 2500V, 50mA, 50H, and if I knew, I suppose four of them would not be cheaper than 500$. So why not a simple MOSFET ccs circuit? Any ideas?

 

[316a]

Re: Re: Re: Leakage

That is indeed true. The majority of the stray capacitance will be between the heater winding and the electrostatic shield in the transformer (it has to have a shield to avoid coupling signal back to the mains over the interwinding capacitance). I'm not sure which capacitance will be larger, this stray C or the ESL.

Anyway, the tapped choke idea seems to be the best way to go. Cascading several tube CCS will create the same problem with the heaters as with the SRPP.

But being openminded is a Good Thing 😀

I must admit I 'cheated' when I used a tapped choke . The choke was a 50VA mains toroid with the secondaries left open . The stators were connected using a pair of those lovely Russian teflon caps straight from the anodes . The output stage was a 7236 (a cross between a 6528 and a 6080) running at 220V so I'm yet to try toroids at higher HT , luckily my treble panels don't need that much juice to keep up with my Jordan JX92VTL's . The good news was it sounded lovely with the toroids 🙂 The bad news was that I had to tear the amp down to build another one but I'll be rebuilding the amp using 6S41 in push pull with the same toroid over the next week .

316a

 

[SY]

I admit that I cheated on my choke, too. I used an ESL drive transformer's secondary, with the primary left open. I'm experimenting with networks across the old primary to tailor the impedance. Hmmm, maybe that could help make a mains transformer work better? The main issue will be the degree of coupling between the two halves of the winding.