LCC design with these capacitors throughout: 80uF 500vAC motorstart caps
C1 is comprised of two of those motorstart caps in series, so is C2, four caps in total just for voltage doubling, does voltage doubling take kindly to voltage deviders parallel to each cap? or is this another wild goose chase?
If so what is a good value for this design? I'm not experienced enough to calculate it myself.
C3 are two of those motorstart caps in series, so is C4.
I know I'm not anywhere near experienced enough to design something so dangerous, but I can't help just fumbling about with theoretical designs.
Check out what PSUD had to say anyway:
C1 is comprised of two of those motorstart caps in series, so is C2, four caps in total just for voltage doubling, does voltage doubling take kindly to voltage deviders parallel to each cap? or is this another wild goose chase?
If so what is a good value for this design? I'm not experienced enough to calculate it myself.
C3 are two of those motorstart caps in series, so is C4.
I know I'm not anywhere near experienced enough to design something so dangerous, but I can't help just fumbling about with theoretical designs.
Check out what PSUD had to say anyway:
Attachments
Is there a suitable source for the choke?
I need about 1.2kV minimum, 1.5kV preferred max DC voltage headroom of the choke and preferrably about 50-150 ohms resistance.
I can easily pull down the final voltage just by choosing another power supply transformer, Sowter can help out here with their vast selection of fairly small voltage increases in all their models. ie, one model is 320v, another 340v, another 360v and so on.
I just need a choke...I could use a hammond with a 1kV limit but that limits me to about 850v, I wouldn't mind going up to about 1.2kV.
I need about 1.2kV minimum, 1.5kV preferred max DC voltage headroom of the choke and preferrably about 50-150 ohms resistance.
I can easily pull down the final voltage just by choosing another power supply transformer, Sowter can help out here with their vast selection of fairly small voltage increases in all their models. ie, one model is 320v, another 340v, another 360v and so on.
I just need a choke...I could use a hammond with a 1kV limit but that limits me to about 850v, I wouldn't mind going up to about 1.2kV.
You are having quite a conversation with yourself.
> preferrably about 50-150 ohms resistance
Why? 100 ohms 200mA is only 20V drop, quite negligible in the face of 1,000V. Either pay a whole lot for the choke or wind a slightly higher transformer to cover choke loss: simple economics say you probably want about 10% loss in the choke. The extra 100V from the tranny is cheaper than a low-R choke.
> a hammond with a 1kV limit
Put the choke in the ground leg. That requires a little better insulation in the transformer, and the first caps, but the choke insulation becomes unimportant.
1,000V at 200mA, 200 watts.... what are you going to build????
> I can't help just fumbling about with theoretical designs.
Here's the real sticking-point. Vacuum tubes up to about 500V-1,000V can use soft oxide cathodes. Lots of emission for the heater power.
But above 500V, secondary emission bombards the cathode and strips-off the thorium layer faster than it can self-heal. Generally you don't use oxide cathodes at 1,000V and up.
You can run much higher voltages with a tungsten filament. Very tough, secondary emission won't erode it. But very much less emission current per watt of heater power.
And also, generally, capacitors are cheap to 450V, not-expensive for 800V (series electrolytics), but over that range electros are too much trouble while oil/paper/film is much more expensive. Yes, mass-produced motor-caps do change that economic.
So a tungsten cathode working at 1,000V on the plate is usually a less efficient more expensive amplifier than an oxide cathode tube working at 500V-700V. If you can't get enough power at that level, the next step is to move to 1,500V-2,500V and make up with voltage what you lack in cathode emission.
You can easily get 100, 200, 300 Watts audio from 6550 at 600V. And 600V through your little finger will, at best, leave it numb for decades. A local electrician crossed 440V, burned 80% of his body and died. Not sure I want to mess with 1,000V at 200mA.
> preferrably about 50-150 ohms resistance
Why? 100 ohms 200mA is only 20V drop, quite negligible in the face of 1,000V. Either pay a whole lot for the choke or wind a slightly higher transformer to cover choke loss: simple economics say you probably want about 10% loss in the choke. The extra 100V from the tranny is cheaper than a low-R choke.
> a hammond with a 1kV limit
Put the choke in the ground leg. That requires a little better insulation in the transformer, and the first caps, but the choke insulation becomes unimportant.
1,000V at 200mA, 200 watts.... what are you going to build????
> I can't help just fumbling about with theoretical designs.
Here's the real sticking-point. Vacuum tubes up to about 500V-1,000V can use soft oxide cathodes. Lots of emission for the heater power.
But above 500V, secondary emission bombards the cathode and strips-off the thorium layer faster than it can self-heal. Generally you don't use oxide cathodes at 1,000V and up.
You can run much higher voltages with a tungsten filament. Very tough, secondary emission won't erode it. But very much less emission current per watt of heater power.
And also, generally, capacitors are cheap to 450V, not-expensive for 800V (series electrolytics), but over that range electros are too much trouble while oil/paper/film is much more expensive. Yes, mass-produced motor-caps do change that economic.
So a tungsten cathode working at 1,000V on the plate is usually a less efficient more expensive amplifier than an oxide cathode tube working at 500V-700V. If you can't get enough power at that level, the next step is to move to 1,500V-2,500V and make up with voltage what you lack in cathode emission.
You can easily get 100, 200, 300 Watts audio from 6550 at 600V. And 600V through your little finger will, at best, leave it numb for decades. A local electrician crossed 440V, burned 80% of his body and died. Not sure I want to mess with 1,000V at 200mA.
PRR said:
A stereo set of 813's in SE, A rehash of Pete Millet's SE 813
The 813 is Thoriated Tungsten...
You're right it is dangerous especially without any CT, I've still got to research how to properly ground this amp, one step at a time I don't want to muck it up.
I believe even 100mV ripple on 1200Vdc for the 813 SE amp is acceptable.
If it is a 10k to 8 ohm OPT, the hum voltage at the output would be extremely small.
In case you really want to get absolute no ripple, you may add another L-C filter.
Make sure that the capacitors and chokes you use can stand at such high voltage. You may find chokes which can stand 1.5kV/ For capacitors, you probably need to put two in series.
Mind you that you need to be very careful when working/testing around the components. The wires must be wrapped on the choke/capacitor terminals before soldering.
Good luck.
Johnny
If it is a 10k to 8 ohm OPT, the hum voltage at the output would be extremely small.
In case you really want to get absolute no ripple, you may add another L-C filter.
Make sure that the capacitors and chokes you use can stand at such high voltage. You may find chokes which can stand 1.5kV/ For capacitors, you probably need to put two in series.
Mind you that you need to be very careful when working/testing around the components. The wires must be wrapped on the choke/capacitor terminals before soldering.
Good luck.
Johnny
C1 and C2 need to be big. Way big. Fanarkalingly big.
Tim
An externally hosted image should be here but it was not working when we last tested it.
Tim
Indeed but the cheapness of this power supply quickly fades as you add another choke, if you wanted to though you could easily add a second LC filter and change both L1 and L2 values to 1 henry, this will give quite good results.kmtang said:
The caps are rated for approximatley 1400vDC or 1000vAC, the choke I'm hoping is rated for 1.5kV but I'm not holding my breath on that one.
Sch3mat1c said:
Yes the capacitance in a vdub is directly related to the amount of current available for the load, at about 60uF and below, then voltages begin to sag for 200mA.
We can take advantage of the fact that voltages are half at C1 and C2, so we don't need a 1.4kV capacitor, not even a 600vAC capacitor, the capacitors I've selected will do just fine by themselves for C1 and C2 with a max voltage of 500vAC or 700vDC!
So C1 and C2 doesn't need to be two 80uF 500vAC caps in series, they can be two 80uF 500vAC caps in parallel, or 160uF each, or more.
The final tally is, C1 = 80uF 710vDC, C2 = 80uF 710vDC, C3 = 2x 80uF in Series for 40uF 1400vDC, C4 = Same as C3.
Someone pinch me quick!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.