ESL bias supply the GE way

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hey

So I was looking for some computer motherboad capacitors and found a source. Liked what I saw so I looked further. I found 1N4007's for $0.02 each for 100. I'm thinking of making a 25 stage multiplier bias supply run directly from "you ess aaa" mains (120 volts RMS 60 Hz). I made a 6 stage multiplier with 1N4006's and 1n4007's and 0.01uf 630volt caps before and it worked, but the source for those parts sucked.
So I want to know, do you think I should use bigger caps at the bottom of the ladder to stiffen the circuit?

Jim
 
Yes, using bigger caps at the bottom of the ladder is more effective. Interestingly, as you add more stages to a conventional multiplier (all caps equal), you eventually reach a point where adding more stages actually decreases the output voltage.

Attached is a text file of notes I made a couple of years ago when I looked into this stuff. Can't remember where I got the info, but maybe it'll help with your calculations / design.

Good luck - Godfrey
 

Attachments

  • V-mult.txt
    1.7 KB · Views: 179
Hi,

25 stages are absolute overkill and won´t give the calculated output voltage of ~8.5kV. The problem beeing that the caps appear in series connection, so their overall capacitance value becomes small and their impedance becomes high. This reduces the current capability of the circuit and the voltage stability under load seriously.
I wouldn´t opt for much more than 5 stages.
I would opt for a galvanic barrier, by using two trannies back-to-back and to achieve a higher voltage level at the multipliers input.
I don´t know the american legislative situation, but in a lot of countries a galvanic coupling of a circuit with the powersupply lines is not allowed and in case of an accident it gives the insurance company a well defined reason not to pay.

jauu
Calvin
 
When I look in my own audiostatics (commercial brand) they did make their HV-supply directly run from the mains, so without any transformer at all.
These speakers are pretty aged (though completely refurbished) and possibly audiostatic may use a different aproach these days.
I don't know to what extend this may lead to dangerous situations, to itself or to other equipment. But these hv-supples seem to work very reliable.
 
Two motivations for using a transformer:

Galvanic isolation - Wikipedia, the free encyclopedia

"Galvanic isolation is used in situations where two or more electric circuits must communicate, but their grounds may be at different potentials. It is an effective method of breaking ground loops by preventing unwanted current between two units sharing a ground conductor. Galvanic isolation is also used for safety considerations, preventing accidental current from reaching the ground (the building floor) through a person's body."
 
One way to make a long ladder with smallish capacitors is to increase the frequency. As others pointed out, at 50/60 Hz you can only do so many stages before the ladder voltage doesn't increase anymore. But, if you make a small inverter with two transistors and a small ferrite transformer (or scavenge it from a CCFL inverter), you can get pretty far with 10nF caps...

If you power the inverter from a +24V wall wart, safety is taken care of, too...

Kenneth
 
When I look in my own audiostatics (commercial brand) they did make their HV-supply directly run from the mains, so without any transformer at all.
These speakers are pretty aged (though completely refurbished) and possibly audiostatic may use a different aproach these days.
I don't know to what extend this may lead to dangerous situations, to itself or to other equipment. But these hv-supples seem to work very reliable.


I rebuilt a set of ES50's and they did NOT hook directly to the mains. They used the ever-polular filament transformer backwards driven from an AC output wall-wart.

http://quadesl.com/pdf/es50_schem.pdf


Sheldon
 
Thats the method I had used here.


http://www.diyaudio.com/forums/plan...tor-insulation-mylar-coating.html#post2531218


Just make sure that recovery time of the diodes is suitable for the frequency that you are driving them them with. jer

That's about it. But there's really no need for the 555. Just drive the bases of the transistors from a feedback winding (mind the phase ;-) ). That way the inverter will automatically adjust its frequency to the load, so you don't need to trim the frequency for maximum output voltage.

Kenneth
 
That an excelent way to solve the oscillator issue,Kenneth.

The only reason I did it the way I did was because my transformer had a rather sharp
Q and I wanted it to have a variable output voltage and I wasn't sure if the output regulation and oscillator startup would be relieble if I varied the supply voltage feeding the FET's.

Another method that maybe more relieble once the resonate frequency of the transformer is detrimined is to drive the transformer with a small power opamp such as an LM2003 or LM1875 driven by an opamp sinewave oscillator fed through a volume control to determine the output voltage level. jer
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.