0
\$\begingroup\$

YES, I understand corona, arcing, safety, and Ohm's law...

I am building a high power AC/DC variable power supply: Microwave oven transformers, CW voltage multipliers, flyback transformer, variac... Yes, the bits are potted, and in a massive heat sink with active air-flow. Adequate grounding, etc...

My question is: I would like to use a Raspberry Pi2 as a function generator/oscilloscope. I would use the analog GPIO for the function out, a bitscope for the imaging.

Do I control the upstream supply? The downstream power, or somewhere in between?

What is the best way to use the Pi to generate the signal (sine, square, etc.. vari freq from <1Hz to <1 GHz) then match that signal with the enormous values I get from the supply - a HVLC side, and a HCLV side isolated, and read the output on the scope image

I have a guy who will be coding the function generator, and the scope, so that I have preset wave forms, and display will be one period of the signal. The frequency change will be done via pot, the voltage with a Variac, and the current.. maybe a rheostat.

NOTE: The DC side will have signals: ON/Pulse - the pulse with variable dwell and delay, I thought to drive that with a simple relay.

Using 5vdc to control 0-1MvAC/DC (Not at the same time, and not necessarily that value for each mode)

It runs on 115vAC @ 60Hz. I had considered an LC for the variable freq, but this does nothing for different wave forms, and I do not believe that I can get the frequency range I am looking for with this method.

Thanks!!

\$\endgroup\$
  • \$\begingroup\$ Have a look at how cheap 40W CO2 laser powersupplies are made. They use a switch mode IC to drive transformer isolated transistors driving what is almost a fly-back transformer. They have a DC input voltage for control. A MOT is not suitable for frequencies far outside 50 to 60 Hz and varying your output voltage on the HV side is tricky and practical to about 0.8 kV for mere mortals. \$\endgroup\$ – KalleMP Sep 26 '16 at 11:57
0
\$\begingroup\$

I use cascaded string of R dividers for regulating feedback for <25kV and capacitive dividers for feedback of regulating the primary up to 1GV which is done by several companies incl. Chinese and ABB. Since both caps charge up during power on the voltage DC or ac ratio is based on C ratio.

CUrrent and voltage sensing can be at suitable stages in primary and on DC supply for arc detection, and fault protection. Partial Discharge sensing is advised when contamination is an issue on insulation. DC ratings must be derated from BIL ratings for lightning surge due to ionic contaminants.

Marx generators are the typical mode for generation of UHV low power, using leather friction belt motors and stepper actuated gaps to,parallel charged caps and triggered by a pulse to make all the spark gaps as a series switch for each cap. I have used these up to 200kV for testing, using two well cleaned BIL200 bushing insulators for both impulse and DC.

For very low currents like on non-ionizing LASER Lexmark printers, I developed a resonant stepup XFMR oscillator using LM358's with R feedback and PWM control from +/-15kV within 1% with no shielding required for $15 BOM cost in volume

| improve this answer | |
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.