Simplistic Mosfet HV Shunt Regs

[CSlee]

Originally Posted by Salas

What changed in the full circuitry, at what point they started acting up? Are all semis in the regs showing working voltages like VGS, VBE? Was there a certain incident or revision, or an active part likely to have changed behavior in the client? A rewiring? Have you actually seen AC waveforms riding the DC line on a scope? When oscillating weird things happen all over, so its not unlikely for oscillation actuallly happening. Are there any capacitors from rail to ground or series connected parts on the rail after the reg's output? If yes, what are they?

Did some more testing this morning. Both regs work fine on thier own, but they go instable when running together. When both regs are running together I can not get anywhere near the voltage required for the circuit with the trim pots.

There is definite LF oscillation, when inputs are shorted there is large instability in the low end when reading a live trace. And there is a pulsing AC waveform on the DC rail. But this is not present when using one reg at a time.

The setup is;

CRC - individual RC per reg - SSHV2 - circuit. There are no filters after the regs just OPT and input stage anode load.

This is a new issue as I have been testing for a couple of weeks with no trouble at all, but after a break and coming back to it this has happened. I have not made any major changes to the ciruitry, only adjusting bias.

All transistors have working voltages. Is there anything else I can check?

Thanks

[Salas]

Only thing I can theorize is as the impedance of the regs rises towards DC due to finite capacitive filtering (10uF) and semis characteristics, some weird loop takes form in the specific circuitry and/or build. What values is the CRC?

[Salas]

Try to differentiate value of the last capacitor in one CRC for a test. By adding one extra at one reg's input as much close to it as you can.

[CSlee]

Values are 47UF - 1K - 220uF - 1K

I placed a final 47uF and small R right infront of the boards and they appear to be working fine. I will do the same with the second channel and hopefully that issue will not come back.

Thanks for your help!

[rclourenco]

Hi Salas, wishes of happy new year.

I need your help if possible, yesterday working late and tired, I did a stupid mistake with one working SSHV2 board, I connected my meter probes set to measure on the 20A range, to the output of the board, wanting to measure voltage

Of course this produced a nice small flash on the terminals/probes, and the regulator stopped working. The led is still on, but now I get only around 6,4V on the output.

What I need is for you to tell me in your opinion what should be changed to avoid taking all transistors out.
Many thanks.

[bmxmen]

Hi i have build my salas, but i have another board, little smaller. But similiar to the org salas. When I have attached power suply R1 burned, when i mesure the R1 on the board it has 97R, when I measure R1 in another board it have 120R( i have two salas's) i have 280vdc power suply, my target is 260vdc, for now my dummy load is 17k(15mA in 260vdc)What is going on? please help me.A have add zip with the eagle files.

[bmxmen]

I have search for the bugs, but nothink found. So i add a pictures.

[merlin el mago]

Why you have these logn cables for IRF840? If I remember correctly the gate resistor have to be near the IRF840.

[nicoch58]

clean well the pcb....

[bmxmen]

Sorry for pcb, for now dont have cleaner. I have measured thad R1 and R11 are domaged. Long calbes becouse i dont have site for radiator near the pcb. It is pasible that salas dont work coz this gate resistor? I will power on next salas.