Simplistic Mosfet HV Shunt Regs

[buzzforb]

Should the CCS portion work regardless of the V setting portion? I am currently redrawing the my PCB layout and double checking as I go.

[Salas]

Originally Posted by buzzforb

I am reading about 100mv across R5, so that means the CCS is passing about 10ma. Great, right. Unfortunately the R in my prefilter CRC is passing almost 45mA of current, measuring about 10V across the 2W 220r. MY question is, where can this extra current be going if not through R5. FYI, the layout of my boards is an exact copy of the master, with the exception of being reversed.

That is odd. You should be passing on the filter resistor exactly as much as the reg draws. R5 is the only road for current to pass from the CCS part to the shunt voltage part in a correct working reg. Thus, you must check Q2, D2, R2 paths and orientations, health.

[Salas]

Originally Posted by buzzforb

Should the CCS portion work regardless of the V setting portion? I am currently redrawing the my PCB layout and double checking as I go.

If the CCS has a path to earth it can draw as a current drain alone.

[Salas]

Originally Posted by buzzforb

Zener? I am pretty sure the orientation is correct!

CCS sets top current, so its a form of protection against output shorts also, if the heatsinks can withstand for long. Two zeners in the CCS protect DMOS gates, one zener in the shunt part protects the Vref during start up, while the filter cap charges up. DMOS gate resistors should be really near their gates with rudimentary free lead length towards gates.

[buzzforb]

THanks Salas.

[Lazybutt]

Originally Posted by garbage

figured that out with the help of the sshv2 spreadsheet.

i can drop R5 to 1R for TP purposes, and use a R4 pot of 20R to adjust current.
R11 can be a pot of 200R to adjust voltage.

Q3 will dissipate lots of heat. a large heatsink is required.

Good calculation, and good choice of parts values ! R3 would also probably benefit from being a little smaller (50 to 100 Ohms)

There's one thing I'm not sure though...In fact my calculator does not consider Q1 effect since it can be neglected for usual currents... with cascoded Q1, you might not be able to go as high as 150mA.

=>From the datasheet, Rds of DN2540 is around 30 Ohm @150mA, so Vds of Q2 would be ~4.5V which puts Q2 Vgs way beyond the cutoff voltage.... if it's not enough you may still use SSHV1 type of CCS (very solid performance)

[buzzforb]

Got the CCS portion up and running. Set at 40mA. Now got to work on the voltage section. If the LEd is not lighting up, I have it backwards , correct.

[Salas]

Either a backwards LED or a dead FET, whatever. It goes by checking.

[buzzforb]

I suspect the LED. Skiddish with all this HV.

[buzzforb]

THe Led is OK. I have no Vdrop across the gate resistor for the 840. This means something has to be wrong with either the bipolars or the Jfet. If the LED is OK, and i have checked, then it must be that the Jfet is not passing any current, correct? If it was, wouldnt the LED light up even if there were problems elsewhere? If the Jfet was OK and the IRf840 was sttill not turning on, then it would be the bipolars. I have checked the F+/s+ and S0/F0 connection and it is good. I am not measruing any voltage at the output of the regulator. Sorry, I am trying to think my way through.

Additional INfo. There is no Vbe on the top bipolar.