Simplistic mosFET HV Shunt Regs

[Salas]

Dear Salas,
Assume the master regulator retain in the power supply, what would happen if replacing the slave regulator (the one close to signal path) with your SSHV2?

What the differences between slave regulator and your SSHV2 for this application?

Thanks!

If done well with no oscillations and no additional interference noise, then some subjective differences between series with chip and a simple shunt I would guess. Don't try that without a scope.

 

[merlin el mago]

Its little CCS range for Q1 & Q2 nonetheless. They usually go up to about 100mA in this arrangement. Measure R5. Is it 10R indeed? If no and its higher by mistake restore to 10R. If yes its 10R, weird CCS result if so, try 1R (1mV=1mA TP). Also check if R4 trimmer & R3 resistor are good.

All CCS measured good: resistors, trimmer, semis & diodes.

Changed 10R for 1R and now can reach 30mA but max Vout is 107V wiuth 4K7 dummy load.

 

[Salas]

What Vgs you got at Q1 & Q2?

 

[merlin el mago]

Black led probe to S
Q1 -0.368V
Q2 -1.562V

 

[merlin el mago]

Salas I think the SSHV2 is fine, attached simulation PSUD II of my PSU.

 

[merlin el mago]

Attached simulation with only 1 choke and input capacitor.

 

[Salas]

Your Q1 Vgs is low. Maybe that has to do with losses in the raw supply.

 

[Salas]

Try RCLC to make the curve climbing only and smooth up to 20V > target Vout

 

[merlin el mago]

What R value? Target Vout SSHV2 is 150V so 170V is OK.

 

[merlin el mago]

CCS only reach max 26mA so max Vout 126V both measured

 

[merlin el mago]

Salas I bought my DN2540 to RS Amidata but the manufacturer says Microchip in place of Supertex, link to specs DN2540N5-G | MOSFET, DN2540N5-G, N-Canal, 500 mA, 400 V, Reduccin, 3-Pin, TO-220 | Microchip

 

[merlin el mago]

Salas what can I do?

 

[Salas]

Make the CCS only on a breadboard and test Mosfets in LV that give normal SSHV2 larger adj range i.e. up to 100mA plus together. Move them to your board. If they don't repeat that on the board then the board or another component have a problem.

 

[merlin el mago]

One picture speaks more than 1.000 words.

 

[merlin el mago]

Now CCS

 

[Salas]

Was it limited due to a "suspicious" Mosfet sample?

 

[merlin el mago]

Not, the problem is AZ1 hybrid graetz with two UF4007, as you know AZ1 needs the first cap max rated 60uF and not 200uF, also I don't have clear if AZ1 needs more voltage on the anodes because data sheet tell 2 x 300VAC to have 100mA and increasing voltage decrease intensity and I'm using 125V transformer so possible AZ1 is giving 125 or 150mA or more.

 

[merlin el mago]

One picture speaks more than 1.000 words.

Now CCS

Used bridge UF4007 in place of AZ1 + two UF4007.

 

[Tubenstein]

Dear Salas,
Write to you on my friend's behalf. My friend unit has the attached regulator in the power supply. What will happen if replacing the existing one with SSHV2??

Any improvement? In what way?

Thanks!

 

[Vbenonisen]

I've changed a Solen 10uF output capacitor on the SSHV2 with a ClarityCap CMR and the results are absolutely amazing. Forget bypassing (trust me I've tried everything including silver foils and teflons). ClarityCap CMR is THE capacitor to rule them all. However they are big so place has to be made either above or under the circuit board.