Simplistic MosFET HV Shunt Regs

diyAudio Chief Moderator
Joined 2002
Paid Member
Hi Salas,

What changes /or possible/ to SSHV2 for 180V / 120mA out?

Use 56K instead of 68K for the two in series Vref resistors so it starts lower for Vo. As for 120mA consumption I am afraid its not possible with the DN2540s cascode. You will need 140mA CCS for that and the Vgs of the upper DN is not large enough to open the lower one that much. Maybe if you will find an alternative depletion mode MOSFET for the upper one (input connected Q1) with same pin out but higher Vgs spec.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Thaks!
Then 120mA is the max value of the DN2540 ccs?

Depending in combo samples Q1 & Q2 the usual max is 90-100mA. You could check them on a breadboard before stuffing them, with low voltage something like 25V across, swap them for one another in a same CCS cascode circuit with the kit's same peripheral parts also. Find the one for Q1 that will yield the more current from the pair.
 
There are many examples how to do that. For example, for DN2540:
Look at page 4 of datasheet "Power Dissipation vs. Temperature", taking 2W dissipation (as mentioned by Salas) on vertical axis, got around Tc= 130°C/W on horizontal axis. Look at page 2 "Thermal Characteristics" for TO-220: θjc = 8.3 °C/W. Lets take max ambient temperature as 40 °C. Max temperature rising is 130-40=90 °C. Total temperature coefficient max: 90 °C/2 W = 45 °C/W, from which 8.3 °C/W is internally spent and around 1°C/W should be reserved for some losses in case-to-sink connection, so the max heatsink's coefficient should be 45-8.3-1=35.7 °C/W. Following Salas' suggestion of 30% degradation the final number should be 25 grad.C/W. SK104 heatsink has Thermal Resistance=14°C/W (less than counted), so in this case should be OK.
For IRF840 I got Vishay datasheet. 3.75W dissipation (from Salas). RthJC (Page 2, THERMAL RESISTANE RATINGS) = 1 °C/W, Linear Derating Factor (Page 1, ABSOLUTE MAXIMUM RATINGS) = 1 W/°C, Tj (in the same table) =150 °C. Max temp rising = 150-40=110 °C. Total temp.coeff.= 110/3.75=29 °C/W, heatsink capability = 29-1-1=27 °C/W -30% = 19 °C/W. Should be checked against derating: Total counted coefficient 19+1+1=21 °C/W at the 3.75W the temperature rising should be 21*3.75=79 °C. The datasheet mentions 25 °C as the nominal case temperature, we started at 40 °C, so for counting we should use 79+(40-25)=94 °C of temperature rising. At given derating factor there is max power dissipation of 125 - 94*1 = 31W - well over our planned dissipation. So even for this case the SK104 is still suitable.
 
Last edited:
diyAudio Chief Moderator
Joined 2002
Paid Member
then 1" q1 and 2" q3 SK129 are safer? Yes? How did you calculate this?

Thanks a lot

The dissipation I calculated by (200V-150V)*40mA for the DMOS CCS and 150V*(40mA-15mA) for the output MOS. The think of your sink as 30% smaller rule of thumb is because it is not driven evenly all over its surface as manufacturers spec but from a hotspot (MOSFET). So think of a 11C/W sink as 14C/W and so on and so forth. For the rest Poty wrote well.
 
Great performance in practice, tried to keep em practical with easy to source components. 2N6520 is a -350V PNP and KSP94 is an alternative -400V PNP for the higher voltage shunt, as 2SA1625 is. That is if you wanna go to that 360V mark or go to a higher range. The IRF840 works up to 500V.

Hi

I red most of the thread in last past days, and I have question regarding Shunt regulator, can this circuit be adopted for higher voltage than 360V?
I need quality power supply for my PSE amp with 300B, 400V I=160mA (~200mA)

In mean time I will build one circuit as you provide for my Aikido preamp, DC=250V, so as you're suggesting DC in should be ~300VDC to regulate, is this right?

Thanks
 
diyAudio Chief Moderator
Joined 2002
Paid Member
SSHV2 goes to 400V but not over 100mA. It can regulate over 20V DCin-out mains play safety difference. Good for your Aikido. For the amp you would be better off with a series MOSFET reg because of the heat a shunt reg would create even if you make a high voltage safe and high current 400mA CCS for it. Or if you even settle for a ballast resistor instead.