Simplistic MosFET HV Shunt Regs

Jump to Latest
P

poty

Member
Joined 2011
Abszero said:
Thank you Andrew for make things "easy" to understand...
Click to expand...
I think the initial drop of the word "raw" played bad role in the following discussion. :) Andrew was advising the properties of the power supply as a whole and you thought about your question about raw power supply.
If you ask my opinion - I'd go as simply as possibly: C(RC)^n. SSHV consists of two parts and the first part (working in its technical boundaries certainly) isolates raw power supply from the rest of the circuit. Why do not to save the original idea of CLC for example? CLC produce very low frequencies by its nature. While the first part isolates the frequencies well (actually - the lower, the better), the shunt (second) part gradually decreases its effectiveness after around 100Hz into lower frequencies. So I'd prefer to stay, say, in the range of 100Hz where both first and second part has their almost best properties. The simple raw power supply also technically consistent with the Simple Shunt Regulator in design. :)
 
P

poty

Member
Joined 2011
Well, not exactly in such terms. "Stab part" (actually - constant current source) gradually looses its effectiveness with frequency increase (I hope I said it right in English. :) ). It does not have exactly 100Hz turning point or something like that. On the other hand the "shunt part" definitely shows the area of most effectiveness: from something like 50Hz to 30-50kHz in the rough estimation (the graphs appeared in the thread several times).
 
Abszero

Abszero

Member
Joined 2012
poty said:
I think the initial drop of the word "raw" played bad role in the following discussion. :) Andrew was advising the properties of the power supply as a whole and you thought about your question about raw power supply.
If you ask my opinion - I'd go as simply as possibly: C(RC)^n. SSHV consists of two parts and the first part (working in its technical boundaries certainly) isolates raw power supply from the rest of the circuit. Why do not to save the original idea of CLC for example? CLC produce very low frequencies by its nature. While the first part isolates the frequencies well (actually - the lower, the better), the shunt (second) part gradually decreases its effectiveness after around 100Hz into lower frequencies. So I'd prefer to stay, say, in the range of 100Hz where both first and second part has their almost best properties. The simple raw power supply also technically consistent with the Simple Shunt Regulator in design. :)
Click to expand...
Poty,
Thanks for your thoughts. Is there any advantages to go CLC over CRC?
CLC could have a lower impedance compared to crc. Or that low Z for the first part(raw part) is not very important because the SSHV will do the job?
Cheers,
Radu
 
Salas

Salas

diyAudio Chief Moderator
Joined 2002
Paid Member
The CCS part will do a good ripple isolation job in general over a wide enough practical frequency band since its a cascode of low capacitance depletion mode MOSFETs. But EMI/RFI are nasty things that have a tendency to couple all over the place in many ways, so pre-filtering is a good thing. CCS is high Z by nature so it does not ask to be fed by low Z.
 
P

poty

Member
Joined 2011
Abszero said:
Is there any advantages to go CLC over CRC?
CLC could have a lower impedance compared to crc. Or that low Z for the first part(raw part) is not very important because the SSHV will do the job?
Click to expand...
As it is already mentioned - the low-Z output impedance is the job of the SSHV.
CLC may be needed for the power transformer to work properly. There are always some choices in the balance between current and voltage. Theoretically we could have the amplitude voltage of the secondary(~1.4*Vrms) at the output of a power supply, but then we expect that current must be very small. We can have the current at the output equal to the RMS current of the secondary minus the losses in the rectifiers and filtering, but the output voltage will be definitely lower than secondary RMS. In practice most power supplies are somewhere in between the extremes. Chokes (versus resistors) allows to save some energy which otherwise is lost in filters and use the power capability of a transformer more efficiently.
 
B

bekim

Member
Joined 2004
I have received SSHV2 boards and I have couple concerns regarding assembly and capacitors, such:

1. Will it cause any problem if LED on board is connected on panel with leads approximate 10cm? (Oscillations, not recommended, etc.)

2. I believe that is matter of "taste" when comes to capacitors, and I can say that I went through this thread to see what other guys are using, but again your input is appreciated as you'll are experienced with this circuit.

What is your opinion on these caps?

- 10uF/400V Cornell Dubilier 935C4W10K (ebay ~$9, mouser ~$24)
- 0.33uF/400V WIMA MKP10 (ebay $4-$6)
- 10uf/400V MKP Solen (ebay ~$8)
- 0.33uF/400V MKP Solen (ebay ~$5)

- Jantzen Cross-Cap MKP 5% 400V - anyone used them on SSVH2?

3. Will MKP caps be "better" solution than MKT caps - C1 & C2?

Thanks
 
A

AndrewT

R.I.P.
Joined 2004
What is your range of mains supply voltage?
What effect will that have on the input voltage to your regulator?
What will be the input voltage when mains is at lowest?
What will be the dissipation in the CCS when mains is at highest?

These are the standard checks for a LV regulator.
For an HV regulator they become even more important. For a shunt regulator at HV they are doubly important.
 
You must log in or register to reply here.
Top Bottom