Simplistic Mosfet HV Shunt Regs - Page 306

albertli · 27th May 2012, 04:53 AM

Quote:

Originally Posted by Salas

My mind went to the tiny trimmer wiper now. Maybe a resistor in trimmer's place after you check the set value?

The idea did go over my mind once, but I gave up the thought for the reason that both channels with this symptoms the same time. I checked the temperature on the 170 with a laser gun. It showed almost 40c at the flat surface. The R4 100k 2w is rather hot at 60c and kind of close to the 170. Don't know if the R4 reflects the heat to the 170 or..............other reasons. Is R4 supposed to be this hot.

triode_al · 27th May 2012, 08:17 AM

Quote:

Originally Posted by Salas

Albert, is it a PCB or P2P SSLV1?

Very P2P.
Agree to this weirdness (I can't stand things I don't understand

)
I changed the trimmer 500 ohms (NOS) to new Bourns 200 ohms multiturn but that was no solution; replacing the MJE350 gave the improvement (that went away a bit later again).
My high voltage supplies are in the lab supply just 16uF/L/6uF/4uF and in my RIAA preamp it is 4 uF/L/30uF. Could it be these race from 250V DC to zero faster than the reference capacitor bleeds down? (But then, I have now installed a zener over there, of which the reverse diode is very effective in protecting the MJE350 B-E junction.)

So I did touch the SK170 (experiment somewhat like Albertli did): immediately the voltage drifted a bit. It went down tens of mV (on the scope it drifted).
If thermal responsiveness of the 2SK170 is a cause, I can check it with a 2SK30, which is known precisely for that good point: no or little thermal drift. (I choose that jFET for my lab supply for that reason .. even though the 2SK170 has a lower noise).

The output behaviour itself is quite acceptable now, just I know that this type of noise is transmitted to an output quite quickly (standard anode loaded triodes have low PSRR), while 50 Hz hum goes rather undetected, these DC-shifts can be experienced more.
albert

Salas · 27th May 2012, 11:25 AM

Albert & Albert:

So the pots are good.
40C on the K170 is high to be self produced in that bias it uses. It should be radiated from a nearby object. Either a more indifferent fet or some way to protect it from varying heat. A 5W R4. I have experience with 2SK30ATM, problem is in this app it should not have more than 0.6V VGS(OFF) so the under 1mA IDSS ones from a bag should be used. You may need higher Rvref to reach previous Vo. Its noise figure is low, but at very high impedance. We still give it ~100K here so it should be quiet. A low side IDSS BF245A is a good option too.

triode_al · 27th May 2012, 03:11 PM

I've been looking lately at the 2SK246 too; this is very linear at low voltages (better at a lower Vds at low Id than both 2SK30 and 2SK170).
The other trick could be to hide the main jFET with a cascode. It makes the jFET more temperature stable and might have an improvement too of the input capacitance of the gate. [In fact, a cascode is what I used a tube for in my HV supply]. I would still call this simplistic, as the main error amp is the MJE350/powerFET.

I'll try this asap

triode_al · 27th May 2012, 04:41 PM

Oops, of course a cascode won't work at 0,6 V

Sorry for that, won't even try it...

JOIMONF · 3rd June 2012, 08:45 PM

After very good results with SSLV1.0 to power a Tripath, I am trying to run now my first SSHV2, but after several attempts, still no success. I have burned some zeners, one KSA1381 and three DN2540 yet.
Now I have only one DN2554 healty and miss another one to continue amusing this way.
My cuestion, do is possible to place as Q1 or Q2, IRF9610 ?
Input voltage is 310v and output should be 225v/60mA.
Thanks in advanced.

Salas · 3rd June 2012, 08:52 PM

Lower the input-output drop down to 20V-50V. Use CRC in your rectification for that. Try a capacitor directly across SSHV2's DC input for decoupling.

merlin el mago · 3rd June 2012, 09:00 PM

Let's go Jordi

Salas · 3rd June 2012, 09:02 PM

P.S. No, those DN are 400V depletion Mosfets, you can't sub them with 9610. Some info about your rectification stage and your load application could help assess.

JOIMONF · 3rd June 2012, 09:56 PM

On pic attached I hope it appears the circuitery on rectification stage. Yellow marker are components deleated, and blue color marker is 4k7 resistor value I changed to raise up the voltage (now I see that's wrong) There is 320v on secondaries, then CRC 100uF/4k7/100uf after modification.
This powers a pair ECC81, on a buffer stage, eating 225v./60mA.
I placed a dummy load resistor: 13k7/8w.
And expect to fix 800mV. on TP but never got start the module.

27th May 2012, 03:11 PM	#3054
triode_al diyAudio Member Join Date: May 2004 Location: Westland Blog Entries: 1	I've been looking lately at the 2SK246 too; this is very linear at low voltages (better at a lower Vds at low Id than both 2SK30 and 2SK170). The other trick could be to hide the main jFET with a cascode. It makes the jFET more temperature stable and might have an improvement too of the input capacitance of the gate. [In fact, a cascode is what I used a tube for in my HV supply]. I would still call this simplistic, as the main error amp is the MJE350/powerFET. I'll try this asap __________________ DAC TDA1541 S1; MC30-Super --> two stage RIAA; SP-6 clone ; 300B PP: 6N7/ECC82, Metalimphy output.

27th May 2012, 04:41 PM	#3055
triode_al diyAudio Member Join Date: May 2004 Location: Westland Blog Entries: 1	Oops, of course a cascode won't work at 0,6 V Sorry for that, won't even try it... __________________ DAC TDA1541 S1; MC30-Super --> two stage RIAA; SP-6 clone ; 300B PP: 6N7/ECC82, Metalimphy output.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Regs - Shunt vs Series	richie00boy	Solid State	42	14th March 2013 06:45 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

27th May 2012, 11:25 AM	#3053
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	Albert & Albert: So the pots are good. 40C on the K170 is high to be self produced in that bias it uses. It should be radiated from a nearby object. Either a more indifferent fet or some way to protect it from varying heat. A 5W R4. I have experience with 2SK30ATM, problem is in this app it should not have more than 0.6V VGS(OFF) so the under 1mA IDSS ones from a bag should be used. You may need higher Rvref to reach previous Vo. Its noise figure is low, but at very high impedance. We still give it ~100K here so it should be quiet. A low side IDSS BF245A is a good option too.

3rd June 2012, 08:45 PM	#3056
JOIMONF diyAudio Member Join Date: Nov 2008	After very good results with SSLV1.0 to power a Tripath, I am trying to run now my first SSHV2, but after several attempts, still no success. I have burned some zeners, one KSA1381 and three DN2540 yet. Now I have only one DN2554 healty and miss another one to continue amusing this way. My cuestion, do is possible to place as Q1 or Q2, IRF9610 ? Input voltage is 310v and output should be 225v/60mA. Thanks in advanced.

3rd June 2012, 08:52 PM	#3057
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	Lower the input-output drop down to 20V-50V. Use CRC in your rectification for that. Try a capacitor directly across SSHV2's DC input for decoupling.

3rd June 2012, 09:00 PM	#3058
merlin el mago diyAudio Member Join Date: Sep 2009 Location: Catalonia new Europe state	Let's go Jordi

3rd June 2012, 09:02 PM	#3059
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	P.S. No, those DN are 400V depletion Mosfets, you can't sub them with 9610. Some info about your rectification stage and your load application could help assess.

3rd June 2012, 09:56 PM	#3060
JOIMONF diyAudio Member Join Date: Nov 2008	On pic attached I hope it appears the circuitery on rectification stage. Yellow marker are components deleated, and blue color marker is 4k7 resistor value I changed to raise up the voltage (now I see that's wrong) There is 320v on secondaries, then CRC 100uF/4k7/100uf after modification. This powers a pair ECC81, on a buffer stage, eating 225v./60mA. I placed a dummy load resistor: 13k7/8w. And expect to fix 800mV. on TP but never got start the module.