Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Simplistic Mosfet HV Shunt Regs
Simplistic Mosfet HV Shunt Regs
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
Reply
 
Thread Tools Search this Thread
Old 12th December 2017, 05:11 PM   #5761
DualTriode is online now DualTriode  United States
diyAudio Member
 
Join Date: Aug 2009
Location: Sacramento
Quote:
Originally Posted by Salas View Post
That's a really good harmonic noise result already I would think for a tube build considering its unavoidable extensive wiring, its big various components, the glass encased big metal elements exposed to probable fields, plus their wide area heaters circuit. Almost all of it at -135dB or under and the mains frequency at -120dB its no mean feat. Maybe by adding grounded shield caps on the tubes and by paying best attention to heaters PSU and its loop area can do little better. Heaters are already on DC with return side referenced to chassis I presume.
Salas,
After poking at this thing even more the line frequency ripple remains. *!Xit.

I unplug the heater supply transformer and the line frequency ripple spike disappears. So the ripple is related to the heater.

Trying a few things including a Tentlabs heater regulator and a different analyzer the shunt regulator magic really shines.

DT

12B4 amp
Attached Images
File Type: jpg 12B4 with regulators FFT.jpg (63.8 KB, 186 views)
  Reply With Quote
Old 12th December 2017, 09:56 PM   #5762
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Simplistic Mosfet HV Shunt Regs
We were right that the heater stuff was also suspicious for that little 60Hz spike then. Can even be just the magnetic field of the heater's trafo if its near. That's a very clean noise floor for a tube build shown in your picture. Since it sounds nice too its a winner.
  Reply With Quote
Old 17th December 2017, 09:22 PM   #5763
LuvValve is offline LuvValve  Australia
diyAudio Member
 
Join Date: Jul 2009
Salas,

I've finished building two SSHV2 with 420V output and 445V input. I've done following changes as you suggested in a previous post.

I replaced R10, R9 with 82K/2W resistor, and used large enough heatsink for Q5, Q4, and of course Q3. The load is less than 5ma per channel, so I adjusted R4 to have around 27ma through R5.

Everything looks ok, however I experience large voltage drifting from high to low. Under room temperature I have 435V output when firstly turned on then drop to 412V after 20min. I'd rather have lower voltage output when turned on then increase and stay stable after 30min

Maybe Q6 is the problem ? what would you suggest me to look at? Thanks in advance
  Reply With Quote
Old 17th December 2017, 11:14 PM   #5764
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Simplistic Mosfet HV Shunt Regs
Blow over Q6 area and if the voltage changes fast then that's the one plus the BJTs Vbe drifts. Its a very high I/V conversion ratio from 2.56mA to 420V so only 0.14mA IDSS thermal drift can give you that 23V delta.

Short the LED and see if the warm up drift behavior changes at all first.

There are several Watt burning on the nearby sinks and radiating heat. Adjust the CCS at 20mA to save on some heat.
  Reply With Quote
Old 18th December 2017, 04:02 AM   #5765
LuvValve is offline LuvValve  Australia
diyAudio Member
 
Join Date: Jul 2009
Thanks Salas. I replaced the Q6 from another batch, it looks I get some improvement. The voltage goes to 428V then drop to 420 after 5 min then stay stable. With the LED being shorted it does not change drift behavior.

I also found that heatsinks for Q4, Q5 are a lot hotter than Q1

I'm wondering why the output voltage goes all the way up then drift down, not other way around?
  Reply With Quote
Old 18th December 2017, 10:53 AM   #5766
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Simplistic Mosfet HV Shunt Regs
Could be the Vbe of the hot cascode BJTs which is Vds to the JFET (Q5's) changes with warm up. In so high a Vout they have more voltage across than in other builds so they dissipate more power. If you can boost the Q4 & Q5 sinking somehow it would possibly help. I have also wondered about replacing Q6 with several J201 JFETs in parallel up to enough Idss and better noise because they are 5nVrtHz FETS which also have far lower Gos and very low Idss individually, that makes them steadier as CCS units, but the whole thing could go chaotic when many (?) plus they are also NOS with lots of fakes floating around except in SMT or in expensive InterFet brand, there will be selection needed and a stamp size board to hold them either way. Big experimental trouble to take I would think for bettering (if successful) an 8V drop thermal drift in the scale of 428V after just using another K117 low noise Q6 sample with possibly different Vgs (off).
  Reply With Quote
Old 20th December 2017, 08:03 AM   #5767
LuvValve is offline LuvValve  Australia
diyAudio Member
 
Join Date: Jul 2009
The only parts that are running very hot are Q4 and Q5 in my build because of much higher output voltage. So, this has caused the output voltage to drift lower as temperature goes up.

If I increase the value of R9, R10, would this reduce power consumption across Q4, Q5? I have replaced a pair of relatively lager heatsinks for Q4 and Q5, they only helped a little. I also tried to blow air on Q6, I don't see big change at the output.
  Reply With Quote
Old 20th December 2017, 09:02 AM   #5768
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Simplistic Mosfet HV Shunt Regs
R8 determines the bias current in the cascoded BJTs not the Vref resistors. Try R8 2.7k.
  Reply With Quote
Old 21st December 2017, 01:28 AM   #5769
LuvValve is offline LuvValve  Australia
diyAudio Member
 
Join Date: Jul 2009
With R8 being 2.7K, it reduces about 30~40% of the heat on Q4 and Q5. The output still goes up to 428V when turning on but quickly goes down to 420V within around 1 min, then stays stable even after 1 hr. it's not too bad ! Thanks Salas.
  Reply With Quote
Old 6th January 2018, 03:20 AM   #5770
tim845 is offline tim845  Hong Kong
diyAudio Member
 
Join Date: May 2004
Location: hk
Would anyone tell DN2540 be destroyed by electrostatic charge easily? I bought 4 of them and totally broken after welded to the PCB.
  Reply With Quote

Reply


Simplistic Mosfet HV Shunt RegsHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Regs - Shunt vs Series richie00boy Solid State 59 17th June 2017 03:59 PM


New To Site? Need Help?

All times are GMT. The time now is 04:19 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki