The simplistic Salas low voltage shunt regulator

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
diyAudio Chief Moderator
Joined 2002
Paid Member
Its a juggling act since the available gate drive current reduces, so while the OLG goes up lower in freq due to more voltage drive it also starts falling earlier. That ups Zo before 100kHz and it could be associated with your preference for carbon film. All in conjunction with M2 type, Zobel, & specific active load. Also the phase margin changes so checking with oscope is essential as you mentioned. Its a tolerant topology but tweaking that far without visual test gear is not recommended to novice builders I guess.
 
Its a juggling act since the available gate drive current reduces, so while the OLG goes up lower in freq due to more voltage drive it also starts falling earlier. That ups Zo before 100kHz and it could be associated with your preference for carbon film. All in conjunction with M2 type, Zobel, & specific active load. Also the phase margin changes so checking with oscope is essential as you mentioned. Its a tolerant topology but tweaking that far without visual test gear is not recommended to novice builders I guess.

Salas thank you for explaining the effects of increasing R5. I am fully aware of the risks involved, but High End audio is often trail and error, and about pushing the limits. Sonically the circuit reaches a level of performance I have not heard before.

Peter
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Those were the last, yes. Two Reflektors in floating double polarity arrangement would be easier to implement for layout, are anti ground loop, also absolutely the same for semiconductors between the two polarities. Which can be many similar types. The CFP in the 1.2R can be sensitive and give pain in the making and deployment. I will show you a 24V example ASAP.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Here you are. This can go to 35Vout safely also. Spot on Vo its all about using the proper R8. Or a 10K trimmer. Not best choice for ppm but handy. For finding the right value including current and VF tolerances in a build and then substituting with a good R8 at least. Keep 10V drop over the reg for the 9610 to be on its low capacitive.
 

Attachments

  • Reflex24.png
    Reflex24.png
    28.3 KB · Views: 510
Hi, I encountered a problem with my BiB-based reg:

The CCS-LEDs light up, but the Vref LEDs stay dark (2 red LEDs). They are oriented properly and work if I apply external voltage to them.

The output voltage is 0V and the current draw from my lab PSU is not measurable (so below around 10mA), whereas it should draw around 100mA.

Did someone have this behavior before?
I have only a limited amount of FETs, so any educated guesses before I pull out random components are very appreciated :)
 
Hi, I encountered a problem with my BiB-based reg:

The CCS-LEDs light up, but the Vref LEDs stay dark (2 red LEDs). They are oriented properly and work if I apply external voltage to them.

The output voltage is 0V and the current draw from my lab PSU is not measurable (so below around 10mA), whereas it should draw around 100mA.

Did someone have this behavior before?
I have only a limited amount of FETs, so any educated guesses before I pull out random components are very appreciated :)
disconnect the shunt regulator part.
You then have only the CCS part.
Short the output of the CCS.
Check that the CCS is delivering and controlling the current into the shorted output.
Now insert a dummy resistor, 1r0 or 10r and repeat the CCS test.
The current into the dummy load will be a little bit less than the total current draw. The CCS uses a little bit of current to power itself.

Once you know that the CCS is working, you can put the shunt back into circuit.


For beginners this is the way to build up the Salas reg. Populate just the CCS part and test it. Then populate the shunt part and test it.
 
disconnect the shunt regulator part.
You then have only the CCS part.
Short the output of the CCS.
Check that the CCS is delivering and controlling the current into the shorted output.
Now insert a dummy resistor, 1r0 or 10r and repeat the CCS test.
The current into the dummy load will be a little bit less than the total current draw. The CCS uses a little bit of current to power itself.

Once you know that the CCS is working, you can put the shunt back into circuit.

Thank you, this is good advice, I´ll try.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.