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.

iko

Ex-Moderator
Joined 2008
Tham, here it is. 3A idle current through shunt mosfet on 0A load. Go hunting for a huge heat sink. Maybe you need a fan too.

Rset_current will set the current through the shunt. Higher value, higher current. You can use a variable resistor there, start with about 100R and go up.

There is a new feature, R4 + C3. You can adjust the output impedance by changing R4. Higher value of R4 means lower impedance. At the moment it is set to be sub milliohm all across the frequency range of interest. If you want, you can play with it, bring it up to 6k8, 7k5, 8k2, 10k, 15k, 20k, 27k, 33k, etc. I doubt that it will stay stable when the impedance is a few micro ohms, as it is capable of a lot of gain, that loop.

C1 and C2 are the high frequency compensation caps. See what works in reality. No need to go too low on those. The values shown in the schematic give a nice flat output impedance well into the high frequency. If unstable, choose larger values.

And, finally, the 2sk170, 7.5mA Idss might be ideal. Those are the values I simulated. Especially the J1, J3, and J4.

Good luck, and please let us know how it works out.

Edit: forgot, make C6 = 1000uF. I only use 100uF in simulation because it takes too long with large caps in that position. The 1000uF cap will pretty deal better with the zener noise.
 

Attachments

  • reg-tamp-v2p-2a.png
    reg-tamp-v2p-2a.png
    13.9 KB · Views: 1,785
diyAudio Chief Moderator
Joined 2002
Paid Member
Is that the 10W chip that needs about 13V? If that is, I have mentioned I think that there is one running from 2008 on V1. We had it on 3A, driving Hemp 8 inch, 8 Ohm FRs, then we got it down to 2.5A, no compression. My friend's comment is that it is better than his 300B with such a reg. Its hot though, because T-Amps are asking for nothing when idle. So we have a cold amp and a hot reg. Life's irony.:cool:
 

Attachments

  • t-reg.gif
    t-reg.gif
    6.6 KB · Views: 1,843

iko

Ex-Moderator
Joined 2008
Salas said:
Is that the 10W chip that needs about 13V? If that is, I have mentioned I think that there is one running from 2008 on V1. We had it on 3A, driving Hemp 8 inch, 8 Ohm FRs, then we got it down to 2.5A, no compression. My friend's comment is that it is better than his 300B with such a reg. Its hot though, because T-Amps are asking for nothing when idle. So we have a cold amp and a hot reg. Life's irony.:cool:

http://www.kafka.elektroda.eu/pdf/tripath/TA2020.pdf

Sometimes I do shudder when I think how much current goes through the thing while it does nothing at all. But then I remember about filaments on tubes and don't feel that bad :D

Quite impressive that it sounds better than the 300B. Wow.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
So don't worry it surely works.;) And it works for long enough, must have been Dec 2008 or Jan 2009. The bass, the 3D, background silence, and liquidity gains are amazing, it really transformed. Like a turbo 300B. Use a big output cap, sounds better on an amp. I used 2200uF Pana FC. I don't know if it can disturb V2 though.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Try V2, its the same, only better. Cascoded and buffered. I don't see any reason not to work and be even a bit more cleaner. If you encounter any practical problems, revert to the V1 I have shown. That has been tested on a T-Amp. (Its 2X6-8W really).
 

iko

Ex-Moderator
Joined 2008
I got oscillation with a 2200uF on the output. I don't trust the simulator that much. Probably best to try some values in practice and see what's best. I've been thinking, and I'm not sure that for a power amp extended bandwidth on output impedance is something that one would want. Perhaps best would be a flat output impedance and phase in the audible frequency range?
 

iko

Ex-Moderator
Joined 2008
Tham said:
Iko, just to confirm if C6 is 100uF and NOT 1000uF as pointed out by Salas. Is R1 at 0.22ohm?

C6 is 1000uF; you can omit that resistor. I use it just to measure the shunt mosfet current.

Edit: I'll post later a simplified schematic. One of the side effects of running so much current is that the output impedance is very low even without the cascode, so that the top transistor in the cascode can be taken out.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.