The simplistic Salas low voltage shunt regulator - Page 549

tpark · 21st October 2012, 07:53 PM

I have read a lot of this forum but not all, so I apologize if already answered. I have been trying to build a +/- V1.2R beta for a pre-amp I am building. Using a version of the F5 from Nelson only reduced currents to act as a pre-amp. Basically hope to replicate a B5 buffer as well all in one chassis. Have built a similar pre-amp for my son with a capacitance multiplier type of regulator which sounds great. I wanted to use a shunt regulator as have used before and liked their sound. To be honest, I have made some mods and use an SK170 instead of the 2N5457, dropped the 47k to 10k and use 100uF electrolytic to by-pass the 10k. I omitted the follower Q6 as the sk170's I have are around 8ma. I tried a follower and didn't make any difference. The current source part works fine. My problem: I cannot get the regulators to stay in regulation. The negative version is the more stable and starts up fine by itself but shortly afterwards the base of Q8 drops from .6 to a lower number and the source and gate of the pass device 9240 become the same value instead of the 4 volts difference. The positive side rarely regulates properly with the same issue and doesn't even start normally. They are conducting though as the current source is passing 333ma and both are well heat-sinked. I have replaced all the transistors several times to no avail. I use 750 ohms instead of the 1k but doubt that should be a problem. I use 4401 and 4403's instead of the BC series as I have lots of them. I have 120uF electrolyic on the output. and have tried various zobels. In or out they make no difference. I have been fighting this for several weeks so I am asking for some help. I suspect oscillation and my scope recently died so that is no help. I use the ground plane frog style for my work which has never been a problem for me. I do have a load attached and makes no difference whether in or out so doubt that circuit is the problem. If you can point me to a post that would help. Any suggestions. I don't really want to go to the Reflector but may have to.

Salas · 21st October 2012, 08:03 PM

Can you post a schematic of what you made?

tpark · 21st October 2012, 10:19 PM

Not good at drawing schematics but basically the same as your V1.2R with Q6 removed. Using 560 ohm as gate stopper on the Fet. 750 instead of 1K for R10 and 10k and 100 for R11 and C1. Moved R6 to the unregulated B+ to make sure the current source would start. Will move back when stable. Used .1uF PP at the unregulated as still bread-boarding and using aligator clips from PS. PS pretty much a stock Pass with CRC and R .22 with 40,000uF. Thanks for the quick reply. Suspect in need some pf somewhere and sometimes get about 7mv of AC when both positive and negative connected. < than .1mv when only use one-side. Must be some positive feedback somewhere.

Salas · 21st October 2012, 11:29 PM

Now that's a very different thing for OLG, phase margins, everything. What is the output voltage aim?

tpark · 21st October 2012, 11:34 PM

28volts in and 20 volts regulated. Do you ever sleep?

Salas · 22nd October 2012, 12:59 AM

To cut a long story short, that config you got will not work. Try this mix with the parts you described. R7 can be your trimmer.

Salas · 22nd October 2012, 02:14 AM

If it ticks, next step is you try restore the driver and the original termination, to see if that will work too. M2 can be IRFP9240 also.

RCruz · 22nd October 2012, 10:46 AM

If you do not connect R6 (Now R4 on latest salas schematic) to the regulated output you will be missing most of the magic of this design.

If it does not start up the ccs when you do so, there is a trick to cure this.... I will be following

tpark · 23rd October 2012, 12:19 AM

Thanks for your quick reply and also RCruz for your interest. This is a most helpful forum. I certainly intend to reconnect the bias chain to the regulated output when I get it stable. For the record I use an LM336-2.5 Instead of the led and by-pass it with a 1uF PP. I am pretty sure the CCS part of this is ok. I am busy with work most of this week but when I get a chance i will put in the degeneration resister. I plan to continue to use the IRFP9240 as I have quite a few of them. I use a 500ohm trimmer and when it works is around 100ohms with my 170. Again thanks for your interest and I will let you know how it goes.

orelli · 23rd October 2012, 08:11 PM

Hi submediterraneum guys, and rest of the world, of course, I've just tried an R core tx from Selectronic to power a TPA Metronome (upsampling device). The tx feed a hexfred diodes bridge and a 12r at 13Vin and 7Vout.The R core tx, respect to
Talema toroid, is more clear,trasparent, dinamic and with more silence between the musical passages. An upgrade on all the line! Then I tried the R core also to power my Buffalo II (also with 12r), but this time the suond was boring, without dinamic,uninvolving at all. A very strange thing! Anyone have compared toroids vs R core?

22nd October 2012, 10:46 AM	#5488
RCruz diyAudio Member Join Date: Sep 2006 Location: Saxon Blog Entries: 1	If you do not connect R6 (Now R4 on latest salas schematic) to the regulated output you will be missing most of the magic of this design. If it does not start up the ccs when you do so, there is a trick to cure this.... I will be following __________________ RC

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

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

21st October 2012, 07:53 PM	#5481
tpark diyAudio Member Join Date: Oct 2005	I have read a lot of this forum but not all, so I apologize if already answered. I have been trying to build a +/- V1.2R beta for a pre-amp I am building. Using a version of the F5 from Nelson only reduced currents to act as a pre-amp. Basically hope to replicate a B5 buffer as well all in one chassis. Have built a similar pre-amp for my son with a capacitance multiplier type of regulator which sounds great. I wanted to use a shunt regulator as have used before and liked their sound. To be honest, I have made some mods and use an SK170 instead of the 2N5457, dropped the 47k to 10k and use 100uF electrolytic to by-pass the 10k. I omitted the follower Q6 as the sk170's I have are around 8ma. I tried a follower and didn't make any difference. The current source part works fine. My problem: I cannot get the regulators to stay in regulation. The negative version is the more stable and starts up fine by itself but shortly afterwards the base of Q8 drops from .6 to a lower number and the source and gate of the pass device 9240 become the same value instead of the 4 volts difference. The positive side rarely regulates properly with the same issue and doesn't even start normally. They are conducting though as the current source is passing 333ma and both are well heat-sinked. I have replaced all the transistors several times to no avail. I use 750 ohms instead of the 1k but doubt that should be a problem. I use 4401 and 4403's instead of the BC series as I have lots of them. I have 120uF electrolyic on the output. and have tried various zobels. In or out they make no difference. I have been fighting this for several weeks so I am asking for some help. I suspect oscillation and my scope recently died so that is no help. I use the ground plane frog style for my work which has never been a problem for me. I do have a load attached and makes no difference whether in or out so doubt that circuit is the problem. If you can point me to a post that would help. Any suggestions. I don't really want to go to the Reflector but may have to.

21st October 2012, 08:03 PM	#5482
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	Can you post a schematic of what you made?

21st October 2012, 10:19 PM	#5483
tpark diyAudio Member Join Date: Oct 2005	Not good at drawing schematics but basically the same as your V1.2R with Q6 removed. Using 560 ohm as gate stopper on the Fet. 750 instead of 1K for R10 and 10k and 100 for R11 and C1. Moved R6 to the unregulated B+ to make sure the current source would start. Will move back when stable. Used .1uF PP at the unregulated as still bread-boarding and using aligator clips from PS. PS pretty much a stock Pass with CRC and R .22 with 40,000uF. Thanks for the quick reply. Suspect in need some pf somewhere and sometimes get about 7mv of AC when both positive and negative connected. < than .1mv when only use one-side. Must be some positive feedback somewhere.

21st October 2012, 11:29 PM	#5484
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	Now that's a very different thing for OLG, phase margins, everything. What is the output voltage aim?

21st October 2012, 11:34 PM	#5485
tpark diyAudio Member Join Date: Oct 2005	28volts in and 20 volts regulated. Do you ever sleep?

23rd October 2012, 12:19 AM	#5489
tpark diyAudio Member Join Date: Oct 2005	Thanks for your quick reply and also RCruz for your interest. This is a most helpful forum. I certainly intend to reconnect the bias chain to the regulated output when I get it stable. For the record I use an LM336-2.5 Instead of the led and by-pass it with a 1uF PP. I am pretty sure the CCS part of this is ok. I am busy with work most of this week but when I get a chance i will put in the degeneration resister. I plan to continue to use the IRFP9240 as I have quite a few of them. I use a 500ohm trimmer and when it works is around 100ohms with my 170. Again thanks for your interest and I will let you know how it goes.

23rd October 2012, 08:11 PM	#5490
orelli diyAudio Member Join Date: Aug 2008 Location: Correggio, Italia	Hi submediterraneum guys, and rest of the world, of course, I've just tried an R core tx from Selectronic to power a TPA Metronome (upsampling device). The tx feed a hexfred diodes bridge and a 12r at 13Vin and 7Vout.The R core tx, respect to Talema toroid, is more clear,trasparent, dinamic and with more silence between the musical passages. An upgrade on all the line! Then I tried the R core also to power my Buffalo II (also with 12r), but this time the suond was boring, without dinamic,uninvolving at all. A very strange thing! Anyone have compared toroids vs R core?