The simplistic Salas low voltage shunt regulator - Page 13

Melon Head · 13th May 2009, 09:42 PM

Have a look at what I got (see attachment)

ikoflexer · 13th May 2009, 09:55 PM

Yes, that's what I got too. 33 milliohms almost all across. If you're happy with that then I'm happy too that you're happy. It could very well be that I'm chasing wild geese here. BTW, we got lots of geese here. I love how they gaa-gaa around the city. Many people don't like them because the s**t all over the place. As far as I'm concerned, I say long live the geese!

Melon Head · 13th May 2009, 09:57 PM

I just want to go back to my previous question.
If the max dynamic current draw of a preamp is 100mA then couldn't you bias it for 500mA. Then the draw from the powersupply would never change from 500mA.
Does this work?
Is there a problem with this?

Melon Head · 13th May 2009, 10:01 PM

Thanks Ikoflexer
You seem to know what you are doing and I am trying to learn as much as I can.
I appologise to everyone If I am hijacking this thread, but I am sure there are others as brainless as me around here that could learn along the way.

Melon Head · 13th May 2009, 10:13 PM

Well I would be interested in PCBs for both V1 and V2.
I also have access to the best measuring equipment money can buy, so if anyone wanted me to do some real measurements as opposed to simulated ones then I would be happy to do them.

ikoflexer · 13th May 2009, 10:30 PM

First things first. As far as I'm concerned, if you have questions, I see no reason why you should not ask them. salas said earlier something that a lot of people forget: these forums are here for some time, and people come after years something was posted, and learn something from your question or the answer to your question.

Now, about your question regarding the class A amp that draws 500mA at idle. You are assuming that this is a constant 500mA. The word "constant" is the key here. In real life it will not be as constant as you think. If it varies somewhere around the 500mA mark, and the psu output impedance is on the high side, the input voltage to your amp will vary too. By how much, that depends. Do you care about it? I don't know, that's something everyone has to decide for himself. Have you noticed something about salas? He seems to accept a circuit if it makes an audible difference. A lot of people will scorn this as subjective. I disagree with those people, because I think it's salas' right to decide for himself what sounds good. And it's your right to disagree. So, if you intend the regulator/psu to be part of some device that you ultimately will listen to, the ultimate test is to actually build it, listen to it, and then decide if it's worth it. Sorry to be long winded about this.

About measuring these circuits, I would certainly be curious about real life measurements. If anything, I encourage you to build v1. Many people are using that and are happy with it, after they compared it to alternatives. As far as v2 is concerned, that's very experimental still, and I'm still working on it. However, if anyone wants to go ahead and build it, who am I to disagree?

ikoflexer · 13th May 2009, 11:36 PM

I was checking out another regulator which is being worked on currently. It is an interesting shunt regulator too, and I was trying to help with its development a bit earlier. Today I managed to run it next to v2 and this is the output impedance plot.

We should not read too much into such plots, but v2 does look promising.

Salas · 14th May 2009, 01:23 AM

Try a run with IRF9510 on V2.

Then with IRF9610.

ikoflexer · 14th May 2009, 01:39 AM

A tiny bit worse with the irf9540. Here's something that surprised me with v2. The 9240s seem to work quite a bit better in this circuit than other mosfets. Even more of a surprise: using a darlington to drive the mosfets (both) gives a significant boost in performance. In particular I tried mpsa77. And this is while all other features stayed about the same (meaning it didn't start oscillating).

Edit: 9510 a lot worse (20-30db worse); 9610 also worse, but a bit better than 9510. Does it surprise you? Because it does me. In v1 9610 actually seems better (if I remember correctly) than 9240.

Salas · 14th May 2009, 02:12 AM

No it does not surprise me. I just know that people will say try ''faster'' Mosfets for V2 at a point.

Just see the transcoductance curves of faster TO-220 Mosfets, see the numbers they hover around, and watch how they diminish with heat. There lies the key. 9240 is 5 to 6 times stronger and it has a much bigger die, and body to sink.

9240 I did not pick just so. Its an all rounder, truly robust, cools off better, fastens easier. It will not oscillate, and gives better bass. Plus it scales up a lot. Its 12A. I use 3 times less than max for no problems with constant current. It can work up to 4A constantly, breaking no sweat with proper sinks. It already works for many months in V1 shunt @ 3A on a 2X10W T-Amp making it sound like a 300B SET but with better bass and analysis, and no DHT hum of course. Total background and buzz silence.

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

13th May 2009, 09:55 PM	#122
ikoflexer diyAudio Moderator Join Date: May 2008 Location: Toronto	Yes, that's what I got too. 33 milliohms almost all across. If you're happy with that then I'm happy too that you're happy. It could very well be that I'm chasing wild geese here. BTW, we got lots of geese here. I love how they gaa-gaa around the city. Many people don't like them because the s**t all over the place. As far as I'm concerned, I say long live the geese!

13th May 2009, 09:57 PM	#123
Melon Head Formerly Thanh1973 Join Date: Nov 2006	I just want to go back to my previous question. If the max dynamic current draw of a preamp is 100mA then couldn't you bias it for 500mA. Then the draw from the powersupply would never change from 500mA. Does this work? Is there a problem with this?

13th May 2009, 10:01 PM	#124
Melon Head Formerly Thanh1973 Join Date: Nov 2006	Thanks Ikoflexer You seem to know what you are doing and I am trying to learn as much as I can. I appologise to everyone If I am hijacking this thread, but I am sure there are others as brainless as me around here that could learn along the way.

13th May 2009, 10:13 PM	#125
Melon Head Formerly Thanh1973 Join Date: Nov 2006	Well I would be interested in PCBs for both V1 and V2. I also have access to the best measuring equipment money can buy, so if anyone wanted me to do some real measurements as opposed to simulated ones then I would be happy to do them.

13th May 2009, 10:30 PM	#126
ikoflexer diyAudio Moderator Join Date: May 2008 Location: Toronto	First things first. As far as I'm concerned, if you have questions, I see no reason why you should not ask them. salas said earlier something that a lot of people forget: these forums are here for some time, and people come after years something was posted, and learn something from your question or the answer to your question. Now, about your question regarding the class A amp that draws 500mA at idle. You are assuming that this is a constant 500mA. The word "constant" is the key here. In real life it will not be as constant as you think. If it varies somewhere around the 500mA mark, and the psu output impedance is on the high side, the input voltage to your amp will vary too. By how much, that depends. Do you care about it? I don't know, that's something everyone has to decide for himself. Have you noticed something about salas? He seems to accept a circuit if it makes an audible difference. A lot of people will scorn this as subjective. I disagree with those people, because I think it's salas' right to decide for himself what sounds good. And it's your right to disagree. So, if you intend the regulator/psu to be part of some device that you ultimately will listen to, the ultimate test is to actually build it, listen to it, and then decide if it's worth it. Sorry to be long winded about this. About measuring these circuits, I would certainly be curious about real life measurements. If anything, I encourage you to build v1. Many people are using that and are happy with it, after they compared it to alternatives. As far as v2 is concerned, that's very experimental still, and I'm still working on it. However, if anyone wants to go ahead and build it, who am I to disagree?

13th May 2009, 11:36 PM	#127
ikoflexer diyAudio Moderator Join Date: May 2008 Location: Toronto	I was checking out another regulator which is being worked on currently. It is an interesting shunt regulator too, and I was trying to help with its development a bit earlier. Today I managed to run it next to v2 and this is the output impedance plot. We should not read too much into such plots, but v2 does look promising.

14th May 2009, 01:23 AM	#128
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	Try a run with IRF9510 on V2. Then with IRF9610.

14th May 2009, 01:39 AM	#129
ikoflexer diyAudio Moderator Join Date: May 2008 Location: Toronto	A tiny bit worse with the irf9540. Here's something that surprised me with v2. The 9240s seem to work quite a bit better in this circuit than other mosfets. Even more of a surprise: using a darlington to drive the mosfets (both) gives a significant boost in performance. In particular I tried mpsa77. And this is while all other features stayed about the same (meaning it didn't start oscillating). Edit: 9510 a lot worse (20-30db worse); 9610 also worse, but a bit better than 9510. Does it surprise you? Because it does me. In v1 9610 actually seems better (if I remember correctly) than 9240.

14th May 2009, 02:12 AM	#130
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	No it does not surprise me. I just know that people will say try ''faster'' Mosfets for V2 at a point. Just see the transcoductance curves of faster TO-220 Mosfets, see the numbers they hover around, and watch how they diminish with heat. There lies the key. 9240 is 5 to 6 times stronger and it has a much bigger die, and body to sink. 9240 I did not pick just so. Its an all rounder, truly robust, cools off better, fastens easier. It will not oscillate, and gives better bass. Plus it scales up a lot. Its 12A. I use 3 times less than max for no problems with constant current. It can work up to 4A constantly, breaking no sweat with proper sinks. It already works for many months in V1 shunt @ 3A on a 2X10W T-Amp making it sound like a 300B SET but with better bass and analysis, and no DHT hum of course. Total background and buzz silence.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)