DC Power supplies for audio gear

marty

Well-Known Member
Apr 20, 2010
1,634
707
275
United States
#1
I didn’t want to take the OP on the SGM server off topic, but I do wish to comment on the topic of power supplies in general as mentioned in post #131 of this thread

https://www.whatsbestforum.com/thre...edition-in-hong-kong.26130/page-7#post-548235

There are some areas where I think audiophiles are often underserved and even misguided when it comes to after-market modifications and in some cases OEM power supplies for audio gear. I have seen power supplies that command ridiculously expensive prices, but what is particularly disappointing is that one almost never sees the specification for these supplies or modification. Yes, they may use fancy components and may be assembled by some guru blessed by an expert or even the Pope. But performance of power supplies have been a serious staple in the science world for decades. It’s not rocket science (actually, some of it may be used for that purpose), but rather a matter of good solid engineering.

Before you spend a zillion dollars for something that has value only an ignorant audiophile might appreciate, ask the manufacturer to show you specifications. It’s not hard. Accuracy and drift of the desired voltage, the amount of current provided, and most importantly, the stability under load as well as ripple and noise of any DC supply should be readily provided.

For example, here’s a list of specs for a fantastic series of HP/Agilent Power supplies.

Screen Shot 2019-01-04 at 5.54.55 PM.png

Even better, I recently picked up an E3616A for my Goldmund turntable for $150 at www.valuetronics.com, an excellent source of certified lab surplus power supplies. If you can show me a device that performs similarly (or better!) at any price, I'd like to know about it.

Honestly folks if you are paying more money than a few hundred bucks for a DC power supply unit made in somebody’s garage, at least ask to see the specs. Audiophiles are often far too naive when it comes to cost and quality of genuine laboratory gear compared to “audiophile” products. Power supplies are a prime example.
Marty
 
Last edited:
May 30, 2010
16,306
1,257
420
Portugal
#2
I didn’t want to take the OP on the SGM server off topic, but I do wish to comment on the topic of power supplies in general as mentioned in post #131 of this thread

https://www.whatsbestforum.com/thre...edition-in-hong-kong.26130/page-7#post-548235

There are some areas where I think audiophiles are often underserved and even misguided when it comes to after-market modifications and in some cases OEM power supplies for audio gear. I have seen power supplies that command ridiculously expensive prices, but what is particularly disappointed is that one almost never sees the specification for these supplies or modification. Yes, they may use fancy components and may be assembled by some guru blessed by an expert or even the Pope. But performance of power supplies have been a serious staple in the science world for decades. It’s not rocket science (actually, some of it may be used for that purpose), but rather a matter of good solid engineering.

Before you spend a zillion dollars for something that has value only an ignorant audiophile might appreciate, ask the manufacturer to show you specifications. It’s not hard. Accuracy and drift of the desired voltage, the amount of current provided, and most importantly, the stability under load as well as ripple and noise of any DC supply should be readily provided.

For example, here’s a list of specs for a fantastic series of HP/Agilent Power supplies.

View attachment 47263

Even better, I recently picked up an E3616A for my Goldmund turntable for $150 at www.valuetronics.com, an excellent source of certified lab surplus power supplies. If you can show me a device that performs similarly (or better!) at any price, I'd like to know about it.

Honestly folks if you are paying more money than a few hundred bucks for a DC power supply unit made in somebody’s garage, at least ask to see the specs. Audiophiles are often far too naive when it comes to cost and quality of genuine laboratory gear compared to “audiophile” products. Power supplies are a prime example.
Marty
Marty,

I can't understand your post. What is the point of posting specifications about DC supplies designed to be used for professional use in a post concerning supplies for audiophile computer use? Although very useful for laboratory use - I have used many HP power supplies - IMHO these generic specifications are of little meaningl for high-end audio.

The main question concerning power supplies is that we still do not know how to correlate power supply specifications with sound quality. In fact, the real situations are too different to allow any generalization.

A Goldmund turntable is a very simple case - IMHO no way comparable with what is being debated.

We all know that stereo sound is not rocket science - it is also a subjective art. Many high-end designers will tell you that the power supply is 90% of the sound quality. But they will not be able to specify their power supplies using the classical parameters you are addressing.

BTW, I use some simple DIY LM317 based linear regulators to power my digital optical link and router components. Probably many of these power supplies using fancy components will make my system sound better and this will go against my scientific mind. But I am prepared to accept it in my audiophile life :).
 

Folsom

VIP/Donor
Oct 26, 2015
4,693
708
220
Eastern WA
#3
Firstly lab supplies are not designed for big swings in power, so they only work on source type stuff.

Next off, what makes you think that the cleanest power is the most desirable? If anything it seems audiophiles have shown with purchasing habits this is not true. They may want a robust supply, but whether or not it’s low noise is personal taste.

I don’t see the point of being so adamant, go by sound you like. Even if you like low noise, you may find the lowest noise are not always the best sound. What’s low noise for an isolated lab use may not be low noise for audio that is interconnected.
 

marty

Well-Known Member
Apr 20, 2010
1,634
707
275
United States
#4
[QUOTE ] Firstly lab supplies are not designed for big swings in power, so they only work on source type stuff.[/QUOTE]
Nonsense. It all depends on what is required. Power supplies can be quite large and massive (several thousand watts and hundred of amps) if necessary. It's clear you really don't have any expertise in this area.

[QUOTE ] Next off, what makes you think that the cleanest power is the most desirable? ...audiophiles have shown with purchasing habits this is not true.[/QUOTE]
These statements are just hyperbole. What is the evidence that clean power supplies are not desirable?? What purchasing habits are you referring to? Examples please.

[QUOTE ] They may want a robust supply [/QUOTE]
Define "robust" please.

[QUOTE ] ...whether or not it’s low noise is personal taste....go by sound you like. [/QUOTE]
On this we agree! Hooray!

[QUOTE ] Even if you like low noise, you may find the lowest noise are not always the best sound. [/QUOTE]
Feel free to use the noisiest power supplies you can find. I won't stop you!

[QUOTE ] What’s low noise for an isolated lab use may not be low noise for audio that is interconnected.[/QUOTE]
Really? Sorry to disagree but low noise is low noise, unless of course your understanding of the laws of physics are different than everyone else.
 
Last edited:

Folsom

VIP/Donor
Oct 26, 2015
4,693
708
220
Eastern WA
#5
Marty,

It may be true there are very large lab supplies, but you don't see those on eBay typically. And even large ones aren't often meant for big swings. Being large doesn't automatically mean they like a changing load.

Grounding boxes are all the rage... I need not say more but; people like certain types of noise, and they'll buy exotic looking stuff that was made not low noise or modified to no longer have low noise with fancy stuff. The examples are pretty endless. Do you recall that old CDP that was extremely expensive from the 90's that someone posted, and inside all there was, was a Marantz with a modified output board to cause types of noise? Ya... very normal. Very consistently measured in the audio world.

Robust, as in no one likes the sound of voltage sag. They call it current limiting, but it's voltage sag. It leaves passages in music flat.

Again, not all low noise supplies can do voltage swing, and nor are they low noise in an interconnected system. Isolated pieces can be lower noise due to the deployment of noise reducing manners that may involve common mode and such. But the moment you introduce other pathways the topology may simply be letting in lots of noise, or letting it go to other parts of a stereo. The most simple of examples is how SE interconnects can transfer more noise parasitically, and leave an opening for RF in an otherwise sealed enclosure. This isn't a big problem, not a particular prevelent one to worry about, but it does exist.
 

marty

Well-Known Member
Apr 20, 2010
1,634
707
275
United States
#6
Folsom
I appreciate your reply. But as I see it, you made an even better case than I did for looking at power supply specifications as voltage stability with change in load is often a well documented spec. I also understand what you are trying to clarify regarding RF leakage and noise, but I don't think it's germane to the topic. Again, we agree on common ground which is letting your ears be your guide.
Marty
 

Taiko Audio

Industry Expert
Feb 10, 2017
1,167
1,309
180
The Netherlands
taikoaudio.com
#7
I didn’t want to take the OP on the SGM server off topic, but I do wish to comment on the topic of power supplies in general as mentioned in post #131 of this thread

https://www.whatsbestforum.com/thre...edition-in-hong-kong.26130/page-7#post-548235

There are some areas where I think audiophiles are often underserved and even misguided when it comes to after-market modifications and in some cases OEM power supplies for audio gear. I have seen power supplies that command ridiculously expensive prices, but what is particularly disappointing is that one almost never sees the specification for these supplies or modification. Yes, they may use fancy components and may be assembled by some guru blessed by an expert or even the Pope. But performance of power supplies have been a serious staple in the science world for decades. It’s not rocket science (actually, some of it may be used for that purpose), but rather a matter of good solid engineering.


Marty
Hello Marty,

Commenting as the designer/manufacturer of the SGM server. We are not trying to hide anything deliberately.

Quoting microstrip above: "The main question concerning power supplies is that we still do not know how to correlate power supply specifications with sound quality. In fact, the real situations are too different to allow any generalization."

I'm sorry to have to tell you I agree to this. Most of the time we end up evaluating different solutions and components by ear.

Anyway as for technical specs: For the high current rails we use a CLCLC filtered supply with a switching regulator. We have developed and evaluated a multitude of linear regulators, the best one being a LT1028 based 20A linear regulator design, but we don't use it as it sounds worse at these high currents (these rails can actually draw up to 15A, my working theory is the power wastage is actually more harmful to the sound then (the very low residual) switching noise). We use a modification of the original Walt-Jung design for low current (1A) regulators. I cannot measure any residual noise myself. I do have Audio Precision measurements on the low current regulators:
No Filter: 5.67uV
A-Weighted: 1.03uV

Further to this we measure electrical field missions for the entire completed product which varies between 20-30 dBuV/m depending on frequency (MHz-GHz).
Hope this helps.
 

marty

Well-Known Member
Apr 20, 2010
1,634
707
275
United States
#8
Dear Taiko,
Please understand that in no way was my post about the Power supply of your SGM server. I have every expectation it is engineered superbly and that voicing your unit was done diligently with the intrinsic PS design being part of the process. My comments pertained much more to add-on 3rd party units, particularly those used for the ubiquitous Mac computers commonly used for audiophile music making. That is what I alluded to in the post I referenced. There must be a dozen or more of these PS's widely available (mostly of Chinese origin), and it is rare to see any specs at all on these, yet they often command very high prices.

I did however, appreciate your notes regarding some of your PS specifications, although when you specify noise measurements, it would be good to know they in reference to a maximum voltage output of what, exactly , so we can get a sense of the percentage of residual noise?
Marty
 

Taiko Audio

Industry Expert
Feb 10, 2017
1,167
1,309
180
The Netherlands
taikoaudio.com
#9
Dear Taiko,
Please understand that in no way was my post about the Power supply of your SGM server. I have every expectation it is engineered superbly and that voicing your unit was done diligently with the intrinsic PS design being part of the process. My comments pertained much more to add-on 3rd party units, particularly those used for the ubiquitous Mac computers commonly used for audiophile music making. That is what I alluded to in the post I referenced. There must be a dozen or more of these PS's widely available (mostly of Chinese origin), and it is rare to see any specs at all on these, yet they often command very high prices.

I did however, appreciate your notes regarding some of your PS specifications, although when you specify noise measurements, it would be good to know they in reference to a maximum voltage output of what, exactly , so we can get a sense of the percentage of residual noise?
Marty
You do make a valid point. This measurement was relative to 5V. The theoretical noise is 0.9uV so you can see there is a difference with the measured noise.

If of interest to you, we value heat / vibration performance over electrical noise performance, I'm not discounting the importance of low electrical noise, not at all, but we have found a strong correlation between heat production and vibration resistance (ringing) and perceived sound quality.

As an example, on our latest product we load our high current power supply for 72 hours at close to 200 watts, at 21 degrees Celsius ambient nothing is allowed to go over 40 degrees Celsius in this period. If it does go over we check torque of all heatsink / chassis bolts or we replace the "offending" component. We us a 200W electronic load for this, picture attached, as the actual load will never be this high continuously. This procedure does a very good job at guaranteeing consistent sound quality.

file-1.jpeg
 

About us

  • What’s Best Forum is THE forum for high-end audio, product reviews, advice and sharing experiences on the best of everything else. A place where audiophiles and audio companies discuss existing and new audio products, music servers, music streamers and computer audio, digital to audio converters, turntables, phono stages, cartridges, reel to reel, speakers, headphones, tube amplifiers and solid state amplification. Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing