The ultimate Power Supply Units for music servers (and other devices for cleaner power source)

[Chopin75]

This is a new thread created as a subtopic or side branch from the massive thread : "A novel way to massively improve the SQ of computer audio streaming"

 

This forum is to focus on all types of Power supply units (PSU) used specifically or (adapted) for audio use, mainly music servers but can also apply to other devices like DACs. 

 

Few things I want to discuss include:

What are the few affordable PSU on the markets: 

HDPLEX

Uptone

Mojo

Sean Jacobs - Statement and his own http://www.custom-hifi-cables.co.uk/home/power-supplies/dc3-power-supply

Paul Hynes SR7 etc -- but apparently he is not take available for more commissions ?

Farad - https://www.faradpowersupplies.com

 

How about those cheap Chinese R Core and toroidal ones? Are they any good even to use for small devices and USB bridges

It appears that they are quite different in terms of designs as well as specs ,e.g. most are < 5A and mainly HDPLEX has 10A due to the need to boot its DC-ATX

 

Single rail vs multi-rail

 

What are the differences between a toroidal vs R Core 

 

[Chopin75]

I will start with a more technical questions. The other thread created early listed has many brands, may be better for brand comparison.

 

So what are the main difference between Toroidal and R Core LPSU ?

 

IS one better than other or it depends ? 

 

 

 

[Chopin75]

18 minutes ago, k27R said:

I have a linear solution reference 1 that’s pretty good.  This latest design is supposed to close the gap the original had with the SR7, or so I’m told.

 

I’ve  only compared it to a js-2 and  lps-1.2, and it’s quite a bit better than both.  It’s quite good.

THe Linear solution looks like a toroidal, does not say much else. So it is supposedly similar to SR7?

 

Uptone JS-2 is promising, it looks like R-core. Can boost to 10A which is what i need for my PC that has the HDPlex DC-ATX. Though it can only do 12 V, I need 19V

 

[Chopin75]

2 hours ago, Ralf11 said:

Isolate the music server from the DAC and don't worry about a PS for the music server.

 

For the DAC, you first want to buy a high quality one with a top notch analog stage - it may have multiple power supplies inside it.  A battery would be the ultimate PS for a DAC.

isolate the music server is one way but isolation may have its draw back. I chucked away my ifi Igalvanic as it tends to "color" the sound.  Currently I just use an optical USB cable from Corning that does the galvanic isolation without reclocking, seems to work better.

I think getting rid of the source of the problem is more effective. Besides, isolating the music server would not solve the dirty  current and noise already inside the music server. If there is less noise feeding into the server in the first place, that would be better than trying to filter out the noise downstream

Hence we should work from upstream, starting with hospital grade receptacle at the wall. ( I have not bothered doing that yet since I am renting right now)

 

 

[Chopin75]

2 hours ago, Superdad said:

 

Surprised to see you perpetuate such a myth Fred.  Since you also wrote:

 

And @auricgoldfinger stated this--though he seems to let slide Paul Hynes' even worse communication and delivery track record:

 

So many people report giving away their money and waiting forever with no updates and mostly excuses:

 

I honestly do not understand why people patronize such unprofessional (perhaps not a stretch to say unethical) businesses.  I spend more than 20 hours per week doing nothing but returning client e-mails, and I have never once accepted payment for product without promising and meeting a firm delivery date--typically either immediate of within 2-weeks shipment.  

Guess that is why there are now about 800 JS-2s in the field and close to 3,000 UltraCap units. 

 

My apologies if such discussion of business and reputation is off-topic.  Perhaps the TLS power supply is a fine unit, but nobody knows what is actually in the box.  (And there was this funky incident:

https://audiophilestyle.com/forums/topic/38316-the-linear-solution-lps912-design-for-sotm-products/?do=findComment&comment=772398

 

Yah, do  try to remain neutral. We do want unbiased opinions not affected by business interest.. we should also keep criticism as toned down as we can. Thanks!

[Chopin75]

50 minutes ago, Superdad said:

 

Really sorry to burst your bubble, but the Corning optical USB cable does not provide galvanic isolation.  It has copper wires running alongside the fiber cable—to power its far end transceiver. If you put a continuity meter on the plug shells at each end you will get a beep. 

The copper wire is to convert optical back to electrical. It is not used for data transfer nor giving vBus. and yes it can cause noise contamination. I use your USPCB vbus off to block any leakage though not sure if it matters.  I am merely talking regards to isolation with the optical part. Now is it truly galvanically isolated, not sure. It works better than my copper shielded  USB cable. Seriously, despite being 10M long! Even playing u-tube from my macbook sounds pretty good

[Chopin75]

1 hour ago, Ralf11 said:

 

optical good

 

hospital grade receptacle not important - I have one but it was free

 

dirty  current and noise already the music server is inconsequential if it is isolated from the DAC

 

are you saying that such isolation is more expensive than buying a low noise music server?  I'd likely disagree as a general purpose computer works fine

Isolation devices can be costly, without OT, i am merely saying the dirty current will still affect the Servers as  it is doing the important software process of decoding the music files, so this disruption would have affected the quality of data transfer to the DAC.it is like the CDplayer reading has lots of errors and so even even u reclock or galvanic isolate, the original data is screwed up, polluted and affects everything downstream. That is why most music servers,commercial or DIY always come with some LPS , and hence the birth of this new thread. 

[Chopin75]

7 hours ago, marce said:

https://www.audioxpress.com/article/Power-Transformers-for-Audio-Equipment

Good info. But all it says about R core is that it is easier to implement at lower cost and somehow not commonly used in US. I see my own Sony SACD player uses R core and my DAC also uses them (3 of them)  i guess u can put R-core in a smaller space as it only radiates EMR in certain direction. China has many products with R core. They probably have many factories making them 

[Chopin75]

1 hour ago, Ralf11 said:

 

can you explain this further?

 

not sure what you mean by a server - if it is acting like a DAC

I mean the PC that is having the OS/Software playing the music, not the DAC.  E.g  a NUC that has Audio linux + J River playing the music files (but not doing the DAC stuff), and the sent to the DAC separately via USB etc... I am not familiar with roon/NAS etc so I can't give example of that.I am a single box guy but I presume the LPS would matter in every single box used. 

[Chopin75]

BTW, one should remember it also depends on what you use the LPS for.  Mainly I want to concentrate on LPS that can do the job of running a PC music server/streamer or whatever you call them, sorry I get confused with terminology like server, streamer, renderer etc...

 

The LPS has to have enough power to drive the desired CPU one uses. It seems that for a larger more power hunger CPU, a LPS that can supply the correct Volt and Amp would be most important (or it won't work at all!). So naturally there would potentially be more noise issue if one needs a LPS that does say 19V/7-10 A,  vs with a 5V/2A (if all else equals) correct ?  

But it seems we have lots of low power LPS to choose from but not many at >12V or > 5A

We all assume LPS is best, though Teddypardo is doing something different, (I have not looked into that in detail) but apparently it is not LPS nor SM

And  Supracap or ultra caps (I presume they are same) are very popular these days. So are they superior to Toroidal or R Core ??

 

 

[Chopin75]

Just now, sandyk said:

 

 They will still need power supplied by a transformer of some kind, or a high A.H. battery.

oh yah, I forget. Thanks,

[Chopin75]

 

Has anyone tried such a filter ?

 

http://www.custom-hifi-cables.co.uk/home/mains-filters-surge-protection/black-box-super-conditioner

[Chopin75]

32 minutes ago, Jud said:

If you reconsider the above, you may see that it isn’t correct. Continuity from one end of a connector to another is unnecessary to run optical-electrical converters at either end. That is why such converters can be used as isolators, because continuity isn’t necessary. If there is continuity, there isn’t isolation.

 

A careful reading of the material on Corning’s web site shows the wires are there to provide the specified 5v USB power. So no galvanic isolation exists. No one can argue with your preference for the cable, but galvanic isolation isn’t the reason.

 

In general regarding power sources, folks will want to think of not only what is in the direct signal path, but what is in the “power path.” What’s being kicked back into the power from the wall or whatever other source through power cabling or other connections in the ground paths of the equipment?

However, the 5vBus is not carrying the audio signal, but to power equipment so that is irrelevant. The actual audio signal is still galvanically isolated but you can argue that the 5Vbus can cause interferences, though only at the distal and proximal end of the wire, which can be blocked with  Regen USPCB connector by blocking Vbus (I did that though I did not do A/B comparison)  . MY DAC does not need Vbus, so that can be blocked. The copper wire is causing some EMF etc which can be issue with regular USB but since this is optical that effect is not there or at least minimalized. The only real criticism of the Corning is that it is too long and probably the plastic used for optical is cheap so jitter is there. Reclocking at the DAC end would help.   

 

Also I don't know where you get that the corning has adequate VBus. It clearly states that the Vbus is not adequate for power. It may be true you get a little bit of power and apparently there is continuity  of the copper wire, (a flaw design for audio) which can be blocked externally as stated above.

 

From Corning web:

 

Are USB 3.Optical™ Cables by Corning compatible with all USB devices?

Though no USB standard currently exists for optical cables, USB 3.Optical Cables by Corning have been successfully tested with more than a hundred devices spanning all major consumer operating systems. In general, you can expect USB 3.Optical Cables by Corning to work with most self-powered USB 3.0/2.0 compliant device.

I want to charge my portable device (like iPhone or iPad) with USB 3.Optical™ Cables by Corning. Is that possible?

USB 3.Optical Cables by Corning are designed to be used with self-powered peripherals only. No significant power is available after powering its internal opto-electronics. A self-powered USB 3.0/2.0 hub can always be used at the peripheral (B-side) to provide power if desire

[Chopin75]

On 6/28/2019 at 12:14 PM, marce said:

https://www.audioxpress.com/article/Power-Transformers-for-Audio-Equipment

 

 

Now any preference of suprecap or ultracaps over traditional R core or Toroidal LPS? 

 

[Chopin75]

16 hours ago, Le Concombre Masqué said:

Loved the Corning SQ too. But it was maddening and one day it definitely died

IT can be fragile so I don't spread it out in case one can yank it, many reviews state its fragile/vulnerability. It me it has the least color or maybe no color, giving the true sound, veil lifted, very revealing. but does not introduce more harshness.

It does have very weak Vbus power but too weak to do anything, as a demo DAC that I am testing now cannot detect my PC probably due to its lack of true VBus and when I switch to a regular USB cable with Vbus it works. 

[Chopin75]

13 minutes ago, Le Concombre Masqué said:

you can plug in a powered hub

oh then the USB may have enough Vbus? Not sure if the thin copper wire has adequate transmission ability. 

[Chopin75]

36 minutes ago, 80levan said:

The hub may ruin the point of using a Corning. I'd go for a USB disruptor or idefender then power it with dxpwr power regulator and then some. 

YEs, I just remember I have the idefender so I can feed LPS into it to handshake the DAC. (dumb DAC). I would try lowest 3.3V first. Hopefully that is adequate. 

[Chopin75]

3 minutes ago, 80levan said:

I already did the 3.3v trick. No bueno. Haha. Huhu.

This also proves that the tiny copper in the Corning is really too weak to do much contamination, as it can't really communicate with the downstream DAC. The Regen Vbus block would do the final trick 

[Chopin75]

On 7/2/2019 at 1:34 AM, Superdad said:

 

 

Sorry, but this really is not true.  USB is D+ and D- plus ground.  The Corning connects PC ground to its distal end for its circuitry and that is the path for leakage.  Blocking of the 5VBUS+ does not break that path.  Therefore no galvanic isolation is provided by the Corning cable.  You have continuity from the PC ground to the DAC ground.

 

The primary benefit of the Corning cable (aside from its ability to greatly extend USB length) is that its thin wire introduces a few ohms of resistance to the ground line—and that sounds good.  We discovered this shortly after we introduced the very first USB REGEN back in 2015.  We wrote about it then and incorporated similar ground line resistance into what became the “amber” USB REGEN (which then sold over 4,000).  It is all chronicled (including discussion of the Corning) at this blog post way back then: https://uptoneaudio.com/blogs/news/20068483-usb-regen-updated-amazing-bass-all-unshipped-orders-will-be-the-latest

 

Of course the ISO REGEN—with its true galvanic isolation (plus other enhancements)—eliminates the need for those extra resistors.

You may be right it is not a true galvanic isolation technically but the main signal component is only contaminated at both ends during conversion from optical to electric and vice versa. The optical part (where the audio signal is) is immune to EMI/RF etc. However I did try using the Regen adaptor that blocks the 5Vbus and it made a huge difference. I found this out by accident when I forgot to block the Vbus (after using a DAC that required it) and then suddenly I realized something was wrong. And finally I realize the issue of not blocking VBUs and when I did the sound became much better with more analogue and natural sound..

 

Please refrain from promoting your own products quoting the numbers sold or the comments will be deleted. 

[Chopin75]

13 hours ago, Jud said:

Reclocking doesn’t “corrupt.” It may reduce noise by giving the DAC input circuitry an easier signal to deal with, though that hasn’t been proved.

Reclocking is only beneficial if the source is corrupted with jitter etc and that the upstream has an inferior clock (e.g poor streamer clock), The reclocking afterwards, if worse than the clock of the source will "corrupt" the chain. (except if you use a crapy USB cable or whatever cable after the source and so by the time it reaches the DAC it has become crap and so it needs reclocking, but still it won't be as good as the source's clock. If the source has a superb clock and the path to the DAC is very short reclocking may not be needed.  Though this may be a simplistic view, because it also depends on what clocking the DAC uses, if it DAC is the master clock or not ..so I am not sure if there is one correct answer. 

Ideally a clock that runs from the DAC to the source or vice versa (depending on where the master clock is) would  be best. Or may have an external master clock that controls the source and DAC.  I think the Brooklyn DACs have such features where they have external master clock to control the DAC and possibly the source perhaps).

So I would only try reclocking if really needed. Sorry I am now going off topic.  

The commercial world seems to sell lots of gadgets that are often redundant or not needed if the source is really improved (and that includes a great clean Power supply!) . There are also problems created to make more money. Usb is a crappy way to deliver audio but has become the norm. They kind of abandoned Toslink or other optical cables. SO we are now left to get many gadgets just to get the USB more tolerable. It becomes totally silly. 

But I still use USB and that is because: 1, I do DSD/DXD that requires USB interphase., and also I do multi-channel, as the Coax, Toslink AES won't be able to transmit such data. 

I am not saying USB is totally junk, but it can become great for music but at a greater cost than needed... 

What I am really saying is that, if the source is really inefficient,  reclocking can give limited help, it is counterproductive. Get a better clock in the source itself!

 

Sorry for rambling on off topic (maybe I need to delete my own post!)

 

[Chopin75]

2 minutes ago, sandyk said:

 

That depends on the quality of the source, which may have even come from streaming.

 I don't have too much to disagree with in your last paragraph though.

Exactly my point, if I did not make too clear. The source is of utmost importance:

3 most important considerations in Audio gear: Source, source, source

or to define them  in 3 components as:

1. Recording quality

2. Power supply/power source from the receptacle out of the wall.

3. The device/services (e.g internet streaming) OS, h/w involved in playing the music.

 

(Ok, sometimes the device has attached internal power supply so there may be overlap of 2 & 3)

[Chopin75]

5 minutes ago, Nenon said:

What if the source is Tidal/Qobuz? 

I don't know, I don't do internet streaming but that would take into account of how they play the music e.g how does Qobuz play the music, and how do they transmit the music (compressed data or full 24 bit etc) so I guess it could be a combination of 1 + 2 I mentioned.

I suppose even the internet service matters, of it is efficient in transferring the data, are the optical cables they use of high quality ??

(for e.g. my comcast give shxty looking TV services, you can see jitter on the screen, though occasionally when I listen to internet radio it sounds fine, I guess there is much less data)

If you stream large files like DSD (if that is possible ?) then there will be other issues of difficulty sending huge files, with signal loss etc I would presume ??  I don't think tidal/Qobuz do DSD anyway, Their upper limit is 24/192 correct ?

 

[Chopin75]

30 minutes ago, 80levan said:

I was committed to my windows laptop +  windows 2012 + audio optimizer + fidelizer pro and was sharing my experience in some threads but was dismissed as an 'old school way' because network streaming is the way to go. Well I tried the Hifiberry Digi+ pro and even the Allo USBridge + Corning but both has an inferior sound. I tried all combinations on the USBRIDGE including a switch running on a 12v power bank and decoupling the modem after device handshake and using cat 8 cables. The sound is still inferior to my surprise. Also, more importantly to this conversation, playing the music files from the laptop on a flash drive that is sharing the USB hub with an Ethernet port sounds more inferior. The best combo I've achieved is having the USB flash drive on the USBRIDGE, run a playlist then unhook the USBridge from the switch. But in the end, the laptop still has the best sound. Wider, better timing and more linear.

 

So after much time and money spent, I'm swearing off a device that gets it's music files through an Ethernet cable or use anything Ethernet at all. I'm swearing off the ultraRendus and such.  Im still mildly curious on the Matrix Element X if I were to play files off it and decouple the Ethernet port on playback.

Simplicity is the key for me. Now with additional devices, we are introducing potential dirty current into the chain, and more variables are added. External devices may have pros but the con of bringing additional jitter, RF/EMR etc all can degrade SQ. I presume you have already used good LPS for these devices, but results are still inferior?  

 

[Chopin75]

26 minutes ago, Nenon said:

 

I was just pointing out that you don't always have access to the source to fix it. I guess in those cases good recklocking of the signal helps. I don't stream, because the quality can't compare to my local files, but I have noticed that as I improve my network and server streaming Qobuz sounds better and better. Maybe one day streaming quality would get close enough that convenience would win :). 

But that seems to be offtopic. 

 

BTW, on the topic of power supplies, I just noticed some good power supplies for sale in the "Buy & Sell" forum - JS-2, HDPlex 400W, and a Sean Jacobs LPS. We have quite a few choices like never before. 

I don't know much about streaming/NAA/Roon but I presume the streamed data will eventually reach some server or PC (the endpoint?) and that box would be where you need to have a nice LPS and clean power source e.g. from a good power conditioner.  (some do regenerative power), it may not be as essential as a device that actually plays the music files but would be next most important I gather. 

Further you can have a USB audio bridge that also has nice LPS as well etc... 

[Chopin75]

3 minutes ago, Jud said:

 

SPDIF DACs have sounded just fine for a very long time.  I'm simply pointing to the physical fact that when we are talking about picoseconds or even femtoseconds, the local clock must always be more precise than control from any remote location, *all else being equal*.  This doesn't mean you can't have damn fine remote clocking, or bad local clocking.

Do u mean the local clock ie the clock in DAC? And remote clock is the music streamer or server or CD player etc... ?