A novel way to massively improve the SQ of computer audio streaming

[atxkyle]

Well, I'm clearly very late to the party.  But today had the chance to set up the network bridge and am happy with the improvement!  Agree with all the comments about a "veil lifted" - more of a live, present sound.

 

I'm currently just using a standard Dell PC (with 1TB SSD drive) as a Roon server ... has me thinking about what tweaks need to be done if the server also has a big impact on SQ ... 

[dgarretson]

I'm close to break-in  of the combo SOtM SMS-200Ultra/ts-USBultra/SPS-500, together with the SOtM-sourced D-Link switch timed by the SMS-200Ultra via a 6" SMB cable from Digikey.  All SOtM pieces are synchronized to a Stanford Research Systems Perf-10MHz rubidium clock via Canare 75ohm BNC cables, with the Perf-10 powered at 12 Vdc by a Hynes SR-7. This is definitely an improvement over mR powered by Hynes SR-7.  The gestalt is enhanced with short DIY USB cables made of shielded 3 nines dead soft silver USB cables.  Building a good silver USB cable is fairly simple.  I wouldn't waste $$ in that area.      

[tims]

Nice setup! Your system must be up there with Romaz's.

Do you have a link/recipe to the DIY silver USB cable you built?

[austinpop]

3 hours ago, dgarretson said:

I'm close to break-in  of the combo SOtM SMS-200Ultra/ts-USBultra/SPS-500, together with the SOtM-sourced D-Link switch timed by the SMS-200Ultra via a 6" SMB cable from Digikey.  All SOtM pieces are synchronized to a Stanford Research Systems Perf-10MHz rubidium clock via Canare 75ohm BNC cables, with the Perf-10 powered at 12 Vdc by a Hynes SR-7. This is definitely an improvement over mR powered by Hynes SR-7.  The gestalt is enhanced with short DIY USB cables made of shielded 3 nines dead soft silver USB cables.  Building a good silver USB cable is fairly simple.  I wouldn't waste $$ in that area.      

 

Welcome to CA! 

 

Please do share your listening impressions, especially in comparison to whatever you were using previously. Also, what are you hearing in terms of improvements specifically attributable to the Stanford Research Rubidium 10MHz reference clock.

 

Nice setup indeed!

[austinpop]

21 hours ago, atxkyle said:

Well, I'm clearly very late to the party.  But today had the chance to set up the network bridge and am happy with the improvement!  Agree with all the comments about a "veil lifted" - more of a live, present sound.

 

I'm currently just using a standard Dell PC (with 1TB SSD drive) as a Roon server ... has me thinking about what tweaks need to be done if the server also has a big impact on SQ ... 

 

Welcome to the discussion!

 

Glad you experienced the improvement due to bridging. What endpoint are you using (mR/sMS/...)?

 

Do read the index in the first post for pointers to other areas of exploration beyond bridging.

 

I also use a Dell desktop system as my bridged network/Roon server, and one easy improvement I got was by installing AO (Audiophile Optimizer) and Process Lasso (PL), even without any other HW changes.

[hurka]

7 hours ago, dgarretson said:

I'm close to break-in  of the combo SOtM SMS-200Ultra/ts-USBultra/SPS-500, together with the SOtM-sourced D-Link switch timed by the SMS-200Ultra via a 6" SMB cable from Digikey.  All SOtM pieces are synchronized to a Stanford Research Systems Perf-10MHz rubidium clock via Canare 75ohm BNC cables, with the Perf-10 powered at 12 Vdc by a Hynes SR-7. This is definitely an improvement over mR powered by Hynes SR-7.  The gestalt is enhanced with short DIY USB cables made of shielded 3 nines dead soft silver USB cables.  Building a good silver USB cable is fairly simple.  I wouldn't waste $$ in that area.      

So the ~7000$ sotm chain better than mr+SR7.

[Deusvult]

Has anybody compared the Mutec MC3+ USB to the SOtM dX-USB HD Ultra? In my setup the sMS-200 Ultra is connected to the Mutec MC3+ USB, which then feeds my Devialet 400 through AES (which should be the best input for the Devialet). Would the presumably better clock of the dX make a significant difference?

 

I am also considering to at one stage add a master clock (e.g. Mutec REF 10) to my setup. As the clock signal is then coming from this device, does the original clock boards of the components it feeds matter or would the end result be the same regardless of it’s a Mutec or a dX being fed?

[One and a half]

3 hours ago, Deusvult said:

Has anybody compared the Mutec MC3+ USB to the SOtM dX-USB HD Ultra? In my setup the sMS-200 Ultra is connected to the Mutec MC3+ USB, which then feeds my Devialet 400 through AES (which should be the best input for the Devialet). Would the presumably better clock of the dX make a significant difference?

 

I am also considering to at one stage add a master clock (e.g. Mutec REF 10) to my setup. As the clock signal is then coming from this device, does the original clock boards of the components it feeds matter or would the end result be the same regardless of it’s a Mutec or a dX being fed?

The SOTM doesn't feature galvanic isolation on the USB input whereas the Mutec does. It's also likely, the AES3 has transformer to isolate as well, but not 100% sure.

[dgarretson]

6 hours ago, hurka said:

So the ~7000$ sotm chain better than mr+SR7.

Yes, but I won't have the full picture until I build a DC Y-cable for the SR7 to connect to the SMS200ultra and USBultra-- enabling comparison of SPS-500 to SR7.

10 hours ago, austinpop said:

 

Please do share your listening impressions, especially in comparison to whatever you were using previously. Also, what are you hearing in terms of improvements specifically attributable to the Stanford Research Rubidium 10MHz reference clock.

 It still needs break-in, but it is clear that the master clock improves pretty much everything: more resolution, more natural pitch & timbre, better bass articulation,  reduced digital smearing.  The comparison is to a mR with hard adapter powered by two DC regulation stages in SR7 to mR via DIY silver umbilical.  DAC is Esoteric K-01X, also synchronized to Perf 10.

6 hours ago, hurka said:

 

Nice setup! Your system must be up there with Romaz's.

Do you have a link/recipe to the DIY silver USB cable you built?

I made two silver USB cables: one 10" shorty between the SOtM units(which doesn't require a Vcc conductor), and a 30" cable from USBultra to K-01X DAC, which does require Vcc.

 

Short Cable between SOtM Devices:

 

Twisted data pair is made of 24awg 4-nines dead soft silver inside 2mm cotton tubes, both from VT4C Studio on-line.  The pair is twisted around 1/2 turn per inch. Twisted bundle is wrapped in teflon plumber's tape. Bundle is wrapped in spiral (A) of 20awg 3-nines dead soft ground wire, 1/2 turns per inch, from Rio Grande on-line.  Bundle is wrapped completely in spiral of 1" wide alum foil shield.  Foil-encased bundle is wrapped by spiral (B) of 20awg 3-nines dead soft silver. Bundle is pulled through 7mm cotton or silk tube from VT4C.  Spiral (A) is connected to ground pad of both USB connectors.  Vcc pad is not used. Spiral (B) is soldered to drilled holes in both USB metal connector bodies.  USB connectors Type A & B are high-quality gold-plated copper alloy from Norne Audio on-line.

 

Long Cable with VCC between USBultra and DAC

 

24awg shielded twisted data pair and 20awg shielded twisted power pair are constructed as completely separate runs so that the two can be pulled apart for isolation.  Each has a spiraled 20awg silver drain wire inside its respective alum shield, but without the above second silver drain wire on the outside of the alum shield.  As above, the drain wire is soldered to a hole drilled in the metal connector body.   

 

You need a good magnifier lamp to solder at this level of detail.  Tin the connector pads first with silver solder. Then just push the wires gently into place with the hot iron and use an Exacto knife to hold the wires down as the iron is withdrawn.

 

All I've compared these wire to so far is a generic USB cable and the standard cable provided by SOtM. The improvement is huge.  More fleshed out and detailed against blacker background.

 

Photos below.  I later added Ferrite clamps that are not shown.  Also, I  brought out the drain wire at the Type B end to support future addition of an external copper braid shield.

 

[atxkyle]

23 hours ago, austinpop said:

 

Welcome to the discussion!

 

Glad you experienced the improvement due to bridging. What endpoint are you using (mR/sMS/...)?

 

Do read the index in the first post for pointers to other areas of exploration beyond bridging.

 

I also use a Dell desktop system as my bridged network/Roon server, and one easy improvement I got was by installing AO (Audiophile Optimizer) and Process Lasso (PL), even without any other HW changes.

Thanks!  Although after I made this post I started getting blue screens of death and haven't been able to resolve it yet.  It was a driver error and I ordered a new PCI ethernet card to try instead of the USB-ethernet adapter I was using, so I'm hoping that will do the trick. 

 

I'm using microRendu as an endpoint.  Still the original 1.3 board but considering the update to the 1.4 board.  Haven't looked into other areas of tuning beyond bridging yet, but will look forward to doing so. 

 

Re: the Dell PC, I guess we Austin guys are loyal to the local Dell brand huh? (I live in Austin too if you didn't gather that from my handle).  My PC doubles as office PC and Roon Server.  From a quick look at the AO manual it sounds like it only intended for a dedicated audio PC?  But if its true that microRendu isn't immune to upstream noise, might have to go the dedicated/optimized Roon server route at some point. 

[Johnseye]

On 8/3/2017 at 0:11 PM, austinpop said:

 

That review certainly matches the findings on this thread. It's too bad they didn't think to try an sCLK-EX modded switch.

 

Thanks for update! The sPS-500 is sounding like a winner. I guess it was only a matter of time before an audiophile SMPS would be going head to head with the best LPSes out there.

 

I don't claim to be an expert and I own LPSUs, but I value this company's expertise.

 

AUDIO MYTH - "SWITCHING POWER SUPPLIES ARE NOISY"

 

Here's a snipit from Benchmark's site:

 

THIS MYTH GOES SOMETHING LIKE THIS:

 

"Switching supplies are noisy."

"Linear power supplies are best for audio."

 

We disagree!

About 5 years ago, Benchmark stopped putting linear power supplies into our new products, and we replaced them with switching power supplies. We did this because linear supplies are too noisy. Yes, you read that correctly, linear supplies are noisy! A well-designed switching power supply can be much quieter than a linear supply.

 

LINEAR POWER SUPPLIES CAUSE HUM

The noise problem is due to the fact that linear power supplies have large transformers and other magnetic components that operate at the AC line frequency (50 Hz to 60 Hz). These line frequencies are audible, and we are all too familiar with the hum and buzz that audio products can produce. It is no secret that this noise is caused by the power supply, but few people understand why it can be so hard to eliminate. Most people think that hum is caused by conducted interference (AC ripple on the power supply rails), but this is rarely the case. Most AC hum is caused by magnetic interference, and this can be very hard to eliminate.

 

HUM IS USUALLY CAUSED BY MAGNETIC INTERFERENCE

Transformers are magnetic devices. Power is magnetically transmitted between a transformer's input and output windings. In a linear supply, power is transmitted from the AC line side of a transformer to the low-voltage secondary side using an AC line-frequency magnetic field. Unfortunately, transformers are never perfect, and some energy always escapes through stray magnetic fields. These stray fields can interfere with virtually every electrical conductor in an audio product. Magnetic shielding is expensive and it has limited effectiveness when sensitive circuits are located in close proximity to a strong field.

 

POWER AMPLIFIERS ARE THE WORST OFFENDERS

The power supplies in high power devices, such as audio power amplifiers, can emit very strong magnetic fields. These strong fields tend to limit the noise performance (SNR) of power amplifiers. These magnetic fields can also cause interference with audio products that happen to be too near the amplifier. Audio cables that enter, exit, or pass near the amplifier may also pick up unwanted hum and buzz. For this reason, it is usually very important to keep the power amplifier well separated from cables and other components in the audio system.

 

[austinpop]

28 minutes ago, Johnseye said:

 

I don't claim to be an expert and I own LPSUs, but I value this company's expertise.

 

AUDIO MYTH - "SWITCHING POWER SUPPLIES ARE NOISY"

 

Here's a snipit from Benchmark's site:

 

THIS MYTH GOES SOMETHING LIKE THIS:

 

"Switching supplies are noisy."

"Linear power supplies are best for audio."

 

We disagree!

About 5 years ago, Benchmark stopped putting linear power supplies into our new products, and we replaced them with switching power supplies. We did this because linear supplies are too noisy. Yes, you read that correctly, linear supplies are noisy! A well-designed switching power supply can be much quieter than a linear supply.

 

LINEAR POWER SUPPLIES CAUSE HUM

The noise problem is due to the fact that linear power supplies have large transformers and other magnetic components that operate at the AC line frequency (50 Hz to 60 Hz). These line frequencies are audible, and we are all too familiar with the hum and buzz that audio products can produce. It is no secret that this noise is caused by the power supply, but few people understand why it can be so hard to eliminate. Most people think that hum is caused by conducted interference (AC ripple on the power supply rails), but this is rarely the case. Most AC hum is caused by magnetic interference, and this can be very hard to eliminate.

 

HUM IS USUALLY CAUSED BY MAGNETIC INTERFERENCE

Transformers are magnetic devices. Power is magnetically transmitted between a transformer's input and output windings. In a linear supply, power is transmitted from the AC line side of a transformer to the low-voltage secondary side using an AC line-frequency magnetic field. Unfortunately, transformers are never perfect, and some energy always escapes through stray magnetic fields. These stray fields can interfere with virtually every electrical conductor in an audio product. Magnetic shielding is expensive and it has limited effectiveness when sensitive circuits are located in close proximity to a strong field.

 

POWER AMPLIFIERS ARE THE WORST OFFENDERS

The power supplies in high power devices, such as audio power amplifiers, can emit very strong magnetic fields. These strong fields tend to limit the noise performance (SNR) of power amplifiers. These magnetic fields can also cause interference with audio products that happen to be too near the amplifier. Audio cables that enter, exit, or pass near the amplifier may also pick up unwanted hum and buzz. For this reason, it is usually very important to keep the power amplifier well separated from cables and other components in the audio system.

 

 

Very interesting! I was the happy owner of a Benchmark DAC2 HDR for a couple of years, so I too value their expertise.

 

It just goes to show that one shouldn't cling to any given technology dogmatically, but rather keep an open mind and ears.

[Always.Learning]

I own a Crayon Audio integrated amp (CFA 1.2) that has a switching power supply, manufactured in China (the amp itself is manufactured in Germany). I've owned amps by Shindo and Bakoon and have had earlier versions of Vinnie Rossi's SS amps in my system. The Crayon is extremely quiet and transparent and does a lot of other things at a very high level. Best not to be dogmatic. 

[Johnseye]

1 hour ago, austinpop said:

 

Very interesting! I was the happy owner of a Benchmark DAC2 HDR for a couple of years, so I too value their expertise.

 

It just goes to show that one shouldn't cling to any given technology dogmatically, but rather keep an open mind and ears.

 

The DAC3 is out which I'm going to have to give an audition.

 

Here's some other interesting info from that page.  The mention of filtering out interference made me think of Lee at SOtM's comments about his preference toward filtering.

 

OUT-OF-BAND NOISE

One major advantage of switching supplies is that the operating frequency is above the range of human hearing. If interference occurs, it will not cause audible interference. This interference can even be removed with a filter without infringing on the audio band. But, the power supply in the AHB2 is so quiet that we do not need to filter the audio output. The AHB2 delivers a 200 kHz bandwidth without evidence of any significant switching noise, to a measurement limit of 500 kHz.

 

and perhaps why SOtM is not using a linear supply:

 

"Please note that magnetic interference is radiated not conducted. This means that is cannot be removed with filter capacitors. Adding filters to a linear supply will not remove the hum and buzz from a power amplifier."

 

After the switching vs linear video is a star quad cable comparison.  Interesting to hear how magnetic interference effects sound.

 

 

[RickyV]

The Meitner MA1 has a smps. Sound  very good anyway.

[ElviaCaprice]

I'm sorry, but after hearing from Rob Watts and his great design for filtering power to perfection for his DAC designs only to find that a good LPS was much better, I don't believe anybody and their smps claims.  Best to get off the direct AC power grid and/or add an isolation transformer.

[RickyV]

I use a 450 VA one.

https://www.toroidal-transformer.com/shop/high-end-audio/audio-safety-transformer.html

[RickyV]

This one i have too for the pre amp.

http://audio-life.nl/product/audio-consulting-isolation-transformer/

[Middy]

 https://airlinktransformers.com/category/conditioning-balanced-power-supply

 

UK version

[beautiful music]

22 minutes ago, Middy said:

 https://airlinktransformers.com/category/conditioning-balanced-power-supply

 

UK version

Did you know US version

[Johnseye]

3 hours ago, sadekkhalifa said:

Did you know US version

 

A lot of people like the Topaz.  The B&K 1604A is another that comes up but with less wattage.

[Middy]

People have said the Elgar is a little quieter transformer  hum but the Topaz Elgar's v low winding capacitance is the strong point.

With second hand models it's always good to get an EE to pass them off as safe....

Airlink do a US version but as for UK to US is the freight cost. 

[InegoMontoya]

Going back to bridging. Anyone using an Airport Time Capsule for bridging?

 

Thanks.

[[email protected]]

Can anyone confirm the use of a Phasure "Lush" cable with the SOTM double stack (SMS-200 Ultra + tXUSBultra) or trifecta (+ switch)? Peter and I are trying to determine whether the cable I received is singularly defective or that the cable fails to work in tandem with the above variables. Upon receiving and inserting the cable last night, I heard, from the first, "snap, crackle and pop(s)". Within 10 or perhaps 20 seconds, a strident and piercing hash would follow, not unlike the spasmodic introduction of white noise. I cross-examined all variables, from power-conditioner to server, source and all components downstream, to no avail. This culminated in a cable swap and all was right with the world again. 

 

So, has anyone used the Lush in a similar setup? Was it a success or did similar problems emerge? I find it difficult to fathom that this cable is somehow unobliging in this combination, but Peter acknowledged this as a point of concern. It should be noted that I have tried four usb cables in lieu of the "lush"  -- a Revelation Audio Labs (RAL) Reference USB Cable (the fussiest I've ever owned),  a Synergistic Research USB Active SE,  an Uptone USPCB, and lastly a run of the mill 6 inch (15.24 cm) USB cable; included with some component or other. They each performed matter-of-factly and without controversy. 

 

In an effort to be comprehensive, the gear used is as follows: BPT 3.5 Signature > Macbook Pro > DLink Switch (SOTM Modified)> sMS-200 Ultra (LPS-1) > tXUSBultra (LPS-1) > TotalDac D1 Dual/Lampizator Gen. 5 LV. 5 > ECBA (Eddie Current Balancing Act) > Focal Utopia 

_____________________________________________________________________________________________________________

 

On a side note, I would like to share some observations on the SOTM trifecta, which perhaps will be of some guidance to someone out there sitting on the fence. Prior to the arrival of the SOTM gear, I was rather happily using an MRendu > Offramp 5 via i2s > Lampizator Gen. 5 Level 5 custom-fitted with an RJ-45 interface.

 

In the interest of being concise, my comparative remarks will be more of the broad and sweeping variety, rather than expounding on the various shades, penumbra and minutiae that most of our ears are tuned for. When I received the SMS-200 Ultra & txUSBUltra, I was not, let us say, afflicted or affected by an an immediate and unforeseen enrichment of my system. I did not forsake food or renounce sleep and "rediscover" my heretofore collection of music. Perhaps the sound was improved, but the word "different" came more readily to mind. Where the staging and locality improved and the layers of detail more readily apparent, an intimacy was gone and so too a brute power that can only be called an addicting directness. The sound was thicker, blunter and more candid or straightforward in the previous setup. 

 

I was astonished to discover, several days later, that it was not until the addition of the switch that the SOTM gear really began to hit its stride and outpace the Mrendu > Offramp (i2s) combination. I've not fully come to terms yet with this trifecta, for it has not been long enough, but as to improvement the answer is no longer illusory. It's better and perhaps considerably better. To what extent I'll discover in time; but I should not be surprised if I come to forsake my dinners and renounce my sleep in the coming days. 

[AmusedToD]

I am using the Lush cable between my sms200Ultra and Singxer SU-1. And yes, there were (and still are from time to time) some intermittent pops and clicks, but they seem to appear randomly. I restarted my PC, the sms200Ultra, the DAC and the Singxer several times, I also changed the resync delay settings in Roon and Eunhasu. It works better now, but there definitely exists some kind of an issue which appears occasionally. I never had the problem with Curious USB in the same configuration, but then again I think it has to do with the new SOTM firmware (v 4.0 or something). I think the pops and clicks started to appear after I updated the firmware several days ago.