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

[jean-michel6]

Hi All ,

 

First a very warm thank you to @romaz for such a solid and extremely good contribution to our hobby and also to @austinpop for keeping this very long thread well organised and so interesting .

 

For the last few month i have been also pursuing the improvement of my server/player source with extremely positive and good results even though i have not yet reach the trifecta set-up .

My DAC being the AMR DP777-SE i am using the AES/EBU for digital input .

 

My set up is made of two Gigabyte MB with Celeron processor( server and player ) and of course fanless , i am now feeding these with 4 separate power supply with 12V lead acid batteries one for each processor and one for each pico psu of each mother board .

The separation of power supply specific for each processor is bringing very positive results . 

To go further i have on order an SR7 PS from Paul Hynes to replace those batteries.

I can confirm also that running at the lowest possible processor speed is a key point to achieve good sonic result , i went from 2.1 Ghz to 800Mhz for both processor and ram with a very positive sonic benefit .

I am using roon and very light linux configuration and each os is on it's own SSD with it's down dedicated low noise power supply . 

Music is on two large SSD also on their own dedicated low noise power supply.

 

The output of the player pc is insured through a tx-USBexp card with its own clock board sCLK-48.0 ( this is fed by its own Lifepo4 battery ).

And last i have a dx USB-HD which has been modified by SOTM with the sCLK-EX clock board( word clock and two clock going to the FMC's) , this is fed by an LPS-1

 

Communication between the two pc's is done trough ethernet with a dedicated optical switch and FMC all that of course on their own specific batteries .

 

Sorry for this long introduction to let you know from where i am coming from .

The sonic result i am achieving with this set up today are the best i have ever had in my system with a very natural sounding source ,enjoyable and much closer to the "real thing " .

 

I have found that the much profitable route is not to keep buying more expensive piece of hardware but to work very hard to remove what is embedded in our music and is not music namely the three worst offenders :

- Jitter everywhere ...

- Vibrations of all kind ... even ssd are sensitive 

- Noise in the power supplies ...

 

I am now on the road to explore the new sCLK-EX route  . I have already  dx USB-H with sCLK-EX board .

I have on loan a SMS 200 ultra with it SPS 500 that i will compare to my own PC player set up .

 

Next steps will be to test the Tx USB ultra and also an sCLK-EX board for PC player and or server .

 

Cheers for our wonderful hobby , it is a great time with all those progress achievable today !

 

 

 

[jean-michel6]

Hi Roy ,

 

I have read your very interesting long last post on your server set-up and i have a few comments and questions .

 

- Did you do listening test to pick up the right MB ? this is a tricky choice . I fully agree with the celeron low horse power pc it keeps the noise lower  . Also much of the hardware on the processor keep thing's efficient  .

 However since you went with one since 12 v power supply , all of the different voltage are supplied through noisy DC-DC  converters including for the processor itself .

I think i will probably compare on my pc player my current set-up with a single mini-its celeron processor as you did

 

-  I see that you are using the Pachinko Reference for the OS data connection to the MB , why you did not went to the Pure Reference which my self i am using ( following i think you recommendations ) . The Pure Reference is a notch better than the Reference mainly on the attack of the notes .

 

- I have picked up from Paul Pang site one thing which work very well . To my surprise SSD are sensitive to vibration and i have attached each of my SSD's to an aluminum bloc( 15 mmm thick ) with very good result . It bring much more clarity to the sound .

In my system it brought a progress similar to the separate power supply for each ssd .

 

- Which power supply are you using to feed the SSD  and also to feed the sCLK-EX board ?

 

- EMI-RFI reduction , you have been using the SOTM paper . I have done the same but with the 3M 5100s pair which is also transform    EMI-RFI to heat . Have you ever compared both since the 3M pair is quite cheaper ?

 

Thank's for everything and i do hop that we can still enjoy from time to time your writings !

[jean-michel6]

2 hours ago, Johnseye said:

 

The 3M AB 5000S series EMI Absorber, including the AB-5100S, is very similar and may in fact be the same as the SOtM eABS-200.  The 5100S is the same 1m thickness.  The stated applications by 3M are applicable.  3M states it has enhanced absorption at lower frequencies, below 1GHz.  If you look at the chart it shows 6GHz and SOtM says on their site, noise removal up to 6GHz possible.

 

However that model number seems out of production.

 

The new model numbers are showing as halogen free versions AB5100HF or AB5100SHF but they are between 200MHz - 10GHz and 50Mhz - 10GHz respectively.

I think I used the 5100SHF from Digikey. It looks very similar to SOTM paper but cheaper 50$ vs 120$

Anyhow results are good with noise reduction you get darker background in the music. 

[jean-michel6]

5 hours ago, Johnseye said:

 

I can only guess covering more of the frequency range between 6GHz and 10GHz is a good thing.  I wonder why Lee was recommending not using too much.  What could the negative effects be of blocking out EMI?

I can't see negative effect by removing unwanted sound pollution from emi-rfi.

Emi-rfi brings glare to the sound . We are so much used to this that sometimes we initially perceive the removal of this glare by a loss of life in the sound. 

However extended and careful listening tells you that the top end sound of your system is more refined and detailed. 

[jean-michel6]

On 15/10/2017 at 6:54 PM, Johnseye said:

 

See my post here:

The board I've identified, and will likely send off to SOtM for modification is the Jetway NF591.  Very similar to the DFI board Roy is using.  The Celeron proc is 6w.  It has a PCIe slot so I can use a tx-USBexp.  It has DC 2.5mm in which means I don't have to use an adapter from my SR7 anymore.  This is a big win.  But it has an M.2 slot which means I can A. use my current low power SSD and B. 

 

The M.2 slot is not capable of using the Optane stick, but I may be able to use it with the mini PCIE.

 

This board will go out to SOtM for modification this week.

Hi @Johnseye

 

I am very interested by your feedback when available about the  Jetway NF591 motherboard modified by SOTM .

Do they told you that they could modify both clock for the processor clock and also the ethernet clock ( for me as i am in a dual pc configuration ) . 

 

I have been looking also to the bios of these motherboard and i understand that there is no way to lock the CPU to a specific speed which usually has a quite significant impact on the sound quality . May be not so necessary for these processors with very low electrical consumption .

[jean-michel6]

4 hours ago, tboooe said:

I've been thinking about modding my FMC as well. Maybe I should contact Sotm and see if they are interested.

I have had two FMC ( MC 220 TP link ) modified by Sotm: clock and dc-dc converter. 

 

I have just received them and as I am away from home , I have not yet tested the benefit. 

[jean-michel6]

13 hours ago, tboooe said:

I would love to hear your feedback once you return.  Can you be a bit more specific about the DC-DC converter mod?  What exactly was done?

They replace the DC-DC regulator in the fmc by component of higher quality which are less noisier .

It is similar i think to what is done on the modified switch. 

Those regulators create a lot of electrical noise which is bad .

[jean-michel6]

On 09/11/2017 at 5:45 AM, Johnseye said:

For those considering modifying their mobo, the Jetway board had the following clocks: Main clock, 2*LAN, USB, Display, & I.O Controller. The IO controller is a 6*UART and appears to be related to the COM ports. I elected to only modify the system and ethernet clocks. I won't be using the on board USB and see no reason for the others. 

 

I went with the tX-USBultra option. I asked SOtM to pull the sCLK-EX out of it and make direct connections to the mobo and tX-USBexp clocks. An SMB connector will then run from the sCLK-EX to the remaining card in the tX-USBultra which is a tx-USBhub. 

 

I will also be using an Intel X-25E SSD with SOtM SATA 2 filter. This was an afterthought inspired by recent discussions here about SSD noise. Roy tried convincing me on this a while ago but I was stuck on an M2 solution. It's a worth while comparison. A no cable,  direct connection, low power SSD vs a low power cable connected and filtered SSD. 

 

I will have a lot of connectivity options and tests to perform. Other than adding some of Barry Diament"s vibration isolation from Ingress, this should be my final system modifications. Well there is an upcoming LPS from a reputable company I'll be getting as well. 

Hi John,

 

Very interesting , i am going a similar path but with a different mobo board the super micro X10SBA  which is also a good board for audio and easier to source in Europe than the jetway.

I am currently testing the tx-USB ultra in my system. Very smart your idea to buy this and move the sCLK-ex board to the mobo. 

How do you will power the sCLK-EX board , txUSB hub .

 

I am going also the same way with the os ssd with used intel X25-SE .

However i have read in jplay forum that there is a much preferred solution than ssd. 

It is to go with cf cards . To implement this you need an inexpensive cf to sata adapter , run your sata in ide mod and use industrial slc cf cards which are bootable. 

I will be trying that also in a few weeks. 

[jean-michel6]

On 22/10/2017 at 2:50 AM, tboooe said:

I would love to hear your feedback once you return.  Can you be a bit more specific about the DC-DC converter mod?  What exactly was done?

I finally took the time to put in my system these 2 sotm modified MC 220 TP link (optical-rj45 converter ). 

Before I speak I talk about the effect of this mod let me mention briefly my system. 

It is a dual pc fed y batteries and running under roon ( Linux ). 

The player PC has an sotm tx-USBexp card who is connecte to a dx-USBhd which has a sCLK-ex clock board inside and 2 of the clocks were used for the FMC .

Both pc are linked with an optical switch. 

The dc-dc mod done by sotm on the fac is simply replacing those converters by less noisy components .

 

Now what about the impact of this mod . In a wor : amazing , i did not expect such a big improvement with this mod. The sound is much more analogic and natural , i have completely lost a very small brightness in the high midrange. 

I am extremely happy. 

I think the mod I did is a little bit similar to what have observed those who have installed à sotm modes switch between their server and renderer .

 

I did not expect to get such a big improvement , basically I think what this mod does is improve the timing of the flow of data between the server and the player.

However if I look at the overall lan connection between the two pc's there is 5 clock (lan server mobo, fmc , optical switch ,fmc , lan player )  and only the two fac have the upgraded sotm ( sCLK-ex) clock. 

 

I can't imagine what I can get once I have all 5 clocks modified ... I will go in that direction. 

 

Computer audio d'iOS is really fun and sometime frustrating but results are stunning. 

[jean-michel6]

7 hours ago, Johnseye said:

 

I've been thinking about this.  How does ferrite "kill the sound"?  Is it then only removing EMI/RF which we previously thought of as part of the sound and something we've become accustomed with digital music?  It can not suppress, or cause a negative result before the analog conversion because it can't affect the data, only the noise on the line.  Putting it on the interconnects after conversion could have an impact if the ferrite somehow absorbs certain frequencies of the music source, but I haven't read of that occurring. It is possible though and more than possible I haven't read about it.

 

I have read of concern with ferrite becoming magnetized.  If this happens the sound can "harden", but that would assume the softening as a result of the ferrite was a positive, which I think it is.  Then you can either demagnetize the ferrite or replace it.

 

Since I'm rebuilding my server tonight I've decided to remove all the ferrite so I can hear what the new build is like without it.  Then I will add it component by component.  I will leave it off the interconnects for a period of time so I'll be able to hear the difference with and without ferrite on analog cables.

Hi

I have also experiment with ferrite . However you need to be careful on which cable you use them.

All current are electromagnetic waves as well as Emi-rfi that we are trying to get rid of. 

Ferrite is effective with positive results ( from an audio perspective ) on digital cables ( USB, lan,...  ) on power cords. 

On interconnect , it has a negative effect as it just kills the sound making it softer and less engaging. 

One very effective way to shield all cables from emi-rfi is the use of mumetal , with two layers to passively cover cable. 

One cable company in france HiFi cable is using this  technology on all their top line cable with very good results. 

I have treated my digital cables with very good results. 

The main difficulty is to find mumetal for this at a reasonable cost. 

[jean-michel6]

2 hours ago, auricgoldfinger said:

 

How difficult is it to wrap the cables?  Is the mumetal fairly pliable?

 

Would it be possible for you to share some photos of your cables?

Hi ,

 

Below is a picture to show as an example , i did put only one layer of mumetal . I am using a 0.1mm mumetal which has adhesive on one side therefore it sticks to the cable .

The way i do it is i put one layer of mumetal as shown on the picture , then a second layer to insure 100% coverage twisted the opposite way of the first layer .

The mumetal layer are connected to nothing .Since mumetal is conductive , i finish with one additional layer of teflon plumber tape .

This makes the cable somewhat stiffer but it remains flexible , it would be better in my opinion to use a thinner mumetal but i have not found a source at reasonable cost .

[jean-michel6]

42 minutes ago, austinpop said:

@Bruce Orr, @limniscate, and I just concluded a very entertaining and enjoyable digital audio meet on Eric's wonderful setup. Bruce is a CA'er from Dallas - so he practically lives in our back yard; this is Texas! He was gracious enough to fly down for the day, and brought with him some goodies which we were very interested to compare - the UltraRendu, and the sPS-500 PSU with SOtM silver Y-cable. Between his gear, Eric's, and mine, we had more digital front end gear than you could shake a stick at. Funny expression, that. Why shake a stick at digital gear? What am I saying! We're audiophiles - if it improves SQ, you can bet we'll shake sticks and more.  But I digress...

 

Since this was a single session of 7-8 hours total, not all of which was listening, all my impressions below are necessarily preliminary. Please bear that in mind. I fully expect Bruce and Eric to jump in with their impressions as well.

 

Since our interest was in the UltraRendu and the sPS-500, and Bruce's in our trifectas, we designed a manageable set of listening comparisons to satisfy most interests. Even so, we inevitably ran out of time. The baseline configuration, which was held constant, and so is implied in the comparison sections below, was:

• Roon Core bridged server > dcBL7 > iSO6 > dCBL7 > endpoint or switch
• In configs with switch, the switch > endpoint cable was dCBL7
• USB cables:
  • Lush into DAC
  • USPCBs for intermediate
• Unless otherwise specified, each device was powered by an LPS-1
• The rest of the chain was held constant to Eric's setup:
  • Yggy DAC (fed via USB) > Audio Research Ref6SE > Sanders Magtech (?) > Magnepan 3.7i with dual Rythmik subs

Comparison 1: UltraRendu vs. sMS-200ultra (standalone)

We compared

• UR > DAC with
• sMS-200ultra > DAC. No reference clock was used. The sMS-200ultra was powered by 2 LPS-1s in series.

This one was very close. My preference varied with the track selection. One of our go-to tracks that emerged that day was the 5th movement (Rondo - Finale) of the recent BIS release of Mahler's 5th performed by the Minnesota Orchestra:

 

On the Mahler, I found the sMS-200ultra to have just that extra air, timbre, and refinement. However the UltraRendu was no slouch, and on some other jazz and blues tracks, had  a bit more pleasing weight and slam. Let's put it this way - if I owned one, I wouldn't be tempted to sell it to buy the other.

 

Comparison 2: UR+IR vs. full trifecta medley

Here we compared the following:

• UR > ISO-R > DAC, with

• modded switch > modded sMS-200 > ISO-R > tX-USBultra > DAC. Cybershaft OP-14 connected to tX-USBultra clock input

This wasn't a fair comparison, given the price differential between the two configurations, and the results were consistent. The full trifecta medley was head and shoulders better. I'll let Bruce give us his impressions here, as this was his first listen to the full trifecta.

 

The one observation I will make here is this. Depending on how you count, the price difference between the 2 configurations is $2600-3000. What I have found is that the SQ bump I hear from the $3k additional investment in this part of the digital audio chain is far greater than the difference I heard by stepping up to a DAC that was $3k more expensive than mine. Indeed in the cases of the DAVE and the QX-5, the difference was more like $10k and $7k respectively, and I don't feel the difference in SQ was as stark as what I am hearing in this region of the chain.

 

I know I keep harping on this point, but I'm a fan of value for money, and this area - for now - has proved to be a rich vein for me. FWIW.

 

Comparison 3: LPS-1 vs sPS-500

We used the full trifecta configuration: modded switch > modded sMS-200 > ISO-R > tX-USBultra > DAC. Cybershaft OP-14 connected to tX-USBultra clock input. We varied the PSU used on the tX-USBultra, because from past experience, the tX-USBultra in this chain seems to be the most sensitive to PSU quality. We compared:

• LPS-1 with Ghent starquad DC cable, set at 7V
• sPS-500 with SOtM silver Y-cable (only one output used), set at 7V
• sPS-500 with SOtM silver Y-cable (only one output used), set at 9V.

This one was just too close too call. Perhaps with more time and extended listening, subtle differences would emerge, but in the limited time we had, we just could not discern enough to tell these apart. We did wonder how much of a boost the silver DC cable was providing, but since the sPS-500 uses the funky Hirose connectors, we had no standard alternatives to swap in.

 

On the one hand, one could say the sPS-500 sounded wonderful, for an SMPS. On the other hand, one could say that the LPS-1, once again, demonstrated the sonic gem it is at its unbeatable price point.

 

Summary

I cannot overemphasize how much fun these meets are. I'm just amazed at how easy it is to click with fellow audiophiles. I hope to do this again with more local CA'ers.

Thank you for this very interesting feedback which clearly demonstrate the interest of the trifecta and clock upgrade .

 

I noticed that you did used a ghent dc cable with oyaide connectors . Did you compare those cable to anything else ? If yes did they sound better ?

Also do you think that the connectors , cable ... used by Ghent are genuine ? 

I have been living seven years in Shanghai were is located Ghent audio and i am also very suspicious about "genuine" product coming from china , they are so good at making copies of almost anything .

On the other hand i do agree they have also some very good products .

[jean-michel6]

6 hours ago, dgarretson said:

I haven't personally worked with mu-metal, but my understanding is that it loses much of its permeability when bent and needs to be heat-annealed thereafter to restore permeability.

 

From Wiki:

 

""Mu-metal objects require heat treatment after they are in final form—annealing in a magnetic field in hydrogen atmosphere, which increases the magnetic permeability about 40 times.^[4] The annealing alters the material's crystal structure, aligning the grains and removing some impurities, especially carbon, which obstruct the free motion of the magnetic domain boundaries. Bending or mechanical shock after annealing may disrupt the material's grain alignment, leading to a drop in the permeability of the affected areas, which can be restored by repeating the hydrogen annealing step."

Hi ,

 

The only thing i can say is that it does work . The impact of mumetal on cables is a quieter background , noise floor is decreased .

Even though bending may decrease it's permeability .

[jean-michel6]

12 hours ago, Johnseye said:

Intel X25 E and SATA II filter

 

Hi 

Thank's for ''tis very interesting feedback. 

I have also x25 E ssd that I should get in a few weeks. 

I am interested if you can test in your system how the sotm Sata II filter improves the sound. As I have sotm  sata III filter only.  

[jean-michel6]

9 hours ago, Johnseye said:

Tonight I compared the X25 E drive with and without the SOtM SATA II filter. First I listened to a handful of songs, then listened intently to two in particular for sounds to key in on. I pulled the filter out and repeated. 

 

The difference is very noticeable. It is the same as I previously experienced when comparing it to the M.2 drive. By recollection alone, I don't think there was much of a difference between the X25 and M.2.  What this filter does is eliminates the noise coming from the drive which contributes to a harsher digital edge sound. By eliminating this noise the sound becomes much more organic, natural and analog like. The soundstage expansion of width, height and depth or dimensionality remained without the SATA filter. I believe that is from the tX-USBexp and tX-USBultra.  The SATA filter tames the digital beast. It's the best $65 you can spend.

 

I have to say what you all probably have guessed. Roy was right. That guy is not only sharp as a tack but he has a great ear. His recommendations were spot on. That doesn't mean I would blindly follow, but I've tried many of his suggestions and haven't been misled. Right now my digital system is sounding incredible.  My recommendation is to try the SATA II filter for yourself. I don't think you need the X25 drive. Other SATA II drives will work. It's the filter that makes the difference. 

 

 

Hi ,

 

Thank's so much to tae the time and trouble to run this test . I will definitely go ahead and order some sata II filter .

 

You are right Roy was spot on when he mentioned that this filter had the biggest impact improvement on  SQ , better than the impact of sac ssd or a better sata cable .

 

By the way in my system a dual pc set up running roon i have replaced all sata cables with Pachanko pure reference . They are extremely good and do a very good job for the ssd with my music library . For the OS ssd it does also a good job but the impact is not as important as on the music library ssd.

I use to have a PPA sata red they are just much much better than the PPA red.

 

On my OS SSD on the player PC ( samsung Evo sata III ssd ) i have a SOTM sata III filter . My ssd are each individually powered by their own dedicated 5v power supply .

I have seen no benefit with the use of the sata iii filter . I recommend going the sata ii ssd and sot filter .

[jean-michel6]

On 15/12/2017 at 8:15 AM, austinpop said:

 

 

I think I've said this before, but it bears repeating from time to time, so people reading this thread are clear about all the confusing clock mods that get tossed around here!

 

The whole approach being discussed here with sCLK-EX and reference clocks is NOT ABOUT replacing clocks IN THE DAC! Why? Because most DAC designers - like Ted explains in his video - already take extraordinary care to use high-quality, low-phase-noise clocks in their DACs. It takes an extraordinary clock to improve on this. 

 

No - the clock mods being discussed here apply to the usually crappy oscillators found in switches, mobos, Ethernet and USB interfaces, and the like. Replacing these with high-quality, low-phase-noise clocks like the sCLK-EX, and further improved by OCXO reference clocks like the Ref 10, Cybershaft, etc, is what is yielding the SQ improvements so many of us are hearing.

 

I also want to be upfront about the fact that we do not really understand why we are hearing these SQ improvements. The SQ improvements undoubtedly exist. Our ears don't lie, and so many here have now heard the benefit. But the fact remains that there is not (yet) a clear scientific explanation for why we are hearing an SQ improvement. This is the usual point of divergence between the subjectivists and the objectivists. The former will accept and enjoy the benefit, while the latter will question whether, absent an explanation, the SQ benefit actually exists.

 

I like the approach @JohnSwenson is taking. Rather than being dogmatic, he's digging into the "why" and appears to be making some real headway into understanding what is going on.

 

Even though I am a subjectivist, I am not blind to the fact that this is a puzzle. So - what is the puzzle, really? Primarily, it relates to buffers - or buffering. Ted refers to this in the video. Simply put, the question is this: why would the "quality" of an upstream clock matter, when the data being clocked is flowing into a buffer? What is a buffer? In the most abstract sense, it is an area of storage (in this context, in a device's memory) where data is staged before being sent on. In a buffer, data can arrive and leave at different (clock) rates. Think of the buffers involved in a Tidal stream playing through Roon:

• The stream flows in from your ISP into your modem/router
• Switches and routers often have buffer memory at each port
• data flows into the Roon server over the network interface, into a memory buffer in the Roon application, from which it then flows out over the network interface
• data flows into a streamer (like the sMS-200ultra) over the network interface into a memory buffer in the Roon Ready app, before then flowing out over the USB interface
• many DACs implement a buffer into which data flows in over USB, say, and is then internally clocked in via the DAC's clock.

At every such buffer, if the data is received without error, then it is reasonable to ask - why on earth does the phase noise or jitter characteristics of the clock upstream of the buffer have any effect downstream of the buffer? It's a reasonable question, and is ofter the point of contention when people start arguing about this stuff.

 

Pending breakthroughs by the people actually digging into this - like John - I only have some conjectures. 

 

My strong suspicion is that what we are hearing is the effect of low phase noise clocks on good ol' fashioned analog noise. Let's not think of our long spaghetti chains in terms of their digital functionality, but rather as a connected chain of electronic components, through which analog noise can propagate. Perhaps the impact of low phase noise clocks upstream of the DAC is not in the digital domain, but rather to somehow reduce or mitigate analog noise, either in the data path, the ground plane, or both. Again - this is just a conjecture on my part.

 

One of the primary reasons for this conjecture is our other observation of what drives massive SQ improvements in this upstream chain - PSU quality. I've said before that clock quality and PSU quality are independent axes of optimization. We have ample evidence that extreme optimization on either of these axes has a marked effect on SQ.

 

I'll leave you with an intriguing thought. Do extreme optimizations on the PSU axis reduce the impact of clock optimizations? Roy's latest findings (over on head-fi) with the Zenith SE suggest this might be a distinct possibility. The Zenith SE does not have any heroic clock optimizations that we know of. What it does have are extreme PSU optimizations. The fact that Roy found it equalled or surpassed his reclocked server - at least in some areas - is highly intriguing. 

 

I expect to explore this further soon. But it does underscore the fact that there is much more research and analysis needed to explain why we are hearing these SQ improvements.

Hi Rajiv ,

Thank's for your writing on what do you think could explain why clock's upgrade improve sound . Your explanation makes a lot of sense . I agree with you that with all the buffering happening in the digital audio chain as long as bits are transferred properly , only the last clock in the chain should impact sound quality .

On my system i have made the following observation .

I have a PCie SOTM USB board and also a dx-USB HD SOTM usb to SPDIF converter,  on those two pieces of equipment who had originally good clocks .

For the PCie USB board i did an upgrade with sCLK-48.00 board and the Converter which had the optional good clock board was upgraded with an sCLK-ex board  .

With those two clock upgrade i have observed only a small sound quality improvement in my system .

 

To take advantage of the two available clock in the sCLK-ex board i had the two TP link MC 220 ( modified by SOTM ) .

My set up is a two pc's set up running roon and linked by an optical switch.

The improvement brought by this modification on TP link clock was very significant and much more important than the previous clock upgrade .

 

My "proposed"understanding is that the average clock that is on pc board or other computer type equipment are extremely noisy and replacing those by much more quiet and precise clock bring sonic improvement .

Also replacing crystal which are a source of EMI/RFI noise by a cable and connection ( if you use sCLK-ex board ) may be also help 

My two cents !

[jean-michel6]

5 hours ago, hurka said:

Some Lt 3045 boards arrived, Sotm sps 500 9v regulated to 2x5v feed singxer f1,and iso regen usb input,

same 9v connected to iso regen ,and tx usb ultra.Sparky sbc feed 5v 3A ultracaps,play from CF card,upsampled to 385khz.

Sound:

much better mids(stronger,accurate,clean) M a g i c a l !!!

better highs(some recordings too much,its not psu fault)

more airy than ultracaps(little)

bass more controlled(little)

 

Ultracaps warmer or just less clarity.

I need listening longer,this is just one  first impression.

Interesting results. I have also SPS 500 feeding SOTM dx USB hd and tx USB ultra. 

I have LT 3045 boards on order. 

Which LT 3045 boards did you use ? Did you modify them to add larger capacity ?

 

About ultra caps , could you give some info on what did you use ?

 

Thank's

 

 

 

 

 

[jean-michel6]

59 minutes ago, welldone said:

As May from SOTM said to me that LPS1 is not juicy enough to power Tx USBUltra.

Now, my Tx USBUltra has been feeding by a LPS1, and i really want to get your information about this.

 

Can anyone here using USBUltra with LPS1 share your own experiences?

Hi I have both SPS-500 and LPS-1 and also a Tx USB ultra and a dx USB hd ( with an sCLK ex clock board inside) .

From a power consumption i think that my dxUSB hd draw slightly more current than than the tx USB ultra .

The dx USB  accept 7~9v PS.

 

My dx USB hd run with the LPS1 which has no problem to drive it , i try the SPS-500 on this unit but i prefer the LPS-1 because it is a little bit more dynamic . 

 

For the tx USB ultra i have not try the LPS1 because my unit is set up for 9V and it won't run with the 7v LPS1 , i xill ask sotm to see which jumper should be modified .

There is three possible PS voltage for the tx USB ultra  6.5~8.5V , 9V , 12V  if yours is 6.6 to 8.5 you should be fine with the LPS1 as long as you do not the 5V on the USB connector from the tx USB ultra to power an other device

 

[jean-michel6]

1 hour ago, hurka said:

regulator picture alredy posted,

1a juice enough this time,dont trialed parallel

ultracaps:2.7v 100F PowerStor green cap

 

2.7 V ultracaps  ?  What do you power with such low voltage ?

 

Do you put them in serial for higher voltage ? 

[jean-michel6]

17 minutes ago, beautiful music said:

 

This is the answer i got it from Roy as well for his thoughts regarding to both Zenith SE and his custom server

 

"" With the Zenith SE combined with the tX-USBultra + Ref10, this combination is extremely close to my custom server. Because I need a 2nd server for my other listening room, I have purchased a Zenith SE""

Interesting information which shows that the zenith SE is a good product and the tx-USBultra a very good ad on to many systems .

 

One question keep puzzling me about Roy server system , he had obviously tick all the good thing's to get probably one of the best one board server system . Do you know which player he is using to run his system . Is it Roon ?

 

 

[jean-michel6]

1 hour ago, ElviaCaprice said:

Actually it doesn't tell us anything much.  The fact that the Zenith SE sounds better with the tXUSBultra, means it is deficient in some way to need a fixer/reclocking.from a master clock.  What would be more curious is how does the Zenith SE stack up against the sCLK-EX server, neither incorporating any fixers/added components.

 

1 hour ago, ElviaCaprice said:

Actually it doesn't tell us anything much.  The fact that the Zenith SE sounds better with the tXUSBultra, means it is deficient in some way to need a fixer/reclocking.from a master clock.  What would be more curious is how does the Zenith SE stack up against the sCLK-EX server, neither incorporating any fixers/added components.

I have a different view , I have a 2 pc set up ( server and player ) running with roon. 

The pc player output is with the SOTM pcie USB with it's own sCLK-48.0 clock board.  

 

Adding the txUSB ultra in the chain brings a valuable improvement to the sound even though the USB output of my player is already very very good. 

 

Other people ( Mozes ,may be ? ) are running two txUSB ultra in serial and got also more improvement. 

 

It it just shows the complexity of dealing properly with this crazy USB signal ....

[jean-michel6]

4 hours ago, ElviaCaprice said:

Actually it doesn't tell us anything much.  The fact that the Zenith SE sounds better with the tXUSBultra, means it is deficient in some way to need a fixer/reclocking.from a master clock.  What would be more curious is how does the Zenith SE stack up against the sCLK-EX server, neither incorporating any fixers/added components.

 

Deleted ...

[jean-michel6]

2 hours ago, seeteeyou said:

 

Awesome! That would a great service to give us a much better idea about where CHC would stand. Hopefully you're gonna swing for those Mundorf caps as well.

 

BTW, there's so much we could learn from Innous since they're really pushing the envelope IMHO

 

https://devialetchat.com/showthread.php?tid=3853&pid=60742#pid60742

 

https://devialetchat.com/showthread.php?tid=3853&pid=68044#pid68044

 

http://www.digitalaudioreview.net/2017/10/richer-sounds-with-the-innuos-zenith-mkii-se/

 

 

In other words, it's really all about the meticulous details as well as the road less traveled if you will. Otherwise @romaz wouldn't be so impressed by its performance even though he's got his own custom server (slightly better than ZENith SE that is) already.

Thank you .... Very interesting reading as i am in the process to develop a new server using this super micro board .

[jean-michel6]

9 hours ago, seeteeyou said:

 

You're very welcomed and that sounds exciting.

 

BTW, here's a little tip from Roy and maybe you'll find it helpful

 

https://www.computeraudiophile.com/forums/topic/30376-a-novel-way-to-massively-improve-the-sq-of-computer-audio-streaming/?page=50&tab=comments#comment-647362

 

He also compared 2GB memory modules to 8GB ones below

 

https://www.computeraudiophile.com/forums/topic/30376-a-novel-way-to-massively-improve-the-sq-of-computer-audio-streaming/?page=47&tab=comments#comment-644453

 

If you're able to run 2GB of RAM at 1,066 MHz, definitely get HMT425S6CFR6A from SK Hynix (it's like $9.95 a piece on eBay now) and you're gonna get much lower (maybe it's pretty much the lowest?) current draw to further reduce the noise

 

https://www.skhynix.com/products.view.do?vseq=1740&cseq=75

 

Thank's for this info. 

I agree that less current means less electrical noise and on our pc server set up we need to hunt for everything that can reduce current or Amp consumption :

- pick the right board , i spent quite some time before finding the super micro board

- shutdown all the processes and hardware not needed for audio

- minimize the ram size.

-.....

I am now using one 2gb memory module on the X10SBA. I am using crucial memory modules. 

I will look in the memory you recommend as lowest current draw. From my reading I got the info that Samsung memory modules were very energy efficient.  

 

 

 

[jean-michel6]

5 hours ago, sandyk said:

 

Somewhere there  has to be voltage regulators used in conjunction with the Ultracaps.

The Ultracaps are just large storage devices with much higher capacitance than the usual electrolytic capacitors used after a voltage regulator. and act  more like a battery. 

Basically, in this case they charge one bank of Ultracaps while the other bank supplies the load until the need to automatically change over to the charged bank.

This helps with isolation from the A.C. mains supply.

Can you use just ultra caps after a regulator just to replace the electrolytic cap. 

Is it possible ? And is there a benefit to do that ?