My (very) mini review of the Singxer SU-1 DDC

[SolidCore]

 

Hello Mogos

 

I understand what you say. I can remove SMD parts, but cannot solder it. If the test without the SMD caps bad,

i must live without them. Thats why i dont try it. I believe  it can make a difference.

I am so happy with my SU-1, i wont do any more Mods. I have also two older Highend CD-Transport,

a Mission Discmaster with PSX-R (contains a Philips CDM9Pro), and a Theta Transport. They both are much better as the SU-1 without Mods, but now they Sound in the same League. Now i can hear Musik from PC on the same Level.

 

Greetings

Stephan

 

[Horace28]

Can someone explain how the LPS-1 connects to the SU? I just bought one and can't figure it out. The plug in this picture doesn't look like it would fit anywhere into the back of my SU.

 

https://cdn.shopify.com/s/files/1/0660/6121/products/UltraCap_LPS1_kit.jpg

[SolidCore]

Hello Horace28

 

You need a Adapter to the Plug which fits on the SU-1 Mainboard. You must configure your LPS-1 with 5V.

I can say, that your LPS-1 sounds much better than the original PSU. I have also tried it.

 

The connect to the SU-1 you can see in this threat on page 2.

 

Greetings

Stephan

 

 

[Horace28]

5 hours ago, SolidCore said:

Hello Horace28

 

You need a Adapter to the Plug which fits on the SU-1 Mainboard. You must configure your LPS-1 with 5V.

I can say, that your LPS-1 sounds much better than the original PSU. I have also tried it.

 

The connect to the SU-1 you can see in this threat on page 2.

 

Greetings

Stephan

 

 

 

I'm not sure I understand. I have to open it up and solder some wires to it? If so, I'm going to send it back. I thought it would work out of the box and I'm terrible at soldering.

[roberto2]

I think that you need that:

 

https://kitsunehifi.com/product/singxer-su-1-dc-power-conversion-kit-kitsunehifi-2-1mmx5-5mm/

[BigAlMc]

Hi @Horace28,

 

I bought the Kitsune kit and it was really easy to set up. 

 

Likewise I'm not skilled enough to even consider soldering or anything that technical. 

 

Regards, 

Alan 

[motberg]

I agree it is an easy mod and in my case was absolutely necessary to get acceptable sound in my system from the SU-1..

But please remember that there could be potentially lethal voltages inside.

[SolidCore]

 

Hello together

 

Here is another SU-1 with Toroidal Transformer for Audio, with 6N Copper and Earth-Shielding between primary and secondary Coil.

It has a sheed of MU-metal, so you have no interferences with magnetic fields.

The Toroidal Transformer sounds with a little more "Attack" or "Speed". While the R-Core sounds more "Sweet" and "warm".

 

 

Greetings

Stephan

 

[drmike]

can the small power supplies made by alexi on ebay, who makes the lt3045 regs.,  be used for the mod shown by solid core?

thanks,

drmike

[Mogos]

Solid Core thank you  for sharing with us your findings.

They ave very valuable. Especially that you can test odifferent configurations and their impact on sonics of the su-1

[elan120]

Just completed another SU-1 mod and this one turned out to be very effective and not so subtle on the improvement.

 

After working on down streaming side for a short while, I decided to replace the voltage regulator on the USB input side to see if the lower noise output regulator (<6.5uVrms/PSRR 82dB@1KHz) will help improve the overall SQ, and apparently better spec'ed regulator does help cleaning up the playback, which I speculate lowered the jitter from the 24.000MHz oscillator that is supplying the clock signal to the XMOS 208 chip.

 

If you are comfortable working with small surface mount device, and this one is a little on the small side at 2.90mmx1.60mm, I would highly recommend to give this a try.  The end result is a more detailed, clean, focused music, and the separation between instruments also improved by good margin, quite similar to the 3.3V regulator change on the down streaming side that many people have experienced, but this regulator take the improvement up another step.

 

Pictures below show both old regulator removed and new regulator installed.

[Duy Le]

10 minutes ago, elan120 said:

Just completed another SU-1 mod and this one turned out to be very effective and not so subtle on the improvement.

 

After working on down streaming side for a short while, I decided to replace the voltage regulator on the USB input side to see if the lower noise output regulator (<6.5uVrms/PSRR 82dB@1KHz) will help improve the overall SQ, and apparently better spec'ed regulator does help cleaning up the playback, which I speculate lowered the jitter from the 24.000MHz oscillator that is supplying the clock signal to the XMOS 208 chip.

 

If you are comfortable working with small surface mount device, and this one is a little on the small side at 2.90mmx1.60mm, I would highly recommend to give this a try.  The end result is a more detailed, clean, focused music, and the separation between instruments also improved by good margin, quite similar to the 3.3V regulator change on the down streaming side that many people have experienced, but this regulator take the improvement up another step.

 

Pictures below show both old regulator removed and new regulator installed.

 

Could you please provide the product link on Mouser/Digikey/Farnell ?

 

Thanks.

[Energy]

31 minutes ago, elan120 said:

Just completed another SU-1 mod and this one turned out to be very effective and not so subtle on the improvement.

 

After working on down streaming side for a short while, I decided to replace the voltage regulator on the USB input side to see if the lower noise output regulator (<6.5uVrms/PSRR 82dB@1KHz) will help improve the overall SQ, and apparently better spec'ed regulator does help cleaning up the playback, which I speculate lowered the jitter from the 24.000MHz oscillator that is supplying the clock signal to the XMOS 208 chip.

 

If you are comfortable working with small surface mount device, and this one is a little on the small side at 2.90mmx1.60mm, I would highly recommend to give this a try.  The end result is a more detailed, clean, focused music, and the separation between instruments also improved by good margin, quite similar to the 3.3V regulator change on the down streaming side that many people have experienced, but this regulator take the improvement up another step.

 

Interesting modification. What is the specifications for the new regulator versus the old one?

Could I get a part number?

Have you done the oscillator mod?

 

I think I will do this for the fun of it. It's hard to say that the overall jitter has been improved considering asynchronous technology but I reckon there may be some other influence it has that makes up the chain.

[elan120]

39 minutes ago, Duy Le said:

 

Could you please provide the product link on Mouser/Digikey/Farnell ?

 

Thanks.

 

15 minutes ago, Energy said:

 

Interesting modification. What is the specifications for the new regulator versus the old one?

Could I get a part number?

Have you done the oscillator mod?

 

I think I will do this for the fun of it. It's hard to say that the overall jitter has been improved considering asynchronous technology but I reckon there may be some other influence it has that makes up the chain.

Forgot to write down the regulator I used...it is LP5907MFX-3.3/NOPB made by TI.  Mouser and Digikey both carry this item, but Mouser was out of stock couple weeks ago when I was looking at their website.  This regulator is very inexpensive, so the cost of this experiment will likely be lower than the shipment cost, very much worth to give this 15 minute mod a try and see if it will help in your system.

 

I replaced the two down streaming oscillators (22.5972MHz and 24.576MHz), and they also help improved the overall SQ by good margin in my system, which is another highly recommended mod for SU-1 (picture below).  As far as the 24.000MHz oscillator on the up streaming side, it didn't seems to change much for me (could be just placebo) or perhaps I need to find a better spec'ed oscillator.

 

The main motivation for replacing this up stream regulator was based on the improvement I got from using ISO Regen, the improved signal also is on USB input side prior to entering XMOS 208, so I thought why not give the regulator change a try.

 

Concerning the stock regulator spec., unfortunately, I am unable to identify what it is based on the device marking, so I basically use the best spec'ed regulator in the same physical size I can find and went for it.  Fortunately, the result is very positive.

 

The simple spec for LP5907MFX-3.3/NOPB was mentioned earlier, the general spec show 82dB of PSRR @1KHz / 20mA, 6.5uVrms of output noise at 250mA.  For detailed spec., use the link here:  http://www.ti.com/lit/ds/symlink/lp5907.pdf

 

Please report back your result if you end up going for this mod.  I will be very interested in your input and hopefully help improved your system as well.

[Duy Le]

22 minutes ago, elan120 said:

 

Forgot to write down the regulator I used...it is LP5907MFX-3.3/NOPB made by TI.  Mouser and Digikey both carry this item, but Mouser was out of stock couple weeks ago when I was looking at their website.  This regulator is very inexpensive, so the cost of this experiment will likely be lower than the shipment cost, very much worth to give this 15 minute mod a try and see if it will help in your system.

 

I replaced the two down streaming oscillators (22.5972MHz and 24.576MHz), and they also help improved the overall SQ by good margin in my system, which is another highly recommended mod for SU-1 (picture below).  As far as the 24.000MHz oscillator on the up streaming side, it didn't seems to change much for me (could be just placebo) or perhaps I need to find a better spec'ed oscillator.

 

The main motivation for replacing this up stream regulator was based on the improvement I got from using ISO Regen, the improved signal also is on USB input side prior to entering XMOS 208, so I thought why not give the regulator change a try.

 

Concerning the stock regulator spec., unfortunately, I am unable to identify what it is based on the device marking, so I basically use the best spec'ed regulator in the same physical size I can find and went for it.  Fortunately, the result is very positive.

 

The simple spec for LP5907MFX-3.3/NOPB was mentioned earlier, the general spec show 82dB of PSRR @1KHz / 20mA, 6.5uVrms of output noise at 250mA.  For detailed spec., use the link here:  http://www.ti.com/lit/ds/symlink/lp5907.pdf

 

Please report back your result if you end up going for this mod.  I will be very interested in your input and hopefully help improved your system as well.

I don't know how much LDO help to improve the sound because I fed the USB with Elfidelity USB Booster w/iPower, it isolated VBus and ground from PC. But I will try this mod when I have opportunity.

Another question: could you please describe the improvement after modding clocks ? I read somewhere the NZ2502SD does not sound harsh such as CCHD-575 but it is less detail.

[Energy]

1 hour ago, elan120 said:

Forgot to write down the regulator I used...it is LP5907MFX-3.3/NOPB made by TI.  Mouser and Digikey both carry this item, but Mouser was out of stock couple weeks ago when I was looking at their website.  This regulator is very inexpensive, so the cost of this experiment will likely be lower than the shipment cost, very much worth to give this 15 minute mod a try and see if it will help in your system.

 

I replaced the two down streaming oscillators (22.5972MHz and 24.576MHz), and they also help improved the overall SQ by good margin in my system, which is another highly recommended mod for SU-1 (picture below).  As far as the 24.000MHz oscillator on the up streaming side, it didn't seems to change much for me (could be just placebo) or perhaps I need to find a better spec'ed oscillator.

 

Interesting. I will get one to try out however at this point I find it hard for me to do A/B tests.

 

I did the 24.000MHz modification and found some super small improvement but again, I don't know if it's placebo at this point so I'll just say it was only put in for good measure.

 

Where did you source your downstream oscillators? Is it better specification than the Crystek CCHD-575?

Did it come preloaded on those carrier boards or did you have to solder it yourself?

 

I might as well go the extra mile and have all these mods done to my unit as well. 

[SolidCore]

5 hours ago, Duy Le said:

I don't know how much LDO help to improve the sound because I fed the USB with Elfidelity USB Booster w/iPower, it isolated VBus and ground from PC. But I will try this mod when I have opportunity.

Another question: could you please describe the improvement after modding clocks ? I read somewhere the NZ2502SD does not sound harsh such as CCHD-575 but it is less detail.

Hello together

 

If you feed the USB Side of SU-1 with iPower, it will improve the Sound. But the Voltage from iPower goes on the Mainboard to the LDO,

which feeds the XMOS-Chip. The LDO adds much Noise by himself. So it is a great idea to change this LDO to a much better, with lower noise. I am sure it improves the Sound much more as the iPower.

But why change the LDO with a better? You can also change it with a Superregulator like Belleson SPZ17 3,3V, or a Pulsar. The little Pulsar Board can fixed near the old LDO, and solder with short cable.

http://www.pulsarclock.com/Power.html

Or any other Superegulator like NewclassD and so on.

 

Greetings

Stephan

 

[elan120]

6 hours ago, Energy said:

 

Where did you source your downstream oscillators? Is it better specification than the Crystek CCHD-575?

Did it come preloaded on those carrier boards or did you have to solder it yourself?

 

I got mine from the link below (via message), and these sorted oscillators have a phase plot that show -112 to -113dBc/Hz @ 10Hz

http://www.ebay.com/usr/analog.research

 

These oscillators neither come mounted nor with carrier board.  I soldered them to the carrier boards myself and then solder the carrier boards to the main board.

[roberto2]

Hello! I have tried what Stephan (Solidcore) suggested: to put Elna Silmic 2 instead of Panasonic capacitors.

 

I have purchased yesterday 4 x Elna Silmic 2 220uF 35V (did not find 25V in my reseller, but it is the same size) and 1 x Elna Silmic 2 100uF 50V for the output of Sparkos Labs 3,3 regulator.

 

I dont have the picture of this first mod. The sound at the beginning was clearly quit horrible! A lot of basses, no air, highs were disappeared ... 

I red that this kind of capacitors take a lot of time to break in (50 to 150 hours!)

In the evening things sounded better but still I had the impression to have "loudness" in my system.

 

Today I wanted to understand which was the role of every capacitors so I have decided to change one Elna with Panasonic FR.

 

Firstly I have changed the USB side capacitor:

High fréquences were back but there was something not coherent.

 

So I have decided to put the second in 5V input before the Sparkos Labs regulator:

Now the sound was better and transparency was back and could say it was more similar to 5 Panasonic FR configuration but with more weight on notes and with more meat around the bone.

 

Finally I decided to put again the Elna on USB side thinking I was going to loose transparency again but...no, finally it sounded very well!

 

So I don't know if the Elna are breaking in, if my ears are getting addicted to this sound or if it is very important to have at least a Panasonic FR in the 5V input before the 3,3 regulator to avoid to loose to much dynamics and life.

 

I will keep this last set up for a while to break in the Elna capacitors. It seems to me that there is more weight and meat then with all Panasonic FR capacitors.

 

 

[Energy]

On 9/27/2017 at 9:44 AM, elan120 said:

I got mine from the link below (via message), and these sorted oscillators have a phase plot that show -112 to -113dBc/Hz @ 10Hz

http://www.ebay.com/usr/analog.research

 

These oscillators neither come mounted nor with carrier board.  I soldered them to the carrier boards myself and then solder the carrier boards to the main board.

 

There are many different ones listed, which one be the one?

Also where did you get your carrier board?

 

7 hours ago, roberto2 said:

Hello! I have tried what Stephan (Solidcore) suggested: to put Elna Silmic 2 instead of Panasonic capacitors.

 

I have purchased yesterday 4 x Elna Silmic 2 220uF 35V (did not find 25V in my reseller, but it is the same size) and 1 x Elna Silmic 2 100uF 50V for the output of Sparkos Labs 3,3 regulator.

 

I dont have the picture of this first mod. The sound at the beginning was clearly quit horrible! A lot of basses, no air, highs were disappeared ... 

I red that this kind of capacitors take a lot of time to break in (50 to 150 hours!)

In the evening things sounded better but still I had the impression to have "loudness" in my system.

 

Today I wanted to understand which was the role of every capacitors so I have decided to change one Elna with Panasonic FR.

 

Firstly I have changed the USB side capacitor:

High fréquences were back but there was something not coherent.

 

So I have decided to put the second in 5V input before the Sparkos Labs regulator:

Now the sound was better and transparency was back and could say it was more similar to 5 Panasonic FR configuration but with more weight on notes and with more meat around the bone.

 

Finally I decided to put again the Elna on USB side thinking I was going to loose transparency again but...no, finally it sounded very well!

 

I mean you shouldn't be using analog capacitors (Elna RFS) on a digital board to begin with. Their ESR is much higher to where it's not even specified and the material dampening doesn't help for digital domain thus why you hear removal of high frequency detail and or bloating of the bass.

 

For all Elna Silmic II that I use on analog domains such as amplifiers, the bass does increase and tighten itself over time. The overall sound smoothens and jarring treble softens up. Again, I would not use this within a DAC or Digital Transport.

[elan120]

9 hours ago, Energy said:

 

There are many different ones listed, which one be the one?

Also where did you get your carrier board?

 

The best way is to contact Pat using the message from the link and let him know you are looking for a matched pair of 22.5972MHz and 24.576MHz oscillators.  He will hand sort the oscillators, and the lead-time typically been one to two weeks pending the order size.

 

I made these carrier boards from blank PCB's I purchased through general PCB supplies.

 

On 9/27/2017 at 2:29 AM, Duy Le said:

I don't know how much LDO help to improve the sound because I fed the USB with Elfidelity USB Booster w/iPower, it isolated VBus and ground from PC. But I will try this mod when I have opportunity.

Another question: could you please describe the improvement after modding clocks ? I read somewhere the NZ2502SD does not sound harsh such as CCHD-575 but it is less detail.

I also use dual head USB cable to separate VBus and USB signal, but this is through iFI micro iUSB3.0, and I speculate the improved result likely due to having cleaner output 3.3VDC feeding to the 24.000MHz oscillator resulting in lower jitter clock signal going into XMOX 208, but this is only my hypothesis, and is purely based on the 3.3VDC regulator change from the down-streaming side.  The good thing is, this is a very low risk experience if you are comfortable with small size surface mount devices.

 

The normal production NZ2520SD is good, but by just looking from the spec., it isn't necessary better then CCHD-575, the sorted pair I installed has phase noise plot -112 to -113dBc/Hz @ 10Hz.  In my system, these oscillators provided better detail, impact, quicker and more defined bass, cleaner treble.  Overall presentation got wider and deeper, and layering is also improved as well.  As always, this is very much system dependent and YMMV.  In my case, it was well worth the effort changed to these oscillators.

[roberto2]

10 hours ago, Energy said:

I mean you shouldn't be using analog capacitors (Elna RFS) on a digital board to begin with. Their ESR is much higher to where it's not even specified and the material dampening doesn't help for digital domain thus why you hear removal of high frequency detail and or bloating of the bass.

 

For all Elna Silmic II that I use on analog domains such as amplifiers, the bass does increase and tighten itself over time. The overall sound smoothens and jarring treble softens up. Again, I would not use this within a DAC or Digital Transport.

 

I found that the sound was not good in the first hours but now I like it a lot!

 

Maybe I will try other options like Elna Cerafine or Nichichon Polymer that I'm waiting. But I think that the most important thing is to listen with your ears and not only see technical specifications.

 

I see that Kitsune use Nichicon FG 470uF 16V. If it was bad sounding why would they do it?

 

[roberto2]

And why Light Harmonic use Elna Silmic II in their DAC? :

 

https://www.audiostream.com/content/light-harmonic-da-vinci-dac-page-2

[Energy]

On 9/29/2017 at 1:10 AM, roberto2 said:

 

I found that the sound was not good in the first hours but now I like it a lot!

 

Maybe I will try other options like Elna Cerafine or Nichichon Polymer that I'm waiting. But I think that the most important thing is to listen with your ears and not only see technical specifications.

 

I see that Kitsune use Nichicon FG 470uF 16V. If it was bad sounding why would they do it?

 

I answered these questions 10-20 pages ago.

You are more than free to listen to your ears.

Technical specifications are more important in digital design than you think.

If what you're using sounds good to you, stick to them.

 

Nichicon FG is fine. It's not bad. They're just not optimal. 

On 9/29/2017 at 5:45 AM, roberto2 said:

And why Light Harmonic use Elna Silmic II in their DAC? :

 

https://www.audiostream.com/content/light-harmonic-da-vinci-dac-page-2

 

Elna Silmic II (RFS) is one of the most popular audiophile capacitor. They have their benefits when used in certain parts of the circuit. A DAC, even while being digital, has some analog areas of it's own where ESR doesn't play a big role and material composition becomes more beneficial.

[pkane2001]

I see that there is now a v4.11 Class-2 driver published for SU-1 in the last week or two. I installed it but couldn't see (or hear) any difference with the previous version. Any idea what's different in the new driver?