High quality power supplies for single board computers

[barrows]

Linux is a very flexible computing platform.  Really, the OS used for our (Sonore) products, and most Linux based servers as well, is as you describe: designed from the ground up for audio purposes.

[nbpf]

5 hours ago, barrows said:

Linux is a very flexible computing platform.  Really, the OS used for our (Sonore) products, and most Linux based servers as well, is as you describe: designed from the ground up for audio purposes.

In which sense is Sonicorbiter designed from the ground up for audio purposes? Doesn't it rely on MPD, upmpdcli, alsa-utils just as Volumio, DietPi, etc. do? Basic latency improvements can be achieved by simple OS tuning as documented in https://wiki.linuxaudio.org/wiki/raspberrypi. I am actually not aware of any documented OS that has been designed from the ground up for audio purposes but I might be mistaken, of course. Projects like  http://wtfplay-project.org/index.html#about go quite a long way towards optimization and minimalism but they do not attempt at designing any OS from the ground up, I believe.  

[nbpf]

Back to the topic: has anybody compared UpTone Audio's JS-2, Custom HiFi Cables' DC3, Paul Hynes power supplies, etc. to less expensive wall wart power supplies like the ifi iPower and the Allo 5V/3A for powering SBCs?

[Qhwoeprktiyns]

1 hour ago, nbpf said:

In which sense is Sonicorbiter designed from the ground up for audio purposes? Doesn't it rely on MPD, upmpdcli, alsa-utils just as Volumio, DietPi, etc. do? Basic latency improvements can be achieved by simple OS tuning as documented in https://wiki.linuxaudio.org/wiki/raspberrypi. I am actually not aware of any documented OS that has been designed from the ground up for audio purposes but I might be mistaken, of course. Projects like  http://wtfplay-project.org/index.html#about go quite a long way towards optimization and minimalism but they do not attempt at designing any OS from the ground up, I believe.  

Agreed

[vortecjr]

2 hours ago, nbpf said:

In which sense is Sonicorbiter designed from the ground up for audio purposes? Doesn't it rely on MPD, upmpdcli, alsa-utils just as Volumio, DietPi, etc. do? Basic latency improvements can be achieved by simple OS tuning as documented in https://wiki.linuxaudio.org/wiki/raspberrypi. I am actually not aware of any documented OS that has been designed from the ground up for audio purposes but I might be mistaken, of course. Projects like  http://wtfplay-project.org/index.html#about go quite a long way towards optimization and minimalism but they do not attempt at designing any OS from the ground up, I believe.  

SonicOrbiter is one part closed source and one part open source. The closed source is 100% developed by SGC and Sonore. This part of the OS is the software on top of the software that controls and configures the system. The closed source is Fedora GNU/Linux. We do not develop this part of the operating system, but as can be expected we interact with some of the developers on a regular basis. There is also some third party apps that are closed source that run on the unit. Also SonicOrbiter is meant to be a flexible "system" and it can be configured to act as a music server / NAS or a renderer. These work well independently or concurrently of each other. For the music server we are using off the shelf hardware. For the renderer we are developing our own hardware and the software is highly optimized (in ways I don't care to explain in a public forum) to work with each other. We feel this approach gives us greater flexibility to control the vertical and horizontal.

[vortecjr]

1 hour ago, nbpf said:

Back to the topic: has anybody compared UpTone Audio's JS-2, Custom HiFi Cables' DC3, Paul Hynes power supplies, etc. to less expensive wall wart power supplies like the ifi iPower and the Allo 5V/3A for powering SBCs?

If by "compare" you mean measure...I have measured various power supplies with our gear. I have our power supply, JS-2, iFi and a few others here. 

[Qhwoeprktiyns]

A lot of what has been outlined above is described in this white paper from Digi International, whose boards are probably a good basis for audio as well. 

 

http://ftp1.digi.com/support/documentation/0220069_e.pdf

[vortecjr]

26 minutes ago, hopkins said:

A lot of what has been outlined above is described in this white paper from Digi International, whose boards are probably a good basis for audio:

http://ftp1.digi.com/support/documentation/0220069_e.pdf

Do they have linear regulators and high quality oscillators....probably not.   

[nbpf]

1 hour ago, vortecjr said:

SonicOrbiter is one part closed source and one part open source. The closed source is 100% developed by SGC and Sonore. This part of the OS is the software on top of the software that controls and configures the system. The closed source is Fedora GNU/Linux. We do not develop this part of the operating system, but as can be expected we interact with some of the developers on a regular basis. There is also some third party apps that are closed source that run on the unit. Also SonicOrbiter is meant to be a flexible "system" and it can be configured to act as a music server / NAS or a renderer. These work well independently or concurrently of each other. For the music server we are using off the shelf hardware. For the renderer we are developing our own hardware and the software is highly optimized (in ways I don't care to explain in a public forum) to work with each other. We feel this approach gives us greater flexibility to control the vertical and horizontal.

My post was not meant to be in any way dismissive of the work involved in selecting and integrating existing software components and in providing customization options for those components, e.g., via new and well designed web interfaces. It is just that I do not believe that such developments warrant the outcomes to be denoted as  "designed from the ground up". It is finally only a matter of taste in the way language is used and we can agree to disagree on this point, I believe! Thanks for the clarifications on SonicOrbiter!  

[nbpf]

1 hour ago, vortecjr said:

If by "compare" you mean measure...I have measured various power supplies with our gear. I have our power supply, JS-2, iFi and a few others here. 

Sure, careful comparisons can involve measurements. But they do not have to and I am also very interested in the subjective findings of users who have powered their DigiOne, USBridge, HiFiBerry, etc. devices with high quality power supplies. Have such PSUs brought improvements in sound quality over power supplies like the ifi iPower? Did they have any impact at all?

 

As put forward in the title, I am interested in the impacts of high quality power supplies on SBCs that are deployed (typically as hosts for devices like the Allo DigiOne) as server and/or network renderers. The impacts of high quality power supplies on devices like the ultraRendu, the sMS-200ultra, etc. have been discussed in detail in other threads in this forums. 

[vortecjr]

No worries. I understood your point and thus the reason for my clarification. The product pages also state that Sonicorbiter is closed source and open source GPLv2. 

[vortecjr]

4 minutes ago, nbpf said:

Sure, careful comparisons can involve measurements. But they do not have to and I am also very interested in the subjective findings of users who have powered their DigiOne, USBridge, HiFiBerry, etc. devices with high quality power supplies. Have such PSUs brought improvements in sound quality over power supplies like the ifi iPower? Did they have any impact at all?

 

As put forward in the title, I am interested in the impacts of high quality power supplies on SBCs that are deployed (typically as hosts for devices like the Allo DigiOne) as server and/or network renderers. The impacts of high quality power supplies on devices like the ultraRendu, the sMS-200ultra, etc. have been discussed in detail in other threads in this forums. 

We can lump these all together and talk in generalities since I don't like to pick on other products. I mostly measure USB to analog and USB to SPDIF devices. The devices I use are powered from the SBC. So the SBC and the converter remain the same and the power supply varies. I can describe what I see if you feel that relates to your question.        

[Qhwoeprktiyns]

I certainly would be interested as well on your feedback concerning those tests.

[nbpf]

11 minutes ago, vortecjr said:

We can lump these all together and talk in generalities since I don't like to pick on other products. I mostly measure USB to analog and USB to SPDIF devices. The devices I use are powered from the SBC. So the SBC and the converter remain the same and the power supply varies. I can describe what I see if you feel that relates to your question.        

Sure, go ahead. The more we understand the better. Thank for making you expertise available to this discussion!

[vortecjr]

Comparing SPDIF, USB, and a DAC hat with different power supplies would be to complex right out of the gate. I'm most familiar with USB to analog, but I don't feel that the DAC hat will be much different with respect to the results with different power supplies. What are you actually using right now? 

[vortecjr]

2 hours ago, hopkins said:

A lot of what has been outlined above is described in this white paper from Digi International, whose boards are probably a good basis for audio as well. 

 

http://ftp1.digi.com/support/documentation/0220069_e.pdf

So I had a look at these boards and I see the usual stuff I would want to remove from my audio board:)    

[nbpf]

2 minutes ago, vortecjr said:

Comparing SPDIF, USB, and a DAC hat with different power supplies would be to complex right out of the gate. I'm most familiar with USB to analog, but I don't feel that the DAC hat will be much different with respect to the results with different power supplies. What are you actually using right now? 

I am using an ifi 5V iPower to power a RPi + DigiOne and Teddy Pardo 12V/4.5A and 9V/2A units to power fitPC3 and M2Tech hiFace Evo, respectively. The Schiit Eitr is powered by its own AC/AC wall wart. I am particularly interested in whether a JS-2, DC3 or SR4 could possibly improve the sound quality of the RPi + DigiOne combo. I understand that the best results would likely be obtained by powering RPi and DigiOne separately, but this is not the focus of this thread. Thanks, nbpf

[Qhwoeprktiyns]

24 minutes ago, vortecjr said:

So I had a look at these boards and I see the usual stuff I would want to remove from my audio board:)    

 

Good thing you did before everyone started buying it

[vortecjr]

24 minutes ago, nbpf said:

I am particularly interested in whether a JS-2, DC3 or SR4 could possibly improve the sound quality of the RPi + DigiOne combo. I understand that the best results would likely be obtained by powering RPi and DigiOne separately, but this is not the focus of this thread. Thanks, nbpf

There is a simple test you can conduct. iFi Audio sells a device called the Groundhog that allows you to connect the ground of your AC power source  to the negative terminal of the DC output of your iFi iPower supply. Depending on your electrical skills you can also make the cable.    

[tapatrick]

On 5/11/2018 at 4:36 AM, barrows said:

...But, I would caution folks to understand that one is never going to get the best performance for audio by using the commercial computer gears suggested here.  Even with an ultra low noise pre-regulated supply, these boards are going to have noisy DC/DC switching regulators onboard, so they are going to make a lot of noise on the board....

On 5/11/2018 at 3:27 PM, rikhav said:

 

Do laptops have same noisy dc/dc converters ?

If yes then what would kind of computer would be best for audio for someone who can't spend big bucks on music servers or likes of it 

 

Sorry for sidetracking the main topic of the thread but I thought I'd add my 2 cents.... It's true using a SBC like a Rpi or a Sparky on its own, even with low noise power inputs will not result in a high quality sound BUT I have been working on a Sparky Roon Ready I2s streamer chain for a few months and just got it all working together today after many hiccups. The results are no less that stunning and I am not exaggerating. This set up could also be used as a source but I currently run it as a streamer fed by my Music server running Roon core.

 

I wanted to make use of the I2s input on my PS Audio Junior DAC as I believe it is the best input. It's taken a lot of trial and error and I highly recommend checking out the build and components as it is VERY good value for the end result.

 

My new streamer chain is: Sparky Roon end point (network connected running customised Dietpi software with Roon bridge on a 4 GB micro card) I2S outputs connected to Ian Canadas McFifo Isolator board, then connected to his McDualXO reclocking board with upgraded Crystek CCHD-957-25-45.1584 & CCHD-957-25-49.152 clocks, then to an Audio gd HDMI module to I2S input on my PS Audio Junior DAC. It's all fed by LifePo4 batteries, LT3045s and SR4 power supplies. Everything is running superbly and sound is stunning even though all the inputs of the Junior DAC are excellent this takes it to another level with a detailed, sweet, ultra smooth, rich and harmonious sound.

 

I shared the journey here but there wasn't a lot of interest, what can you do..:)

 

 

[chauphuong]

3 hours ago, nbpf said:

I am using an ifi 5V iPower to power a RPi + DigiOne and Teddy Pardo 12V/4.5A and 9V/2A units to power fitPC3 and M2Tech hiFace Evo, respectively. The Schiit Eitr is powered by its own AC/AC wall wart. I am particularly interested in whether a JS-2, DC3 or SR4 could possibly improve the sound quality of the RPi + DigiOne combo. I understand that the best results would likely be obtained by powering RPi and DigiOne separately, but this is not the focus of this thread. Thanks, nbpf

What matter most for a SBC like Pi, as indicated by Barrows, is the switching dc-dc converter onboard. 

[nbpf]

3 hours ago, vortecjr said:

There is a simple test you can conduct. iFi Audio sells a device called the Groundhog that allows you to connect the ground of your AC power source  to the negative terminal of the DC output of your iFi iPower supply. Depending on your electrical skills you can also make the cable.    

I am not sure I understand: I have no problems with humming or buzzing components in my system. The only device connected to the amplifier (a SN2) is the Naim DAC and its grounding is set to chassis as per manual. I understand that if I had to connect another grounded device to SN2 I would have to set the grounding on the DAC to floating to avoid ground loops. But the Groundhog seems to address cases in which the grounding is missing (no device in the chain providing a grounding) which is not the case here. Am I missing something? 

[vortecjr]

38 minutes ago, nbpf said:

I am not sure I understand: I have no problems with humming or buzzing components in my system. The only device connected to the amplifier (a SN2) is the Naim DAC and its grounding is set to chassis as per manual. I understand that if I had to connect another grounded device to SN2 I would have to set the grounding on the DAC to floating to avoid ground loops. But the Groundhog seems to address cases in which the grounding is missing (no device in the chain providing a grounding) which is not the case here. Am I missing something? 

The iFi iPower Supply does not have a ground pin.

 

Have a look at this post for details on what this ground trick does: 

 

[nbpf]

1 hour ago, vortecjr said:

The iFi iPower Supply does not have a ground pin.

 

Have a look at this post for details on what this ground trick does: 

 

Got it! I am going to try it tomorrow, thanks! 

[nbpf]

20 hours ago, vortecjr said:

There is a simple test you can conduct. iFi Audio sells a device called the Groundhog that allows you to connect the ground of your AC power source  to the negative terminal of the DC output of your iFi iPower supply. Depending on your electrical skills you can also make the cable.    

I have now done the test with a simple cable as suggested in John Swenson's original post. It is difficult to say whether there is an improvements or not. Perhaps a little bit more clarity and separation in certain complicated string quartet passages, but it could be just my imagination. I'll leave the grounding for a while and then try to revert. Assuming the grounding does yield slight improvements: which conclusions shall I draw from the test? Does the result imply that it would be worth upgrading to a JS-2, SR4, DC3, etc.? Does it imply the opposite? Thanks, nbpf