Do I need clean USB power when using DAC with its own power supply?

[visigt]

Hello all,

 

I have a DAC that takes power from a wall socket and the audio signal input uses USB.

 

Will the "dirty PC power" in the USB cable still adversely affect the signal line even though the DAC may not be using the power?

 

Will I benefit from PCI USB cards with separate power supply (eg. Matrix element H)? Can I simply block the "dirtu power"simply by blocking the power connection in the USB plug?

 

Thanks!

 

 

[audiobomber]

On 7/7/2021 at 3:20 AM, visigt said:

Hello all,

 

I have a DAC that takes power from a wall socket and the audio signal input uses USB.

 

Will the "dirty PC power" in the USB cable still adversely affect the signal line even though the DAC may not be using the power?

 

Will I benefit from PCI USB cards with separate power supply (eg. Matrix element H)? Can I simply block the "dirty power"simply by blocking the power connection in the USB plug?

 

 

IME it depends on the specific DAC.

- My Audiolab DAC was powered from the wall outlet, did not not need power from USB. I blocked the 5V pin with electrical tape and heard a cleaner sound.

- The Benchmark DAC3 needed 5V power from the USB feed, otherwise it would not play. 

- My Schiit Modius DAC can be powered from the USB cable, or from the supplied wall wart. If the wall wart is used, minimal power is drawn from the USB cable. I found no difference when powering it from the wall wart, or from a clean USB feed via sMS-200 and LPS, as predicted by the Schitt FAQ.

 

You will need to investigate your particular situation. Could be any of the above.

[audiobomber]

I should have mentioned, the Benchmark also plugs into a wall outlet. The Schiit does not have a power cord, it is powered externally.

[visigt]

On 7/10/2021 at 11:39 AM, audiobomber said:

IME it depends on the specific DAC.

- My Audiolab DAC was powered from the wall outlet, did not not need power from USB. I blocked the 5V pin with electrical tape and heard a cleaner sound.

- The Benchmark DAC3 needed 5V power from the USB feed, otherwise it would not play. 

- My Schiit Modius DAC can be powered from the USB cable, or from the supplied wall wart. If the wall wart is used, minimal power is drawn from the USB cable. I found no difference when powering it from the wall wart, or from a clean USB feed via sMS-200 and LPS, as predicted by the Schitt FAQ.

Interesting experience with very different results, esp the Benchmark DAC which still needs USB power despite having its own wall plug power supply.

 

I guess I'll go with feeding my USB clean power and hope it will improve the sound. At least the DAC won't refuse to work like if the USB power is blocked.

 

Thanks.

 

[audiobomber]

44 minutes ago, visigt said:

Interesting experience with very different results, esp the Benchmark DAC which still needs USB power despite having its own wall plug power supply.

 

I guess I'll go with feeding my USB clean power and hope it will improve the sound. At least the DAC won't refuse to work like if the USB power is blocked.

 

The Element H looks like it would help. I've not tried anything similar.

 

The USB bus on both my Raspberry Pi 3B+ and Dell Inspiron laptop is very dirty. There are two solutions I've used to feed clean power to a DAC that needs USB 5V.

1. I have an iFi iDefender, which interrupts USB 5V power when you connect an external power source to its micro USB port. 
2. I have some USB cables with two male USB A connectors one male USB B connector. I plug USB B into the DAC, USB A to the computer, and the second USB A leg goes to a cleaner power supply.

If I were you, I would try blocking the USB 5V pin first, as this is a no-cost solution.

https://community.octoprint.org/t/put-tape-on-the-5v-pin-why-and-how/13574

 

[botrytis]

So far the answer is no. There is a thread in the objective forum her that shows a well designed DAC should not allow noise from a PC go through to the analog output.

[botrytis]

Disagree if you want @One and a half it is fine. I have asked people to show me noise measurements and they always tell me, I can't measure it but I can hear it. Well, that may be 'Expectation bias'. Meaning if you think the noise is there, your brain will add it in for you.

 

A link to the noise measurement thread. 

 

PC Noise measurments

[opus101]

42 minutes ago, botrytis said:

There is a thread in the objective forum her that shows a well designed DAC should not allow noise from a PC go through to the analog output.

 

Those measurements were made in the normal mode. But common-mode is where the noise over USB shows up.

[audiobomber]

55 minutes ago, botrytis said:

Disagree if you want @One and a half it is fine. I have asked people to show me noise measurements and they always tell me, I can't measure it but I can hear it. Well, that may be 'Expectation bias'. Meaning if you think the noise is there, your brain will add it in for you.

 

A link to the noise measurement thread. 

 

PC Noise measurments

The measurements purportedly show that the PC output doesn't change under load. That is a totally different premise than saying my RPi or laptop USB power makes my DAC sound bad, all the time. 

[botrytis]

3 minutes ago, audiobomber said:

The measurements purportedly show that the PC output doesn't change under load. That is a totally different premise than saying my RPi or laptop USB power makes my DAC sound bad, all the time. 

 

Actually, THAT is the point. One of the computers he used was a laptop so yes it is the same.  Maybe it is your DAC. Have you tried another DAC or just blaming the laptop or RPi? The point being that noise did not transfer from the computer THROUGH the DAC.

 

Heck, one of the PC's I used to use before it died was a small purpose built AMD mini PC. No noise out of that what so ever with either my iFi NANO ( first version) or my TEAC UD-501. 

 

Just saying.

[audiobomber]

58 minutes ago, botrytis said:

 

Actually, THAT is the point. One of the computers he used was a laptop so yes it is the same.  Maybe it is your DAC. Have you tried another DAC or just blaming the laptop or RPi? The point being that noise did not transfer from the computer THROUGH the DAC.

 

Heck, one of the PC's I used to use before it died was a small purpose built AMD mini PC. No noise out of that what so ever with either my iFi NANO ( first version) or my TEAC UD-501. 

 

Just saying.

Here's the list of DAC's I've tried with my laptop: iFi iOne, Sabaj Da2, Korg DS-DAC-10R, Korg DS-DAC-100M, XTremPro X1, Schiit Modi and Modius. All benefitted from external power. I used an Audiolab 8200CD with the laptop but blocked the USB 5V pin and heard cleaner sound.

 

I did not try the Benchmark DAC3 or my current exaSound e32 with the laptop. I used the Benchmark with an sMS-200. The sMS-200 uses a linear power supply, a common upgrade among owners. I use a Teddy Pardo 12/2 PSU with the e32, an upgrade recommended by exaSound. I use a Zerozone 2020 S.S. LPS with the Playpoint streamer. ExaSound does not recommend upgrading the PSU of the Playpoint. The Zero-Zone provided a barely perceptible improvement over the stock supply, but I like that my main audio system contains no noisy SMPS PSU's. 

 

 

Quote

Actually, THAT is the point. One of the computers he used was a laptop so yes it is the same. 

 

No, not the same at all:

Linked test: Measure USB noise under light load and heavy load, no identified difference

My situation: Play music from laptop via USB. Replace USB 5V power from the laptop with cleaner power from an external PSU. Load is irrelevant. Whether working hard or just playing music, USB power from the laptop, and RPi, affects sound quality of the DAC. 

 

Did you ever try bypassing or dead-ending the laptop USB power to your DAC? 

[One and a half]

6 hours ago, botrytis said:

Disagree if you want @One and a half it is fine. I have asked people to show me noise measurements and they always tell me, I can't measure it but I can hear it. Well, that may be 'Expectation bias'. Meaning if you think the noise is there, your brain will add it in for you.

 

A link to the noise measurement thread. 

 

PC Noise measurments

This statement is the reason why there's an objective forum to nut out these details, in the 'open' the discussions can get very tense. 

I thought to raise the disagree flag and leave it at that, but no, the same tired arguments come into play, which I will ignore. Expectation bias over several weeks listening for cable or a different source, doesn't apply, can also be applied equally that you don't expect any change, cause the measurements say it isn't so - laughable. Be honest and say that's good enough for me instead, have no problem with that, but don't hide behind measurements.

 

Measuring single sine waves and how they work won't cut it for 2Ch music with rapid dv/dt all over the place. Stage height, depth, location of instruments relies on critical timing. With every USB cable change, most members here can hear the change for better or worse as far as that image is concerned. EMI is easy to transmit over very small capacitance, and USB especially when, not if, the potentials of the grounds of the source and target a different even by a few microvolts, which would be nice, but from a PC? Hardly attainable by any stretch of the imagination, with over a dozen SMPS/PWM controllers let alone the millions of gates in the CPU, bus controllers and the like. Each transistor has a gate, source and drain, there's capacitance between the gate and both when the transistor is switched on/off it emits EMI, that's fact. 

[cab33]

6 minutes ago, One and a half said:

  Expectation bias over several weeks listening for cable or a different source, doesn't apply, can also be applied equally that you don't expect any change, cause the measurements say it isn't so - laughable.  

 

Can somebody translate this into English ?

[audiobomber]

3 hours ago, cab33 said:

 

Can somebody translate this into English ?

Yes, it means expectation bias goes both ways:

• I expect a difference, therefore I will hear one
• I expect no difference, therefore I will hear none

[kumakuma]

5 minutes ago, audiobomber said:

I expect no difference, therefore I will hear none

 

Probably true but I believe that differences that small probably don't matter... 

[audiobomber]

1 hour ago, kumakuma said:

 

Probably true but I believe that differences that small probably don't matter... 

Unless you pile them on top of one another.

 

[kumakuma]

16 minutes ago, audiobomber said:

Unless you pile them on top of one another.

 

 

I don't think that all subtle differences are additive and combine to make a larger difference. 

[AudioDoctor]

On 7/9/2021 at 10:39 PM, audiobomber said:

- My Schiit Modius DAC can be powered from the USB cable, or from the supplied wall wart. If the wall wart is used, minimal power is drawn from the USB cable. I found no difference when powering it from the wall wart, or from a clean USB feed via sMS-200 and LPS, as predicted by the Schitt FAQ.

 

 

The Unison USB input is pretty awesome. I wish the Modius was Multibit as well.

[MaxBuck]

10 hours ago, AudioDoctor said:

 

The Unison USB input is pretty awesome. I wish the Modius was Multibit as well.

What specific benefit would you seek to gain by "multibit" configuration? 

[AudioDoctor]

1 hour ago, MaxBuck said:

What specific benefit would you seek to gain by "multibit" configuration? 

 

sounds better in my opinion.

[audiobomber]

On 7/15/2021 at 11:42 PM, AudioDoctor said:

 

The Unison USB input is pretty awesome. 

Unison is very good, but still benefits from a better USB source. Connecting the Modius to the sMS-200 provided much improved sound vs. connecting to my laptop. 

[Archimago]

On 7/15/2021 at 5:31 AM, botrytis said:

So far the answer is no. There is a thread in the objective forum her that shows a well designed DAC should not allow noise from a PC go through to the analog output.

 

I agree @botrytis.

 

I haven't looked at the objective forum thread but all these years of testing and reading comments by people that say a linear power supply makes a difference (or silly USB filters like the JitterBug), I'm left scratching my head.

 

Yes, one should try to keep one's DAC away from a noisy computer to reduce interference especially from busy CPU/GPUs. Yes, USB ports can be noisy. Indeed, we can use battery power to reduce noise.

 

But in the big scheme of things, the differences are really tiny unless one does something silly like put the DAC right on top of the computer case running at 100% load!

 

This week, I published some tests of the Topping D10s which is just a USB DAC with no power supply of its own. I measured it connected to the Raspberry Pi 3 B+ which @audiobomber claims is "very dirty" (which it might well be, who knows), yet the noise floor and resolution including multitone intermodulation testing doesn't show anything unusual/noisy. Likewise, I don't hear any issues - dead quiet when I connected it to my main system in the basement with low ambient noise and great playback resolution.

 

Bottom line is that I see a lot of words by people who complain about noise, yet in all these years, where's the evidence? For those who have "issues" with switch mode power supplies, what exactly are you hearing? Is this something you've measured?

 

Remember years ago Benchmark released their video of linear vs. switching power supply showing that there are times when the switching supply is in fact better. Take heed audiophiles because I think there's just too much nonsense out there with many people talking generalizations that are simply unsubstantiated.

 

[fas42]

39 minutes ago, Archimago said:

 

Bottom line is that I see a lot of words by people who complain about noise, yet in all these years, where's the evidence? For those who have "issues" with switch mode power supplies, what exactly are you hearing? Is this something you've measured?

 

 

 

What one hears is a degradation of SQ - it's almost never an actual 'noise', which is audible over the speakers; or prominently visible on a scope, say.

 

A simple test: a recording which is marginal to listen to - that is, at times it can sound awful; or, other times it's quite pleasant to experience, nothing untoward that stops one from enjoying the performance ... a noise interference issue can well and truly send the listening right down the negative end - you are ready to hit the stop button almost immediately ... that's the nature of the beast.

[kumakuma]

20 minutes ago, fas42 said:

 

A simple test: a recording which is marginal to listen to - that is, at times it can sound awful; or, other times it's quite pleasant to experience, nothing untoward that stops one from enjoying the performance 

 

 

I have never had this experience. I wonder if anyone other than Frank has.

[botrytis]

39 minutes ago, kumakuma said:

 

I have never had this experience. I wonder if anyone other than Frank has.

 

Frank and  computer in the same sentence?  🤣