Topping D10 measurement issue

[brydon10]

Hi guys, I'm not sure what is going on with this Topping D10 of mine. I've done some rudimentary measurements that are not to be taken seriously, but clearly show some issue. I've used the line in on my motherboard for this.
I was trying to get some basic measurements of a Schiit Vali 2 amp. First one using the D10 DAC. Second one using the optical input of the SMSL Sanskrit 10.

For some reason the D10 is measuring very poorly. I'm wondering if it is because it is from 2018 and probably has about 3 years of powered on time.
I have tried different USB cables and different USB ports, and a different amp with the D10, same performance every time.

Vali 2 with Topping:

Vali 2 with SMSL over optical:

Any thoughts? Thanks.

 

[olieb]

Any thoughts? Thanks.

Maybe some kind of ground loop effect? Did you check with the USB input of the Sanskrit?

 

[audio_tony]

Your first measurement is indicative of a ground loop somewhere. The Sanskrit probably measures ok because you're using an optical connection.

My D10 still measures fine and it has been on pretty much since I bought it (February 2019).

 

[staticV3]

Use a USB isolator (not filter!) between the computer and D10 and the issue will most likely disappear.

That's not a fault of the D10.
If you connect the SMSL via USB, then it too will measure like that.

 

[brydon10]

Use a USB isolator (not filter!) between the computer and D10 and the issue will most likely disappear.

That's not a fault of the D10.
If you connect the SMSL via USB, then it too will measure like that.

Here is the SMSL connected and powered by USB. It's still quite a bit better than the D10. My question too, since it seems like a ground issue from the D10, is this measured performance indicative of what I am actually getting when using it for audio, or does the ground loop get "fixed" when it's just being used as a regular DAC and not trying to measure it? Thanks.

 

[brydon10]

Maybe some kind of ground loop effect? Did you check with the USB input of the Sanskrit?

Yes, just the post above. It's not fantastic but it's better than the D10. I used the same USB port as the Topping.

 

[audio_tony]

.....is this measured performance indicative of what I am actually getting when using it for audio, or does the ground loop get "fixed" when it's just being used as a regular DAC and not trying to measure it?

The ground loop "distortion" is still present even when using it for audio. It makes no difference if it's being measured or being used for audio - the noise will still be present. The noise is low enough that you probably can't hear it anyway.

Here's some noise measurements I did a while back - the last graph on the page shows what a ground loop does.

USB noise measurements

USB noise measurements

gtkc.net

 

[brydon10]

The ground loop "distortion" is still present even when using it for audio. It makes no difference if it's being measured or being used for audio - the noise will still be present. The noise is low enough that you probably can't hear it anyway.

Here's some noise measurements I did a while back - the last graph on the page shows what a ground loop does.

USB noise measurements

USB noise measurements

gtkc.net

Yea, that's unfortunate. I'm going to read through your information. I thought I was using a pretty good spec for all these years, when turns out my DAC was not doing so well. I usually use a lower distortion headphone amp than the Vali 2, but it got me curious to do some measurements so here we are. The D10 "distortion" is not audible to me so I guess at the end of the day it didn't really matter.
Here is a measurement of the (USB) Schiit Fulla v2. I don't see any ground loop problem here really, I still wonder why my D10 in particular is so bad. It must just be super sensitive for some reason.

 

[MaxwellsEq]

I thought I was using a pretty good spec for all these years, when turns out my DAC was not doing so well.

You are incorrectly blaming the DAC for a poor installation setup. The earth loop or noise being injected is caused by the setup.

 

[DVDdoug]

The D10 "distortion" is not audible to me so I guess at the end of the day it didn't really matter.

Right! There's no point in improving the measurements if he sound doesn't need improving.

I think it's noise rather than distortion. Background noise is most noticeable with silence.

 

[audio_tony]

Yea, that's unfortunate. I'm going to read through your information. I thought I was using a pretty good spec for all these years, when turns out my DAC was not doing so well. I usually use a lower distortion headphone amp than the Vali 2, but it got me curious to do some measurements so here we are. The D10 "distortion" is not audible to me so I guess at the end of the day it didn't really matter.
Here is a measurement of the (USB) Schiit Fulla v2. I don't see any ground loop problem here really, I still wonder why my D10 in particular is so bad. It must just be super sensitive for some reason.

View attachment 477889

That measurement contains a fair bit of noise, the shape of which suggests a ground loop to me.

The ground loop is no fault of the DAC, and it will depend what the audio output of the DAC is connected to.

If the amplifier / preamp / headphone amp that the DAC is connected to** has any sort of ground reference, then a ground loop will be formed.

**and the source

In the case of your measurements, the ground loop is being formed between the source that the DAC is connected to, and the ground for your PC.

Note: this may not necessarily be a ground reference as such - it may even be noise induced through a switched mode power supply powering the source that the DAC is connected to, or even the power supply of your PC / amplifier etc. etc.

When carrying out any kind of measurements, you have to be sure that a ground loop (or other potential noise source) has been eliminated.

Hence, any USB connection is best made via a USB isolator (Topping HS01 / HS02 or similar). Similarly, if your audio source is a PC or other device that connects to the DAC vis USB - an isolator can be particularly helpful in eradicating ground induced noise.

The noise I mentioned from switch mode power supplies is not always ground induced - a lot of these supplies have a very small value capacitor connected between the primary and secondary for EMI / RFI suppression - you know if your power supply has such a cap, as often touching the DC output and mains ground can give you a little tingle. In the case of a two pin plug connection, this can even change according to which way round the plug is connected (meaning live / neutral is changed according to plug orientation).

 

[brydon10]

Thanks for the replies. I'm definitely confident it is a good DAC, it's just unfortunate my setup is causing issues. If I can find a reasonably priced USB isolator I might go that route, or just use the SMSL for now. The D10 is super handy for it's optical and coaxial outputs anyways which are always in use.

 

[olieb]

The ground loop is no fault of the DAC, and it will depend what the audio output of the DAC is connected to.

To me there are two ways to look at this.
If a dac is only supposed to produce great output under good to perfect conditions then yes, it is not the fault of the dac, and the setup is to blame.
But if ground loops are considered a common problem and therefore this holds:

Hence, any USB connection is best made via a USB isolator (Topping HS01 / HS02 or similar).

Then a dac with a USB input should put isolation in place to pass the test. Sadly this is by no means what happens on a regular basis. Why would I have to buy an additional device if the problem happens all the time? Manufacturers seem to be much more inclined to improve sinad from 123dB to 127dB instead.

EDIT:
And it is well possible that the ground loop is not (or much less) there with audio. This happens if the measurement part of the setup is creating the loop. The adding of the measurement part (and the unplugging of the audio chain if that was done) can both increase or decrease the noise and the measurement again is a different thing. It is not so simple to get a valid measurement with ground loops.
Obviously the noise is different for the two dacs and that is no surprise, every device is different here. It's all hit and miss.
I use a D10 to convert to Toslink and then I connect optical to DSP and to DAC.