SMSL M500 DAC Only “Thermal Issue” not “Golden Sample”

[WolfX-700]

Thanks. Meanwhile I am testing mine starting from cold and measuring THD+N once a second to see if I can replicate your scenario.

I actually put it on my balcony-it was only about 32-40 ° F.

Next I will run the test and record the THD + N curve with a recorder

 

[amirm]

I actually put it on my balcony-it was only about 32-40 ° F.

Next I will run the test and record the THD + N curve with a recorder

Thanks. That is exactly what I am running as I type this. Please be sure to use the dashboard setting for THD+N recorder. And set the THD+N notch tuning to Generator Frequency. The frequency counter in AP can get confused if it sees too much noise generating false THD+N readings (not an issue here though).

 

[WolfX-700]

Thanks. That is exactly what I am running as I type this. Please be sure to use the dashboard setting for THD+N recorder. And set the THD+N notch tuning to Generator Frequency. The frequency counter in AP can get confused if it sees too much noise generating false THD+N readings (not an issue here though).

OK, and by opening the doors and windows of my workshop, the current indoor temperature is about 53 ° F

Here is the first measurement. I hope I didn't make a mistake.

 

[amirm]

It takes a while for the guts to cool off. Please let it do that.

 

[WolfX-700]

 

[amirm]

I am definitely seeing thermal effects after 30 minutes. I will let it run until stabilizes.....

 

[amirm]

View attachment 41679

Our posts crossed. So yes, that is what I am seeing. But my starting point is a lot better than yours.

 

[amirm]

Confirming that you did use 22.4 kHz bandwidth and 44.1 kHz sampling for the thermal graph.

 

[WolfX-700]

Confirming that you did use 22.4 kHz bandwidth and 44.1 kHz sampling for the thermal graph.

These parameters can be seen in the lower right corner of the picture

Measurements are still in progress ...

 

[stunta]

You guys should get on Skype or something These live updates are creating so much suspense though!

Appreciate the effort to keep the manufacturers honest! This is a truly awesome website.

 

[patrickdmiller]

The suspense!!!

 

[MusicNBeer]

I knew I was hearing distortion products above -110 dBFs!

 

[restorer-john]

I've been shouting about pre-conditioning for amplifiers. Why should D/A converters and headphone amplifiers get a free pass and be characterized when cold? It's hardly a typical usage scenario- these things are on for hours and hours.

Load them (say 5/10K for D/A converters and 16/32 ohms for headphone amplifiers) and run them for an hour at rated output @ 1KHz before the tests.

 

[WolfX-700]

 

[DDF]

There are many things that can lead to measurement differences at such sensitive levels. Unit to unit manufacturing variation, temperature, susceptibility to RF, conducted RF on the interconnects, radiated RF and how they affect other gear in the set up, even the cleanliness of the AC.

@amirm and @WolfX-700 take care with their test set ups, and so represent a more ideal case. Buyers of gear need to recognize that their particular situation may note be so ideal, and this level of performance may not be possible. For example, as Amir has shown, so much consumer gear doesn't go through proper RF (FCC or otherwise) testing so there's no knowing how your other gizmos are bombing each other with RF, or how good they are at rejecting being bombed. Rf can modulate down to the audio band and raise the noise and add spurious low level tones. Ground loops between pc and other gear can cause noise floors to raise and USB frame rates (8kHz) to be visible on the plots.

For another example, my home has bad AC feed, and is in a RF hot zone. I characterized the RTX6001 in my home environment:
https://www.audiosciencereview.com/...udio-measurement-gear.113/page-12#post-153972

Just changing the RTX's power cord to a lower gauge with slightly poorer shielding increased the power line harmonics in loopback by 5 to 10 dB.

Temperature dependence is also a gremlin, especially when a device is using highly sensitive I/V converters (current noise is tough to stabilize over temp). Here are two measurements I took of a Lite Dac Ah back in '06 (using just a SB card), one at room temp, one after 20 mins in the freezer (not that cold).

By all means it'll be interesting to see where this leads and maybe it'll lead to a lesson learned to not accept gear from manufacturers (especially after the cherry picked pot). But I bring this up to hopefully ease minds to not to worry so much about such small measured differences and how it'll affect your listening pleasure (it won't).

 

[Feyire]

View attachment 41683

This should be a new reference chart to include in all future reviews. Starting with the DAC at room temperature and playing over 1 hour or so.

 

[restorer-john]

Once the thing has reached operating temperature and with your AP still reading, you can quickly whip off the casework and hit the usual suspects with freezer spray to locate the source of the trouble. Start with the D/A converter itself.

 

[Veri]

This should be a new reference chart to include in all future reviews. Starting with the DAC at room temperature and playing over 1 hour or so.

That does seem wise

 

[renaudrenaud]

Well, I thing it is unfair to accuse SMSL before providing strong evidence and let explain their point of view.

So it's necessary to cool down and maybe change the wordings.

I am very interested by the results, but I do not want conclude before having all the elements in hand.

 

[RayDunzl]

I sense a new category of audiophile tweakery coming on...

Refrigerant TypeR-290
Temperature Range 33 - 40 Degrees F