• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Review and Measurements of SMSL SU-8 DAC

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Q is only applicable to 2nd order filters.

If you plot the values, you can easily see the filter shape.

I'm sorry, I still don't understand. How do I plot the values? Perhaps you have a link that explains it? Or @Krunok who seems to understand could explain it perhaps?
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
If you can't do a simple Cartesian plot by hand, drop the numbers into Excel and plot them. Conventionally, we use the x axis for frequency and the y axis for amplitude.
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
I still don't understand if anyone else wouldn't mind chipping in to explain, please?

I thought you were giving some kind of data for different filters at each specific frequency you listed.
 
Last edited:

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
Oh hold on a second, you're only describing the bandwidth for one filter at 200Hz. Makes perfect sense now.
 
  • Like
Reactions: SIY

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
In this case it is very likely that power supply spuriae at 180Hz and 250Hz would come into play, wouldn't you agree @Krunok?
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
The D to A conversion doesn't care about frequency, within reason. So you get no more information by running linearity at multiple frequencies. What you do by that is to conflate D to A nonlinearity with the analog linearity. You're best using a relatively low frequency ( like 200 Hz, or 1 kHz, or...) to determine D to A linearity, then a separate sweep at a fixed amplitude (preferably near the max) of THD vs frequency. That keeps the two mechanisms separate.

Ok, I see what you mean regarding the D to A conversion - you're right.

But wouldn't the analog output stage be challenged to maintain low THD throughout the whole freq range at lower amplitudes as well?
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
In this case it is very likely that power supply spuriae at 180Hz and 250Hz would come into play, wouldn't you agree @Krunok?

Not really, especially not with that odd shape of the red curve (starting above 0dB and then going lower). What's your say on this @SIY ?
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Yeh, but we can never hope to have the complete picture, just good enough data to make educated inferences about the complete picture, surely?

Surely not - as we are paying 100EUR monthly access fee to participate on this forum we have every right to demand as detailed measurement as can possibly be done! :D:D
 
Last edited:

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
But wouldn't the analog output stage be challenged to maintain low THD throughout the whole freq range at lower amplitudes as well?

Yes, but high levels really challenge them, especially as the open loop gain starts to roll off. If memory serves, there was a paper by Bob Cordell about this.

Not really, especially not with that odd shape of the red curve (starting above 0dB and then going lower). What's your say on this @SIY ?

You're supporting my point about the importance of power averaging. :cool:
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Yes, but high levels really challenge them, especially as the open loop gain starts to roll off. If memory serves, there was a paper by Bob Cordell about this.

THD figure measured at 0dB of the balanced output certainly proves your point.

You're supporting my point about the importance of power averaging. :cool:

Yes, indeed it does. :)
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,558
Location
Seattle Area
Ok, I see, thank you. And how narrow is the bandpass filter? Noise peaks from power supply seem to be around 180Hz and 250Hz.
You can see my filter here:
Jude Filtering versus Amir.png
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,558
Location
Seattle Area
On reducing the amplitude of balanced to get rid of its saturation, here are the results as I step from 0 to -10 dB in 2.5 dB steps:

SMSL SU-8 dashed versus balanced level measurement.png


As you see, there is a trade off there. Initially as you lower the amplitude from zero (top blue line going to green), distortion improves as it should. But as you keep making progress there, you start to lose dynamic range due to higher noise to signal ratio now. So it is not possible to get to unbalanced performance. The black line for example is for -10 dB which has no additional distortion but 4-5 dB worse than unbalanced.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
@amirm Thank you for additional effort! Although I believe none of this would be an issue during listening test it obviously shows that some additonal engineering effort is needed to improve this device.

What is your take on not-so-perfect linearity on unbalanced output?
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,033
Location
Tampa Bay
Seems like Topping still reigns supreme, but other mfg's are stepping up to the plate and that's what is important!
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,033
Location
Tampa Bay
Was just wondering regarding the SMPTE 'hump'. Other than the DAC chip what else might be common between them? Other than the D50, DX7s seems to have it too right ?

Did the NX4 DSD exhibit that ? (don't recall seeing that)

Could it be the dual chip implementation or an opamp they use in common that's causing that ?
I posted about this in another thread, its from the 9038Q2M with the "THD compensation" turned on.
 

mindbomb

Active Member
Joined
Nov 11, 2017
Messages
284
Likes
175
I posted about this in another thread, its from the 9038Q2M with the "THD compensation" turned on.
John Siau commented on this in the dac 3 thread. He said it was because the analog section of the problem dacs didn't have high enough common mode rejection. And the problem dacs all seem to use the opa1612, so it seems like that's the culprit? A fully differential amp needs to be used instead?
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,880
Likes
2,033
Location
Tampa Bay
John Siau commented on this in the dac 3 thread. He said it was because the analog section of the problem dacs didn't have high enough common mode rejection. And the problem dacs all seem to use the opa1612, so it seems like that's the culprit? A fully differential amp needs to be used instead?
Can you link me to that post?
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,558
Location
Seattle Area
You already know my question;)- is the deviation from "linearity" because of actual deviation from linearity or is it from noise?
Since the deviation is negative, it is not noise. Here is the linearity using very narrowband FFT method:

SMSL SU-8 Balanced FFT linearity Measurement.png


Each peak needs to hit the corresponding horizontal line. We see deviations starting around -100 dB and it gets worse, always showing lesser amount. Not that I measure it past -120 but you can see it is all pretty wrong in that region.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,376
Likes
234,558
Location
Seattle Area
What is your take on not-so-perfect linearity on unbalanced output?
It is fine. We are talking about a $250 device with a lot of functionality. :) For CD playback it is definitely good enough anyway.
 
Top Bottom