• 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!

Worlds Worst USB Interface?!

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
How bad can it be?
is it "transparent"
whats SINAD, THD, Noise and output voltage?
Make a guess and i will later post some Measurements.

photo5244657455491823207.jpg
photo5244657455491823208.jpg
photo5244657455491823209.jpg
 

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,096
Likes
7,570
Location
Bjerringbro, Denmark.
@JayGilb Because it's a piece of silicon substrate soldered directly to the PCB?

BTW, there's a short between two of the USB pins?
 
OP
L

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
1637423560370.png

1637423646517.png

This is the output of the DAC.
SINAD of around 47dB
Output level -1.2dBV
Noise is lower then exacted-
1637423960956.png

-84dbV (-86.5dbV A) noise with a peak from the 1khz USB Clock viable.

With smoothing.
Green: 0dBFS
Orange: Output disabled
Blue:150Ohm reference (DAC noise)
1637424357176.png
 
OP
L

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
1637426088767.png

A THD+N of 38dB at around 300Hz is relay bad...

IMD -25dB
1637426532209.png



IMD AES:
-32dB
Super bad imaging and filter.
1637426734883.png



1/3okt 31 band multi tone:
1637427223018.png



I wonder if you could abx it vs your go-to DAC
I need build a setup to test this.
but looking at the AES IMD test im sure i can hear this giving the right test signal.
with "normal" music it might be harder.
 

Feargal

Member
Joined
Sep 2, 2019
Messages
90
Likes
159
Maybe they could market this as the ADC for electric guitarists - no distortion pedal needed!

[edit - whoops, helps to wait until the ADC graph has been posted!]
 
Last edited:
OP
L

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
The ADC is surprisingly good!
-72dBFS /-79.5dBV noise with 150ohm

1637428573282.png


No clipping till -9dBV input voltage.
THD is exceptionally low with -87dB
So Almost 70dB dynamic range.
SINAD of 51dB is also ok.
1637428849654.png
 

Attachments

  • 1637428702696.png
    1637428702696.png
    78.1 KB · Views: 72

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,703
Location
Hampshire
Can you zoom in on the 1 kHz fundamental? I'd like to see how it compares to this monstrosity:

1637432889507.png
 
OP
L

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
Can you zoom in on the 1 kHz fundamental? I'd like to see how it compares to this monstrosity:
What Window function and FFT length did you use?
1637433944644.png


I think this is Hann and 2M
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,373
Likes
3,318
Location
.de
1637423560370.png

1637423646517.png

This is the output of the DAC.
SINAD of around 47dB
It's clearly Not Happy like this, try reducing the output by 1...3 dB and see what that does. Looks like it's clipping to me. I would expect the output stage to be running on the same supply as the DAC, in which case you will never quite get to 0 dBFS. -1.2 dBV (0.87 Vrms) is almost exactly 2.5 Vpp. Alternatively, similar max output has been observed with misc. Realtek chips on +3.3VA.
 
OP
L

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
It's clearly Not Happy like this, try reducing the output by 1...3 dB and see what that does. Looks like it's clipping to me.
I have tied to find an point of better performance.
1637486941626.png

1637487096486.png

to me this looks like THD+N is failing linearly with the output.

in which case you will never quite get to 0 dBFS. -1.2 dBV (0.87 Vrms) is almost exactly 2.5 Vpp. Alternatively, similar max output has been observed with misc. Realtek chips on +3.3VA.
The output is DC coupled and at 2.5V with no signal.

i also measured this when connecting input straight to the output with a short cable
1637487617510.png

Not sure whats going on.
Looks like comb filtering to me but i cant make sense of it.
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,709
Likes
4,771
Location
Germany
It
I have tied to find an point of better performance.
View attachment 167085
View attachment 167087
to me this looks like THD+N is failing linearly with the output.


The output is DC coupled and at 2.5V with no signal.

i also measured this when connecting input straight to the output with a short cable
View attachment 167089
Not sure whats going on.
Looks like comb filtering to me but i cant make sense of it.
it makes sense if you bring two timedelayd signals togheter.
 

Roland68

Major Contributor
Joined
Jan 31, 2020
Messages
1,449
Likes
1,271
Location
Cologne, Germany
That will be a PCM2912 or something similar in economy circuit.
What does that cost?
 
OP
L

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,791
Likes
1,525
What does that cost?
2$ including shipping.

But there are at least 5 different version with the same caes on ebay and you don't now what version you gonna get.
Some have the chip on the front some on the back of the pcb.
some have solder ic some have epoxy blob
some have decoupling output caps (this one don't)
some have a Quartz (this one don't)

it is one of this products that might started out ok but it got "cost optimized" so often so now you don't even have output capacitors.
 
Top Bottom