• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Review and Measurement of Cheap Ebay CM108 Bare Board USB DAC

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
13,493
Likes
5,662
Location
Seattle Area
#1
This is a Review and Measurement of Cheap Ebay CM108 Bare Board USB DAC. There are a number of these with the cheapest one retailing for just USD $8 (https://www.ebay.com/itm/CM108-TDA1...833370?hash=item467db6a85a:g:9hwAAOSwkJZbEbqj). It is part of a collection of USB dacs to test from a kind member.

Here is what it looks like:

CM108 Bare Board USB DAC review and Measurement.png.jpg


The white solder mask saves the thing from looking crusty and cheap. Let's see how it measures.

Measurements
Starting with our dashboard we see this:
CM108 Bare Board USB DAC Dashboard Measurement.png


Distortion is about 100 times worse than the best products we test. Admittedly this is at max volume (but with 100 kohm load). Still, we are talking bottom of the barrel performance here.

Of note, the frequency counter was jumping up and down. Likely due to use of adaptive USB mode instead of async interface. FFT shows high noise floor and lots of distortion spikes.

Quick look at distortion versus output level produces this:
CM108 Bare Board USB DAC THD vs output level Measurement.png


We see that the performance at 300 ohm load in red is all over the place. Good news there is that it doesn't really clip. It simply stops getting louder.

At 33 ohm, things calm down but at pretty high distortion levels until hard clipping at 1 volt output.

Didn't bother testing more as it is clear there are a lot of performance issues here.

Listening Tests
Quick listening test shows modulated bass frequencies and oddest response. It almost sounds like the output is run through a synthesizer in each bass note. All of this may be in my head, but I think it got there through my ears. :D

Conclusions
You all know that I am always on a hunt for a bargain product that performs excellently. Alas, this one isn't. Definitely NOT recommended.

If you want something cheap, the Speak DAC and its clone are the way to go at $30 or so. See https://audiosciencereview.com/foru...-speaka-usb-dac-and-headphone-amplifier.2423/

As always, questions, comments, corrections, jokes, etc. are all welcome!
----
If you like this review, please consider donating funds for these types of hardware purchase using Patreon (https://www.patreon.com/audiosciencereview), or upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
 

gvl

Senior Member
Joined
Mar 16, 2018
Messages
409
Likes
93
Location
SoCal
#2
It is DACs like this that probably benefit from USB power cleaners and signal purifiers.
 

gvl

Senior Member
Joined
Mar 16, 2018
Messages
409
Likes
93
Location
SoCal
#4
I think it has a lot more problems than that. Inadequate power. Adaptive USB. etc. Hard to bandage these.
Adaptive USB could benefit from a USB re-clocker, e.g. Schiit Wyrd. Of course even if they help the additional cost makes saving money on cheap DACs, well, impractical.
 

maxxevv

Active Member
Joined
Apr 12, 2018
Messages
174
Likes
38
#5
Did you try the readings at say 25%, 50% and 75% ouput ?

Many of these boards have inadequate power leading to huge distortions at near-max or max ouputs but seems alright at more modest levels. Might be worth exploring as it was somewhat like the Burson your measured ?
 
Last edited:

bennetng

Active Member
Joined
Nov 15, 2017
Messages
212
Likes
58
#8
Of note, the frequency counter was jumping up and down. Likely due to use of adaptive USB mode instead of async interface. FFT shows high noise floor and lots of distortion spikes.
In some other reviewed products you measured some USB packet noise, albeit at very low level. Are they running in async mode?
 

dallasjustice

Major Contributor
Patreon Donor
Joined
Feb 28, 2016
Messages
1,190
Likes
577
Location
Dallas, Texas
#9
Quick listening test shows modulated bass frequencies and oddest response. It almost sounds like the output is run through a synthesizer in each bass note. All of this may be in my head, but I think it got there through my ears. :D
Have you considered measuring distortion <1khz playing 50hz sine wave into a 300ohm load? Doesn’t this this test have the potential to show very audible types of distortion?
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
13,493
Likes
5,662
Location
Seattle Area
#10
In some other reviewed products you measured some USB packet noise, albeit at very low level. Are they running in async mode?
They are.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
13,493
Likes
5,662
Location
Seattle Area
#11
Have you considered measuring distortion <1khz playing 50hz sine wave into a 300ohm load? Doesn’t this this test have the potential to show very audible types of distortion?
I don't know. Will experiment to see.
 

restorer-john

Addicted to Fun and Learning
Joined
Mar 1, 2018
Messages
529
Likes
461
Location
Gold Coast, Quuensland, Australia
#12
Consider this D/A converter is from 1994/95 and essentially a device for ultra low cost devices like portable CD players and the like.

Clearly though the implementation with the USB interface, poor local regulation and isolation (I see none) has massively compromised its ordinary specifications even further.

I also wonder whether the chips are actually genuine or possibly old stock rejects- it is after all a 23 year old D/A converter.
 

έχω δίκιο

Active Member
Patreon Donor
Joined
Apr 20, 2018
Messages
231
Likes
236
#14
I also wonder whether the chips are actually genuine or possibly old stock rejects- it is after all a 23 year old D/A converter.
Usually the kind of chips that counterfeiters target are those that are pricey and new. There's no money to be made remarking chips to look like obsolete parts that can be had for pennies in genuine form.
 

restorer-john

Addicted to Fun and Learning
Joined
Mar 1, 2018
Messages
529
Likes
461
Location
Gold Coast, Quuensland, Australia
#15
There's no money to be made remarking chips to look like obsolete parts that can be had for pennies in genuine form.
You'd be surprised. Metal film resistors are faked. Ancient MOSFETs are faked. Very old and obscure STKxxxx hybrids are faked. ICs are faked all the time.

Vast quantities of NOS out-of-spec silicon has found its way to China.

Without actually contacting Philips to determine whether that D/A is still being made, and the batch/week/year codes are valid, we simply don't know.

All I know is Amir's testing makes it essentially a waste of $8.
 

έχω δίκιο

Active Member
Patreon Donor
Joined
Apr 20, 2018
Messages
231
Likes
236
#16
You'd be surprised. Metal film resistors are faked. Ancient MOSFETs are faked. Very old and obscure STKxxxx hybrids are faked. ICs are faked all the time.
I was aware of the obsolete, but still in high demand, parts that were faked. Hadn't known about metal film resistors, but I guess I'm not surprised given how easy they are to mark, the price differential to cheap resistors, the volume used, and how unlikely it is to spot fakes via QA/QC of finished goods.

Vast quantities of NOS out-of-spec silicon has found its way to China.
Well, they have to so something with it now that Radio Shack is out of business. :D I've never understood why they don't destroy the out-of-spec parts. I'd be much more likely to spec parts from a company that committed to destroy its out-of-spec parts.

Without actually contacting Philips to determine whether that D/A is still being made, and the batch/week/year codes are valid, we simply don't know.
Even then, it can be marked with valid codes from older production and the Chinese supplier can claim it's NOS.

It would be interesting to take that exact board and stuff it with known-good parts and see how it retested. Not that I'm volunteering to do that, since I'm a lazy, retired engineer, but if someone with more energy than me (a sloth trained as an EE, for example) were to do it, I'd love to see the results.
 

Jimster480

Addicted to Fun and Learning
Patreon Donor
Joined
Jan 26, 2018
Messages
552
Likes
187
Location
Miami
#17
Adaptive USB could benefit from a USB re-clocker, e.g. Schiit Wyrd. Of course even if they help the additional cost makes saving money on cheap DACs, well, impractical.
Wyrd is literally just a USB hub with one output.
You could use anything but that overpriced garbage to have the exact same effect.
 

Jimster480

Addicted to Fun and Learning
Patreon Donor
Joined
Jan 26, 2018
Messages
552
Likes
187
Location
Miami
#18
I'd be really interested to see the overall performance of one of those $5 USB Sound cards.
I used one literally for years, and I know the quality wasn't great but it worked for VOIP.
Had both headphone output and MIC input.

https://www.ebay.com/itm/USB-3d-Aud...290461&hash=item236ff7822e:g:P~YAAOSwHlRbR4s4

This is what I am talking about. I should have found one and sent it over to amir in the box I mailed to him. But I wasn't even thinking about it at the time.
 
Top Bottom