• 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!

Budget DAC Review and Measurements: Behringer UCA222

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,393
Location
Seattle Area
UCA222 is an extremely low cost DAC/ADC targeted at home recording consumers. For this review I measure its DAC and headphone output performance. This is an ultra low cost device retailing at just $30 with shipping included! As such, it is an "Amazon Choice" product: https://www.amazon.com/Behringer-U-...ie=UTF8&qid=1509659923&sr=8-1&keywords=UCA222

upload_2017-11-2_14-59-47.png

The device is obviously made out of plastic but it is not as cheesy as that sounds. The metal red color makes it look like it is made out of metal. The UCB cord as you see is permanently attached so no tweaking with $1000 UCB cables is possible! :D

Functionality is pretty good given the headphone output, optical S/PDIF and aforementioned analog to digital converter. Unfortunately sampling rate is limited to 48 Khz.

For this testing I did my comparison against two other low cost DACs: the Topping D30 ($129) and Behringer UMC204HD ($79). Later, I also compare it to my laptop native output.

Measurements
Let's get into it with 24-bit/48 Khz J-Test signal which is just a sine wave for the purposes of this graph at 12 Khz.

J-test.png


Clearly the UCA222 underperforms the other two DACs by 15+ db in noise floor. And it adds some salt and pepper in the form of various noise/jitter spikes. Still, not awful. Just not as good.

For awful, we get to run our -90.31 db, 24-bit sinewave test. I first ran this test and got pretty miserable performance out of UCA222. So I thought maybe it is a 48 Khz only DAC so I tested it it at that sample rate. Alas, same sad performance was delivered:

-90 db 48 Khz.png


The well-defined output in red/yellow is that of the more expensive Behringer UMC204HD. The awful, noisy, fluctuating output is from UCA222 DAC. In other words, it has no prayer of reproducing the low order bits of a 16-bit signal. Effective bit depth accuracy is probably in 13 to 14 bits.

Measurements Compared to Laptop Built-in Output
Seeing how the Topping D30 and Behringer UMC204HD set too high a bar, I thought I test it against the native headphone output of my laptop. This is from an older but very high-end Z-series HP laptop which retailed for over $2000 when I bought it a couple of years ago. I have done a ton of critical listening tests using its headphone output. So subjectively I know it is good.

Again, let's look at J-Test output:

J-test vs laptop.png


The laptop's built-in "IDT high definition codec" indeed delivers high-definition performance! It has lower noise floor and total absence of any jitter/noise products! Yes, the noise floor is some 10 db higher than UMC204HD and Topping D30. But given the fact that it is "free", it is remarkably good. Needless to say, it easily outperforms the Behringer UCA222.

Now let's look at low level (bit) resolution again:

-90 db 48 Khz vs laptop.png


My laptop is in yellow and while the output is still a bit noisy, it runs circles around the horrid performance of the Behringer UCA222.

Headphone Listening Test
Using my Sony MDRV6 headphones, the UCA222 produced clean sound that never really got distorted even at max setting. Alas, it has pretty anemic output. Just eyeballing it, its output is half as much as the headphone jack of my laptop. While I can get my laptop to become way too uncomfortable for listening with respect to level, I can listen to UCA222 almost at max setting. Fidelity wise, the laptop seemed more authoritative although this may be just due to its higher output power.

I did not bother testing the RCA outputs subjectively.

Conclusions
The UCA222 is a dirt cheap DAC. It uses older style 48 Khz max sampling rate DAC which probably retails for less than the change in my pocket. Its measured performance ranges from "OK" to pretty bad. Seeing how my laptop beats it in every regard, I see no reason to recommend it. Spend a few more dollars and get the Topping D30 which remains my recommended DAC at these low price ranges.

As always suggestions, comments, corrections, bad jokes, etc. are encouraged.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,393
Location
Seattle Area
As suggested by another member, the previous version of this DAC -- the UAC202 -- was praised by NWAvGuy. I found his review here: http://nwavguy.blogspot.com/2011/02/behringer-uca202-review.html

I think his positive impression and negative version of mine is that of 6 years going by and exposure to better DACs. There are probably 100X more DAC in market now than there was then. It is this wide choice and much higher competition that has led to superior low-cost DACs, albeit not at $30 price range.

He does agree with my finding of the headphone output being too low/too high impedance and offers a mod to remedy that (new op-amp soldered in place).

Edit: fixed the link.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,393
Location
Seattle Area
And for those who car, this device is built on a single chip USB DAC/ADC silicon from "Coolaudio Semiconductor." Part number is V2902. It is retails for $1.50 in 1000 piece quantity. Likely Behringer negotiated even lower cost than this. Here are the bits about it: http://www.coolaudio.com/features-page.php?product=V2902

FYI, it is a clone of a TI PCM2902 which retails for 5X more money! I suspect corners were cut in the fidelity department to get to lower cost.
 

sssn

Member
Joined
Nov 1, 2017
Messages
92
Likes
74
And for those who car, this device is built on a single chip USB DAC/ADC silicon from "Coolaudio Semiconductor." Part number is V2902. It is retails for $1.50 in 1000 piece quantity. Likely Behringer negotiated even lower cost than this. Here are the bits about it: http://www.coolaudio.com/features-page.php?product=V2902

FYI, it is a clone of a TI PCM2902 which retails for 5X more money! I suspect corners were cut in the fidelity department to get to lower cost.

Wait, wait, wait! You may be on to something!

NwAvGuy clearly stated in his review of the UCA202 that it uses a TI PCM29xx:
UCA202 INTERNALS: The UCA202 uses a highly integrated Burr Brown/TI PCM29xx CODEC chip.

So if your UCA222 only has a cheap clone of that, it might explain the worse performance.

Maybe UCA202 and 222 differ after all. Everyone thought the UCA222 is the same only with a bit of software added.
 
Last edited:
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,393
Location
Seattle Area
Just looked up the data sheet on both chips. The performance graphs are identical down to period over the i and cross on the t. :)

So either they completely copied the TI design or the datasheet is a copy and not to be trusted. My money is on the latter. To wit, the PCM2902 has this important footnote in the specs:

"(1) f IN = 1 kHz, using a System Two™ audio measurement system by Audio Precision™ in the RMS mode with 20-kHz LPF, 400-Hz HPF in calculation."​

There is no such note in the V2902 data sheet.

I would think in the price of cost cutting attempt here, there may be other changes to the rest of the circuit.

I opened the unit and found leftover soldering flux on the PCB although the rest was rather clean.

Overall though, in this day and age there is no reason to buy a DAC limited to 48 Khz and without async USB input.
 

sssn

Member
Joined
Nov 1, 2017
Messages
92
Likes
74
Well, I suspect too that Behringer cut (even more) corners with the UCA222 and opted for a cheap, untested clone of the chip used in the UCA202. Or maybe even current 202s have the bad chip now too.
 

Dimitri

Senior Member
Joined
Jan 28, 2018
Messages
365
Likes
426
Location
Valencia California
Well, I suspect too that Behringer cut (even more) corners with the UCA222 and opted for a cheap, untested clone of the chip used in the UCA202. Or maybe even current 202s have the bad chip now too.
For the record , unfortunately you are correct.
I have a UCA-202 with a date code 1611 and sure enough has the V2902 in it; at least for another few days until I get my hands on a TI chip :)
 

Interference

Member
Joined
Mar 14, 2018
Messages
86
Likes
112
As far as I understand, Coolaudio is part of the Behringer group and they are probably licensed from TI to produce their own chips.

@amirm would you say that your findings are comparable to the NwAvGuy measurements or do they look worse? (I am not so good at reading this kind of stuff yet and it's always hard to compare different test setups).
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,393
Location
Seattle Area
@amirm would you say that your findings are comparable to the NwAvGuy measurements or do they look worse? (I am not so good at reading this kind of stuff yet and it's always hard to compare different test setups).
I took a look and it is very hard to correlate the two sets of measurements. Many parameters and presentations of tests are different. His web pages are hard to read too as the graphs are resized and fuzzy until you click on them, making it very tedious to go through them.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,393
Location
Seattle Area
Is "a few more dollars" quite fair? The D30 costs four times what the UCA222 does. :)
$1 is four times 25 cents but neither buys you a cup of coffee anymore. :)
 

ctate

New Member
Joined
Mar 29, 2018
Messages
2
Likes
1
$1 is four times 25 cents but neither buys you a cup of coffee anymore. :)
Sure, but "Instead of spending $30, spend an additional $100" doesn't really feel like "a few more dollars" to me. If you're shopping at that price point, that's probably off the table to begin with.

How does the UCA222 compare around the $30 mark? Looking around I've found it surprisingly hard to find options with USB in and line out at low price points.

(As an aside, I need to work out how exactly to do these analyses myself. I have a couple of Chordette units that are miserably bad, and want to put it on the bench to see what the heck is going on.)
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,393
Location
Seattle Area
You can get Topping D10 for $89 with shipping included on Amazon. It goes on sale for less than that. I don't have recommendations for anything cheaper that performs well.

Ultimately you have to compare your hardware budget to what you spend on music.
 

Dimitri

Senior Member
Joined
Jan 28, 2018
Messages
365
Likes
426
Location
Valencia California

mshenay

Active Member
Joined
Feb 9, 2018
Messages
177
Likes
206
Huh, I've never compared them but I've got a UCA 202 like 5 years ago!!!! I wonder if they stopped using the PCM chip an switched to something else recently?

I've always been impressed subjectively with how it sounds compared to a lot of stuff. But it's limited to 44.1/16bit and the lack of Asynchronous USB is likely problematic!!!

Oh? I just noticed your running 24bit 44.1 files, that's odd. That said I don't use mine with anything above 16bit 44.1, so here's my ask are you in a position to tear apart some older units and/or re run the tests with just 16bit files.

Second, I'm very happy to see that you did this measurement. I have two units, one with a date code of 1607 and a serial of 16XXXXXXXXXXXXX, my second unit has a Serial of 13XXXXXXXXXXXX an the date code is long gone, but all the stickers are yellow and kinda faded, I figure it may be safe to assume the date codes year will match the serials first 2 numbers. Maybe I'm wrong, but let's assume I'm right [I know assumptions are bad] Also to double check this I did a search on eBay and found a second hand unit with a 09XXXX Serial and a date code of 09XX

My point here is the new 16XXXXXXXX sounds terrible. I purchased primary to use with my laptop when I used to use it to record audio for my videos. I mic separate from my camera and even mic into an entirely different system. I no longer do that so the old one sat in a draw. That said reading over this made me pull it out and just give it a listen!

Now the following is purely subjective, so YMMV and I may be full of schiit, that said let's start with some back story. I listen to my 13XXXXX UCA 202 EVERY day in my car, some times I play the same songs. Stuff I like for driving, stuff that helps me get through the grog of my daily commute. I listen at the same volume level's also and with a set EQ Profile.

Now Swapping in the 16XX Serial'd unit into my car system, running the same songs, EQ Preset, Stero Presets ect.... it was pretty obvious the differance between the two. Even my wife noticed, the newer unit is very noisy and lacks clarity. I like to listen to a lot of Heavy metal when I drive [again helps me fight the urge to ROAD RAGE] and almost immediately I noticed how... splashy and unresolved the top end was. On SO many tracks, it as like listening to V4 Lossy file [and I'm running losses mind you] that level of... distortion/noise

That said swapping back into the 13XXXXX I noticed the top end was immediately more resolved, that is to say I could hear the full range on the top hats, that initial peak and the release or the transition to quite, very discernible differences

But with the 16XXX only that initial peak was discernible and what followed was just noise, yuck! What a shame,
 
Last edited:

pedrocaetanos

New Member
Joined
Mar 27, 2019
Messages
1
Likes
1
I know both these little Behringer interfaces for long, and never managed to make sense of the difference between the two (202 and 222). I believe it's the color...
I have a red one that says UCA222 on the front (black sticker). But it says UCA202 on the back (engraved)!! - I don't think they bothered to make a different plastic mold for the 222.
The S/N is S11XXXXXXXAXX, which supposedly predates both those 13XXXXXXX and 16XXXXXXX - I always though it sounds rather decent.
It also has a Date Code of 1104 - I bought mine in September 2011, so I believe it was made in April that year.
By the operating manual dates, I can tell that 202 is being made from before March 2006 (manual version 1.1), and the 222 at least from 2009.

Summing up, I believe somewhere between in time they may have changed the DAC chip, but that should have happened to both 202 and 222, the difference being just the color.
Hope this info helps someone trying to make some sense out of this mess.
 

LeftCoastTim

Senior Member
Forum Donor
Joined
Apr 15, 2019
Messages
375
Likes
757
I have a UCA202, silver colored, that I bought in 2017 on Amazon.

Serial number is S1614xxxxxx, Date Code 1607.

It has a TI PCM2902C chip inside.
 

DDF

Addicted to Fun and Learning
Joined
Dec 31, 2018
Messages
617
Likes
1,355
S/N S14... Date code 1406: PCM2902E
S/N S15... Date code 1506: Coolaudio v2902

Considering using this with REW as a room measurement ADC (high SNR not needed) as I already have a great mic and mic pre but with analog out only. Anyone know of any ADC measures for this puppy?
 
Top Bottom