• 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.

xDuoo TA-20 Tube Balanced Headphone Amplifier Review

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
22,772
Likes
35,923
Location
Seattle Area
#1
This is a review and detailed measurements of the xDuoo TA-20 hybrid transistor and tube balanced headphone amplifier. It was kindly sent by member @xtenik who is a distributor for xDuoo products. The TA-20 costs US $399.

I previously reviewed the xDuoo TA-10 and was impressed by the quality of the DAC implementation. The tube amplifier stage had a lot of distortion of course but subjectively it sounded good. That, and solid build led me to recommend the TA-10. Some people however asked for the balanced version of that unit. Turns out that doesn't exist. The TA-20 for some reason eliminates the DAC which was the bell in that ball.

The TA-20 has the same nice build quality of the TA-10:

XDuoo TA-20 Balanced Headphone Amplifier Review.jpg

The two tube guards are magnetic and snap nicely into their sockets. They make you want to keep taking them out and reattaching them. :) The volume control feels good. It is a rotary encoder and linearity changes values as you turn the volume.

Naturally you have both 1/4 headphone jack and XLR for "balanced" headphones.

The back panel shows three inputs which is nice:

XDuoo TA-20 Balanced Headphone Amplifier Back Panel Connectors Input and Output Review.jpg

For my testing, I only used the XLR inputs.

Inclusion of mains power supply is a great touch to reduce clutter on or around your desk.

In use the tubes barely get warm. The real work is performed with solid state transistors in the output stage. This has the advantage of lower output impedance than a tube transformer-less design would deliver.

The unit came with stock Chinese 12AU7 tubes:
XDuoo TA-20 Tube 12AU7 Balanced Headphone Amplifier Review.jpg

Headphone Amplifier Measurements
I connected my analyzer output to XLR inputs set to 4 volts and volume adjusted for the same 4 volt output (unity gain) and this is what I got:
XDuoo TA-20 Headphone Amplifier Balanced Audio Measurements.png


We see some channel imbalance as indicated by the sine wave graphs and RMS levels.

Distortion is sky high but it is mostly in the form of second harmonic as is often promised, but not always delivered with tube amplifiers.

SINAD is the combination of noise and distortion relative to our 1 kHz signal and suffers greatly due to high distortion:
Best Headphone Amplifier Review 2019.png


Signal to noise ratio is actually good, clearing the 16 bit mark that is 99% of the music you listen to:
XDuoo TA-20 Headphone Amplifier XLR In SNR Audio Measurements.png


Measured at much lower level of 50 millivolts though (right), performance drops severely:
Least Noisy Headphone Amplifier Review.png


That is probably due to the mains hum that is indicated in the FFT spectrum. So I would avoid the TA-20 with sensitive IEMs/headphones.

Frequency response was very wide and flat but not without problems:
XDuoo TA-20 Headphone Amplifier Balanced Frequency Response Audio Measurements.png


The channel in red is clearly having some stability issues in addition to being at a different level than the other. The latter may cause a subtle shift in soundstage which some may take for an improvement. It is not.

Intermodulation and noise versus input level shows copious amount of non-linearities:
XDuoo TA-20 Headphone Amplifier Balanced IMD Audio Measurements.png


Running the same test but examining the THD+N gives us:
XDuoo TA-20 Headphone Amplifier Balanced THD+N vs Input Level Audio Measurements.png


So the optimal input is about 0.1 volt. And the unit starts to clip at a little above 3 volts. So our dashboard at 4 volts input is showing more of a worst case situation than other input levels would indicate. But not much.

THD+N versus power at 300 ohms gives us what we can predict:
XDuoo TA-20 Headphone Amplifier XLR HP 300 Ohm Audio Measurements.png


Lots and lots of distortion but also tons of power. My criteria here is 100 millwatts and the TA-20 busts pass that with 326 milliwatts. As result the TA-20 should have no trouble driving high impedance headphones to deafening levels.

Going to the other extreme of 33 ohm, we get a similar picture:
XDuoo TA-20 Headphone Amplifier XLR HP 33 Ohm Audio Measurements.png


Testing for balanced power, I tested the TA-20 both with its 1/4 Headphone jack and XLR:

XDuoo TA-20 Headphone Amplifier XLR HP 50 Ohm Audio Measurements.png


The XLR output (don't confuse with input) produces much lower noise which was nice. It also has less distortion at the limit. As such, if you have a choice, definitely use the XLR "balanced" output. You get more power, less noise and distortion.

Output impedance was 4.5 ohms which can cause frequency response of low impedance headphones to change (for better or worse):
Headphone Amplifier Output Impedance Measurements 2019.png


For kicks, I also measured that versus frequency:
XDuoo TA-20 Headphone Amplifier Output Impedance Audio Measurements.png


The impedance is more or less frequency independent which is good.

There must be some kind of digitally controlled gain stage for the volume control as it had perfect channel matching:
XDuoo TA-20 Headphone Amplifier Channel Balanced Audio Measurements.png


Headphone Listening Tests
I did not have time for controlled testing so these are subjective listening tests of moderate value.

I started with my Drop Ether CX 25 ohm impedance using XLR output of the TA-20. These are very inefficient headphones and require good bit of current to produce high SPL (loudness). The TA-20 had no trouble driving them as hard as I would dare. Overall dynamics was very good but I thought the highest were a bit exaggerated and vocals a bit lispy.

I then switched to my Sennheiser HD-650 using 1/4 headphone out. I had no trouble waking up the neighbors with the TA-20! :eek: There seemed to be infinite amount of power here and produced very satisfying sound which I could not fault. Even at maximum volume, the combination was dead quiet.

I detected no "tube magic" though. Soundstage didn't' change. Digital did not become analog. Angels did not lift me and take me to a sunny beach either. I just heard good sounding music as I do with other headphone amps of similar power.

All of this may change in a controlled AB test but being lazy and out of time, that is all I have for you.

Conclusions
The measured performance of the xDuoo TA-20 gets a failing grade as one could predict. The build quality, digital volume control and powerful amplifier stage substantially change the final calculus. The cost is high though at $399 especially since you have to add a DAC to it as well.

Overall, I can't recommend the TA-20 but you have all the data to judge for yourself.

--------
As always, questions, comments, corrections, etc. are welcome.

One of the panthers is complaining that he is getting too old to see in the dark and wants night vision scope for night time hunting! I checked and these things are expensive. So I appreciate some donations to get him one for Christmas using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
3,430
Likes
6,125
Location
The Neverlands
#3
@amirm if it is not too much trouble can you show the first dashboard but with -10dB input voltage or -20dB input as those are the levels it would likely get IRL situations.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
22,772
Likes
35,923
Location
Seattle Area
#4
@amirm if it is not too much trouble can you show the first dashboard but with -10dB input voltage or -20dB input as those are the levels it would likely get IRL situations.
All that information is there in this new measurement I ran:

 

pma

Active Member
Joined
Feb 23, 2019
Messages
262
Likes
342
Location
Prague
#5

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
3,430
Likes
6,125
Location
The Neverlands
#6
I know but what one cannot see is how the harmonic spread is and the 'grass' in the higher frequencies as the other plots only show THD.
That is why I asked.
 

pma

Active Member
Joined
Feb 23, 2019
Messages
262
Likes
342
Location
Prague
#7
I know but what one cannot see is how the harmonic spread is and the 'grass' in the higher frequencies as the other plots only show THD.
That is why I asked.
I know. And you know that harmonics quickly disappear only H2 remains. I made a public ABX with almost same distortion profile added, few years ago, and no one was successful. Tube is easily modelled mathematically, as there are almost no frequency dependent non-linearities. I also used 12AU7. You can see my measured plot of 12AU7 preamp is similar to Amir's.
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
3,430
Likes
6,125
Location
The Neverlands
#8
That's why I asked because the harmonic spread at full blast is different from when it is used under normal circumstances.
Would be good to see how it differs which one cannot see from the plots shown now.
I know that, you know that, SIY knows it as well as many others. The casual reader without any knowledge about tube behavior might find it interesting to see though.
But I do realize it takes a lot of time for Amir to connect it again, run tests and post the results.
I know because I do this too all the time (albeit with lesser gear) so added the words 'if it is not too much trouble'.
It could be enlightening to those NOT familiar with response of tubes, not to me I have plenty of plots from different tubes at different levels in certain circuits.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
22,772
Likes
35,923
Location
Seattle Area
#9
That's why I asked because the harmonic spread at full blast is different from when it is used under normal circumstances.
The dashboard is set to unity gain (4 volt in, 4 volt out). The volume as indicated was well below max at 85. I think max is 95.

I have picked this setting as more representative of how the devices may be used. It is not too low as to be noise dominated. But not so high as to represent clipping.
 
Joined
Nov 21, 2019
Messages
44
Likes
9
Location
Nevada
#13
I have to test another product of theirs that was sent in before I ask for another one. :)
@amirm thabk you for this review I was considering this unit to get a different sound then my SMSL thx headphone amp, is there a smaller desktop tube amp you would recommend to get that amazing tube sound at a reasonable price point?
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
22,772
Likes
35,923
Location
Seattle Area
#14
@amirm thabk you for this review I was considering this unit to get a different sound then my SMSL thx headphone amp, is there a smaller desktop tube amp you would recommend to get that amazing tube sound at a reasonable price point?
I don't recommend any tube amplifier. :) I see no positives in them other than nostalgia.
 
Joined
Nov 21, 2019
Messages
44
Likes
9
Location
Nevada
#15
I don't recommend any tube amplifier. :) I see no positives in them other than nostalgia.
I can honestly say that when I listen to my new Smsl 2000 thx headphone amp and the smsl Au8 v2 the standard setting in the dac sounds a bit boring or clinical on some recordings but when I switch the dac to rich 1, 2, 3, or Tube 1, 2, or 3 for some of the recordings the music begins to sound more interesting, this of course depends on the recording and the type of music, in some cases the standard setting from the dac sounds better. Which leads me to my next question, is the DAC more important once I have a decent headphone amplifier such as the new smsl thx 2000 unit? Also is burn in time on an amp, headphones, etc a real thing and they will start sounding even better the more time they run? Thanks for all your guidance, so far I have made many of my purchasing choices based on this website.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
22,772
Likes
35,923
Location
Seattle Area
#16
I can honestly say that when I listen to my new Smsl 2000 thx headphone amp and the smsl Au8 v2 the standard setting in the dac sounds a bit boring or clinical on some recordings but when I switch the dac to rich 1, 2, 3, or Tube 1, 2, or 3 for some of the recordings the music begins to sound more interesting, this of course depends on the recording and the type of music, in some cases the standard setting from the dac sounds better.
Ask someone else to change them for you without knowing to confirm that is what you are hearing.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
22,772
Likes
35,923
Location
Seattle Area
#17
Also is burn in time on an amp, headphones, etc a real thing and they will start sounding even better the more time they run?
Definitely not. I once tested a Schiit Yggdrasil DAC for two weeks and not a thing happened to it during that time.
 
Joined
Nov 21, 2019
Messages
44
Likes
9
Location
Nevada
#18
Ask someone else to change them for you without knowing to confirm that is what you are hearing.
So you are saying there is no difference to the processing of the music by the dac? I can definitely have a buddy test with me, that’s a great call.
 
Top Bottom