• 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!

Woo Audio WA7 & WA7tp DAC and Headphone Amp Review

Tarzan

Member
Joined
Dec 3, 2018
Messages
25
Likes
25
Location
Brisbane, Australia
Any chance of reviewing the WA6?
 

The Mule

Member
Joined
Jan 23, 2019
Messages
29
Likes
20
Glanced at this yesterday (sometimes I just jump to Amir's Listening Tests and Conclusions), but today I read it. That's disappointing. The frequency response is rolled off, but because of the distortion added, it actually brightens the source. I have an early version like this one, and I've never used its DAC, but there are problems in the amp section too.

Just makes me more curious if a Valhalla 2 can ever be reviewed, or I also like the looks of the Pathos Aurium. :cool: There's also the Massdrop Eddie Current ZDT Jr. But I know there are functional issues with that one.

I think it's interesting that I got caught up chasing "tube sound" too, but I still wonder if there's something out there that might be a better example of what some people think "tube sound" is all about.

This one is a little pricey for what I'm personally looking for, but I've seen some good reviews of it too:

https://www.innerfidelity.com/content/icon-audio-hp8-mk2-tube-headphone-amplifier

Maybe not as pretty as the WA7 or the Valhalla 2, but possibly better performing and sounding?

;)
 

Surge

Member
Joined
May 7, 2019
Messages
9
Likes
2
This is a review and detailed measurements of the Woo Audio tube WA7 DAC and headphone amplifier together with its companion WA7tp tube power supply. It is on kind loan from a member. The cost of the pair is USD $1,599 so not cheap. As of this writing, shipping is free but I assume you would otherwise have to pay for it.

Note: this is the generation 1 of the WA7. The generation 2 is supposed to have a better DAC (?).

I remember running into Woo Audio amp a few years ago and drooled over its industrial design. The WA7 series is not quite up there with the rest of the units but still benefits from its massive plastic transparent top to give it elegance and drool appeal:


Strangely, the power supply has a rotary control which does nothing. You push it in and it power cycles on and off. A multi-pin short cable attaches the two. Reading the reviews, there are complaints that no AC power cord comes with WA7tp. What the heck?

As you can imagine, these two boxes are heavy and stay where you put them. There are two headphone jacks: the 3.5mm has lower gain so I focused my testing on the 1/4 inch jack.

The tubes in the unit I tested are an upgrade to Russian made, 6C45Pi Electro Harmonix with nice looking gold pins. Woo sells a matched pair for USD $140.

Unusual in tube amp category, there is an ESS DAC in there with USB input. There is also a set of RCA analog in and a switch for high and low impedance output.

I watched a few video reviews for the WA7 and there were unanimously positive. One reviewer complained about the DAC not being good so in my testing, I went both ways: with the DAC input and analog.

Measurements
With combo DAC and headphone inputs, I always try to test the DAC portion first by setting the output impedance high as to not stress the amplifier and set the voltage to 2 volts. Here is what we get with WA7 combo:

View attachment 23498

Yuck. We have a heap of distortion and a mains peak which I could not get rid of by playing with grounding. The SINAD score (signal over noise and distortion) naturally lands the WA7 at the bottom of the pile:

View attachment 23499

Is the DAC implementation broken? Let's look at the same performance with analog input:

View attachment 23500

Nope. The digital input was actually better by 4 dB or so.

Using digital input again, dynamic range is actually decent for a device with so much distortion:

View attachment 23501

Frequency response was fairly rolled off at both ends:
View attachment 23502

We are down 3.4 dB at 20 kHz. We are OK at 20 Hz but from there on, we lose amplitude there.

Switching to high z (z standard for impedance), raises the output voltage substantially. Similar roll off exists.

Linearity was very poor indicating perhaps 16 bit implementation or just a lot of noise:
View attachment 23503

Between this and high frequency roll off previously, don't bother with any "high-res" content for the sake of bandwidth or resolution.

EDIT: was asked to run multitone tests in the thread and here is that:

View attachment 23549

It is by far the worst results I have seen since I started running this test on DACs. The tall spikes are the 32 tones the DAC is asked to play. It does that but then generates tons of distortion spikes in between tones which can make the music sound bright and reduce effective resolution of music (by masking low level music detail with distortion products).

The all important test for headphone amplifier is power versus distortion and noise. Here is that, with 300 ohm load:

View attachment 23504

I will give you the good news first: there is plenty of power, especially in high-z mode. It is the highest I have measured at nearly 1 watt in that setting!

Using the more sane low-z mode though, the amplifier starts become distortion dominant at just 1 milliwatt. It is downhill from there up to max of 1/4 watt at 0.6% distortion+noise. This is one noise and distortion factory compared to our reference Massdrop THX AAA 789 (in dashed blue) at 1/5th the price.

Here is the picture with 33 ohm load:

View attachment 23505

There is still plenty of power and the same soft rise in distortion. At peak power though, we are now in single percentage numbers with max output at 6%. Fortunately you will go deaf well below that level so not a concern there. The worry is that lowest distortion at noise is at just 0.2 milliwatt. Not 0.2 watt but 0.2 milliwatt! Needless to say, at any listening volume you are going to be experiencing distortion and noise with the WA7.

As if these numbers weren't disappointing enough, we get to measure the output impedance:
View attachment 23506

Even in "low z" mode the impedance is nearly 30 ohm. With high Z you have 80 ohm. You have a built-in equalizer here for just about any headphone that doesn't have a flat frequency response. And it is an equalizer you don't have a control over.

Channel imbalance was pretty good showing a constant level mismatch for most of the volume range:
View attachment 23507

Let's stop here as the advocates of tube amps are screaming as loud as they can: "it is all about the ear; of course tube amps don't measure well."

Listening Tests
I started my testing by playing a 1 khz tone and matching levels between my Topping DX3 Pro combo DAC and Amp and WA7. This proved to be useless as the volume level on the WA7 was content specific due to variations it caused due to its high output impedance. So I resorted to matching levels by ear.

First up was Sennheiser HD-650. On my bass heaving headphone test tracks, the bass on WA7 was more muted. It was also a bit more bloated compared to the tight bass from DX3 Pro. This was subtle. What was less subtle was the brighter signature of the WA7. Highs were exaggerated. The lisp in vocals became worse and high frequencies became grungier. Note that this effect was most audible during periods with strong bass frequencies.

On detail and resolution front, the WA7 lost it there too. The case was loud instruments that vibrated up and down rapidly in level. The detail in between the peaks was nicely rendered by the DX3 Pro but became muddy and lost in WA7. Note that this was not across the board. On content with infrequent transients, the issue was not there.

Switching to Hifiman HE-400i showed similar issues just the same although not always with the same tracks the same way.

These two headphones represent difficult loads for amplifiers since one is voltage hungry, and the other, current. So I thought I throw my AKG K92 headphone in there too. The K92 is extremely efficient compared to the others and is closed back to boot, letting one better examine detail.

The power output was impressive from both devices. I had to turn the levels way, way down for comfortable listening. Level matching became much more difficult here as even during a single track, the WA7 would be louder and less loud. Clearly the frequency response of the K92 was modified by the WA7. Working around that, I heard the same issue with lack of bass impact and resolution in WA7.

Since I had so much power available with K92 I decided to try a new experiment. Try not to laugh but I put in two foam ear plugs in my ear. No, these are not musician ones. They are normal noise reduction yellow foam plugs. What this did was essentially cut off everything from mid to high frequencies and heavily attenuated the bass. I was then able to turn the volume up substantially and could easily hear the bass becoming more distorted in WA7 than Topping DX3 Pro.

Correlating Measurements with Listening Tests
Can we explain the subjective results using the objective data? Sure we can. Take a look at at the FFT spectrum again:

View attachment 23508

The only thing we told the WA 7 to produce was that 1 kHz tone. What we got (ignoring the left side of the graph) was that tone plus a bunch of harmonics. By definition this 1 kHz tone just got "brighter" since we now added more frequencies going up to 20 kHz and above. Now, the brain masks some of the earlier ones such as some of the 2nd and 3rd harmonics. But the rest at some point can become audible. So one would expect harmonic distortion of sufficient level to make everything sound bright. Indeed if you turn up an amplifier beyond its output power to clipping, it does just that. This is what I was hearing in my testing.

Let's consider that in our music we have many tones. Each tone creates that spray of harmonic so together they add up and increase the high frequency content. But they do more than that: the harmonics now land on top of low level detail in music. That would obscure them and potentially mask them completely. Again, this is what I was hearing with detail getting lost between peaks.

Now the audible effects are nowhere as extreme as the measurements. Reason is that there is a lot of masking going on where the brain is throwing away lower level sounds -- whether they are distortion or musical detail. Selecting the right content, and playback level, plus focusing on specific attributes of music was necessary to hear these changes.

Most importantly, you need to have a reference. If I just listened to WA7 by itself, I would have no idea that these changes had occured. Your memory is simply not good enough to hear 10% more high frequency detail during some segments of some musical tracks. It was the instant AB switching with Topping DX3 Pro which allowed me to find and zoom in on them. All the reviewers who just listen to WA7 by itself have no prayer of observing any of this.

I should say that the fidelity I heard was 90% due to the content itself. As such, the WA7 can sound exceptional depending on how good the source content is. So don't think that measurements say it produces garbage. Again, perceptual masking saves us from hearing a lot of what the measurements show.

Conclusions
Yet again we measure a tube amplifier which produces copious amount of distortion. However, unlike some budget products, the WA7 is a quite amplifier and has major conveniences such as built-in DAC. It is also pretty to look at. For me as an audiophile, the WA7 takes a step backward. Last thing I want is the highs getting extra sizzle or the tight, impactful bass losing some of both characteristics.

As I have experienced time and time again in controlled testing, there is nothing euphonic or positive with such tube amplifiers. No soundstage is changed. No instrument is isolated more (it is actually worse with WA7). Detail is actually lost. The sound definitely is not warmer either. People read these attributes into such devices because they think they are there. A simple AB test with a $20 switcher is enough to convince you otherwise.

Lack of fidelity in measurements simply translates into lack of transparency and listening pleasure.

Now, if you are coming from a low power or no headphone amplifier, the WA7 will be a revelation. But remember, you can have your cake and eat it too by buying 100% transparent headphone amplifier with tons of power such as the $99 JDS Labs Atom. Or the $370 Massdrop THX AAA 789. From that vantage point, you will be going backward with Woo Audio WA7 in all respects but looks.


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

I will make this simple: you can never have enough money. So please consider donating 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).

Completely useless review. You would be better off measuring lab equipment, not musical equipment.
The funniest part was when you said ‘it’s not about measurements with tube gear, it’s about listening. Then you switched to listening and what did you do, play a 1khz frequency, and then listen with foam earplugs. BAHAHAHA. Come on Bro.

Do yourself a favour and Listen. To Music.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
Completely useless review. You would be better off measuring lab equipment, not musical equipment.
That's my specialty. Don't you try to compete with me with useless posts!

The funniest part was when you said ‘it’s not about measurements with tube gear, it’s about listening. Then you switched to listening and what did you do, play a 1khz frequency, and then listen with foam earplugs. BAHAHAHA. Come on Bro.
So you thought I sat there listening to 1 kHz tone? You didn't realize that was used for matching levels using an instrument?

How about this bit that followed:
First up was Sennheiser HD-650. On my bass heaving headphone test tracks, the bass on WA7 was more muted. It was also a bit more bloated compared to the tight bass from DX3 Pro. This was subtle. What was less subtle was the brighter signature of the WA7. Highs were exaggerated. The lisp in vocals became worse and high frequencies became grungier. Note that this effect was most audible during periods with strong bass frequencies.

This should have been the give away that real music was used, no?

How about the next paragraph:

On detail and resolution front, the WA7 lost it there too. The case was loud instruments that vibrated up and down rapidly in level. The detail in between the peaks was nicely rendered by the DX3 Pro but became muddy and lost in WA7. Note that this was not across the board. On content with infrequent transients, the issue was not there.

This wasn't a clue either that I was listening to real music Bro?

Maybe the third time would have been the charm with yet another headphone?
The power output was impressive from both devices. I had to turn the levels way, way down for comfortable listening. Level matching became much more difficult here as even during a single track, the WA7 would be louder and less loud. Clearly the frequency response of the K92 was modified by the WA7. Working around that, I heard the same issue with lack of bass impact and resolution in WA7.

Let me guess: you thought 1 kHz tone has bass?

Then there was this:

I should say that the fidelity I heard was 90% due to the content itself. As such, the WA7 can sound exceptional depending on how good the source content is. So don't think that measurements say it produces garbage. Again, perceptual masking saves us from hearing a lot of what the measurements show.

Again you thought I was talking about 1 kHz tone?

Do yourself a favour and Listen. To Music.
I did listen to music. Seemingly if Moses came down and gave you the ten commandments, you would question who he was.

You want to live in audio science and engineer ignorance, do. But please don't insult our intelligence with your inability to read what is done and posted for free for you.
 

Belgarathian

Member
Joined
May 29, 2019
Messages
21
Likes
32
Location
New Zealand
Hello! everyone. this is Mike from Woo Audio. :)

@amirm Thank you for the review. I want to point out that the WA7 in this review is a 1st gen model introduced in 2012 and discontinued in 2016. It does NOT have an ESS SABRE 24-bit/384kHz DAC that our current model 2nd gen has. The 1st gen uses the Texas Instruments 32-bit/192kHz DAC and a Cmedia USB controller. We learned a lot since 2012 and digital technology has advanced significantly. The current 2nd gen WA7 is greatly improved in every way including the DAC and XMOS USB controller. Hope you can update the posted to avoid confusion for the reader.

Our amplifiers are an all-tube design, not a hybrid. Vacuum tubes audio products are not going to measure favorably against solid state products. If you're seeking an analytical/dry sound, buy a product with impressive measurements. For us, we value how our products enhance the music listening experience for the user. WA7 have been in production since 2012 and we have a strong customer following and many unsolicited favorable reviews on this model alone.

Thanks for the clarification... Thoroughly impressed with your comms too.

Love your design ethos for your product lines.
 

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
Well i have demoed sa7 wa22 wa33. They all sound dirty and harsh. Not sibilant but irritating, plasticy. Some people can feel the same sound as juicy and musical. They should train their ears to hear better.
 

cjfrbw

Senior Member
Joined
Mar 5, 2018
Messages
410
Likes
472
Best reply 2019.
Taken down with the flaming sword of objectivity!
Don't let them get a beachhead, Amir, the subjectivists are raising ladders to the ramparts!
 

o2so

Active Member
Forum Donor
Joined
Sep 9, 2019
Messages
250
Likes
143
Location
Sydney, Australia
I am new here and greatly appreciate a forum that provides reviews based on actual measurements rather than subjective listening impressions, which taken individually are hardly helpful (what I like, you may not like) and far too abundant in Hi-Fi.

With that said, I think Mike from Woo Audio might have a point when he says that he is aware of his products not measuring well, that designing something that measured well was not what he wanted to do.

Roger Sanders, who is widely considered an extremely knowledgeable sound engineer and among the world's best speaker/amplifier designers, in a long email once told me that tube amplifiers sound different from solid state amplifiers (of equivalent design) only when they clip, or distort. In that, I am sure everyone here already knows, they produce a type of distortion that sounds much better that clipping solid state amps. According to Roger there was then no point in designing/owning a tube amplifier, because even if they sound better than a solid state amp when they clip, a more powerful, non clipping amplifier (tube or solid state) is much preferable. So, if this is the only difference between solid state and tube amps, I wonder what is the point of even designing a tube amplifier with low distortion? You might as well go for a solid state design which is much more reliable and requires zero to minimal long term maintenance, right?

Here is where I understand what designers such as Woo Audio do. They offer an alternative to great measuring solid state amplifiers, far from neutral, with an amount of distortion that they consider acceptable (other clearly consider this unacceptable) because of what this does to the overall sound presentation. As in, they think the presentation is enhanced in some respects. This is subjective, obviously. However looking at the numerous raving subjective reviews that amplifiers from Woo Audio have received all over the web, one may start to think that actually the average listener sort of likes the way Woo Audio amps sound and does not really notice or care much about the measured distortion.

This is something quite common for tube amps, I found. The Stereophile review of the WA5 is absolutely stellar. Until it was measured.
I own an Hovland power amplifier that received a similarly outstanding review from Micheal Fremer, but did pretty bad in the measurment department.

It would then appear that perhaps, at least for a lot of listeners (including people as experienced as Micheal Fremer), there is a mismatch between how the sound presentation is perceived and the relevant measurements.

Personally, I have owned and own high quality solid state amplifiers that measure beautifully, as well as amps from Woo Audio. I like both, just like one likes a glass of Chardonnary today and a glass of Pinot Gris tomorrow. I owned a WA7 gen 2 for a while and found its presentation significantly warmer, lusher (or "tubey" if you want) than all my solid state gear, including pure class A specimens. Was it equally clear and detailed? Not all all. Was it equally neutral? No. Was it enjoyable to listen to? IMHO, yes, very much so. Was it better or worse? Nonsense question.

Just my two cents.
 
Last edited:

Surge

Member
Joined
May 7, 2019
Messages
9
Likes
2
Measurements cannot capture how real something sounds. Never judge audio gear by graphs and charts. Utter waste of time.
 

o2so

Active Member
Forum Donor
Joined
Sep 9, 2019
Messages
250
Likes
143
Location
Sydney, Australia
Measurements cannot capture how real something sounds. Never judge audio gear by graphs and charts. Utter waste of time.
I think that measurements are the only way to compare the performance of audio gear against an objective scale. This can tell a lot about the gear being measured. However in my experience it is quite possible that something measuring bad will be very enjoyable to listen to.
 

The Mule

Member
Joined
Jan 23, 2019
Messages
29
Likes
20
Surge and o2so - did you even *read* this part of the review?

Listening Tests

Seems to me, the. WA7 doesn't measure well, and it doesn't sound good either.

Read Amir's Schiit Valhalla 2 review. Fares much better than this.

[And I have both of these, btw.]
 

o2so

Active Member
Forum Donor
Joined
Sep 9, 2019
Messages
250
Likes
143
Location
Sydney, Australia
Surge and o2so - did you even *read* this part of the review?

Listening Tests

Seems to me, the. WA7 doesn't measure well, and it doesn't sound good either.

Read Amir's Schiit Valhalla 2 review. Fares much better than this.

[And I have both of these, btw.]
I did read that part. I'm confused. What is the point you are trying to make with your (not so) mildly condescending comment?
Amir did not like the presentation, I do. As simple as that. No?
 

The Mule

Member
Joined
Jan 23, 2019
Messages
29
Likes
20
Ok ... Remind me *not* to subscribe to any blog of yours, where you subjectively review audio gear. :)

Listening Tests

...

First up was Sennheiser HD-650. On my bass heaving headphone test tracks, the bass on WA7 was more muted. It was also a bit more bloated compared to the tight bass from DX3 Pro. This was subtle. What was less subtle was the brighter signature of the WA7. Highs were exaggerated. The lisp in vocals became worse and high frequencies became grungier. Note that this effect was most audible during periods with strong bass frequencies.

On detail and resolution front, the WA7 lost it there too. The case was loud instruments that vibrated up and down rapidly in level. The detail in between the peaks was nicely rendered by the DX3 Pro but became muddy and lost in WA7. Note that this was not across the board. On content with infrequent transients, the issue was not there.

...

Clearly the frequency response of the [AKG] K92 was modified by the WA7. Working around that, I heard the same issue with lack of bass impact and resolution in WA7.

Since I had so much power available with K92 I decided to try a new experiment. Try not to laugh but I put in two foam ear plugs in my ear. No, these are not musician ones. They are normal noise reduction yellow foam plugs. What this did was essentially cut off everything from mid to high frequencies and heavily attenuated the bass. I was then able to turn the volume up substantially and could easily hear the bass becoming more distorted in WA7 than Topping DX3 Pro.

...

Conclusions
Yet again we measure a tube amplifier which produces copious amount of distortion. However, unlike some budget products, the WA7 is a quite amplifier and has major conveniences such as built-in DAC. It is also pretty to look at. For me as an audiophile, the WA7 takes a step backward. Last thing I want is the highs getting extra sizzle or the tight, impactful bass losing some of both characteristics.

As I have experienced time and time again in controlled testing, there is nothing euphonic or positive with such tube amplifiers. No soundstage is changed. No instrument is isolated more (it is actually worse with WA7). Detail is actually lost. The sound definitely is not warmer either. People read these attributes into such devices because they think they are there. A simple AB test with a $20 switcher is enough to convince you otherwise.

...

I've actually known of Amir and his work for a long time. Back to the old HD DVD vs Blu-ray format war. (Yes, there was a competing high def format to Blu-ray, a long time ago.) If any one in audio tells it like it is, it's him.
 
Top Bottom