-
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!
Review and Measurements of DarkVoice 336SE Headphone Amp
- Thread starter amirm
- Start date
Jimster480
Major Contributor
More like your ears got used to the horrible sound. Tubes don't break in that much.
Ok, whatever. I get that you're not a fan of tubes.
Jimster480
Major Contributor
I like anything that sounds good. I have heard some nice tube setups. But tubes don't break in the way you are mentioning. Your ears were conditioned to the horrible sound and now you think it sounds fine. This is scientifically proven to be a thing that happens.
Indeed they don't. They'll change a little over their first few hours of operation, though - the bias will shift slightly. Probably not enough to be audible, though.
Unlike semiconductors tubes do burn in some as their physical properties change initially. It's a known fenomenon that some tubes can hum after they are first put in service but then the hum disappears and not present even from cold start. I experienced it myself when I owned a DV.
Jimster480
Major Contributor
I read that too. There is some burn in/break in time but overall there is not much of a difference in sound in my experiences. I have listened to a few friends tube amps after they swapped tubes the first time and then some months later after they are "well broken in" and my friends who listen to them all the time told me how they are "much better" but really when I listened to their amps; they sounded basically the same.
While I know that tubes do change slightly as they break in; the difference is not as large as people have claimed and bad sounding tubes don't magically sound good again.
I just recently got a DarkVoice. Out of the box the supplied tubes had no hum and the sound produced via my Sennheiser 6XX headphones was more lush and rich than the comparatively more analytical sound provided by my E30/L30 stack.
I replaced the tubes with a Svetlana 6AS7 and Tung Sol 6SN7 (2021) . The hum was unbearable. I narrowed it down to the 6SN7 and contacted the supplier, claiming a faulty tube. He swapped it out for another one (2016), and I purchased a JJ at the same time. Hum still there on the Tung Sol, bit not as bad. No hum whatsoever on the JJ, but I really don't like the sound; it's muddy and distorted.
So, this has been pretty interesting. I'm now "burning in" the 6SN7 for 72 hours as has been discussed on this thread to see what happens. It's frustrating, but if all else fails, the sound from the stock 6SN7, whilst not being as good as the Tung Sol, provides no hum and a pleasant experience.
One thing I'd like to ask here. I have to use a step-down transformer as the model I have is 110V and in Australia we run off 230V. The supplier claims this is part of the problem, and I understand that. If I were to invest in a really good step-down transformer, what would it's specs need to be? Something that is low-noise and will be dependable for a long time. Any input would be appreciated.
I replaced the tubes with a Svetlana 6AS7 and Tung Sol 6SN7 (2021) . The hum was unbearable. I narrowed it down to the 6SN7 and contacted the supplier, claiming a faulty tube. He swapped it out for another one (2016), and I purchased a JJ at the same time. Hum still there on the Tung Sol, bit not as bad. No hum whatsoever on the JJ, but I really don't like the sound; it's muddy and distorted.
So, this has been pretty interesting. I'm now "burning in" the 6SN7 for 72 hours as has been discussed on this thread to see what happens. It's frustrating, but if all else fails, the sound from the stock 6SN7, whilst not being as good as the Tung Sol, provides no hum and a pleasant experience.
One thing I'd like to ask here. I have to use a step-down transformer as the model I have is 110V and in Australia we run off 230V. The supplier claims this is part of the problem, and I understand that. If I were to invest in a really good step-down transformer, what would it's specs need to be? Something that is low-noise and will be dependable for a long time. Any input would be appreciated.
- Joined
- Jan 15, 2018
- Messages
- 295
- Likes
- 676
The problem isn't the tubes, it's the amplifier. The DarkVoice has it's AC heater supply ground referenced on one leg as opposed to a virtual center tap. It's a poor design decision resulting in noise due to heater-to-cathode leakage. The severity will vary from tube to tube, but if the amp were designed well, it would be present in none of them. The fix is to either get a new amplifier or modify the circuit by creating the virtual center tap with a pair of 100ohm resistors, which requires soldering skills. This may not reduce the hum completely though, the best solution is to also elevate the heater supply off ground by say 30VDC, which again requires some knowledge and soldering skills.
zepplock
Active Member
I've now tried 5 different tubes and the only one's that don't hum are the original stock tube and the JJ.
I can't handle the hum, it's loud and obtrusive. I had no idea... I wonder if the Little Dot Mk II has the same issues? It's the other tube amp that people like me with 6xx's buy as a first tube amp.
I can't handle the hum, it's loud and obtrusive. I had no idea... I wonder if the Little Dot Mk II has the same issues? It's the other tube amp that people like me with 6xx's buy as a first tube amp.
I have been enjoying your video reviews for about a year. I have watched you test and rip apart many products that I thought were pretty good...at least. It does take guts to stand up in a crowd, and announce findings that are uncomfortable or worse...I have the DarkVoice Headphone Amp. I also have a couple of SolidState amps as well as a STAX Solid State Class A amp. My DarkVoice sounds GREAT with 11 headphones I tried with it. There is no buzzing, no distortion, and highlights what many of us like about Tubes.This is a review and detailed measurements of the DarkVoice 336SE pre-amp and headphone amplifier. It is on kind loan from a member who has literally been waiting months to get it reviewed! It costs USD $310 on Amazon including free shipping. It has 42 reviews summing to 4.5 stars on Amazon so customer satisfaction must be good.
This is one beast of a headphone amplifier with a massive transformer and large chassis weighing it down:
Up to now we have tested hybrid tube headphone amplifiers that use tubes as the first stage but drive the headphones using transistors. That allows the output impedance to be as low as solid state amplifier yet give you the "benefits" of tubes, whatever they may be. In DarkVoice 336SE, the signal path is all tube. The configuration is OTL meaning Output-Transformer-Less. A tube amplifier needs to block high voltage DC from going out of the unit and also reduce the rather high output impedance of tubes. This is most often done with a transformer but in OTL designs, other methods are used. The benefit is supposed to be getting rid of nonlinearities of the transformer. The drawbacks depend on the design with the most common one being high output impedance.
The big transformer here is for the power supply. I see no safety markings on the unit. I will be opening the unit later to see how it is inside. Until then, I would make sure to not touch the chassis any more than you have to and only use one hand if possible.
Fit and finish is OK but you probably noticed the crooked power switch. The volume control is damped and a bit stiff but otherwise fine.
There are a set of RCA outputs in addition to inputs so you can use the unit as a pre-amplifier if you like.
Let's get into the measurements and see how she does.
Measurements
The owner had given me a few tubes for the first stage (6SN7). I put one in there randomly and boy, was that a mistake. There was massive amount of hum both audibly and in measurements. Indeed in measurements there were more hum than signal! So I switched to the Raytheon 6SN7 which I had from review of Schiit Saga. That reduced the hum substantially although you could start to hear it past 12:00 o'clock or so on the volume control. Dashboard measurements using RCA outs show the same:
View attachment 21262
Yes, this is a total disaster as measurements go. There is really no reason for such elevated power supply hum. They reach up above the largest harmonic of the amplifier itself!
Speaking of the harmonics, they fall off close to perceptual masking so they are less audible than their numbers would indicate:
The Masking threshold on the right drops off exponential and so does the distortion components of the 336SE. So if you are going to have distortions, you want them this way.
Switching to headphone jack, performance remains more or less the same but by loading it down with my standard 600 Ohm, distortion rises substantially:
View attachment 21263
Covering our basics, here is the frequency response:
View attachment 21264
Pretty good actually with just a 0.7 dB drop at 20 Hz.
Warm-up showed quick stability:
View attachment 21265
Problem arises in impedance measurement:
View attachment 21266
Best case impedance is 78 ohm from 300 Hz up. Below that there is a large increase to 163 ohm at 20 kHz. This means two things:
1) With headphones with variable impedance at low values, there will be substantial change in frequency response.
2) The output is going to drop radically with lower impedance headphones.
Speaking of power, let's measure that into 300 Ohm:
View attachment 21267
We have plenty of power here, beating even Massdrop THX AAA 789. Reason for this is that at high impedances, power is limited by voltage. Tube amps run at high voltages easily and hence, are not limited this way. The DarkVoice 336SE delivers double the power of the THX (in single-ended mode). If we increases the output load impedance to say 600 Ohm, the Darkvoice would likely pull ahead even more.
Distortion of course is massively high, reaching up to nearly 5% at max rated power! "Good news" is that it sets in gradually and there is no sharp hockey stick as we see with many transistor amplifiers.
Situation degrades substantially at 33 ohm as we could easily predict:
View attachment 21268
Power output is just 0.2 watts, losing massively to THX's 1.3 watts. Distortion is also sky high at 20%.
Channel matching starts OK but as the volume goes lower and lower, it loses all hope:
View attachment 21269
So if the hum doesn't get you with sensitive IEMs, the channel mismatch will.
No sense in running more tests. We know the nature of this beast.
Listening Tests
Testing the subjective fidelity turned out to be more complicated than one would imagine. Due to high output impedance of the Darkvoice 336SE, the output level you get in your headphones will be highly variable. As such, matching levels with test tones and my analyzer got me close but not close enough. I had to match levels manually against the reference Massdrop THX AAA 789 which is a difficult trial and error. After much work I think I got some useful results but not as reliable as I wanted.
Let's start with easier part of this trial, namely using low impedance headphones like my Hifiman HE-400i. Power output dropped substantially. As such the THX amp would run circles around the 336SE, producing tons more power, and dynamic range. Negating that by lowering the volume on THX 789, we were greeted with severe distortion if you turned up the volume on the 336SE. Pretty unpleasant experience. I read a bunch of reviews online and it was good to see the community's consensus being the same.
What the community did recommend was high impedance headphones such as my Sennheiser HD-650. Boy, did the tables turn. I could easily outrun the the THX AAA 789 (in its single-ended mode). Because of this you could operate the 336SE at lower volumes and hence much reduced distortion. Still, push the 336SE and you were greeted with shrill highs and lack of bass. The accentuated highs did give the impression of "more space" at times and exaggerated high frequency tones on some content was a pleasant addition. Once my ears tuned into them though (through careful AB test with the THX), it became a small annoyance. With a reference to test against, if you are coming from a lower output (or no) headphone amplifier, you are going to find the 336SE a revelation. It has enough power to do justice with high impedance headphones.
I then tested the two amps with my Sony MDRV6 which I think is rated at 60 ohms (?). Here, hum from the 336SE was audible when music was paused, or at times during very quiet parts. The issue was loss of resolution and impact in bass frequencies. That sense of "hi-fi" that makes you feel every note was gone with the 336SE.
Switching to my sensitive AKG AK60 headphones exaggerated the situation with the Sony above. Bass impact was gone and highs become more shrill/hissy.
The lesson here is that what you hear will be highly load, volume and content dependent. It will also depend on your hearing acuity as to whether you can hear small distortions and true resolution.
Conclusions
From objective point of, the DarkVoice 336SE is absolutely horrible. It is worse than it should be but at $310 for all that you get, I guess that is to be expected.
People though get such products for their subjective qualities. There, I can confirm the appeal of using the 336SE with high impedance headphones of 300 ohm and higher. There is copious amount of power here, and what distortion there is, is masked for the most part. In my experience of either power amplifiers or headphone, how much power you have available determines fidelity first and foremost. Lack of power results in anemic sound which seems to lack impact, and even resolution. The high output voltage of DarkVoice 336SE gives it such a strong advantage that it can override its much higher distortion. With that distortion following perceptual masking, it is not as much of a detriment as it may seem at first.
Going to lower impedance headphones is unwise though. There is insufficient current to drive them together with large output impedance which substantially degrades/changes the frequency response of the headphone. You are much better off with a powerful and distortion-free solid state headphone amplifier such as JDS Labs Atom and Massdrop THX AAA 789. They will get louder and will be cleaner to boot.
So do I recommend the DarkVoice 336SE under any conditions and lose my objectivist license with it???With high impedance headphones if you like what comes with tubes as far as maintenance, and can be blind to any safety issues here, sure, you can get the 336SE and I won't be there to hound you. Use it in other scenarios and you and I will have words!
------------
As always, questions, comments, recommendations, etc. are welcome.
They say you should save money for a rainy day. It rains for about 6 months here so I need plenty of money!!! Please consider donating funds 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).
Dont get me wrong, when you review products like the PS Audio Noise Harvester, and warn us that it is nothing less than a scam, you are doing "Gods work" to the Audiophile Community. Paul McGowan has built a beautiful biz, he does NOT need to scam us with snake oil...but I strongly believe that your findings with the DarkVoice 336SE, are not fair. Just like Danny form GR Research, needs to listen to speakers, and not just analyze them...computers do not have ears ....
respectfully,
Mike S.
New Jersey USA
Jimster480
Major Contributor
Hi mike, thank you for signing up to the Forum to share your thoughts. I would like to start off by telling you that you are both right and wrong at the same time.
@amirm findings on this amp are not wrong. However that doesn't make you wrong.
I have been following the audio industry for many years now and while I have been in this scene a lot less time than most other people. I also come with a very open mind about how audio actually is perceived.
See I grew up at the Advent of MP3 and had one of the world's first MP3 players the Rio 600. I transcoded songs that I acquired on the internet and from CDs into low bit rate MP3. My CD sounded pretty much fine at 64 kilobits per second but had a noticeable decline at 56 kbps. Stepping up to 128 kbps per second would not allow me to fit a full album on my little MP3 player at that time... however there was not much difference in audio quality.
Today I have top tier measuring equipment as well as some lower end cheaper stuff that does not measure that well.
I have to say that lower end Hardware often sounds completely fine on music with less dynamic range or which was not recorded in a way that takes advantage of the entire bit depth.
However there is a second Factor at play; this is about what your brain knows and what you have personally experienced. After growing up listening to low bit rate music for many years I graduated up into working at a desk and listening to high quality streaming all day. I listen to literally over 1,000 hours of Music out of my realtek integrated audio chipset on my work computer motherboard. That + my Panasonic RP-HT360 and I thought it sounded pretty good... Even others in my office compared their headphones and a couple people actually bought the headphones I had.
Then I left the company to work at home and ended up buying a 1More Triple driver IEM.... the difference between that and other headphones I had heard was so large that it got me into the HiFi scene...
So if you are used to listening to your dark voice for example then you will think that it sounds great. If you listen to something else side by side especially in a non-sighted test then you might find surprising results. I know that I myself found quite a few surprising results doing tests similar to what I just mentioned...
If you watch a 720p TV all day every day; you think it looks good. Come see a higher end TV and all the sudden it doesn't.... Watch a cheap 4k TV... Then come see a Sony Master series OLED or another flagship TV and all the sudden your cheap TV doesn't look good.
However if you watch 720p YouTube channels.... The master series won't look much better.... After all it's the source content.
Which is the other part of the audio 'mystery' that I think keeps people arguing forever....
Many people in the audiophile / hifi Community are into lots of older music which was not recorded with high dynamic range of any form. Many others are into more recent music which is also not recorded with high dynamic range or even great bit depth... Maybe the ADCs they used sucked or the mastering, etc, etc, etc...
This has gotten quite long so I will make my final point. The performance measured here is typically audible up to a certain amount (around 105DB Sinad assuming linearity is textbook perfect) and after that I find no real differences to note even in the best recorded music I can find....
However as a person who tests products for a living.... I have listened to plenty of mainstream music through cheaper dac and amps I've gotten for test that I know won't be SOTA if measured but still "sound great" with lots of the music I have...
So don't doubt the scientific process here as audio is definitely science in every form. However the understanding of that science is where a lot of the industry and its members go astray.
Similar threads
- Poll
