Review and Measurements of Topping DX3Pro DAC and Headphone Amp

[eziitis]

Yep, the 100uf value used in DX3 is bad, too. 4493 is suffering with all the ripples.

I am afraid changing capacitor value will not be big help here...

 

[Sangram]

kind of OT, but it would be very helpful if Topping would provide schematics. such a request is not so silly as it may sound. for instance Khadas tone board has all schematics public, and I can say from the point of EE they have done very good job. yes there is a switcher for opamp power as well, but it is not so critical, because opamps have high PSRR. another matter is impedance of course .... nevertheless the critical AVCC comes from a dedicated reg and is pretty clean in fact.

Kind of OT but opamp PSRR strongly depends on the frequency of the ripple, the PSRR is very poor once you enter switching frequency and harmonics of the output of a switcher. It is indeed possible to make a very quiet switcher - indeed some switchers can exceed the performance of a linear power supply.

Topping's design is good enough, but that is all - they haven't taken the design to where they could've. You hit most of the issues a little earlier - clubbed supplies, poor I/V design choices, horrible ergonomics, lumped summing instead of proper diff receivers for each set of outputs - improving these would yield a better unit but I'm pretty sure they would have doubled the cost and size also. What is not certain is if all of this is oversight or part of the market plan - given the number of people who have complained about issues I'm inclined to think it's a bit of both.

 

[Yuno]

What is not certain is if all of this is oversight or part of the market plan - given the number of people who have complained about issues I'm inclined to think it's a bit of both.

I think all issues come from the fact that this unit was made to be dirt cheap. Personally I don't really care for most things, dx3 sounds very good to me, measures competently and is very convenient to use. The only thing I can't stand is the reliability. It's been quite a while since we had any update from topping too, quite a few theories on why dx3 pros shut or break down but no official info. Mine so far shut down 6 times, with no permanent damage thankfully, frequency of about every 2 weeks. As much as I like this thing, if I knew it would be like this I would not buy it. Last time I beta test product shipped from China, that's for sure.

 

[eziitis]

the PSRR is very poor once you enter switching frequency

you are right about PSRR vs f of course, but about -40db at switching f is a way better than about 0dB what DAC reference pins usually have (did not find exact no in ak4493 data sheet). secondly, of course the switcher is not the sole root of evil here, in audio band op amps will be more immune though. personally I think switchers are fine to power LPF opamps, if one can design a PS with low and flat z in audio band.

for OPA1611/12

 

[noobie]

good points indeed unfortunately sending mod unit elsewhere for measurements is absolutely impractical for me.
so if the thread is about measurements, and the DAC performance as it is in original state has been measured in very beginning, what more we can measure?
in order to come up with sensible mods one should identify critical points, and at the moment it appears VREF to me. so I am putting back those
TPS54331 switchers, what I prematurely removed in favour to linear PS and will try to measure what noise we have on +5VL/R test points. I would say first one should look for presence of switching frequency and signal frequency. upon the extent of contamination one can decide whether and how things could be changed. anyone with a decent scope willing to join this measurement?

kind of OT, but it would be very helpful if Topping would provide schematics. such a request is not so silly as it may sound. for instance Khadas tone board has all schematics public, and I can say from the point of EE they have done very good job. yes there is a switcher for opamp power as well, but it is not so critical, because opamps have high PSRR. another matter is impedance of course .... nevertheless the critical AVCC comes from a dedicated reg and is pretty clean in fact.

@eziitis no doubt the original unit has been measured by @amirm quite extensively. I also don’t doubt that some on this thread have the means and possibly backing from others to do testing. But as the thread and the whole forum are founded on @amirm measurements, would be appropriate protocol to have confirmatory measurement done by him on improved units...

As I am not aware of your situation I don’t know what complexities are involved in you or others shipping a modified unit to @amirm. However, if it is cost, I and maybe others might volunteer to donate some of the funds necessary to get units shipped if it is not too exorbitant.

In all I note that there are two flavors to issues - improvements to the performance of the unit (what I am referring to) and reliability of the unit. The improvements are subject to relatively rigorous analysis (you, @finneybear, @amirm and others can drink some beer/wine and debate what measurements to do).

The reliability... I am afraid most of what has been discussed on this thread falls into the category of anecdotal evidence. It would be fantastic if Topping were to give us a failure rate number, but I can understand the corporate mindset and realize they are probably not likely to (they are a for-profit enterprise with competitors, after all). Earlier in this modern rendition of “War and Peace” a Topping rep commented that there were no failures returned to them, but I am sure with more units sold and more time they now have some and can calculate their failure rate. But we probably won’t hear a number. If a Topping rep is reading this, maybe they would be willing to tell us if the DX3PRO failure rate is higher, equal, or lower to other units they sell (though cost differences likely skew that).

And my own little unit keeps on chugging along...

 

[eziitis]

... What is not certain is if all of this is oversight or part of the market plan - given the number of people who have complained about issues I'm inclined to think it's a bit of both.

I guess it is a market plan, since the unit has 2xAK4493, and a hi-res sticker )) 2 chips here do not make EE sense, even more having common +5V reg per chip. but dx3pro does not sound bad after all, and more importantly is has HPA as well.
of course, roughly for the same investment one can make much better sounding single chip DAC, but "just a single chip" would have lesser market impact perhaps.

btw Benchmark designed their DAC3 with ES9028 and clearly explained all the engineering reasons and benefits to use it in a stereo DAC, while many other brands were taken by ES9038pro hype...

so, summarising ... do not open your DX3pro unit, because you may find some worms there just relax and enjoy the music!

 

[Killingbeans]

It is not the amount of THD, it is the shape of the curve. After signal going through the cap, the distortion amplitude changes among different frequencies. The impedance curve is another important factor, too. Then after the cap, if the signal gets amplification, the THD will go up again. This THD is just a component of the signal chain, not the final THD number of the whole circuit. From the curve we can see coupling cap would degrade the low frequency signal quality more. How serious the problem is, all depends on what's at the next stage. If the cap is for the final output, things may be OK. But as the output coupling cap for DX3's 4493 output, it can be a problem.

But are you talking about lowering the final amount of harmonic distortion?

Because earlier you wrote:

The mod is mainly to fix some poor engineering decisions in DX3, not to lower noise floor or to reduce distortion, etc.

And I'm getting seriously confused

 

[Veri]

The linear supply will certainly improve the DX3.

Please. stop. with the hyperbole, gosh.

 

[la2ygoo]

@eziitis no doubt the original unit has been measured by @amirm quite extensively. I also don’t doubt that some on this thread have the means and possibly backing from others to do testing. But as the thread and the whole forum are founded on @amirm measurements, would be appropriate protocol to have confirmatory measurement done by him on improved units...

As I am not aware of your situation I don’t know what complexities are involved in you or others shipping a modified unit to @amirm. However, if it is cost, I and maybe others might volunteer to donate some of the funds necessary to get units shipped if it is not too exorbitant.

In all I note that there are two flavors to issues - improvements to the performance of the unit (what I am referring to) and reliability of the unit. The improvements are subject to relatively rigorous analysis (you, @finneybear, @amirm and others can drink some beer/wine and debate what measurements to do).

The reliability... I am afraid most of what has been discussed on this thread falls into the category of anecdotal evidence. It would be fantastic if Topping were to give us a failure rate number, but I can understand the corporate mindset and realize they are probably not likely to (they are a for-profit enterprise with competitors, after all). Earlier in this modern rendition of “War and Peace” a Topping rep commented that there were no failures returned to them, but I am sure with more units sold and more time they now have some and can calculate their failure rate. But we probably won’t hear a number. If a Topping rep is reading this, maybe they would be willing to tell us if the DX3PRO failure rate is higher, equal, or lower to other units they sell (though cost differences likely skew that).

And my own little unit keeps on chugging along...

Some people who have modified dx3 have their own measurements, and they know what they have done.
I think that only those who have not measured and feel that the modified dx3 sounds better, it is necessary to let amirm measure, just for fun, no need to be serious.

 

[Killingbeans]

No need to be delusional either...

Sorry for being a party pooper

 

[Barce]

Which OpAmp swap stops the thread from being overrun with hyperactive weirdoes who make 8 consecutive uninterrupted posts of solipsistic blather in a single hour?

 

[trl]

Finally done with the LPF. Use only one set of outputs for each 4493:

View attachment 22915
[...]
The FC point is now around 130KHz, down from 178KHz.
[...]

Very interesting, also a very clean work. I wonder if you could "inject" a 20KHz square wave somewhere at the input of the LPF and measure the output at the RCA plugs with a scope (of course, repeating the test with 10KHz might be even better). It could be helpful to see both IN and OUT traces in overlay and see the attenuation and phase delay/shift. Of course, before vs. after measurements would be nice to compare.

Funny thing, I usually notice that increasing amplifier's bandwidth might help me in getting more micro-details and cleaner sound, not vice-versa. It's odd that you find the sound better now, with a lower bandwidth, so the above "square wave generator & scope" test might help in determining the actual reason (if any). Usually manufacturers are limiting the bandwidth just to be sure that their devices will not oscillate, especially opamp-swappable equipment, so increasing bandwidth sometimes might help.

 

[Veri]

The real king of mods has spoken

 

[amirm]

Was the test done while knocking on the ceramic caps? If not, then I think the test is valid as piezoelectric effect clearly exist.

The paper is quite vague on detail as I noted but you can read it from the link provided. My sense was that speakers were used to generate noise or something like it.

 

[amirm]

Some rules moving forward:

1. Feel free to discuss mods to DX3 Pro.

2. Please do not make subjective claims regarding what improvements it has made. It is only going to trigger arguments. In this forum we encourage subjective testing very much but they must be controlled/blind. If the mods are for your benefits, then keep them observations to yourself.

3. Do not make technical claims of audio superiority of parts unless you can demonstrate with your measurements as I did in the original review of before and after.

4. Feel free to create threads elsewhere if you want to argue about the above. This thread is for people interested in the product, not arguments.

 

[finneybear]

@eziitis no doubt the original unit has been measured by @amirm quite extensively. I also don’t doubt that some on this thread have the means and possibly backing from others to do testing. But as the thread and the whole forum are founded on @amirm measurements, would be appropriate protocol to have confirmatory measurement done by him on improved units...

As I have mentioned, it is a good idea for Amir to do the measurements for a consistent result, not that I do not have AP to do test. And I can predict the result will be pretty much aligned with what other people have got - much less noise spikes, etc. I plan to swap the XO for lower phase noise ones though.

My fixed DX3 has been running non-stop for 4 days now. Solid stable.

 

[finneybear]

What problem? THD+N is -106 dB. 2nd harmonic distortion is at -110 dB.

And you have another headphone amp stage. The non-linearity of the impedance/distortion is the real problem.

 

[finneybear]

sorry, what kind of HF signals we are talking about in an audio application? I would kindly ask to stop confusing people knowing even less that use of DIP8 opamp package in an audio application is a major design fault. just opposite, it may have even some benefits as it allows more solid power traces.

There are a lot of over 100KHz upsampling and switching noises coming out of DAC and they will have to be removed by LPF. LPF has long been a design challenge for DAC.

 

[finneybear]

I am afraid changing capacitor value will not be big help here...

Datasheet calls for at least 470uf here. Designers at least should follow what datasheet says. Topping uses 100uf here to save money.

 

[finneybear]

But are you talking about lowering the final amount of harmonic distortion?

Because earlier you wrote:

And I'm getting seriously confused

Mmm.... you are confused with two different things. For the capacitors, I was answering to people who claimed all capacitors have the same performance. The THD talk is a follow up discussion between Amir and me.

For the mod of DX3, what I was doing is mainly to resolve the stability problem.
I see Topping has pull DX3 out of the market. Most likely a revised version is coming.
Just wondering how Topping will do with the old versions we have now?