Review and Measurements of Topping DX3Pro DAC and Headphone Amp

[zephyros]

I've been using the device without any driver and no problem so far. Granted it's only a few days, but may be it's worth a shot. (though without the driver the WASAPI event mode have crackling sound for 24/96 music)

 

[777]

One of the biggest problem it is the different classes. Some of them are class II, without earth ground, the other are with earth ground. When You mixed them, It is possible to apear a voltage between them and of course, currents. This can trigger certain digital processes like shutdown.

 

[Yuno]

It would be really good if someone who experiences shutdown problem more often could test with some decent, aftermarket power supply. Mean well for instance. I would do it myself but self shutdown happened to me twice in 2 months, both solved by powering on.
For the record, I have everything connected to APC back-ups at all times.

 

[bleachershane]

It would be really good if someone who experiences shutdown problem more often could test with some decent, aftermarket power supply. Mean well for instance. I would do it myself but self shutdown happened to me twice in 2 months, both solved by powering on.
For the record, I have everything connected to APC back-ups at all times.

See MattG's comment on page 102...

 

[Yuno]

See MattG's comment on page 102...

Thanks, missed this. Disappointing.

 

[Cidious]

I opened it up and decided to replace the SOP8 socket with a swappable DIP8 socket. But obviously there is not much space and I'll be trying to squeeze in a SPARKOSLABS SS3602 discrete opamps instead of the standard (already good) OPA1612 opamp.



Comparing the DX3 Pro to my LKS ES9018 (tuned with jefet class A I/V stage and high-end audio grade components)


I notice that the DX3 has more detail and more forwards mids which I like. But it's lacking in soundstage and the lower frequencies are a bit underwhelming. I think this is mainly because chip flavor and not so much due to output stage differences thought. Still I want to try to make an upgrade to the OPA1612 to see if I can open up the soundstage a bit and get that bass a bit more defined.

First need to figure out how to get it all to fit in that small space and convert SOP8 to DIP8

OPTIONS!



Lowest possible option while still swappable. I might solder it in completely later if I like the change.



Barely fits... Like I said soldering it in completely will be the safer thing to do but first I need to roll some!


Costs are already getting out of hand since I was tired last night and thought I could quickly solder out the SOP8 opamp to solder in the DIP8 socket.... but my eyes couldn't keep up at that time and I broke the coper tracks on the board and it's properly useless now. So I contacted the Topping factory in Shenzen and let them send me a new board for 800 RMB (€100 I think). Next to that I ordered a completely new unit to compare with. I'll modify the replacement board and keep the new unit original.

The Opamp is the most expensive out there at 690 RMB (€80) it's pure overkill for a device that cost me €140 at first, but I have a feeling that this device might be able to sound really bigger than what's its capable of originally. So let's just go for it and do it right this time.

In about 2 days I'll be able to report back after all my orders came in to finish the project.

 

[maty]

Like a lot of DAC, preamp, phono, class D amps ... it has thin aluminum wall. Two millimeters?

 

[Pavel]

View attachment 20614

You can benefit more in sound quality if you replace 8 conventional tantalum capacitors with Kemet KO-CAP.
Cidious, and check you inbox pls.

 

[noel_fs]

I opened it up and decided to replace the SOP8 socket with a swappable DIP8 socket. But obviously there is not much space and I'll be trying to squeeze in a SPARKOSLABS SS3602 discrete opamps instead of the standard (already good) OPA1612 opamp.

Like previous comment said, you'll probably get more benefit from replacing caps rather than replacing opamps, anyway it will be very interesting to see results!

 

[Cidious]

Replacing caps will be the next step. Honestly rolling opamps always is the biggest change unless I really change something radically in the circuitry. Replacing caps and other components for equal values in better quality often just gives a marginal improvement. Changing valued does help sometimes.

Let's discuss possible mods... I hope to find some people that are willing to get their hands dirty

 

[yue]

I posted last week, doing this exact test:



I came in to my office this morning with the unit in sleep mode: powered on, but only a single illuminated dot on the LCD display, and no sound. So it only made it about a week with the alternate power supply (Meanwell GS90A15-P1M).

So per Topping's email above, it appears to be a problem with the power circuit logic. Hopefully this is firmware fixable!

Edit: I sent Topping an email about this.

It is possible it is caused bynbug in Topping's implementation of the auto-standby mode.
To isolate the problem. Is your device running at auto-standby mode? Could you please turn it off and see if the problem continues?

 

[bleachershane]

I opened it up and decided to replace the SOP8 socket with a swappable DIP8 socket. But obviously there is not much space and I'll be trying to squeeze in a SPARKOSLABS SS3602 discrete opamps instead of the standard (already good) OPA1612 opamp.

View attachment 20614

Comparing the DX3 Pro to my LKS ES9018 (tuned with jefet class A I/V stage and high-end audio grade components)
View attachment 20618

I notice that the DX3 has more detail and more forwards mids which I like. But it's lacking in soundstage and the lower frequencies are a bit underwhelming. I think this is mainly because chip flavor and not so much due to output stage differences thought. Still I want to try to make an upgrade to the OPA1612 to see if I can open up the soundstage a bit and get that bass a bit more defined.

First need to figure out how to get it all to fit in that small space and convert SOP8 to DIP8
View attachment 20613
OPTIONS!

View attachment 20617

Lowest possible option while still swappable. I might solder it in completely later if I like the change.

View attachment 20615
View attachment 20616
Barely fits... Like I said soldering it in completely will be the safer thing to do but first I need to roll some!


Costs are already getting out of hand since I was tired last night and thought I could quickly solder out the SOP8 opamp to solder in the DIP8 socket.... but my eyes couldn't keep up at that time and I broke the coper tracks on the board and it's properly useless now. So I contacted the Topping factory in Shenzen and let them send me a new board for 800 RMB (€100 I think). Next to that I ordered a completely new unit to compare with. I'll modify the replacement board and keep the new unit original.

The Opamp is the most expensive out there at 690 RMB (€80) it's pure overkill for a device that cost me €140 at first, but I have a feeling that this device might be able to sound really bigger than what's its capable of originally. So let's just go for it and do it right this time.

In about 2 days I'll be able to report back after all my orders came in to finish the project.

I hate to say it but the moment I noticed you said you were going to install a SOP8 to DIP8 converter to install a discrete opamp in your earlier comments, I knew this wasn't going to end well. I mean no disrespect but could the expense you have ended up with by taking this path ever be worth the cost when we're talking about a sub-£200 all in one DAC/headphone amplifier, with more than a few buggy issues and according to the reports on here its long term reliability is questionable?

I say this as someone that used to spend a lot of time trying to out think designers by pursuing perpetual upgrades, before learning the hard way that 99% of the time, the simplest and most routes usually made the most worthwhile changes.

 

[simonchretien]

It is possible it is caused bynbug in Topping's implementation of the auto-standby mode.
To isolate the problem. Is your device running at auto-standby mode? Could you please turn it off and see if the problem continues?

Whats "A-0" and "A-C" when pressing Auto on the remote? I dont even understand their garbage manual.
Which one disables auto-standby?

 

[yue]

Whats "A-0" and "A-C" when pressing Auto on the remote? I dont even understand their garbage manual.
Which one disables auto-standby?

I read the manual in both English and Chinese and here is my understanding:


AO is to turn on “auto power off” mode and this is the factory default setting.

AC is to disable that feature. this is the setting you want to test.

 

[Cidious]

I've had no issues with the device whatsoever.

For the other part that it doesn't make much sense cost effectively wise. I totally agree ! It's a hobby to keep me busy a bit. I spend over 4000 RMB now on this mod. And that would have bought me a Singxer SDA-1 with same chips and probably more he GH end components. Or wait for the D70. BUT it's fun and I havent got much else to do since it's almost spring festival.

Thanks for the comment anyway. Appreciate it. The discrete opamp I can use in my Lehman too so it's not a total waste. The ruined board is sad but I'll use it as a practice board for future mods and exploration. The new unit will be my control unit and the replacement board will be for modifications. Purely curiosity and educational more than trying to achieve the ultimate DAC sound.. I could have gone another way if I wanted that.

I hate to say it but the moment I noticed you said you were going to install a SOP8 to DIP8 converter to install a discrete opamp in your earlier comments, I knew this wasn't going to end well. I mean no disrespect but could the expense you have ended up with by taking this path ever be worth the cost when we're talking about a sub-£200 all in one DAC/headphone amplifier, with more than a few buggy issues and according to the reports on here its long term reliability is questionable?

I say this as someone that used to spend a lot of time trying to out think designers by pursuing perpetual upgrades, before learning the hard way that 99% of the time, the simplest and most routes usually made the most worthwhile changes.

 

[Pavel]

Let's discuss possible mods... I hope to find some people that are willing to get their hands dirty

I plan to:
1. Use a transformer output instead of an op amp.
2. Replacing the TPS54331 with the TPS54332 (1 MHz instead of 570 kHz) Or simply abandon Step-Down in the DAC's power supply.
3. Simple recloker on flip-flop and NDK between AK4118 and AK4493

P.S. to bleachershane. For me, the DX3Pro is interesting as a donor and acquaintance with the new AK4493 chip. In addition, Topping engineers have done a good job in a tight price range for the product.

 

[Stonetown]

Hi all. Got mine DX3 Pro a week ago and it works as expected for me. Streaming from Apple Music in the watch via Bluetooth to DX3 Pro works just fine. Yes the Apple Music-app is inside the watch so it is not using the iPhone att all to play the music. I did turn off the phone completely and the watch still streams to DX3 Pro. Enjoying music with such good audio quality from just a tiny watch is amazing! However I mostly use USB from the MacMini och sometimes also Toslink from my MacBook Pro. All works flawlessly.

Thank you admin for all the testing here on this site. That helped me a lot to decide about when buying this unit.

A am curious however about how the output spectrum will look like when for example adjusting the computer output (easily done via Apple Audio- Midi-settling app in MasOSX) to send Toslink 24bit/192kHz, and also USB 32bit/768 kHz. If it is better or worse when playing a standard CD (16bit/44,1kHz), ie when forcing DX3 Pro to higher bitrates? Maybe higher bitrates multiple to 44,1kHz is smarter? Any spectrum curves 20-20kHz to would be highly appreciated. Or shall we just leave it to 44,1 kHz?

 

[777]

Cidious, how did you manage to remove PCB ? Did you give your front panel down ?

 

[Pluto]

I read the manual in both English and Chinese and here is my understanding

As you can work in Chinese, there is something you could do that would be rather useful if you can explain it to Topping. On the remote for my Topping DX7s there is (IMHO) an obvious error of logic that should be changed but I fear that it will need someone fluent in Chinese to explain. Here we go....

Buttons such as FIR (filter selection) and AUTO (auto power ON/OFF) serve as both interrogation [tell me what mode am I currently using?] and change (by pressing the button again). Problem is, the first press changes the state and it is necessary to press the button – five further times in the case of FIR – to get back to where you want to be. Logically, the first press should be just interrogation. After a couple of seconds, the display should return to its default state and NOTHING should have changed. It is only a second press – before the display returns to its default state – that should cause the function to change.

Another logical fault is the ease with which DAC mode can be selected. This is the mode in which the volume control is bypassed and the unattenuated DAC output level is sent to the power amp. This has the potential to be harmful (and possibly, dangerous to hearing). The selection of that mode really needs a safeguard involving a second press of a different button ( to safeguard against accidentally double-pressing a single button).

If you could explain the above to Topping, you would be doing their business, and us, a great favor

 

[kukocz]

As you can work in Chinese, there is something you could do that would be rather useful if you can explain it to Topping. On the remote for my Topping DX7s there is (IMHO) an obvious error of logic that should be changed but I fear that it will need someone fluent in Chinese to explain. Here we go....

Buttons such as FIR (filter selection) and AUTO (auto power ON/OFF) serve as both interrogation [tell me what mode am I currently using?] and change (by pressing the button again). Problem is, the first press changes the state and it is necessary to press the button – five further times in the case of FIR – to get back to where you want to be. Logically, the first press should be just interrogation. After a couple of seconds, the display should return to its default state and NOTHING should have changed. It is only a second press – before the display returns to its default state – that should cause the function to change.

Another logical fault is the ease with which DAC mode can be selected. This is the mode in which the volume control is bypassed and the unattenuated DAC output level is sent to the power amp. This has the potential to be harmful (and possibly, dangerous to hearing). The selection of that mode really needs a safeguard involving a second press of a different button ( to safeguard against accidentally double-pressing a single button).

If you could explain the above to Topping, you would be doing their business, and us, a great favor

Exactly the same behavior in DX3 Pro. I even proposed solution for fixed line-out (separate settings available via settings mode in DX3 Pro) but there could be a problem to fix it via firmware. I believe, they'll manage it