Topping A90 Headphone Amplifier Review

[bboris77]

The removal process could have degraded the pot further(I'm very certain it has, and again with reinstalling). (Also the pot is fully customized, you almost are not able to find exact replacement for the pot, 5k 4gang 5% -60db tracking) So this is not recommended for everyone else. Also as I posted earlier it's clearly described and explained. It is the pot itself. But it's caused by external forced applied to the pot during the knob installation process. Some ended up requiring less force some with more. So the end result is some units may exhibit this behavior. But like many others have said, it's acceptable and doesn't affect the performance itself of the volume control.

Your transparency about this minor issue is really appreciated. Most companies would never publicly discuss how this volume knob quirk came to existence. I can only hope that you guys can figure out a way to reduce the occurrence of this in future batches of the A90. As a customer, or a potential customer, I would rather be in the know about things rather than in the dark.

 

[JohnYang1997]

Your transparency about this minor issue is really appreciated. Most companies would never publicly discuss how this volume knob quirk came to existence. I can only hope that you guys can figure out a way to reduce the occurrence of this in future batches of the A90. As a customer, or a potential customer, I would rather be in the know about things rather than in the dark.

Thanks for the support.

 

[JohnYang1997]

It is not necessary to change it, I have already commented on several occasions that it does not affect the fit and the touch is good. I personally do not find it annoying.

Yeah, I acknowledge that. However what I meant was that when you put it back on it will definitely be worse than it was. It's great that you don't find it annoying. But I was concerned that your experience might be worse after this. Well, at least you found out yourself that it's the pot itself. So that's good.

 

[Rigel]

Ok, I did not put the photos as a critic, I hope I have not bothered. I just like to see the inside of the equipment, I suppose that by professional defect, it has happened to me since I was little, wanting to see the inside of anything electromechanical. I suppose there will be people who also find it interesting and that's why I put them on.
I find it a really good amp in sound and finish.
The D90 + A90 combination is the best I have been able to do in a long time in audio and I am not saying this to flatter anyone.

 

[reiserfs]

A90 + D50 has noise on the RCA input for sure like others reported. It seems like the A90 is really sensitive on the RCA inputs. Even the Atom + D50 had no such problems. Thankfully my Modius will be here tomorrow so I can run XLR.

 

[Audimon]

A90 + D50 has noise on the RCA input for sure like others reported. It seems like the A90 is really sensitive on the RCA inputs. Even the Atom + D50 had no such problems. Thankfully my Modius will be here tomorrow so I can run XLR.

Just want to clarify. Is there noise while D50 connected via USB? Have you tried optical? Is there still noise?

 

[CumSum]

Thank you very much to Amir for the measurements and everyone else for all the advice and discussion on here.

Could I please ask for advice whether (assuming my hearing is that of an average 49 year old) I will hear a noticeable difference if I get an A90 to replace my current headphone amp, a Marantz HD-DAC1 (which is a headphone amp + CS4398 DAC) ? I know there are measurements on here from Amir of the HD-AMP1 (an integrated amp + ES9010K2M DAC) which gave not so good results but I don't know how much of those results is down to the amp and how much down to the DAC.

I had a Marantz HD-DAC1 powering Mr. Speakers Alpha Primes and an HE400S for a couple years. The DAC inside of it is okay and honestly sounds like most DAC's. I remember the Marantz DAC sounding very close to the Topping D30, which sounds identical to the SMSL SU-8 V2.

As for the amp, that is where you can see some improvements. I got a JDS Atom as my first ASR device, and it outperformed the Marantz. I then got a THX 789 and that outperformed the JDS Atom but fairly small margins.

To put things short, the Marantz's headphone amp is the shortcoming you will hear most. You've gotten a D90 so definitely ditch the Marantz completely and get a decent headphone amp.

 

[JustAnandaDourEyedDude]

The removal process could have degraded the pot further(I'm very certain it has, and again with reinstalling). (Also the pot is fully customized, you almost are not able to find exact replacement for the pot, 5k 4gang 5% -60db tracking) So this is not recommended for everyone else. Also as I posted earlier it's clearly described and explained. It is the pot itself. But it's caused by external forced applied to the pot during the knob installation process. Some ended up requiring less force some with more. So the end result is some units may exhibit this behavior. But like many others have said, it's acceptable and doesn't affect the performance itself of the volume control.

Is it possible to change the assembly process to decrease the force applied to the fragile part of the pot? Currently, is the little white plastic coupler mated with the black knob first or with the brass shaft of the pot? Is the hole in the case large enough that the white coupler could be mounted on the brass shaft (in a controlled way using a pair of tweezers to grip the shaft while mounting the coupler) and the shaft+coupler pushed through the hole in the case? If this could be done, would the excess force on the pot still be too large when the big black knob is mounted onto the coupler on the outside of the case? [i.e., which fit is very tight and requires so much force?] Would the use of a small amount of a suitable lubricant help in mounting the coupler on the shaft and the knob on the coupler? I am sure Topping has looked into these and more possibilities, but the (minor) issue seems like it might be solvable without big changes or cost.

 

[mt196]

Yes I can confirm that I tried the A90 connected through both XLR and RCA to the DAC, and to the Anaview AMS0100 as speaker amp through RCA, and there is a little noise coming out of the woofers that was not coming out before with the JDS Atom. I will try in some months to modify the AMS to make it XLR and hopefully this will remove the woofers' noise.

I must say that maybe it is better to buy the A90 if you want to do a balanced setup (and given that, I cannot understand why @JohnYang1997 preferred to make the RCA out measuring better than the XLR. I know that one of the two will always measure a bit worse, but since this is a product aimed for people that want to go full balanced, why the XLR worse than RCA, even if just by a negligible amount?) and go for the L30 if you have an unbalanced setup, since the performances will more or less be similar to the A90 with a lot less money

 

[JohnYang1997]

Yes I can confirm that I tried the A90 connected through both XLR and RCA to the DAC, and to the Anaview AMS0100 as speaker amp through RCA, and there is a little noise coming out of the woofers that was not coming out before with the JDS Atom. I will try in some months to modify the AMS to make it XLR and hopefully this will remove the woofers' noise.

I must say that maybe it is better to buy the A90 if you want to do a balanced setup (and given that, I cannot understand why @JohnYang1997 preferred to make the RCA out measuring better than the XLR. I know that one of the two will always measure a bit worse, but since this is a product aimed for people that want to go full balanced, why the XLR worse than RCA, even if just by a negligible amount?) and go for the L30 if you have an unbalanced setup, since the performances will more or less be similar to the A90 with a lot less money

-Why RCA input measures better?
A: According to Douglas Self's small signal book, balanced connection is inherently more noisy than unbalanced. The benefits of balanced connection is to avoid interference and ground related issue. In ideal condition, RCA input outperforms XLR in.

-Atom didn't have the noise with RCA but A90 has
A: This is very common with a device that has ground connection. This is classic ground related issue. It's not severe enough to be a ground loop but it's there. Also SMPS needs ground return path to work the best because of leakage current. So RCA connection + separated two grounding route is the cause of the issue. XLR is recommended.

-Will L30 have the issue?
A: Similar to Atom, external transformer based AC power supply is used. This eliminates SMPS's leakage current, traditional transformer magnetic interference, also the earth connection. Hence its unbalanced connection be more robust to interfacing to other devices. Less than 100nV(70nV typical) of mains noise is what you will get.

 

[jhm]

My A90 should arrive here (UK) today (Friday) which is pretty good when I ordered it from HifiStudio79 in the Netherlands only two days ago.

HifiStudio79 currently have stock and charged me EUR 10 for delivery by GLS (which uses Parcelforce in the UK) compared to Audiophonics in France which is showing 15-25 days for stock and I would have paid around EUR 30 to them for delivery by TNT.

(Apart from ordering the A90 from one and the D90 from the other, I have no connection with either company.)

@JohnYang1997 - thank you very much for what I'm sure will be a great headphone amp and also for your help in this forum. Just one point to mention - I can't find a PDF of the A90 user's manual at http://www.tpdz.net/download but I did find it on the Audiophonics website. Unless I missed it, maybe someone can add it to tpdz.net when convenient. Thanks again!

 

[mt196]

-Why RCA input measures better?
A: According to Douglas Self's small signal book, balanced connection is inherently more noisy than unbalanced. The benefits of balanced connection is to avoid interference and ground related issue. In ideal condition, RCA input outperforms XLR in.

-Atom didn't have the noise with RCA but A90 has
A: This is very common with a device that has ground connection. This is classic ground related issue. It's not severe enough to be a ground loop but it's there. Also SMPS needs ground return path to work the best because of leakage current. So RCA connection + separated two grounding route is the cause of the issue. XLR is recommended.

-Will L30 have the issue?
A: Similar to Atom, external transformer based AC power supply is used. This eliminates SMPS's leakage current, traditional transformer magnetic interference, also the earth connection. Hence its unbalanced connection be more robust to interfacing to other devices. Less than 100nV(70nV typical) of mains noise is what you will get.

Nice, thanks for the reply.
At this point I am wondering why some devices have better XLR measurements than RCA then, if the RCA should always be a bit better in ideal conditions.

The last two parts instead confirm the reasoning I did in the previous post, the A90 is better when used in XLR, if you want to go RCA, then a L30 or A50s will make the user save some money

 

[Veri]

-Atom didn't have the noise with RCA but A90 has
A: This is very common with a device that has ground connection. This is classic ground related issue. It's not severe enough to be a ground loop but it's there. Also SMPS needs ground return path to work the best because of leakage current. So RCA connection + separated two grounding route is the cause of the issue. XLR is recommended.

So even when using RCA, having XLR connected to something to provide a grounding return path is ideal. That's a good tip.

 

[JohnYang1997]

So even when using RCA, having XLR connected to something to provide a grounding return path is ideal. That's a good tip.

I'm not 100% sure on that.
A more traditional alternative solution is to lift the earth on the a90 power supply and use very short RCA cable to connect between the two devices and use the earth connection from the first device for earth.
If the dac is usb powered, then the usb cable can play a role too, shorter the better.
In the end, XLR connection is just hassle free for these devices.

 

[mt196]

I'm not 100% sure on that.
A more traditional alternative solution is to lift the earth on the a90 power supply and use very short RCA cable to connect between the two devices and use the earth connection from the first device for earth.
If the dac is usb powered, then the usb cable can play a role too, shorter the better.
In the end, XLR connection is just hassle free for these devices.

What does "lift the earth" mean? Not to connect it?

 

[reiserfs]

Just want to clarify. Is there noise while D50 connected via USB? Have you tried optical? Is there still noise?

USB. The D50 is USB powered. So you'll have noise regardless of the input or not if you have a especially clean USB port.

 

[JohnYang1997]

What does "lift the earth" mean? Not to connect it?

Yes. But I would not recommend doing it.

 

[bboris77]

What does "lift the earth" mean? Not to connect it?

The only semi-safe way I know of doing this would be to use the Hum-X from EB Tech. Schiit actually recommends this device in their user manuals. The absolute safest way to break the ground loop without affecting the integrity of the signal would be to buy an USB isolator.

 

[mt196]

The only semi-safe way I know of doing this would be to use the Hum-X from EB Tech. Schiit actually recommends this device in their user manuals. The absolute safest way to break the ground loop without affecting the integrity of the signal would be to buy an USB isolator.

I break a great part of a ground loop by using electrical tape on the GND pin of the USB coming from my PC (my DAC does not require USB power to function). That + XLRs = problem solved. Before the taping, I could hear hiss every time I moved the mouse and it increased when the GPU was under stress (Vega 64)

O.T: I don't know why, but the Topping D30 and JDS OL DAC based on AKM chip didn't have problems connected to the same PC, while D50, SGD1, and LA-QXD1 (based on ESS chips) all showed the hiss without the electrical tape on the GND pin of the USB

 

[bboris77]

I break a great part of a ground loop by using electrical tape on the GND pin of the USB coming from my PC (my DAC does not require USB power to function). That + XLRs = problem solved. Before the taping, I could hear hiss every time I moved the mouse and it increased when the GPU was under stress (Vega 64)

O.T: I don't know why, but the Topping D30 based on AKM chip didn't have problems connected to the same PC, while D50, SGD1, and LA-QXD1 (based on ESS chips) all showed the hiss without the electrical tape on the GND pin of the USB

Strange that the D50 exhibited this because it does not have the 3-pin ground power plug. Perhaps you plugged it directly into the PC for the power source. This means that it would then have 2 USB connections to the PC - data and power. Perhaps that created an additional ground loop.