Topping A90 Headphone Amplifier Review

[JohnYang1997]

@JohnYang1997 I did a thing! I graphed the [V vs Z] and [P vs Z] data of the A90 in both balanced and single-ended output. I graphed log(Z) on the x-axis and Vrms and log(Prms) on the y-axis. The two horizontal lines are the output Vrms for balanced and single-ended; the diagonal lines are the ideal (infinite current) power curves based on the output voltage.

I wanted to see how much power the A90 could supply before choking the current supply, and how it stands up against the Benchmark HPA4. Seems like the HPA4 outdoes the A90 in terms of current supply (visible as voltage droop in calculated Vrms at low impedance loads on balanced output), but the A90 wins in pushing more power (A90's 7.6 Wrms vs HPA4's 6 Wrms @ 16 ohms) with that ridiculously high output voltage!

This is confusing....

 

[JohnYang1997]

What current supply do you mean anyway?

 

[ayane]

What current supply do you mean anyway?

I meant total current at the output. I rephrased it to be less confusing. I was trying to make a graph like this one from RME's ADI-2 manual:

 

[JohnYang1997]

I meant total current at the output. I rephrased it to be less confusing

What's the lowest impedance you have tested? With 16 ohm a90 is pushing more power hence higher current.
You can't use voltage drop to calculate anything, really.

 

[ayane]

What's the lowest impedance you have tested? With 16 ohm a90 is pushing more power hence higher current.
You can't use voltage drop to calculate anything, really.

I actually used the data on the product page; I'm calculating V from power and impedance

 

[JohnYang1997]

I actually used the data on the product page; I'm calculating V from power and impedance

"Seems like the HPA4 outdoes the A90 in terms of supplied current"

Where did this happen?

Btw WattRMS is not a thing. Just use Watts.

 

[ayane]

"Seems like the HPA4 outdoes the A90 in terms of supplied current"

Where did this happen?

Btw WattRMS is not a thing. Just use Watts.

Pretty sure I said abbreviated RMS power as Prms not Wrms lol. Also, you're right; the A90 can supply more current. I forgot that the HPA4 doesn't do differential output on XLR. I'm confusing myself too. HPA4's specs say it can do 1.5 amps max current; how much can the A90 do?

 

[JohnYang1997]

@JohnYang1997 (A90's 7.6 Wrms vs HPA4's 6 Wrms @ 16 ohms) with that ridiculously high output voltage!

Ehh?

 

[ayane]

Ehh?

Oh crap, I fixed it. I need to go to sleep...

 

[JohnYang1997]

Pretty sure I said abbreviated RMS power as Prms not Wrms lol. Also, you're right; the A90 can supply more current. I forgot that the HPA4 doesn't do differential output on XLR. I'm confusing myself too. HPA4's specs say it can do 1.5 amps max current; how much can the A90 do?

That's why I asked the lowest impedance you tested.
There are a few things.
Firstly, A90 can do 1.4A.

So it looks it's lower than opa564 which has 1.5A.

At 16ohm the peak current is a bit less than 1A.

Then the THX888 module supposedly limits the current output to 1A for safety.

They don't have external heatsink. From my test it can rush to 80C at the opa564 case when running 16ohm power sweep.

So here I don't think Benchmark would unleash the limit back to 1.5A.

Thus A90 should supply higher current.

 

[ayane]

That's why I asked the lowest impedance you tested.
There are a few things.
Firstly, A90 can do 1.4A.

So it looks it's lower than opa564 which has 1.5A.

At 16ohm the peak current is a bit less than 1A.

Then the THX888 module supposedly limits the current output to 1A for safety.

They don't have external heatsink. From my test it can rush to 80C at the opa564 case when running 16ohm power sweep.

So here I don't think Benchmark would unleash the limit back to 1.5A.

Thus A90 should supply higher current.

Thanks for the detailed info!

 

[bahamot]

Found this post on the other forum, the person cross posted it on headfi too.

Design issues:

1. It is reported that device is very sensitive to ground loops. For engineers it is explained easily. There are two Meanwell SMPS modules. Each module has a built-in safety and EMI circuit (as required by legislation). It makes two, not one source of ground loops. Users have to use XLR connection, as RCA is not working clean.

2. There is no ESD protectors on inputs. It is expected number catastrophic incidents during a lifetime.

3. A reason for using underrated pot is known. They couldn’t find small value of a resistor of a bigger size. Their goal was as usual with Topping to wind up to the top of SINAD, no matter what. The result is a non-standard input impedance, so owners of (by example) Denafrips or SMSL M400 DACs and other devices with non-standard outputs are out of luck. A90 will not pair with these sources.

4. Nested feedback (again). Not a first device of that kind. THX was before and my search revealed that Tom Christiansen tried it already, it is out of production as speak

 

[JohnYang1997]

Found this post on the other forum, the person cross posted it on headfi too.

1, Correct for being sensitive to ground loop but complete bs explanation.
2, Input opamps have built in ESD protection. Never seen an opamp died from ESD.
3, Low input impedance for XLR input is because of the differential amplifier needs to have low noise. Pot is after the opamps. Input impedance for rca input is 10k.
4, As stated before, it's deliberately named to be it. No one has ever thought it to be anything "new". However same as I being challenged by THX designer before, I'll challenge him to make something better. Before that, big mouth.

 

[JohnYang1997]

Well, on something else.
He was aware of Tom. But fail to see that Tom was using the same/similar SMPS configuration for neurochrome HP-1 and TCA HPA-1.
The more sensitive RCA input is because I tried to bring down the power supply noise as much as possible on AP measurements. Yet AP failed to show that in non ideal system, it can be bigger problem. It's the PCB layout, wiring is designed that way not because of the smps module. But to me, D90 is the recommended paring DAC. There's also no issue using XLR input with other dacs and all outputs.
L30 which is fully unbalanced amp is of course needs to be considered to be able to perform in any system. So L30 will have no issue.

 

[Ddd]

Did you take any photo's? It would be nice to see an annotated image of the problem area pointing out specifically what you had to fix - before and after.

Perhaps water under the bridge now, but next time anyone debugs a problem like this caused by manufacturing (or shipping) it would be nice to see the problem before it's fixed, solution, and the fixed problem afterwards.

Thanks for letting us know, I may take a look inside... any more tips on what I am looking for?

Hello, sorry for the so late reply, just saw your message when i re-read the thread, My deepest apologies

But sadly i did not take any picture that time as i wasn't expecting any repair or what,
I just open the back plate of A90 only (i failed to take out the whole A90 out from the case even all screws including bottom screw remove, it's so tight it would'nt slide at all, anyone know how? or is it normal that it is so tighten fitted inside?)
and push tighten all the power supply wire internally hoping to see if it helps.

And it did.. for a while, then the RCA noise returns again.

I guess the A90 are just good to use XLR input only with other grounded device like Dx7 pro D90,
while its RCA input i think are designed meant for use with non-grounded device like their D50s D30.

And i did tried and lift one of the ground connection from either D90 or A90, and the noise problem from the RCA is immediately solved
But since i'm been told lifting the ground is not recommended and grounding it ensured best performance of the internal SMPS so i just reconnect everything back to normal

So i guess the RCA noise came from two grounded device connected together and cause the ground loop noise?
End of the day, i just use the XLR only then.

 

[Ddd]

Update:
Got my replacement A90 in today. First one had a weird feel of the wipers in the pot. A minor issue but Apos replaced the amp absolutely no questions asked. Shockingly good service. Popped it open to confirm topping did indeed use thermal paste on the heatsink this time, applied well too. I tossed on some pads in it's place as that's what I've got on hand. Requires a 10inch key wrench to open so, beware there. Really nice build quality... Here are some PCB shots for the curious.
View attachment 68288
View attachment 68289
View attachment 68290

By the way, @JohnYang1997
Hope you can solve my problem,
Sorry is gonna be a bit long below
But i wish i can explain the whole situation clear for you so maybe you can clear my doubts.

My A90 already experienced fuse burnt for many times when turning on the device using back switch,
I see the stock fuse was using F500mAL 250V Fast Blow
So when i first blew the first fuse, i replaced with the spare one, and it blew less than 24 hours

So i ordered multiple Fast blow 500ma and replaced ( Fuse Eatonn Bussman BK GMA 500R)
After installed. it blew all right away.

Then i come research for the internals image of A90 here and i found it use the Meanwell IRM-20-15 which can be seen in above user posted image.

Input
100 240 VAC 0.6A 50/60Hz
277VAC 0.3A 50/60Hz

My country is 240V, and i checked my house range between 239V - 245V at different days, hours
Also tested at other different location as well

So my question here is
if following above Power supply specs 240VAC 0.6A, used by A90
should'nt the fuse used be at least or higher than 0.6A?
while A90 currently using 500mA fuse?

After blowing many 500mA
i contacted my Seller and he contacted with Topping technician (i hope it is)
The technician recommended me to try higher ampere rating fuse up to Fast blow 1A as they suspected voltage surge or spike.
and said A90 is safe up to T1.0A slow blow fuse.

So i tried 750mA and 1A fast blow
And it blew all the way up to 1A,
750mA blew straight away after installed and turn on,
1A blew after 10 hours of use, which the A90 still usable like normal before i turn off and went out for meal and came back turn on again and it blew.

Then now the technician recommend me to try T0.5A Time Delay or Slow Blow fuse,
currently still using and have not experience any blow yet.

and is not just me who experienced it, another customer of my seller experienced the same issue as me as well after getting the A90 and only used few times and not even burnt in. My unit i used more than 150hours before i experienced this issues.

Do you think my unit is fine? Tried to claim warranty due to worried some internal parts are faulty but they would'nt accept unless the A90 completely failed to turn on even after changing higher fuse ampere rating.

So the two most important concern is
1) Is the fuse used in the A90 (500mA) a bit under ? since the Meanwell PS input rated up to 0.6A at 100 240VAC.
2) The stock fuse is Fast Blow/Fast Acting type, should it be standard install with Slow Blow/Time Delay type if A90 is sensitive like that?

 

[JohnYang1997]

By the way, @JohnYang1997
Hope you can solve my problem,
Sorry is gonna be a bit long below
But i wish i can explain the whole situation clear for you so maybe you can clear my doubts.

My A90 already experienced fuse burnt for many times when turning on the device using back switch,
I see the stock fuse was using F500mAL 250V Fast Blow
So when i first blew the first fuse, i replaced with the spare one, and it blew less than 24 hours

So i ordered multiple Fast blow 500ma and replaced ( Fuse Eatonn Bussman BK GMA 500R)
After installed. it blew all right away.

Then i come research for the internals image of A90 here and i found it use the Meanwell IRM-20-15 which can be seen in above user posted image.

Input
100 240 VAC 0.6A 50/60Hz
277VAC 0.3A 50/60Hz

My country is 240V, and i checked my house range between 239V - 245V at different days, hours
Also tested at other different location as well

So my question here is
if following above Power supply specs 240VAC 0.6A, used by A90
should'nt the fuse used be at least or higher than 0.6A?
while A90 currently using 500mA fuse?

After blowing many 500mA
i contacted my Seller and he contacted with Topping technician (i hope it is)
The technician recommended me to try higher ampere rating fuse up to Fast blow 1A as they suspected voltage surge or spike.
and said A90 is safe up to T1.0A slow blow fuse.

So i tried 750mA and 1A fast blow
And it blew all the way up to 1A,
750mA blew straight away after installed and turn on,
1A blew after 10 hours of use, which the A90 still usable like normal before i turn off and went out for meal and came back turn on again and it blew.

Then now the technician recommend me to try T0.5A Time Delay or Slow Blow fuse,
currently still using and have not experience any blow yet.

and is not just me who experienced it, another customer of my seller experienced the same issue as me as well after getting the A90 and only used few times and not even burnt in. My unit i used more than 150hours before i experienced this issues.

Do you think my unit is fine? Tried to claim warranty due to worried some internal parts are faulty but they would'nt accept unless the A90 completely failed to turn on even after changing higher fuse ampere rating.

So the two most important concern is
1) Is the fuse used in the A90 (500mA) a bit under ? since the Meanwell PS input rated up to 0.6A at 100 240VAC.
2) The stock fuse is Fast Blow/Fast Acting type, should it be standard install with Slow Blow/Time Delay type if A90 is sensitive like that?

I suggest getting a new unit.
There is probably only one A90 had blew fuse since the beginning of the development to today. Usually A90 draws 30mA. If at anytime it would blow(for normal use) it's at start up. Blowing in normal operation is not normal.
And reading your previous post it seems you earth connection is working. But you have to make sure.

 

[Ddd]

I suggest getting a new unit.
There is probably only one A90 had blew fuse since the beginning of the development to today. Usually A90 draws 30mA. If at anytime it would blow(for normal use) it's at start up. Blowing in normal operation is not normal.
And reading your previous post it seems you earth connection is working. But you have to make sure.

Ya i measured the A90 operating current only 30mA as well.
When turning on, highest i can captured is only max 200mA.
Still far from 500mA for the fuse to blow.

And the fuse blew during turning on or start up when turning on A90 own switch
After a successful start up, it won't blow.
But if it fails to start up, means the fuse blew again.

Ya i had checked, the Earth connection is connected, my seller also handed me which device to check for fault in power supply and wall socket.
no reverse live/neutral issues,
Earth connection is present.

I manually checked the Earth by using multimeter to test continuity by touching metal part of A90 to my power socket Earth and the link is present as well.

I wish i can claim and exchange a new unit too to clear my future worried,
but according to my seller, the "topping technician" won't proceed any exchange or warranty until the unit completely not bootable.
And keep suggesting me to change different fuse ampere rating.

Any other suggestions you have?

 

[JohnYang1997]

Ya i measured the A90 operating current only 30mA as well.
When turning on, highest i can captured is only max 200mA.
Still far from 500mA for the fuse to blow.

And the fuse blew during turning on or start up when turning on A90 own switch
After a successful start up, it won't blow.
But if it fails to start up, means the fuse blew again.

Ya i had checked, the Earth connection is connected, my seller also handed me which device to check for fault in power supply and wall socket.
no reverse live/neutral issues,
Earth connection is present.

I manually checked the Earth by using multimeter to test continuity by touching metal part of A90 to my power socket Earth and the link is present as well.

I wish i can claim and exchange a new unit too to clear my future worried,
but according to my seller, the "topping technician" won't proceed any exchange or warranty until the unit completely not bootable.
And keep suggesting me to change different fuse ampere rating.

Any other suggestions you have?

The whole thing lines up except for fusing being blown. I think you can request to get a new unit from the seller and count it as faulty unit since this is obviously not normal.

If that doesn't work out, send back to Topping and we will test it and fix it/replace it. May take time but that the best that we can do if the seller route doesn't work.

 

[Ddd]

The whole thing lines up except for fusing being blown. I think you can request to get a new unit from the seller and count it as faulty unit since this is obviously not normal.

If that doesn't work out, send back to Topping and we will test it and fix it/replace it. May take time but that the best that we can do if the seller route doesn't work.

Alright, thanks for the advice. I'll try to get through my seller. If i can't i'll just have to send back to Topping at GuangZhou according to http://www.tpdz.net/contact ?