Topping A70 Pro HP & Preamp Review

[lollonet]

Have you tried flicking the grounding switch at the back?

Already tried that. Nothing changes.
To be noted that the spurious sound is a little louder on the right channel while on the left is barely audible

 

[Choleryk;-)]

Welcome back. I sent the Topping A70 Pro to the official distributor. I found out that the amplifier would be replaced with a new one. After a dozen or so days, I received the parcel. Of course, I immediately connected the amplifier, everything is fine, it works, I am satisfied and suddenly the message "HP DC abnormal" appears. My anger boiled over and my word reached the pavement :-( Fortunately, after turning the potentiometer, the message disappeared, but the bad taste remained. I really don't know whether it is worth using a device that causes unpleasant surprises and whether after some time the message appears again, and Topping will stop working.
I have a question whether any of my colleagues had a similar case where after replacing the A70 it stopped working?
I use Google translator, I hope the post will be readable ;-)

 

[Great Sun Jester]

Witamy spowrotem. Wysłałem Topping A70 Pro do oficjalnego dystrybutora. Dowiedziałem się, że wzmacniacz został wprowadzony. Po otrzymaniu otrzymanego nowego egzemplarza. Oczywiście od razu podłączyłem wzmacniacz, wszystko jest w porządku, działa, jestem usatysfakcjonowany i nagle pojawia się komunikat „HP DC abnormal”. Moja złość się zenitu, a moje słowo dotarło do bruku :-( Na szczęście po przekręceniu potencjometru komunikat zniknął, ale niesmak pozostał. Naprawdę nie wiem, czy warto używać urządzenia, które sprawia niemiłe niespodzianki i czy po czasie pojawia się ponownie i Topping przestanie działać.
Mam pytanie, czy ktoś z podłączonym do forum miał urządzenie, które jest urządzeniem do urządzenia A70 Pro do dystrybutora?
Korzystam z tłumacza google, mam nadzieję, że post będzie czytelny ;-)

Translated:
Welcome back. I sent the Topping A70 Pro to the official distributor. I found out that the booster had been introduced. After receiving your new copy. Of course, I immediately connected the amplifier, everything is fine, it works, I am satisfied and suddenly the message "HP DC abnormal" appears. My anger boiled over and my word reached the pavement :-( Fortunately, after turning the potentiometer, the message disappeared, but the bad taste remained. I really don't know whether it is worth using a device that gives unpleasant surprises and whether after some time it reappears and Topping will stop working .
I have a question, has anyone connected to the forum had a device that is a device for the A70 Pro device for the distributor?
I use Google translator, I hope the post will be readable ;-)

 

[Roland68]

Hi People
Does anyone owning an A70 pro noticed that there is some low level signal (volume independent) at the SE pre out even when only HPA is selected?
I found that on a first unit I bought. I sent it back complaining about that but the second one I got has the same small issue.
Thanks

This is called crosstalk and you will find it on many devices if you look for it or check it.
It is also due to these small devices, where there is no possibility of keeping components and conductor tracks at a distance.

Have you tried flicking the grounding switch at the back?

Ground Lift cannot help at this point. It is intended to separate or raise the mass in the event of mass disturbances.

 

[Sokel]

Welcome back. I sent the Topping A70 Pro to the official distributor. I found out that the amplifier would be replaced with a new one. After a dozen or so days, I received the parcel. Of course, I immediately connected the amplifier, everything is fine, it works, I am satisfied and suddenly the message "HP DC abnormal" appears. My anger boiled over and my word reached the pavement :-( Fortunately, after turning the potentiometer, the message disappeared, but the bad taste remained. I really don't know whether it is worth using a device that causes unpleasant surprises and whether after some time the message appears again, and Topping will stop working.
I have a question whether any of my colleagues had a similar case where after replacing the A70 it stopped working?
I use Google translator, I hope the post will be readable ;-)

Is is doing it with a specific song?
Can you remember which?

I'm curious to see what is got down low.

 

[Music1969]

I'm curious to see what is got down low.

Likewise, and if it is a streaming service, details of which and quality setting etc

Could be different between Spotify vs Apple Music vs Tidal vs Qobuz

 

[Roland68]

Is is doing it with a specific song?
Can you remember which?

I'm curious to see what is got down low.

That would interest me too. A song like that should cause the protective circuit to respond not only with other HPAs, but also with normal amplifiers.

 

[mike7877]

I have the A70 Pro and L70. Regarding power output... Although you might be inclined to think the A70 Pro would be higher power output into low impedances than the L70, depending on the impedance, this might not be the case.

I made myself these handy contraptions for listening to music at low levels (for many purposes: to save electronics, save power, experiment with sound here and there)

Yes I took the picture in the dark lol


You can probably guess what low impedance drivers I was talking about driving now that you've seen the adapters.

Speakers?! Yes, speakers!! The speakers are a very simple load for speakers, though. First, it's a two way. Then, the two-way is in a sealed box. They're definitely more reactive than a typical headphone, but soo so much less than a ported speaker. If Headphones are 10% difficulty to deal with, and ported speakers are 100%, these speakers are probably 25%, maybe even 20 because they're 8 ohms and only dip to 6.5.

So, yes, I know that the A70 Pro's 17 watt rating is into 16 ohms. The amount of current flowing for 17 watts in that case should be a bit over 1A.
It's 1.031A to be almost exact.

Usually.. what I'd expect from an amp with a <0.1 ohm output impedance and a 17 watt rating into 16 ohms, is exactly half of 17 watts into 8 ohms, or 8.5 watts...

And since the speaker does hit 6.5 ohms at a point, that 8.5 watts may turn into 6.9 watts if a lot of the music power is in the upper-upper bass.

Buuuuuuut

That's not what we get here. We get a watt peak, maybe. And I said peak!

I don't know if there's a complex analysis going on by a chip, and anything under 10 ohms is being severely current limited on the A70 Pro, or what exactly is happening. I do have some 5W resistors of the 2.2 ohm variety... I'll have to start putting them in series with the speakers (I only have a few) to see if I can get a tonne more voltage with a slightly higher impedance. If I find out Topping intentionally gimped the A70 Pro so that the LA90 was needed for any serious listening through speakers, I'm gonna feel betrayed - I've bought so much Topping stuff! Generally I love it, but the A70 Pro sounding ~2dB quieter before cutting out, not good
(I really hate arbitrary lines)

 

[Music1969]

and anything under 10 ohms is being severely current limited on the A70 Pro

This is a headphone amp. That would be normal, no ?

Amir even said goes into protection mode below

 

[Roland68]

I have the A70 Pro and L70. Regarding power output... Although you might be inclined to think the A70 Pro would be higher power output into low impedances than the L70, depending on the impedance, this might not be the case.

I made myself these handy contraptions for listening to music at low levels (for many purposes: to save electronics, save power, experiment with sound here and there)
View attachment 365424
Yes I took the picture in the dark lol


You can probably guess what low impedance drivers I was talking about driving now that you've seen the adapters.

Speakers?! Yes, speakers!! The speakers are a very simple load for speakers, though. First, it's a two way. Then, the two-way is in a sealed box. They're definitely more reactive than a typical headphone, but soo so much less than a ported speaker. If Headphones are 10% difficulty to deal with, and ported speakers are 100%, these speakers are probably 25%, maybe even 20 because they're 8 ohms and only dip to 6.5.

So, yes, I know that the A70 Pro's 17 watt rating is into 16 ohms. The amount of current flowing for 17 watts in that case should be a bit over 1A.
It's 1.031A to be almost exact.

Usually.. what I'd expect from an amp with a <0.1 ohm output impedance and a 17 watt rating into 16 ohms, is exactly half of 17 watts into 8 ohms, or 8.5 watts...

And since the speaker does hit 6.5 ohms at a point, that 8.5 watts may turn into 6.9 watts if a lot of the music power is in the upper-upper bass.

Buuuuuuut

That's not what we get here. We get a watt peak, maybe. And I said peak!

I don't know if there's a complex analysis going on by a chip, and anything under 10 ohms is being severely current limited on the A70 Pro, or what exactly is happening. I do have some 5W resistors of the 2.2 ohm variety... I'll have to start putting them in series with the speakers (I only have a few) to see if I can get a tonne more voltage with a slightly higher impedance. If I find out Topping intentionally gimped the A70 Pro so that the LA90 was needed for any serious listening through speakers, I'm gonna feel betrayed - I've bought so much Topping stuff! Generally I love it, but the A70 Pro sounding ~2dB quieter before cutting out, not good
(I really hate arbitrary lines)

Very bad topping, really. A headphone amplifier that can't drive speakers...
The A70 Pro is a chip amplifier, probably 4 x tpa6120, and has really been pushed to the extreme in terms of development and optimized for headphones.

For what you have in mind, you might want to take a look at the Whammy from the diyaudio forum. You can use it for small speakers, but you may have to adjust a few values.

 

[mike7877]

Very bad topping, really. A headphone amplifier that can't drive speakers...
The A70 Pro is a chip amplifier, probably 4 x tpa6120, and has really been pushed to the extreme in terms of development and optimized for headphones.

For what you have in mind, you might want to take a look at the Whammy from the diyaudio forum. You can use it for small speakers, but you may have to adjust a few values.

This does have implications for some very high end headphones which have impedances in the 9 to 11 ohm range. In my post I said I'd be trying to figure out at which impedance this overly aggressive current limiting begins by adding a couple of my 2.2 ohm resistors in series with the speakers. There's more information in there too, like I have an LA90 Discrete hooked up to the A70 Pro to drive the speakers / the idea is to save power-on hours of the LA90 Discrete when doing things at lower levels for extended periods of time. The A70 Pro, in its current state, is just a small bit too quiet (ie. sometimes the protection kicks in, and that means restarting the whole device - the L70 just clips and keeps going unless the signal really rams into the rail. The L70 does sound worse than the A70 Pro when driving the speakers directly though, so I can't just use it instead. Allegedly both have output impedances of ~0.1 ohm, so I can't fathom the reason, but the L70 seems to have less control over the bass, a bit too much of it in the 40-55Hz range, and the treble has a... characteristic to it. Like there's a bit too much of it and it's slightly smeared. With my 37 ohm SRH-1440s this isn't a problem, nor my 32 ohm earbuds, or any of my higher impedance cans. The L70 and A70 Pro both sound nearly identical as preamps, too.

With that, I don't know if I believe the A70 Pro uses the tpa6120s. The L70 appers to, as well as a couple of their other devices (in topping's own photos), but I'm not sure of the A70 Pro. It's advertised as having their newest and best iteration of their 3 stage "nested feedback composite amplifier", they named it "T'ang-ku-la". The capacitor bank immediately beside the output transistors (whatever they are) is absolutely massive! I forgot, here's a pic lol

 

[mike7877]

This is a headphone amp. That would be normal, no ?

Amir even said goes into protection mode below

Well, not necessarily. The amplifier has a current output capability of 1.03A. It's probably designed to 1.000A The amplifier also has an output impedance of 0.1 ohms. With a good speaker cable and good connections to the 4 pin xlr, you could get a damping factor of 40-50 with the 8 ohm nominal, 6.5 ohm minimum speakers, and 8.5 watts of power. Protection could have been set to 1 ohm or 2 ohms, even 4 ohms, instead of 12 ohms (or whatever it is precisely), and with 8 ohm speakers connected, the amp wouldn't be working any harder making those 8.5 watts than it would be making 17w into 16 ohms.

Basically it's a design choice which I don't understand. I personally wouldn't do it - what if you've got two people listening in parallel, three? There's just no reason to set it at 12 and not 3, especially when I've seen headphones one or two ohms away from single digits (I saw a planar 11 ohm - I'm sure they go lower. Imagine 2x 9 ohm in parallel? The A70 Pro could do it, no problem, give them over 2 watts each, but not the way the protection is configured

And you're right about Amir's chart - I could extrapolate it and it takes you to my position exactly

 

[Music1969]

The amplifier has a current output capability of 1.03A. It's probably designed to 1.000A

Yes and that is fine because it is a headphone amp... Wanting 10A for a headphone amp is wowza !

You said: "and anything under 10 ohms is being severely current limited on the A70 Pro"

A70Pro is 100% fit for purpose

 

[Roland68]

This does have implications for some very high end headphones which have impedances in the 9 to 11 ohm range. In my post I said I'd be trying to figure out at which impedance this overly aggressive current limiting begins by adding a couple of my 2.2 ohm resistors in series with the speakers. There's more information in there too, like I have an LA90 Discrete hooked up to the A70 Pro to drive the speakers / the idea is to save power-on hours of the LA90 Discrete when doing things at lower levels for extended periods of time. The A70 Pro, in its current state, is just a small bit too quiet (ie. sometimes the protection kicks in, and that means restarting the whole device - the L70 just clips and keeps going unless the signal really rams into the rail. The L70 does sound worse than the A70 Pro when driving the speakers directly though, so I can't just use it instead. Allegedly both have output impedances of ~0.1 ohm, so I can't fathom the reason, but the L70 seems to have less control over the bass, a bit too much of it in the 40-55Hz range, and the treble has a... characteristic to it. Like there's a bit too much of it and it's slightly smeared. With my 37 ohm SRH-1440s this isn't a problem, nor my 32 ohm earbuds, or any of my higher impedance cans. The L70 and A70 Pro both sound nearly identical as preamps, too.

With that, I don't know if I believe the A70 Pro uses the tpa6120s. The L70 appers to, as well as a couple of their other devices (in topping's own photos), but I'm not sure of the A70 Pro. It's advertised as having their newest and best iteration of their 3 stage "nested feedback composite amplifier", they named it "T'ang-ku-la". The capacitor bank immediately beside the output transistors (whatever they are) is absolutely massive! I forgot, here's a pic lol

View attachment 365437

To be clear, they are operating these two HPAs outside of their scope and specifications. If a defect occurs, you can only blame yourself and under no circumstances Topping.
In addition, you can operate these devices in a performance range that would not be possible even with Hifiman HE-6/SE, Susvara, etc. Extraordinary wear and tear, as well as defects, cannot be ruled out.

As in the THX HPA circuits, these amplifier chips used are nothing more than more powerful OPAmps. These chips were optimized for 32 - 600 ohms, below this the values and performance become significantly worse. Do you notice anything?
It doesn't matter whether you calculate the values, impose a restriction on topping, or set the protection to 4 ohms, it doesn't change the specification of the components used. And what does the output impedance of 0.1 ohms have to do with it?

And so it's not a design decision from Topping, except that Topping uses very proven components developed for HPA use.
I have never seen the performance that Topping gets out of these components from any other developer/manufacturer.
But feel free to try it yourself.

By the way, there have been countless projects in which someone tried to drive speakers with such OPAmps. And that from really capable engineers and most of them were hired. Either too complex, too expensive, too weak, or too bad.
There have been a few successful attempts with tens of these components in parallel and cascaded, but I am not aware of any successful ones.
Again, no one would stop you, but the L70 and A70 Pro are by far the worst devices for driving speakers.

You need a device with transistors or chip amps for speakers.

 

[Choleryk;-)]

Czy robi to z konkretną piosenką?
Czy pamiętasz które?

Ciekaw jestem, co tam spadło.

Odtworzyłem muzykę z płyty CD, nagraniami była muzyka filmowa Hansa Zimmera z filmu „Gladiator”. Poprzednia blokada miała miejsce prawdopodobnie podczas słuchania płyty Genesis „Nursery Cryme”.

 

[Sokel]

Odtworzyłem muzykę z płyty CD, nagraniami była muzyka filmowa Hansa Zimmera z filmu „Gladiator”. Poprzednia blokada miała miejsce prawdopodobnie podczas słuchania płyty Genesis „Nursery Cryme”.

Ok,if I understood well by the translation:



That's three pieces from Gladiator.
Dynamic stuff,specially the first two with the second showing some signs of extra energy near DC but nothing extraordinary,we have seen much worst.
Now,if that filter interacts so soon (combined with level maybe? ) that needs fixing.

 

[mike7877]

To be clear, they are operating these two HPAs outside of their scope and specifications. If a defect occurs, you can only blame yourself and under no circumstances Topping.
In addition, you can operate these devices in a performance range that would not be possible even with Hifiman HE-6/SE, Susvara, etc. Extraordinary wear and tear, as well as defects, cannot be ruled out.

As in the THX HPA circuits, these amplifier chips used are nothing more than more powerful OPAmps. These chips were optimized for 32 - 600 ohms, below this the values and performance become significantly worse. Do you notice anything?
It doesn't matter whether you calculate the values, impose a restriction on topping, or set the protection to 4 ohms, it doesn't change the specification of the components used. And what does the output impedance of 0.1 ohms have to do with it?

And so it's not a design decision from Topping, except that Topping uses very proven components developed for HPA use.
I have never seen the performance that Topping gets out of these components from any other developer/manufacturer.
But feel free to try it yourself.

By the way, there have been countless projects in which someone tried to drive speakers with such OPAmps. And that from really capable engineers and most of them were hired. Either too complex, too expensive, too weak, or too bad.
There have been a few successful attempts with tens of these components in parallel and cascaded, but I am not aware of any successful ones.
Again, no one would stop you, but the L70 and A70 Pro are by far the worst devices for driving speakers.

You need a device with transistors or chip amps for speakers.

I already went over why it's safe to drive the speakers. Output impedance is low enough and there's a limit on the current - you can't damage the thing driving 6.5 ohms... The manual gives power rating down to 16 ohms and says 8 ohms should be minimum. The average impedance of my speakers is probably over 8 ohms and only dip to 6.5 - I'm not worried.


Are you talking about this chip?

https://www.ti.com/lit/ds/symlink/tpa6120a2.pdf

Topping's pictures do show the L70 using these chips internally, though I've assumed they were used for the second stage of their three stage "NFCA" type amplifier (with the final stage being discrete).

If you look very closely at the TPA6102A's datasheet, you'll see that, even though the chip is purported to be a headphone amplifier good down to 16 ohms /stable to 8 with an exceptional THD+n of -112.5dB, how confident are they about their 16 ohms? Well, the output voltage noise is spec'd at 16 ohms! Excited? Well quit it! When you get to the operating characteristics table on page five (7.6), what... do.. we.... seeeee....

We see that the only 112.5dB THD+n in sight, is at unity gain with the highest supply voltage and a load of with 10k ohms (10,000) {ten-thousand} [100*100] {({in case you missed it})].
"Somebody must've misread 16 for 10k when they typed in the datasheet". While not out of the realm of possibility, the rest of the datasheet doesn't support that ridiculous claim. Why would I make such a ridiculous claim? I don't know! Why would TI state -112.5dB THD+n in the features list of a headphone amplifier while spec'ing noise at 16 ohms right below it? I don't know!! (ie. it makes about as much sense)

Well in that same chart (pictured below) we can see why the 16 ohm figure wasn't chosen for the front page:
If 32 ohms 100mW (1.79V / 56mA) is -90dB THD+n, 16 ohm / 100mW = -87dB THD+n!

-87 just don't have the same ring as -112.5, now does it?

TI is usually better than this...

For fun:
32 ohms 10mW is 0.57V and 18mA
32 ohms 100mW is 1.79V and 56mA
64 ohms 10mW is 0.80V and 13mA
64 ohms 100mW is is 2.53V and 40mA

Interesting how they increase supply to +-15 for the 1.79 and 2.53V output.

I hate when datasheets make you have to guess at the performance of something. All while implying "112.5" TI isn't typically sketchy like this.
Yes, I know the test is standard, but the results of standard test produce results that need some explaining...


Anyway, I have suspicions that the chip only does the second stage of amplification in the A70 Pro especially, maybe the L70, unless people know differently. Another reason they might be using the chip is it's not too expensive and it has a 1300V/us slew rate which really fast, and that quick response might be needed for feedback to work so extremely well

 

[Roland68]

I already went over why it's safe to drive the speakers. Output impedance is low enough and there's a limit on the current - you can't damage the thing driving 6.5 ohms... The manual gives power rating down to 16 ohms and says 8 ohms should be minimum. The average impedance of my speakers is probably over 8 ohms and only dip to 6.5 - I'm not worried.


Are you talking about this chip?

https://www.ti.com/lit/ds/symlink/tpa6120a2.pdf

Topping's pictures do show the L70 using these chips internally, though I've assumed they were used for the second stage of their three stage "NFCA" type amplifier (with the final stage being discrete).

If you look very closely at the TPA6102A's datasheet, you'll see that, even though the chip is purported to be a headphone amplifier good down to 16 ohms /stable to 8 with an exceptional THD+n of -112.5dB, how confident are they about their 16 ohms? Well, the output voltage noise is spec'd at 16 ohms! Excited? Well quit it! When you get to the operating characteristics table on page five (7.6), what... do.. we.... seeeee....
View attachment 365463
We see that the only 112.5dB THD+n in sight, is at unity gain with the highest supply voltage and a load of with 10k ohms (10,000) {ten-thousand} [100*100] {({in case you missed it})].
"Somebody must've misread 16 for 10k when they typed in the datasheet". While not out of the realm of possibility, the rest of the datasheet doesn't support that ridiculous claim. Why would I make such a ridiculous claim? I don't know! Why would TI state -112.5dB THD+n in the features list of a headphone amplifier while spec'ing noise at 16 ohms right below it? I don't know!! (ie. it makes about as much sense)

Well in that same chart (pictured below) we can see why the 16 ohm figure wasn't chosen for the front page:
If 32 ohms 100mW (1.79V / 56mA) is -90dB THD+n, 16 ohm / 100mW = -87dB THD+n!

-87 just don't have the same ring as -112.5, now does it?

TI is usually better than this...
View attachment 365460
For fun:
32 ohms 10mW is 0.57V and 18mA
32 ohms 100mW is 1.79V and 56mA
64 ohms 10mW is 0.80V and 13mA
64 ohms 100mW is is 2.53V and 40mA

Interesting how they increase supply to +-15 for the 1.79 and 2.53V output.

I hate when datasheets make you have to guess at the performance of something. All while implying "112.5" TI isn't typically sketchy like this.
Yes, I know the test is standard, but the results of standard test produce results that need some explaining...


Anyway, I have suspicions that the chip only does the second stage of amplification in the A70 Pro especially, maybe the L70, unless people know differently. Another reason they might be using the chip is it's not too expensive and it has a 1300V/us slew rate which really fast, and that quick response might be needed for feedback to work so extremely well

You can't be helped. TI itself has confirmed the limited performance of 20 and 60 ohms to developers, which is why I find Topping's development achievement all the more remarkable.
But since they know their stuff very well and everything works perfectly, I think everything is fine.

 

[mike7877]

You can't be helped. TI itself has confirmed the limited performance of 20 and 60 ohms to developers, which is why I find Topping's development achievement all the more remarkable.
But since they know their stuff very well and everything works perfectly, I think everything is fine.

I'm completely able to be "helped". Thanks for finally "helping" me by finally telling me the reason you "think" the way you do. Sounded more like an unfounded opinion before

 

[mike7877]

You can't be helped. TI itself has confirmed the limited performance of 20 and 60 ohms to developers, which is why I find Topping's development achievement all the more remarkable.
But since they know their stuff very well and everything works perfectly, I think everything is fine.

I had a huge thing written about what I believe is happening inside these things I meant to re-read and post, but my PC shut down and when I came back to the thread, the forum hadn't saved it like usual :/

They use these chips in their non-headphone amps as well, so they may be the last stage before the headphone stage in headphone amps, and the stage which drives the XLR/RCA in other devices (edit: and in the A70 Pro/L70)could be wrong about this, but if you look at the datasheet, the "-112.5" spec is only for 1k ohm loads. 32 ohm is 87 or something, meaning 16 would be 84 (or worse), and that's definitely not the case in the A70 Pro or L70 - Topping would have to be making the chip perform 36.5dB better than its datasheet, not 7.5, which seems extremely far-fetched