AIYIMA A70 Stereo Amplifier Review

[Jon L]

After 3 years A07 and 8 opamps....the best sounding opamp for me is the phillips NE5532.

https://nl.aliexpress.com/item/1005003071748273.html?spm=a2g0o.order_list.order_list_main.11.50b179d2aR4Ui7&gatewayAdapt=glo2nld

Do you mind briefly sharing which opamps you have compared the Philips NE5532 and how they sounded? TIA

 

[ad_fletch]

For those who have one with the 10A GAN PSU, it seems from Aiyima’s photos (here and on Aliexpress) that it’s actively cooled with a small fan and grille on the DC side. Yet in the review Amir mentions a sealed box. Has it been revised? Also, I wonder whether the fan is triggered by temperature or amperage (it varies in comparable power supplies)…?

(Photo from aliexpress listing as of today)

 

[AIYIMA]

For those who have one with the 10A GAN PSU, it seems from Aiyima’s photos (here and on Aliexpress) that it’s actively cooled with a small fan and grille on the DC side. Yet in the review Amir mentions a sealed box. Has it been revised? Also, I wonder whether the fan is triggered by temperature or amperage (it varies in comparable power supplies)…?

(Photo from aliexpress listing as of today)

The front uses a fan to cool down the temperature, and the fan starts automatically when it reaches a certain temperature.

 

[ad_fletch]

The front uses a fan to cool down the temperature, and the fan starts automatically when it reaches a certain temperature.

Ok thanks - I’ll look forward to tests showing how high the threshold is in practice.

It’s just that some of us are a bit fussy about fan noise (esp higher-pitched noise from tiny fans), and would want it to be silent as much as possible….

 

[bramsat]

Do you mind briefly sharing which opamps you have compared the Philips NE5532 and how they sounded? TIA

The phillips 5532 cost 2.70 euro for one and sounds naturel and musical,the jrc 5532dd,opa 1656, BB5532 and more cheaper 5532 not so good or musical.

https://nl.aliexpress.com/item/1005003071748273.html?spm=a2g0o.order_list.order_list_main.15.684579d2cYyDQs&gatewayAdapt=glo2nld

 

[GPJ]

I just bought an A70 based off of this review. Yet another superb one from Amir. I was also reminded how much money Amir saves me and other people and so headed right over to Paypal and made a donation to ASR as my appreciation. Thank you!

The amp was received and I am quite shocked with the audio. Does it compete with my $1700 Yamaha A-S1100. Unfortunately and fortunately .... YES it does and it has balanced whch the Yammy doesn't. I've always wanted a balanced amp and for $200 driving my Klipsch RF7 II ...... holy cow.

I just ordered the one from Amazon with 48V-5a. Volume gets very loud at 11 o'clock in a 15 x 25 room. I don't want anymore beyond that, 48V-10a is totally unnecessary and the power brick is cool to touch (so good!). I did make a set of nice XLR to TRS (Mogami/Neutrik) to connect to my Topping E70V. This could be my audio purchase of the decade. Just incredible. The only question at this point is reliability, but at this price it hardly matters (unless of course it's a quick death).

Trigger works great with E70V.

 

[spendormania]

Honestly, this amp is disappointing. I wanted to drive an 8 Ohm SB Acoustics 20 cm Bass, but the bass amount and quality was far away from what I got driving it with an Icepower 125ASX. Sealed enclosure, same input and voltage via MiniDSP SHD Preamp, active lowpass at 500 Hz.

No recommendation, in addition the enclosure went very hot. Never experienced this with another Class-D amplifier.

 

[Jon L]

bass amount and quality was far away from what I got driving it with an Icepower 125ASX. Sealed enclosure, same input and voltage via MiniDSP SHD Preamp, active lowpass at 500 Hz.

No recommendation, in addition the enclosure went very hot. Never experienced this with another Class-D amplifier.

I mean, Icepower specialty is bass. Do you think you could try the A70 on some treble/midrange speakers and let us know how it sounds there?

 

[AIYIMA]

Honestly, this amp is disappointing. I wanted to drive an 8 Ohm SB Acoustics 20 cm Bass, but the bass amount and quality was far away from what I got driving it with an Icepower 125ASX. Sealed enclosure, same input and voltage via MiniDSP SHD Preamp, active lowpass at 500 Hz.

No recommendation, in addition the enclosure went very hot. Never experienced this with another Class-D amplifier.

Honestly, this amp is disappointing. I wanted to drive an 8 Ohm SB Acoustics 20 cm Bass, but the bass amount and quality was far away from what I got driving it with an Icepower 125ASX. Sealed enclosure, same input and voltage via MiniDSP SHD Preamp, active lowpass at 500 Hz.

No recommendation, in addition the enclosure went very hot. Never experienced this with another Class-D amplifier.

May I ask if you have a subwoofer connected?

 

[Zoli64]

Hello! I bought an A70 the other day, with the factory 48/5A power supply. I like it so far. I am planning to replace the OPAs with 1656, based on the recommendations of a local hi-fi developer. Has anyone tried it? As I looked at the data, Amirm's measurements, I don't understand why the rca Sinad is so much weaker than the XLR. Also, I reduce the peak performance a little. As I was examining the power supply in my hands, it occurred to me: could the output power cable on the secondary side be too thin? And does it lose some performance because of this? I would like to replace it with a thicker one, but if I understand correctly, the 48V/5A Gan power supply cannot be disassembled without damage? (if it breaks, trash?). I also thought of shortening the output wire to 1/3 the length and putting a new DC plug on it. Nonsense? By the way, what hardware modifications are there that improved the sound quality, SINAD, etc.? Thank you, welcome Zoli, Hungary

 

[AIYIMA]

Hello! I bought an A70 the other day, with the factory 48/5A power supply. I like it so far. I am planning to replace the OPAs with 1656, based on the recommendations of a local hi-fi developer. Has anyone tried it? As I looked at the data, Amirm's measurements, I don't understand why the rca Sinad is so much weaker than the XLR. Also, I reduce the peak performance a little. As I was examining the power supply in my hands, it occurred to me: could the output power cable on the secondary side be too thin? And does it lose some performance because of this? I would like to replace it with a thicker one, but if I understand correctly, the 48V/5A Gan power supply cannot be disassembled without damage? (if it breaks, trash?). I also thought of shortening the output wire to 1/3 the length and putting a new DC plug on it. Nonsense? By the way, what hardware modifications are there that improved the sound quality, SINAD, etc.? Thank you, welcome Zoli, Hungary

Power cables are standard, sustained full-power operation will not be a problem, do not recommend replacing. Improve the sound quality of this can replace some coupling capacitors and op-amps. SINAD this in the current known circuit and device conditions, basically has been to the limit of the value, there is no room to improve the RCA will be weaker than the performance of the XLR, because the XLR is a differential signal, differential signal noise suppression is higher than the single-ended signal of the RCA itself, while the RCA is a single-ended signal, the noise suppression effect will be a little worse, and after the op-amp to do single-ended to differential, after each level of op-amp will produce corresponding noise and noise suppression. RCA is a single-ended signal, the noise suppression effect itself will be poorer, and after the op-amp to do single-ended to differential, after each level of op-amp, it will produce the corresponding noise and distortion.

 

[spendormania]

May I ask if you have a subwoofer connected?

Of course not. This is a classic three-way design, driven by two amps thanks to its passive mid/high crossover, picture attached

@JonL: Hypex Ncore delivers a much more defined bass than the 125 ASX, so I wouldn't point out the bass as a particular strength of the Icepower modules. They are clearly no high end, but very good for solid and inexpensive applications. Therefore it was a surprise for me to see how much the Aiyima lags behind, I didn't expected that.

 

[MCH]

.

Of course not. This is a classic three-way design, driven by two amps thanks to its passive mid/high crossover, picture attached

@JonL: Hypex Ncore delivers a much more defined bass than the 125 ASX, so I wouldn't point out the bass as a particular strength of the Icepower modules. They are clearly no high end, but very good for solid and inexpensive applications. Therefore it was a surprise for me to see how much the Aiyima lags behind, I didn't expected that.

Do your two amps have the same gain? Aiyima XLR input is only 19db

 

[Zoli64]

Power cables are standard, sustained full-power operation will not be a problem, do not recommend replacing. Improve the sound quality of this can replace some coupling capacitors and op-amps. SINAD this in the current known circuit and device conditions, basically has been to the limit of the value, there is no room to improve the RCA will be weaker than the performance of the XLR, because the XLR is a differential signal, differential signal noise suppression is higher than the single-ended signal of the RCA itself, while the RCA is a single-ended signal, the noise suppression effect will be a little worse, and after the op-amp to do single-ended to differential, after each level of op-amp will produce corresponding noise and noise suppression. RCA is a single-ended signal, the noise suppression effect itself will be poorer, and after the op-amp to do single-ended to differential, after each level of op-amp, it will produce the corresponding noise and distortion.

Thanks for the reply. I only thought about it because if the power supply was built in, the approx. Instead of a 0.2 ohm supply line, there would be a significantly lower resistance connection. Another question: does this GAN power supply accept an extra buffer capacitor on its output, for example 4700 uF/63V? Also, doesn't the amplifier itself use some trick so that if I turn on the switch, it starts immediately, without pressing a button? Then the sound system could be started with a switch, I wouldn't have to press the volume button.

Third question: Is the a70's bandwidth guaranteed to be straight from 20Hz to 20kHz? So doesn't the bass drop off sharply around 30-35 Hz? Has anyone replaced the coupling capacitor?

And other question: all my speakers are 4 ohms. Is it possible to push the upper frequency limit on the pffb circuit a little with a simple change so that the high pass does not drop up to 20 kHz? If so, and it is not a serious professional secret, which component is responsible for this and what is the recommended intervention? Thanks to whoever helps me with these questions. Have a nice day!

 

[Rover]

Of course not. This is a classic three-way design, driven by two amps thanks to its passive mid/high crossover, picture attached

@JonL: Hypex Ncore delivers a much more defined bass than the 125 ASX, so I wouldn't point out the bass as a particular strength of the Icepower modules. They are clearly no high end, but very good for solid and inexpensive applications. Therefore it was a surprise for me to see how much the Aiyima lags behind, I didn't expected that.

Check the phasing of the acoustics again.Perhaps this is the reason.

 

[Rotgrunblau]

Honestly, this amp is disappointing. I wanted to drive an 8 Ohm SB Acoustics 20 cm Bass, but the bass amount and quality was far away from what I got driving it with an Icepower 125ASX. Sealed enclosure, same input and voltage via MiniDSP SHD Preamp, active lowpass at 500 Hz.

No recommendation, in addition the enclosure went very hot. Never experienced this with another Class-D amplifier.

That‘s to bad to hear!

I wanted to do something similar, driving a Linkwitz LxSub with this amp were each channel would drive a Seas L26ROY in an open baffle design.

My first try was powering this with a Topping PA5 II which was basically impossible even at super low volumes, power was just not sufficient. Which also surprised me, since just from the specs it’s really similar to my Alesis RA150 which worked perfect.

Too bad to hear that this also happens with this one. Maybe it’s something TPA3255 related?

@AIYIMA did you do some tests with powering a passive (dual) subwoofer with this amplifier? Just from the data sheet it sounds super promising so for me it’s really unfortunate to hear that there might be some problems regarding this kind of setup.

 

[spendormania]

Perhaps we should accept that a profound, defined and beefy bass response need some effort from the amplifier too. It's a reason that a lot of pros and audiophiles still swear on some bulky and really heavy class AB-Designs to drive the lowest octave.

And once you experienced an amp which dictates the rhythm based on the bass, you don't want other solutions.

Old AB International amplifiers worked this way and were unprecedented pushing bass and pace even out of small bookshelf speakers...

If you want overwhelming bass quality via class D, you have to spend too, e.g. for the Hypex Ncore 400. There's no free lunch.

 

[Zoli64]

Thanks for the reply. I only thought about it because if the power supply was built in, the approx. Instead of a 0.2 ohm supply line, there would be a significantly lower resistance connection. Another question: does this GAN power supply accept an extra buffer capacitor on its output, for example 4700 uF/63V? Also, doesn't the amplifier itself use some trick so that if I turn on the switch, it starts immediately, without pressing a button? Then the sound system could be started with a switch, I wouldn't have to press the volume button.

Third question: Is the a70's bandwidth guaranteed to be straight from 20Hz to 20kHz? So doesn't the bass drop off sharply around 30-35 Hz? Has anyone replaced the coupling capacitor?

And other question: all my speakers are 4 ohms. Is it possible to push the upper frequency limit on the pffb circuit a little with a simple change so that the high pass does not drop up to 20 kHz? If so, and it is not a serious professional secret, which component is responsible for this and what is the recommended intervention? Thanks to whoever helps me with these questions. Have a nice day!

Does anyone have anything to say about this? Also, I need a recommended modding list, what would you change? So far: opa1656, some bipolar electrolytic capacitors are my idea, along the lines of the Aiyima A07. Give me a tip, please.

 

[Rotgrunblau]

Perhaps we should accept that a profound, defined and beefy bass response need some effort from the amplifier too. It's a reason that a lot of pros and audiophiles still swear on some bulky and really heavy class AB-Designs to drive the lowest octave.

And once you experienced an amp which dictates the rhythm based on the bass, you don't want other solutions.

Old AB International amplifiers worked this way and were unprecedented pushing bass and pace even out of small bookshelf speakers...

If you want overwhelming bass quality via class D, you have to spend too, e.g. for the Hypex Ncore 400. There's no free lunch.

Maybe you are right, at least this what I noticed when testing these small class D amps.

And my hope was to get rid of the rather bulky, heavy and inefficient class AB systems.

But when this is really the case, do all these measurements like SINAD, load (in) dependency, power output per channel do have any real world consequences?

Because when just going with the specs and measurements it should be possible to drive for example my seas drivers(even with the topping PA5). But in reality it’s just not the case.

Actually Hypex and Icepower were the brands I was familiar with but I didn’t want to use them because they are way more expansive and I wanted to give some new players to the market a chance. But in the end… yeah no free lunch ‍

Did someone else here noticed a similar behavior? That the real world output of small class D systems (especially TPA based systems) fall behind their Class AB counterparts? And why is this so?

 

[DavidM1]

Maybe you are right, at least this what I noticed when testing these small class D amps.

And my hope was to get rid of the rather bulky, heavy and inefficient class AB systems.

But when this is really the case, do all these measurements like SINAD, load (in) dependency, power output per channel do have any real world consequences?

Because when just going with the specs and measurements it should be possible to drive for example my seas drivers(even with the topping PA5). But in reality it’s just not the case.

Actually Hypex and Icepower were the brands I was familiar with but I didn’t want to use them because they are way more expansive and I wanted to give some new players to the market a chance. But in the end… yeah no free lunch ‍

Did someone else here noticed a similar behavior? That the real world output of small class D systems (especially TPA based systems) fall behind their Class AB counterparts? And why is this so?

It’s probably due to headroom. The measurements show that class D typically has no headroom beyond its max power where’s class A/B allow peaks for small amounts of time beyond rated power.