AIYIMA A05 Review (Budget Amplifier)

[Doodski]

Thank you for your opinion.
I meant not the lack of power, definitely smps has a lot, but the typical smps issue - noise.

Do you mean mechanical whine/buzzing noise from the power supply itself or noise induced into the amplifier audio output?

 

[Ralferator]

Hello All,



The blind tests have been done. The SINAD testing is complete. The threshold of human hearing is documented.

I have seen @amirm post plots of SINAD measurements with the threshold of human hearing overlaid on the same plot. I would point out this plot if I could recall where here on ASR it is posted. (If someone recalls please point it to the rest of us.)



Thanks DT



Note:

The above is a KISS answer. If you dig deeper it is not so simple.

I would like to see this too

 

[rammster]

Do you mean mechanical whine/buzzing noise from the power supply itself or noise induced into the amplifier audio output?

The induced noise. Those chips all have some Power Supply Ripple Rejection, but I never had chance to quantify performance degradation depending on PSU quality.
Sneaky Chinese friends offer improved/advanced power supplies as an upgrade, I was guessing how much improvement would it give.

 

[Doodski]

The induced noise. Those chips all have some Power Supply Ripple Rejection, but I never had chance to quantify performance degradation depending on PSU quality.
Sneaky Chinese friends offer improved/advanced power supplies as an upgrade, I was guessing how much improvement would it give.

The operating frequency of the SMPS these days is far above your hearing range. By the time the filter caps smooth the SMPS output there will be little to nothing remaining in the form of high frequency ripple and what is remaining is not audible anyway due to it being maybe 50KHz or higher. Just a guess on the operating frequency of 50KHz or higher because I don't have the power supply to measure the switching frequency. The power supply that comes with the amp is probably more than within spec guidelines for high frequency ripple and regulation. But keeping in mind that we are being subjective because we can't measure the power supply circuitry. I've measured and repaired thousands of SMPS in car amps and they operated from about 30KHz to 60KHz and the SMPS noise was not audible in the audio amplifier output. What was audible was the hissing and squeeling of the SMPS torroidal transformers. They can be very annoying and noisy when they are being overloaded or are faulty. As per a linear supply they will be larger for the same power output and the ripple will be within your hearing range and could affect the audio signal you listen to. To get the ripple down to low low levels requires lots of capacitance and that means bigger size again. I would run the SMPS that is provided with the amplifier and look elsewhere for improvements that will make a satisfactory improvement. I don't remember all the specifics of your details and so going off memory I recommend getting some EQ/PEQ into your system for a major improvement that anybody can hear and appreciate.

 

[restorer-john]

I would run the SMPS that is provided with the amplifier and look elsewhere for improvements that will make a satisfactory improvement.

The provided SMPS is 160W. The combined left and right power outputs into 4R before shutdown was very close to that rating. The way to find out if there is more performance (power) to be had would be to run a single channel instead of both. Perhaps then the output may approach its rated 105W specification and if so, a better power supply could make a difference.

We don't know at this point if it was amplifier limited or power supply limited.

 

[DualTriode]

While we are poking at this thing, what does the FFT noise floor look like with the 1kHz signal turned off?

With a MeanWell power supply what does the FFT look like?

Changing the Aiyima 07 power supply made a big difference.

Thanks DT

 

[deprogrammed]

Some of think Bluetooth is useful, great in fact. Make the stereos more accessible to others who don't care about gear at all. In fact right now my daughter is listening to Lorde via her phone. That is a good thing. Accessibility is a wonderful thing if you want to share music.

 

[Toku]

Isn't that last observation literally impossible given the measurements and the same volume level? I ask because the A07 has far better SINAD. BTW and FWIW, my A07s don't "pop" on external power-off/power-on sequence -- my entire system, including two A07s, is connected to a single power strip, and I use the switch on that strip to turn the whole shebang on and off with no popping whatsoever.

The pop noise issue talks about A05. There is no problem with A07.
The measurement conditions are slightly different between the A07 measurement and the A05 measurement. Since A05 is measured with the volume set to MAX, THD + N will be slightly worse.

 

[Bruce Morgen]

The measurement conditions are slightly different between the A07 measurement and the A05 measurement. Since A05 is measured with the volume set to MAX, THD + N will be slightly worse.

If that were true, wouldn't that make Amir's SINAD chart useless as a comparison tool?

 

[restorer-john]

If that were true, wouldn't that make Amir's SINAD chart useless as a comparison tool?

S/N is traditionally measured with the volume control at maximum and the generator input level varied to give the rated output. The generator is then switched off, the input shorted and the residual measured. Distortion is measured by notching out the fundamental and analyzing everything left over.

This little amp needed 295mV to give 4.51V out, a gain of 23.688dB. If Amir backed off the volume control to give a 2V in, 4.51V (5W@4R) out, the performance may have been a lot better, a lot worse or much the same. We don't know.

All I know is amplifiers in general have sweet spots on their volume range where the noise is lowest, the levels are such that THD on the front end isn't an issue and overall gain is suitable for the source connected.

 

[rammster]

The operating frequency of the SMPS these days is far above your hearing range. By the time the filter caps smooth the SMPS output there will be little to nothing remaining in the form of high frequency ripple and what is remaining is not audible anyway due to it being maybe 50KHz or higher. Just a guess on the operating frequency of 50KHz or higher because I don't have the power supply to measure the switching frequency. The power supply that comes with the amp is probably more than within spec guidelines for high frequency ripple and regulation. But keeping in mind that we are being subjective because we can't measure the power supply circuitry. I've measured and repaired thousands of SMPS in car amps and they operated from about 30KHz to 60KHz and the SMPS noise was not audible in the audio amplifier output. What was audible was the hissing and squeeling of the SMPS torroidal transformers. They can be very annoying and noisy when they are being overloaded or are faulty. As per a linear supply they will be larger for the same power output and the ripple will be within your hearing range and could affect the audio signal you listen to. To get the ripple down to low low levels requires lots of capacitance and that means bigger size again. I would run the SMPS that is provided with the amplifier and look elsewhere for improvements that will make a satisfactory improvement. I don't remember all the specifics of your details and so going off memory I recommend getting some EQ/PEQ into your system for a major improvement that anybody can hear and appreciate.

Got the idea, thanks.
My concern with SMPS was about possible degradation of the noise floor, but according to you it does not matter. I hoped that sinad would rise with better psu, but looks like it is just marketing offer from sellers.

 

[Doodski]

Got the idea, thanks.
My concern with SMPS was about possible degradation of the noise floor, but according to you it does not matter. I hoped that sinad would rise with better psu, but looks like it is just marketing offer from sellers.

What is the expense for said better power supply? Where does MODing and rolling it end? What are you looking to do? Myself I would have stopped some ways back already because maybe there are options available. What is the total expense for the amp and these MODs that you wanna get? Can you buy a more permanent type of amp that does not require MODs? Or would you MOD that too?

 

[rammster]

What is the expense for said better power supply? Where does MODing and rolling it end? What are you looking to do? Myself I would have stopped some ways back already because maybe there are options available. What is the total expense for the amp and these MODs that you wanna get? Can you buy a more permanent type of amp that does not require MODs? Or would you MOD that too?

Probably I expressed myself wrong. I am used to hearing about crappy_chinese_psu's, but never saw a detailed explanation or research about how they affect the device. Most logical and understandable parameter to me is the noise floor. And I want to understand how serious is the problem of bad_psu vs good_psu in that perspective.

 

[Doodski]

Probably I expressed myself wrong. I am used to hearing about crappy_chinese_psu's, but never saw a detailed explanation or research about how they affect the device. Most logical and understandable parameter to me is the noise floor. And I want to understand how serious is the problem of psu_crappiness vs good_psu in that perspective.

I modified a 12VDC power supply for use on a 3 way car crossover and I compared the power supplies and the before and after ripple was miniscule. I did the same for a workbench regulated power supply and same result the ripple change was small in both instances. I never measured for SINAD or harmonic or any other stuff other than is it linear input to output? Can it hit rated power at 1KHz till it gets very hot and run it there testing for thermal stability for awhile monitoring the current closely, is it 4 Ohm capable and test it there too, what is the Bias and idle setting and are they good? I near never worried about SINAD.

 

[Deleted member 9685]

* AIYIMA A07 was better than this.
* You should use BFA banana plugs. They are thin enough to fit densely packed speaker terminals easily.
Here's the picture of BFA banana plugs.

 

[Deleted member 9685]

It seems 3205000 power supply leaked mains hum into AIYIMA A07, but not into AIYIMA A05.
What is going on? AIYIMA A05 is better at refusing mains hum? Or, you got a better unit of 3205000 power supply?

 

[rammster]

I modified a 12VDC power supply for use on a 3 way car crossover and I compared the power supplies and the before and after ripple was miniscule. I did the same for a workbench regulated power supply and same result the ripple change was small in both instances. I never measured for SINAD or harmonic or any other stuff other than is it linear input to output? Can it hit rated power at 1KHz till it gets very hot and run it there testing for thermal stability for awhile monitoring the current closely, is it 4 Ohm capable and test it there too, what is the Bias and idle setting and are they good? I near never worried about SINAD.

That is why I try to rely on objective measurement results and scientific approach.

 

[Doodski]

That is why I try to rely on objective measurement results and scientific approach.

I rely on ASR for all my data.

 

[rammster]

So, does the less noisier SMPS improve noise floor or not?

 

[Doodski]

So, does the less noisier SMPS improve noise floor or not?

If I remember correctly you where also looking at a linear supply idea to improve sound quality. This web page is maybe a 3-5 minute read and it answers multiple questions for you about SMPS and linear power supplies comparatively. Of course any reduction in radiated noise, conduction noise and component noise is better.