• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

AIYIMA A07 TPA3255 amplifier with MW RS-100-24 SMPS power supply (measurements)

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
AIYIMA A07 TPA3255 Amplifier with MW RS-100-24 SMPS Power Supply

I have just received the Mean Well RS-100-24 SMPS power supply, connected it to AIYIMA A07 and made a set of measurements and some listening. As the amplifier must be paired with a power supply and these 2 units make a complete amplifier, and the behaviour of the class D amplifier with a SMPS power supply is quite special, I have decided to open this new thread and not to make even bigger mismatch in my original review/measurement thread,

https://www.audiosciencereview.com/...-measurements-and-review-lm4562-option.19155/

The power supply used in this new test thread is switched mode power supply Mean Well RS-100-24, rated at 24V/4.5A (108W). As there is a trim pot to fine tune output voltage, I have set it to 27.0V output voltage, which is close to maximum, which I measured 27.45V (OCV).

There are 3 screw terminals to connect mains input (230V, L, N, PE) and -V and +V output DC terminals, each having 2 screws. I have used 1 pair of output terminals to supply the amplifier, and to the other pair I connected 47nF/1000V capacitor to reduce output HF noise a bit. This capacitor was also used to connect oscilloscope probe for ripple measurements.
SMPS_terminals.JPG


The unit is small, 159 x 97 x 38mm.
SMPS.JPG


It should be connected and operated by a qualified personnel, for the reason of bare screws connected to dangerous mains voltage. It should be covered in a vented box in case that supposed to be used with the amplifier, with safety standards fulfilled (diameter and placement of vents).

Okay, I first measured the PSU ripple when connected to the amplifier. Without any input signal, the ripple looks like this
nosignal-4R_SMPS_ripple.JPG


The ripple is about 42mVp-p and sawtooth frequency is about 70kHz.

When both channels of the amplifier were driven to 2 x 37W/4ohm, the ripple looked like this
2x37W-4R_SMPS_ripple.JPG


About 250mVp-p and the SMPS switching frequency is modulated by H2 of the input 1kHz signal. Still not bad.

Then I measured the usual
THD and THD+N (SINAD) at 1kHz/5W/4ohm (both channels driven)
A07_4562_5W_best_4R_30VregPSU.png


We can see quite low THD=0.0014% and SINAD=86.74 dB and very good suppression of mains hum multiples.

THD and THD+N at 1kHz 2x27W/4ohm
A07_27W_4R_27V-SMPS.png


Decent spectrum, considering the chip used.

THD and THD+N at 1kHz 2x54W/4ohm
A07_54W_4R_27V-SMPS.png


Still nice!
However going a bit higher to 2x57W/4ohm, the power supply switched off itself, most probably because its output current limit has just been exceeded. So this is the limit of A07+RS-100-24 combo.

THD vs. amplitude at 100Hz, 1kHz and 6kHz / 4ohm, both channels driven, SMPS
A07_thdampl_multi_4R_27V-SMPS_s.png


So far so good up to some 2x55 – 2x60W/4ohm. Then, the SMPS PSU overcurrent control is activated and the DC supply voltage falls to zero. We can see the “knee” at about 65W, the output voltage is dramatically decreased and the distortion is dramatically increased.

If we make the same test with my 25V/270VA unregulated power supply, we do not see this collapse, however the output power is a bit lower and the spectrum is less clean, especially the mains power components (and the PSU is huge).
A07_4562_thdampl_multi_2x4R_25V_270VAunregPSU_s.png



Conclusions

I think that the amp behaves nicely with this Mean Well SMPS, except for the moment when the overcurrent protection might be activated. It is also necessary to bear safety rules of the PSU implementation in mind. For the amplifier, this lower supply voltage is an advantage re inside temperature and life time expected. The amp itself is not able to give much more than 2x50W, because after some time it is switched off by its temperature protection and with e.g. 48V power supply this happens much much earlier.
 
Last edited:

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
Can you measure the hiss or hum it produces in speakers?

My Aiyima is a bit noisy in that regard (with Glob Tek 36 V / 5 A PSU).
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Can you measure the hiss or hum it produces in speakers?

My Aiyima is a bit noisy in that regard (with Glob Tek 36 V / 5 A PSU).

Yes I can and I did so, it was posted in another (my review) thread, the noise voltage is -74.9dBV integrated over 24kHz bandwidth, this translates as 180uV/BW=24kHz noise voltage, or S/N = 83.9dB re 2.83V (1W/8ohm). With my speaker that has 85dB/2.83V/1m sensitivity, I can hear a slight hiss (no mains bzz or hum) if I put my ear close to the tweeter, but going a little bit from the tweeter there is no hiss audible. As a comparison, many preamps or dacs connected to the power amplifier with 27dB gain produce more hiss in my tweeter.
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Listening test

I have been happily surprised by the sound of this little amplifier, and I have to admit that I was distrustful first.
The sound is very good, with very well controlled and solid bass, excellent imaging and very good soundstage. There is no melting of the sound. This is an impression with my CNO-T25 speakers.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
Yes I can and I did so, it was posted in another (my review) thread, the noise voltage is -74.9dBV integrated over 24kHz bandwidth, this translates as 180uV/BW=24kHz noise voltage, or S/N = 83.9dB re 2.83V (1W/8ohm). With my speaker that has 85dB/2.83V/1m sensitivity, I can hear a slight hiss (no mains bzz or hum) if I put my ear close to the tweeter, but going a little bit from the tweeter there is no hiss audible. As a comparison, many preamps or dacs connected to the power amplifier with 27dB gain produce more hiss in my tweeter.
Thanks.
Is your power supply grounded? Mine isn't. Would there be a benefit if mine was grounded?
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Thanks.
Is your power supply grounded? Mine isn't. Would there be a benefit if mine was grounded?

If you look at his photo
https://www.audiosciencereview.com/forum/index.php?attachments/smps_terminals-jpg.107091/
you see that there is a screw with "GND" symbol, the third from the left. This screw is connected to mains protective earth (PE) and to the metal case of the power supply, but nowhere else. It serves for safety and also against EMI radiation. Output terminals -V and +V are not connected to this grounding point, as well as the amplifier is also not "grounded".
Now, Mean Well datasheet defines the so called "leakage current"
1611076130381.png

<2mA/240Vac and this would be the capacitive current that might flow from primary to output terminals through a stray capacitance of the PSU. You do not want this current to flow through the shielding of the SE signal link cable, as it might create a buzz voltage on the shielding that would add to the useful signal. So best if the signal source is not grounded as well. I have verified it when doing measurements today, if the amplifier was driven from a grounded soundcard, the residual 50Hz and multiples became about 20dB higher.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
If you look at his photo
https://www.audiosciencereview.com/forum/index.php?attachments/smps_terminals-jpg.107091/
you see that there is a screw with "GND" symbol, the third from the left. This screw is connected to mains protective earth (PE) and to the metal case of the power supply, but nowhere else. It serves for safety and also against EMI radiation. Output terminals -V and +V are not connected to this grounding point, as well as the amplifier is also not "grounded".
Now, Mean Well datasheet defines the so called "leakage current"
View attachment 107150
<2mA/240Vac and this would be the capacitive current that might flow from primary to output terminals through a stray capacitance of the PSU. You do not want this current to flow through the shielding of the SE signal link cable, as it might create a buzz voltage on the shielding that would add to the useful signal. So best if the signal source is not grounded as well. I have verified it when doing measurements today, if the amplifier was driven from a grounded soundcard, the residual 50Hz and multiples became about 20dB higher.
Ok. To be sure to understand: your PSU is not grounded, and it is better so. Right?
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Ok. To be sure to understand: your PSU is not grounded, and it is better so. Right?

We may simplify it this way. But sometimes it is necessary to know the grounding and wiring schematics of the whole audio chain.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
We may simplify it this way. But sometimes it is necessary to know the grounding and wiring schematics of the whole audio chain.
Absolutely, your explanations are very welcome.
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Can you measure the hiss or hum it produces in speakers?

Yes I can and I did so, it was posted in another (my review) thread, the noise voltage is -74.9dBV integrated over 24kHz bandwidth, this translates as 180uV/BW=24kHz noise voltage, or S/N = 83.9dB re 2.83V (1W/8ohm). With my speaker that has 85dB/2.83V/1m sensitivity, I can hear a slight hiss (no mains bzz or hum) if I put my ear close to the tweeter, but going a little bit from the tweeter there is no hiss audible. As a comparison, many preamps or dacs connected to the power amplifier with 27dB gain produce more hiss in my tweeter.

I have just measured for you the noise at the A07 output when operating with the RS-100-24 SMPS, and the output of the amp is connected to the speaker. The result is almost same as I have described earlier, -75.2dBV for 24kHz bandwidth. No hum, no buzz, just noise.

A07_noise_27V-SMPS.png
 

bigx5murf

Addicted to Fun and Learning
Joined
Nov 6, 2018
Messages
522
Likes
343
I'd be interested in seeing the measurements of this amp running off a battery or two 12v batteries in series.
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
I'd be interested in seeing the measurements of this amp running off a battery or two 12v batteries in series.

I have no 12V batteries here (now), however I expect a little difference compared to this SMPS and/or the regulated linear supply that I tried before. There is quite a difference when we compare results of SMPS/linear regulated PSU to classical unregulated trafo + diode bridge + filter cap, especially in hum components and also due to modulation by signal current. This is important, when the signal current modulates the power supply output voltage, then it is transferred to distortion according to amplifier PSR.
 

HorizonsEdge

Active Member
Joined
Dec 20, 2019
Messages
223
Likes
315
I appreciate the efforts you have made to investigate the A07! I have one and am quite happy with it. I am using the PSU supplied with the amazon order. Here is the thing. If I can buy a different PSU and get better performance I would be ecstatic BUT like many folks I need a turnkey solution. Playing with the unfinished MW unit is just a recipe for disaster for me.
 
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
I appreciate the efforts you have made to investigate the A07! I have one and am quite happy with it. I am using the PSU supplied with the amazon order. Here is the thing. If I can buy a different PSU and get better performance I would be ecstatic BUT like many folks I need a turnkey solution. Playing with the unfinished MW unit is just a recipe for disaster for me.

No arguing from my side, I show these results because I think they are interesting. I like to go in deep and to find consequences. As the ASR declares itself as a forum with scientific approach, I hope it is fine to try to show something more than a cook-book ;).
 

bigx5murf

Addicted to Fun and Learning
Joined
Nov 6, 2018
Messages
522
Likes
343
I have no 12V batteries here (now), however I expect a little difference compared to this SMPS and/or the regulated linear supply that I tried before. There is quite a difference when we compare results of SMPS/linear regulated PSU to classical unregulated trafo + diode bridge + filter cap, especially in hum components and also due to modulation by signal current. This is important, when the signal current modulates the power supply output voltage, then it is transferred to distortion according to amplifier PSR.

With the right battery, you might not be current limited though.
 

Bruce Morgen

Addicted to Fun and Learning
Forum Donor
Joined
Aug 3, 2019
Messages
865
Likes
1,241
The amp itself is not able to give much more than 2x50W, because after some time it is switched off by its temperature protection and with e.g. 48V power supply this happens much much earlier.

Pardon me, but are you seriously claiming this? I run my main L+R speakers (two vintage 8 ohm JBLs wired in parallel on each side -- i.e. nominal 4 ohm loads on both channels) with an A07 powered by a 48VDC 10A SMPS. I often play that system seriously loud, and have NEVER had the amp get "switched off," even after several hours at that volume. Steady test tones and scope traces are fine for what they are, but they will not tell you how sophisticated protection circuitry like the TPA3255's will behave in actual use, e.g. playing loud music. For that, you need temperature measurements under both formal testing and actual listening conditions at the actual voltage(s) under consideration.
 
Top Bottom