• 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 Review (Amplifier)

trungdtmc

Active Member
Joined
Mar 9, 2022
Messages
123
Likes
78
I want you to look at the TPA3255 datasheet at figure 8:

View attachment 223786

This shows you how much power the amplifier can output at specific input voltages with THD+N of 1%.
Now, let's do the math for all the possible candidates, but first let's assume the power supply are 100% efficient(which is impossible, in reality they are 80-90% efficient) meaning that for every watt consumed from the main it produces 1watt DC.
with a 4ohm speaker:
24V 14.5A: 60W/ch, =>60W/24V=2.5A/ch=5A. We can see this power supply is not limited by current and is able to provide 60W/ch
32V 11A: 110W/ch, => 110W/32V=3.43A/ch=6.875A. Again this power supply is not limited by current and is able to provide 110W/ch
36V 9.5A: 135W/ch, => 135W/36A=3.75A/ch=7.5A. Again this power supply is not limited by current and is able to provide 135W/ch
48V 7.5A: 230W/ch, => 230W/48V=4.79A/ch=9.58A. This power supply does not provide the required current to provide 230W/ch.

Now the amplifier is not 100% efficient, it produces heat:
View attachment 223790
View attachment 223791
at 100W(50w/ch) it produces 22W of heat, and at 300W (150W/ch) it produces 53W of heat. And I don't think that the small heatsink on the amplifier is able to provide enough cooling on a closed case with no exhaust air or venting. At best it provides enough cooling for 5W.

At 50W(25W/ch) the amplifier has an efficiency of 80%, which means 50W*0.2=10W of heat is produced. This is twice the amount of heat the heatsink is able to dissipate.

So, my final questions to everybody are: are you planning to cook something with the amplifier? Why do you want 300W if the amplifier is not able to handle that amount of power/heat? And how much power do you really need? 10W max per channel is all you need.
Power supplies are most efficient at 50% of the load. This is their sweet spot. So if you are going to consume 20W(in reality) then use a 40W power supply.
In summary, don't bother with "big" power supplies. It is all about clean power not raw power. A clean 24W 7A power supply with less than 10mV ripple and noise will sound better than a 48V 10A noisy power supply.
You can help cleaning the power supply output with a combination of capacitors to clean the noise(small capacitor) and ripple(large capacitor)
View attachment 223798

This is why amplifier have those bulk capacitors to help cleaning in supply voltage, but we can add small capacitors to help cleaning the noise also at opamp supply pins:
View attachment 223800View attachment 223801
Hi, what r the values of these capacitors sir?
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
I've got an FX Audio and an Aiyima tube preamp. One with tone control, the other without. I run Dala Spektor 2s and have them up sometimes at 90-92Db at a 40" LP. My thought was the higher amps would be better when pushing higher power. I only have a multimeter and I'm unsure about measuring the output that way. From what I've read there's nothing negative to running 8-10a since the amp will draw the current it needs. The voltage seems to be the main concern with heat, that's why I've kept mine at 36v.
Sure, there is nothing negative by using a 10A 36V power supply, but also nothing positive if you are only using 0.25A(that's what I calculate you might be consuming in total including all power losses, but only 2-4 watts in actual audio output signal).
Voltage alone is not a problem. You can feed the amp with 48V but if you are only consuming 0.25A then your total power consumption is 12W. And at that power the heat generated by the amplifier is around 30%.
30% of 12W is 3.6W. That's a considerably amount of heat for such small heatsink with no fans or vents. It will feel very hot to the touch.At that power it will be probable playing very, very loud at 6w per channel. But if you play at 1W per channel then the heat produced will be 60% 2W*0.6= 1.2W. That's much better and won't feel as hot.
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
Hi, what r the values of these capacitors sir?
You mean the ones on the back of the PCB? I believe that for the bulk power supply capacitors the bypasses are 10, 2 and 0.1uF. The bulk capacitors (1000-3000uF) help to smooth the ripple, for the small spikes(noise) you need small capacitors with a very quick discharge time instead.
 

Midnight Audiophile

Active Member
Joined
Jun 30, 2022
Messages
143
Likes
226
I'm not disagreeing with anything folks are saying but I haven't seen complaints of a heat problem with the A07 in reviews or various forums. I've got two of them I've been using for over a year and they've not run hot. I have them vented now but not due to experiencing any problems before.
IMG_4216.jpg
 

trungdtmc

Active Member
Joined
Mar 9, 2022
Messages
123
Likes
78
You mean the ones on the back of the PCB? I believe that for the bulk power supply capacitors the bypasses are 10, 2 and 0.1uF. The bulk capacitors (1000-3000uF) help to smooth the ripple, for the small spikes(noise) you need small capacitors with a very quick discharge time instead.
yup. thanks so much
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
Hi, what r the values of these capacitors sir?
I'm not really sure about which capacitors you refer. Some electrolytic capacitors are in series to the audio path(those before and after the opamps). For those you use 0.1uF (1% of the capacitor you want to bypass) ceramic NP0 or C0G dielectric type 50V. For the other electrolitycs you can use ceramic X7R type using approximately 1% of the capacitor you want to bypass.
If the capacitor is parallel to ground it helps to remove noise on the supply voltage.
If the capacitor is in series to the audio path then adding a bypass here helps with the high frequencies. Again, small capacitors charge and discharge faster, this helps to transmit the high frequencies better. The signal through the capacitor follows the input signal more purely without distortion/degradation.
The principle is similar to why we use tweeters and woofers on speakers. A big woofer is not able to reproduce high frequencies the same way as tweeters do, small drivers move faster.
 
Last edited:

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
I'm not disagreeing with anything folks are saying but I haven't seen complaints of a heat problem with the A07 in reviews or various forums. I've got two of them I've been using for over a year and they've not run hot. I have them vented now but not due to experiencing any problems before.
View attachment 224267
The reason is very simple my friend. Because not much power is been drawn by the amplifier. Most of the people use 1-2W most of the time. That's not enough to heat up too much the heatsink.
And that has been my whole point since the beginning. We don't use as much power as we think.
 

trungdtmc

Active Member
Joined
Mar 9, 2022
Messages
123
Likes
78
I'm not really sure about which capacitors you refer. Some electrolytic capacitors are in series to the audio path(those before and after the opamps). For those you use 0.1uF (1% of the capacitor you want to bypass) ceramic NP0 or C0G dielectric type 50V. For the other electrolitycs you can use ceramic X7R type using approximately 1% of the capacitor you want to bypass.
If the capacitor is parallel to ground it helps to remove noise on the supply voltage.
If the capacitor is in series to the audio path then adding a bypass here helps with the high frequencies. Again, small capacitors charge and discharge faster, this helps to transmit the high frequencies better. The signal through the capacitor follows the input signal more purely without distortion/degradation.
The principle is similar to why we use tweeters and woofers on speakers. A big woofer is not able to reproduce high frequencies the same way as tweeters do, small drivers move faster.
346D0AF6-B23A-488D-B0B3-7353CA24D920.jpeg

If i not wrong, reds are 104, yellows are 224, so what value of greens sir?
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
View attachment 224268
If i not wrong, reds are 104, yellows are 224, so what value of greens sir?
The top red ones are 0.1uF C0G or NP0 type.
The yellow ones below are 2uF X7R ceramic type.
The small green ones below are 1uF X7R ceramic type. The NP0 are bigger even being smaller in capacitance.
On the power supply bulk capacitors I think I used 2x 2uF X7R and 1uF NP0 ceramic on each big capacitor.
On the opamps. I used 10uF Elna Silmic II, 10uF X7R and 0.1uF NP0. The notch on the adapter is on the wrong side. I was too lazy after I realized I silvered wrong. You can see the squared solder pad for the pin 1 on the top left.
 
Last edited:

Midnight Audiophile

Active Member
Joined
Jun 30, 2022
Messages
143
Likes
226
The reason is very simple my friend. Because not much power is been drawn by the amplifier. Most of the people use 1-2W most of the time. That's not enough to heat up too much the heatsink.
And that has been my whole point since the beginning. We don't use as much power as we think.
Right on. In simple terms how does one go about measuring the output when playing at high volumes? I have a cheap multimeter but thought I had read that wouldn't do the job. I'm also interested in this with respect to level matching when testing between my two amps and different speakers.
 

delta76

Major Contributor
Joined
Nov 27, 2021
Messages
1,615
Likes
2,460
Right on. In simple terms how does one go about measuring the output when playing at high volumes? I have a cheap multimeter but thought I had read that wouldn't do the job. I'm also interested in this with respect to level matching when testing between my two amps and different speakers.
Get a wattmeter, or a smart plug with energy consumption monitoring.
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
Right on. In simple terms how does one go about measuring the output when playing at high volumes? I have a cheap multimeter but thought I had read that wouldn't do the job. I'm also interested in this with respect to level matching when testing between my two amps and different speakers.
The problem is that music is very dynamic but you can see the multimeter and determine a max voltage value and an "average" from your observation.
Don't try tones. Tones are way louder than real music. Not helpful to calculate average power for music.
Now, for level matching you need to use tones, like a 1kHz tone for example. Because of the higher voltage helps you to be more accurate.
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
Get a wattmeter, or a smart plug with energy consumption monitoring.
Wattmeters only tell you how much power is drawn from the mains. You need to measure the amp On with 0% on the pot and then playing music. Also remember the power supply has power losses, the amplifier to operate consumes power for the other functions. Even the chip consumes power while on without playing any sound(quiescent current). All this makes the measurements more complicated and less accurate.
I don't know how accurate are those wattmeters, I think the ones I know don't have decimal values, integers only. And one watt is most of the time the power we use.
Yes, there many ways to measure the output power. You can measure current with a clamp meter on one of the speaker wires, positive or negative(not the pair, both currents will cancel each other).
 
Last edited:

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
I'm using the 500-2000Hz pink noise method on the tones. This looks like it may work, it says it gives instantaneous and peak values with a meter accuracy: Class 1.0; 0.5%-1.5%.
That's nice. I might even buy one for myself. My old Kill-a-watt is almost dying.
Just remember to take a baseline measurement which you will substract from your future measurements while playing music or pink/white noise to calculate your actual power used as output power.
 

sonci99

Active Member
Joined
Feb 11, 2018
Messages
123
Likes
76
Is there any audible difference between this Aiyima and the Topping PA5?
I usually listen at low volume.
 

iLoveCats

Active Member
Joined
Dec 8, 2019
Messages
258
Likes
292
Location
USA
Is there any audible difference between this Aiyima and the Topping PA5?
I usually listen at low volume.
I have two A07 and one PA5. They all sound great on Topping DAC's. I have the PA5 on my best speakers. Its more than likely a mental bias that I do not simply have 3 A07's. The PA5 is paired with an EX5. That setup has a near perfect signal path which satisfies my mental needs.
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
210
From this video
, and then this
I think heat is really problem with tpa32xx and other class D chip
Nice videos. I haven't watched them before.
Interesting that his amplifier comes with thermal pads, because my Aiyima A04 comes with a copper shunt 1.5-2mm aprox and the 2 plastic washers/standoffs.
I replaced the white thermal grease it came with and replaced it with a Thermalright one. Just for peace of mind, because the amplifier never gets hot while playing real music even at uncomfortable levels.
But if one would like to use the Aiyima to play at a concert or stadium (50W or more), then my recommendation is to remove the top cover and cut a hole for a small 8 or 10 cm computer fan(exhausting air) and add some vents to the bottom and sides to allow fresh air to enter and hot air to be exhausted at the top. This will multiply by 4 the cooling capability of the heatsink and will allow the amplifier to handle more power without cooking itself.
 

Midnight Audiophile

Active Member
Joined
Jun 30, 2022
Messages
143
Likes
226
I have two A07 and one PA5. They all sound great on Topping DAC's. I have the PA5 on my best speakers. Its more than likely a mental bias that I do not simply have 3 A07's. The PA5 is paired with an EX5. That setup has a near perfect signal path which satisfies my mental needs.
That's a nice setup with the PA5 and EX5. I'd like to test that against an A07 and FX Audio DAC-SQ3 (roughly $700 system vs. $140 which seems to be about as cheap as you can get with separates that measure decent).
 

Walter

Addicted to Fun and Learning
Joined
Aug 25, 2020
Messages
856
Likes
1,242
Sure, that will work. The same way the stock power supply will. It would be better to spend the money on a cheap preamp or tube preamp like the Fosi Audio, FX audio or Suca Audio. It will allow you to increase the input audio signal.
It would absolutely not be better to add a tube preamp, especially a cheap one. That will drastically increase both noise and distortion. As will many cheap solid state preamp.
 
Top Bottom