AIYIMA A07 TPA3255 Review (Amplifier)

[Doodski]

just for studying purpose have been reading TPA3255 datasheet. LINK. With various implementations and varying power supply voltages and amps, am finding it little confusing. Though at 51V it achieves (~100W, 8 ohms approx. 0.1% thd) It seems those voltages can be lowered to be on safe side and for longetivity of the chip as 51v is maximum.
So what would be ideal power supply requirement (volts and amps) to have 0.1 % distortion(Standard i suppose) and what maximum watts (in to 8 ohms and 4 ohms) one would get with those safe side voltages ?
thanks.

We don't know the relationship values between the input voltage and the power supply output rails supplying the output amp power. Measurements would be req'd to get those details. Otherwise for the rails these are the parameters req'd:
8R @ 140 Watts RMS @ 0.1% THD = rail voltage req'd @ 47.3 VDC
4R @ 240 Watts RMS @ 0.1% THD = rail voltage req'd @ 43.8 VDC
- The pdf does not state if the amp output power is peak or RMS so I am supplying both here -
8R @ 140 Watts @ 0.1% THD = rail voltage req'd @ 33.5 VDC
4R @ 240 Watts @ 0.1% THD = rail voltage req'd @ 31 VDC

 

[grogi.giant]

just for studying purpose have been reading TPA3255 datasheet. LINK. With various implementations and varying power supply voltages and amps, am finding it little confusing. Though at 51V it achieves (~100W, 8 ohms approx. 0.1% thd) It seems those voltages can be lowered to be on safe side and for longetivity of the chip as 51v is maximum.
So what would be ideal power supply requirement (volts and amps) to have 0.1 % distortion(Standard i suppose) and what maximum watts (in to 8 ohms and 4 ohms) one would get with those safe side voltages ?
thanks.

6A@36V is a very sensible compromise.

We don't know the relationship values between the input voltage and the power supply output rails supplying the output amp power.

Theoretically not, but from practical point of view, this is the same value.

There is no transformer inside the amp - so no upping the voltage. Doing stepdown could possible - but traditional voltage regulator will however generate a lot of heat, while PWM doesn't make much sense, as the amp chip is PWM modulator itself.

The input voltage from PSU is fed directly to the amp chip, just with a bit of filtering.

 

[Doodski]

6A@36V is a very sensible compromise.


Theoretically not, but from practical point of view, this is the same value.

There is no transformer inside the amp - so no upping the voltage. Doing stepdown could possible - but traditional voltage regulator will however generate a lot of heat, while PWM doesn't make much sense, as the amp chip is PWM modulator itself.

The input voltage from PSU is fed directly to the amp chip, just with a bit of filtering.

I was wondering about that. I don't know this specific A07 model and so I am just flying on numbers with no physical details.

 

[Hiten]

Thanks gents.
So 36v / 6A would probably give around 80watts in 8 ohms and somewhat nearing double of that in 4 ohms. Pardon to think convservatively as I think for long listening sessions and with typical speakers it would give longevity to the whole system.
regards.

 

[grogi.giant]

Thanks gents.
So 36v / 6A would probably give around 80watts in 8 ohms and somewhat nearing double of that in 4 ohms. Pardon to think convservatively as I think for long listening sessions and with typical speakers it would give longewvity to the whole system.
regards.

Not really, with 4 Ω you are very much current limited. Realistically you'll get more-less same into 4Ω and 8Ω.

 

[grogi.giant]

I'm trying to calculate the theoretical power figures for various power supplies, but I get ridicolously low numbers. Where do I make mistake?!

Assumptions: power per channel, with both channels driven. THD 10%.

Example:

• PSU: 32V, 5A
• Impedance 4Ω
• Voltage limited
  • Maximum power per channel, as per TI Datasheet: 130W
• Current limited
  • Maximum total current from PSU: 5A
  • Maximum current from PSU, per channel: 2.5A
  • Power per channel, with 2.5A current and given impedance: 2.5A * 2.5A * 4Ω = 25W
• Actual max power (min. voltage and current limited)
  • 25W / channel

This is very low, and proven empirically it is much higher than that. Where do I make mistake?

I guess, in BTL mode the chip might reverse polarity when needed, which would double the effective voltage (Vpp = 2 x Vcc) and current, quadrupling the power...

 

[HappyMetalGuy]

I'm trying to calculate the theoretical power figures for various power supplies, but I get ridicolously low numbers. Where do I make mistake?!

Assumptions: power per channel, with both channels driven. THD 10%.

Example:

• PSU: 32V, 5A
• Impedance 4Ω
• Voltage limited
  • Maximum power per channel, as per TI Datasheet: 130W
• Current limited
  • Maximum total current from PSU: 5A
  • Maximum current from PSU, per channel: 2.5A
  • Power per channel, with 2.5A current and given impedance: 2.5A * 2.5A * 4Ω = 25W
• Actual max power (min. voltage and current limited)
  • 25W / channel

This is very low, and proven empirically it is much higher than that. Where do I make mistake?

I guess, in BTL mode the chip might reverse polarity when needed, which would double the effective voltage (Vpp = 2 x Vcc) and current, quadrupling the power...

 

[ModDIY]

I'm trying to calculate the theoretical power figures for various power supplies, but I get ridicolously low numbers. Where do I make mistake?!

Assumptions: power per channel, with both channels driven. THD 10%.

Example:

• PSU: 32V, 5A
• Impedance 4Ω
• Voltage limited
  • Maximum power per channel, as per TI Datasheet: 130W
• Current limited
  • Maximum total current from PSU: 5A
  • Maximum current from PSU, per channel: 2.5A
  • Power per channel, with 2.5A current and given impedance: 2.5A * 2.5A * 4Ω = 25W
• Actual max power (min. voltage and current limited)
  • 25W / channel

This is very low, and proven empirically it is much higher than that. Where do I make mistake?

I guess, in BTL mode the chip might reverse polarity when needed, which would double the effective voltage (Vpp = 2 x Vcc) and current, quadrupling the power...

 

[grogi.giant]

View attachment 284909

That chart shows you maximum power per channel, but not when you are current limited! 48V 3A and 48V 10A will give vastly different power figures, and the difference depends on the impedance as well.

 

[grogi.giant]

I did not ask for the figures, but where did I make a mistake... You never wondered how those are calculated?!

BTW: those numbers directly contradict the TPA3255 datasheet - there there is no way it could generate 60W into 8Ohm THD 10% with 24V supply - if not current restrained, maximum here is 40W. Similarily at 32V 8Ohm - max here is 70W, not 105W... So sorry - I call this table BS.

 

[NTK]

I did not ask for the figures, but where did I make a mistake... You never wondered how those are calculated?!

I'll give you a hint.

5 A power supply into a 4 ohm load gives 4 * 5^2 = 100 W.

You "split" the 5 A into 2, and you ended up with 2 * 4 * 2.5^2 = 50 W. Where did the other half go?

 

[grogi.giant]

I'll give you a hint.

5 A power supply into a 4 ohm load gives 4 * 5^2 = 100 W.

You "split" the 5 A into 2, and you ended up with 2 * 4 * 2.5^2 = 50 W. Where did the other half go?

We have two identical loudspeakers, each 4Ω. They are connected in paralel - as they would in a stereo system. That effectively generates 2Ω load. 5A into 2Ω is 5A*5A*2Ω = 50W for both combined channels, so 25W per channel.

And that precisely is why I am asking - where is the conceptual mistake?! If I'm driving both channels at the same time (thus sharing the current supply), napkin calculations suggest we would only get 25W per channel into 4Ω speakers (so 50W total). Measurements suggest much better figures... What is the magic sauce that allows an amplifier generate 77W per channel, both channels driven into 4Ω speaker (as measured by @amirm) from 5A supply?!

Yeah - 32V@5A does give us 160W. That's enough to cover 2x77W. But 32V @ 4Ω would require 16A of current to drive both speakers at the same time with that voltage. That is not something that the PSU can do.

We would need to have 9A @ 18V (162W) to get 162W out of two paralel 4Ω: 9A * 9A * 2Ω = 162W.

 

[xpop]

Hello, the TPA3255 accepts a power stage supply (max) 53.5V.
Mine is fine, running @53V with a 19A power supply (perhaps a little hot)

 

[ModDIY]

View attachment 284909

That chart shows you maximum power per channel, but not when you are current limited! 48V 3A and 48V 10A will give vastly different power figures, and the difference depends on the impedance as well.

It's not the maximum power, but at 1% distortion. It is a table, of course it must be based on the impedance of the speakers and the voltage. Then the available current of the SMPS in A.

There is another table with maximum power at 10% distortion, I do not take it into account.

 

[grogi.giant]

It's not the maximum power, but at 1% distortion. It is a table, of course it must be based on the impedance of the speakers and the voltage. Then the available current of the SMPS in A.

There is another table with maximum power at 10% distortion, I do not take it into account.

You're being pedantic, but you're right.

TI TPA3255 spec gives two charts - power available with THD 1% and THD 10%, both depending on the voltage. Those depend only on the voltage and will be irrelevant if the amplifier suffocates for current (and as the result the supplied voltage would drop in uncontrolled manner, generating massive distortion in result).

 

[ModDIY]

You're being pedantic, but you're right.

TI TPA3255 spec gives two charts - power available with THD 1% and THD 10%, both depending on the voltage. Those depend only on the voltage and will be irrelevant if the amplifier suffocates for current (and as the result the supplied voltage would drop in uncontrolled manner, generating massive distortion in result).

Why are you calling me a pedant?

You repeat the same thing as me. Are you doing well ?

 

[ModDIY]

It's not the maximum power, but at 1% distortion. It is a table, of course it must be based on the impedance of the speakers and the voltage. Then the available current of the SMPS in A.

There is another table with maximum power at 10% distortion, I do not take it into account.

 

[-Jim-]

My youngest son has been bugging me from time to time for over a month (since my birthday) to give him a list of something to buy me as a birthday gift.
I really don't need anything, but he kept pushing, and said he was going to buy me something really weird soon if I didn't give him something.

So a while ago I was on Amazon, and as I had been looking at this AIYIMA Amp for a cyberbuddy, I went to see it again. Amazon.ca had it on a 15% discount to $86.82 CAN ($64.33 USD) so I sent my son the link.

My Atmos system is currently configured as a 5.2.4 due to the Denon AVR "only" having 9 channels of amplification, but it will process 11 channels. During our Reno, I pre-wired for 7.2.4 and wall mounted the surrounds even though I never powered up the Rears since. I've been toying with the idea of getting a separate Amp just for the rears for ages, but the present configuration seems great just the way it is, and I could never justify it to myself.

Amazon delivered it at 3:15 the next day but I didn't know for sure what it was as it was addressed to my son; and he often gets stuff as he's got a Prime membership. He gave it to me after dinner.

I hooked up the AIYIMA Amp to a pair of 8 ohm bookshelf speakers we've had in storage since the big Reno of 2018. For input I connected it to a NUC (PC) I put together last summer to get familiar with Windows 11. I plugged in a portable hard drive into the NUC with all my FLAC and MP3 back-up tunes on it. It's pretty near field but sounds pretty good considering the gear.

I've been driving it continuously at pretty decent volume levels for well over an hour (multiple times) and the AIYIMA has been flawless and isn't even warm. According to my Thermal Gun the temperature rise is 5.2° C starting from a 20° C ambient.

I'm happy with it. Enough to look for a window to install it to drive the Rear Surrounds in the Atmos System in the Games Room. But for now I'll keep listening / testing out in the Man Cave.

Thanks for the review and all the comments.

 

[Toku]

Hello, the TPA3255 accepts a power stage supply (max) 53.5V.
Mine is fine, running @53V with a 19A power supply (perhaps a little hot)
View attachment 284929
View attachment 284929

The TPA3255 can be used at Max 53.5V from the datasheet. However, the AIYIMA A07 uses a power supply capacitor with a withstand voltage of 50V. Therefore, the limit of the power supply voltage is 48V, which is lower than that. In order to use A07 more safely, it is desirable to use the power supply voltage at 40V or less.

 

[grogi.giant]

The TPA3255 can be used at Max 53.5V from the datasheet. However, the AIYIMA A07 uses a power supply capacitor with a withstand voltage of 50V. Therefore, the limit of the power supply voltage is 48V, which is lower than that. In order to use A07 more safely, it is desirable to use the power supply voltage at 40V or less.

TI says that the capacitors should be selected according to best engineering practices: to tolerate voltage spikes etc. That applies not only to the big ones, clearly see even by a layman like me, but the small ones burried under the radiator.

The spec also mentions that 50v supply would require 80v for the 'big capacitors'. That's 60% overshoot.

Aiyima is using 50v (or 63V capacitors with some newer batches) for PVDD; the same ratio of 1.6 would give us recommended supply voltage ~32V (or ~40V accordingly). I personally would not venture above 36V, there simply is no need to.