Ampapa D1 Stereo Amplifier Review

[ICIETDIYEUR]

@ Amirm: do you think that a 36V/8A PSU, or even a 24V/10A could be suitable with this amplifier, whether for 4 or 8 Ohms loaded per channel, based on your measurement results ?
(This follows my reflection just before regarding the 'heating' of the amplifier in IDLE)

 

[Bleib]

Some of you overthink it, the temperature is nothing compared to say most AVRs
And if you get the new version it even goes idle so it won't constantly put out heat

 

[ICIETDIYEUR]

I'm not one of the 'worried' people, I'm rather one of those who think that we can somehow 'optimize' to improve sustainability in this specific case

 

[Bleib]

I'm not one of the 'worried' people, I'm rather one of those who think that we can somehow 'optimize' to improve sustainability in this specific case

In that case I see no reason to cut down on its power. The whole unit functions like a heatsink, just like Fosi V3 mono. Well, that makes it rather easy to modify... attach heatsinks to it

 

[ICIETDIYEUR]

It seems to me that you are intervening without having followed my reasoning and following the YouTube video.

In addition, 48V/5A = 240W and 36V/10A (even 8A) = 360W (288W): I don't see where I reduce the power?

And lastly, I am not a 'do-it-yourselfer' who sticks heatsinks everywhere while the amplifier is basically very well designed: the problem must be dealt with 'at the source' when there is one, it should not be moved: it is your proposal which is useless, don't you think?

I think we can draw inspiration from the following table by drawing some 'lessons' ->

In the Youtube video, it seems that the person notices that the power supply 'regulates', or even is not sufficient, which can be a factor in increasing the measured distortion, etc...

NB: It also seems important to me to remember that the SINAD result obtained by the 3e audio A5 is so good because the volume circuit is bypassed:
those who design class D amps know full well that the volume and preamplification circuits (OPAMPS for example) constitute a very important part in the final distortion measured.

The AMPAPA D1 is, in its case, rather very good in SINAD:
I would be curious to know the SINAD of the 3e Audio A5 without bypassing its circuits...

 

[john61ct]

You can have infinite current amps available, current is not "pushed". The amplifier will only pull more current when needed, mostly from LF content in a high SPL context with speakers dropping in impedance.

So, if you get concerned and measure chips getting "too hot" for your taste, just turn down the volume (or get more efficient, flat impedance load speakers).

Voltage is a different story, going up to just below the reco max will get higher potential SPL, again subject to your self control. This is the main reason 3e A7 delivers so much more power at 50-51V. IMO going right up to max spec 52V (48V for D1) will impact reliability / longevity. And the voltage drawn is not what varies as current does.

Ideally your PSU has some adjustability so when high output is not needed, you can back off to well under 48V.

 

[ICIETDIYEUR]

Hi.

Your explanations are incomplete or even erroneous regarding the voltage of a PSU.

You mention the 3rd audio A7, but the latter has two TPA3255 chips which operate in PBTL mode: this doubles the possible power to be achieved, it is not only the voltage of 52V for this device....

TPA3255 chips operate in PWM: the latter uses a switching frequency which is adjustable by the value of a resistance of the circuits surrounding this chip.

Concerning the power required for the PSU also, your explanation is incomplete: firstly, as indicated by TI's DATASHEET, the maximum admissible voltage for the TPA 3255 chip is 53.5V and not 52V.
The condition which allows this voltage to be used or not will depend on the admissible voltage per circuit of all the circuits concerned which will be 'crossed' by this voltage, not just the chip.
Secondly, with regard to determining the power of the PSU, it is necessary to understand the operating mode of the AMPAPA D1 amplifier:

The TI's DATASHEET indicates for a BTL mode (the one used for this amplifier) and for a distortion of 1% of 150W stereo (i.e. 300W in total) for an impedance of 8 Ohms and 260W stereo (i.e. 520W in total) for an impedance of 4 Ohms.

Taking these indications into account, compare the difference in available power depending on the voltage of the PSU used, always according to the TI datasheet ->

We indeed notice a drop in available power by switching from a 48V power supply to a 36V one BUT we must not lose sight of the fact that the powers previously announced are for a distortion of 1%!

So-called 'Hi-Fi' listening usually considers an acceptable value of 0.1% maximum which we are certainly not going to contest with our desire to obtain devices with the best 'SINAD'.

The power required to 'limit' this distortion will drop drastically if we wish to maintain this desire otherwise what is the point of getting the best DAC which has the best SINAD if you come with your amplifier to significantly reduce this SINAD (and I haven't even touched on the subject of speakers)?

Let's also not stray from the maximum domestic listening level (you were talking about SPL) that we are actually going to use, which is why I suggested Amirm use a clamp meter during these tests, as you would be greatly surprised at how little power we actually use.

Some of you will 'counter' me by talking about the impulse power needed to 'move' the speakers but class D amplifiers are not concerned by this because they do not work at all like class A/B amplifiers which have accustomed us to large transformers and large capacitors.

Also, a class D amplifier has an efficiency of at least 90%, so the power needed for the PSU will be only 10% more than the amplifier will need to reach its maximum power at 0.1% distortion.

This power must be in continuous mode and at a temperature acceptable by the PSU to avoid any regulation phenomenon of the latter and even less to be made safe by one of these devices.

It is at this moment that the choice of the maximum intensity of the PSU comes into play since it will condition the maximum power in continuous mode for a given operating temperature according to its datasheet and indeed according to the type of load which will be associated at the amplifier output (speaker impedance).

It is all these conditions combined that will determine the power of the PSU and not preconceived ideas wrongly peddled everywhere on internet forums.

Now I invite you to observe the measurements taken by Amirm to get a better idea of my comments and/or to ask Amirm for his opinion as well as his explanations if necessary.

In summary, either you understand my words, or you ignore them and enjoy listening to distorted music.
NB: In the Youtube video that I presented, the person indicates a power of 35W RMS for 0.1% distortion.

If you have a clamp meter, push the volume of your amplifier up to 35W RMS to see the SPL obtained in your room (I do not recommend this if you have speakers with 100 or more decibels of output per Watt per meter !)

You will tell us how high you turned your amplifier up before you noticed the sound was too loud.

With all due respect, have a nice day

 

[KL....]

People hi !

@amirm: I modified the part which indicated SMSL A300 by AMPAPA D1 for the 8 Ohms trace ->

View attachment 514664

I also found another review of this amplifier on Youtube which has some pretty interesting comment sections from the author ->

I noted in particular the notion of operating voltage with regard to the heating produced by the SAGAMI's inductors when the amplifier is at rest:
would it not have been wise to use a 36V/10A power supply rather than 48V/5A in this regard ?

NB: The author also seems to mention the 'sizing' of the latter, which also seems to me to be an interesting 'point'...

Discussions are open as well as reflection for designers

Load dependancy mesurements in the Youtube's video ->

View attachment 514666

View attachment 514667

Power VS distortion (Blue line for 8 Ohms and red line for 4 Ohms) ->

View attachment 514670

My thoughts are that it would be interesting, in my opinion, to compare measurements with the original 48V/5A power supply with another series of measurements with a 36V/10A power supply, both for the power vs distortion and the heating felt with the device when it is in IDLE ->

Indeed, the SAGAMI inductances reach 155°F, or 68.3°C while the amplifier is in IDLE

With a 48V PSU, we obtain 9W and with 36V we obtain 6.5W: something to think about because the temperature of the inductors supplied with 36V drops to 122°F, or 50°C, which is much better

Also, we should be able to attenuate the FFT peaks in the low frequencies (at least) by adding capacitors as recommended by TI in its datasheet ->

View attachment 514671

See you later

Thank you @ICIETDIYEUR, Yes, to ask.... would you suggest, for lower/usual listening level with the Ampapa D1, that Higher current availablety (with Stable/Lower Voltage feed, Lower D/N, and lower but cleaner PVDD feed) will be more enjoyable (even desireable, especially as it would be more Load invariant, for low Impedance Loads), although output volume/W will be lower, won't it?

 

[ICIETDIYEUR]

Good morning.

In response to your question, I'll let you discover a new YouTube video to form your own opinion and this does not exclude comments from other members on this subject, of course, it is even desired ->

Here are two graphs where I have reported points which seem important to me knowing that the voltage indicated on them is for a single channel: it will therefore be necessary to multiply the latter by two to obtain the total voltage for a PSU ->

You will note that I have clearly highlighted the THD which is just as important as the SINAD (since the latter depends on it) of the different devices tested on this site and allow you to appreciate to what level this distortion seems acceptable to you for a so-called 'Hi-Fi' device.

You will thus be able to agree on which PSU you will need knowing that its voltage directly conditions the rest temperature of a class D amplifier with PFFB.

 

[NTK]

Amir had previously tested TPA3255 amplifiers with 36 V and 48 V power supplies. The degradation in performance of the 48 V power supply is inconsequential whilst you get ~1.8x in power.

 

[T&T]

Amir had previously tested TPA3255 amplifiers with 36 V and 48 V power supplies. The degradation in performance of the 48 V power supply is inconsequential whilst you get ~1.8x in power.

That A07 Max dated 2019, was the worse sounding tpa3255, after that Fosi v3 mono, 2 Fosi ZA3 mono , now 3e A7 with 2 tpa3255

 

[ICIETDIYEUR]

Hi

I understood perfectly but what do you do with the distortion greater than 0.1% and the heating of the device (with PFFB circuit) when it is in IDLE ?

You will note that in the example you show us, we still see a difference of 5dB more in distortion with the 48V power supply, which is far from being 'insignificant'...

Are we going to use more than 75W RMS per channel during domestic use even at (very) high levels ?

Also I would like to point out that people who think that the 'sound is wider' and with more impact when using a 48V power supply are certainly not using measuring instruments to ensure that they are comparing at exactly the same level, which inexorably leads them to assert erroneous information

That being said, I have been using my AMPAPA D1 amplifier with its 48V-5A power supply for more than 2 hours of listening without interruption, volume at 50, temperature of my room 26.1°C and using a laser thermometer, I only obtain a maximum of 41.4°C which is really very correct and should certainly not worry anyone who knows the operation of this type of electronic circuit and the 'behavior' which follows: this device is very successful

In addition, a signal absence detection function allows the device to go into standby mode after 5 minutes, which completely avoids unnecessary heating of the amplifier

@ T&T: let's compare like with like, why not mention the TOPPING B100 and B200 in this case (although they are not class D amplifiers) ?

 

[john61ct]

The maximum voltage of the CHIP is higher than any built implementation IME.

3e A7 may in practice allow over 52V, but IMO that would foolishly risk damage, hurt longevity.

I was not comparing A7 to D1, but in doing so of course there are other factors besides the higher voltage in its much greater power output

Those relative differences hold across the board whether measured at high distortion or low.

Please stop using the word "intensity" for electricity - if you mean current / Amps, use those terms to avoid confusion.

And "power" only means Watts, not something else.

Having PSU current AVAILABLE more than anyone will actually use does no harm, there are use cases where I want SPL to go to "safe maximum" and I prefer having the overhead and never needing to use it, rather than v/v.

You should also not assume the music is playing indoors.

With all due respect, have a nice day

You also, sincerely...

 

[john61ct]

would you suggest, for lower/usual listening level with the Ampapa D1, that Higher current availablety (with Stable/Lower Voltage feed, Lower D/N, and lower but cleaner PVDD feed) will be more enjoyable (even desireable, especially as it would be more Load invariant, for low Impedance Loads), although output volume/W will be lower, won't it?

No, at lower volumes the amp will never draw high currents from the PSU, in that case 5A is plenty. There is no reason to go to a lower voltage though so long as what is actually delivered is below the maximum rating for the amplifier.

 

[low_art]

• Engineered for True 2.1 Systems: The adjustable 30–200Hz high-pass filter removes low frequencies from the main output, letting your stereo speakers focus on mids and highs with better clarity, while PRE-OUT offers full-range signal to active subwoofers, amplifiers, or active speakers.

At last, a way to protect my 60 year old EMI SPEAKERS and my baby linkwitz without spending 2000 on a denon with decent sinad. Come on , balanced inputs , meters, tone controls...next thing you gonna tell me is that tariffs have been lifted and prices are falling....do u have any bridges for sale??

DOUK RAIDER....May the force be with you.

Which EMIs are you using? I've passed on a couple pair of the DLS 529s (mismatched or bad cosmetically, etc) but I'm intrigued by them.

 

[confucius_zero]

Forgive me if I don't understand but is this +0.5 db deviation (in red) and +1.5 deviation (in green) in high treble the cause of subjective "digital glare" in classD amps? Seems alarming that we're not getting a straight line of gain.

 

[ICIETDIYEUR]

Hello john61ct.

Sorry because I'm using a translator because I'm just a simple Frenchman (no one is perfect).

In addition, for your information, knowledge of electricity and electronics are two fields among many others that I have used in my profession for more than 35 years...

That said, you have obviously not followed (or read) the content of my interventions and are just pointing out that you are 'peddling' some errors: rest assured, you are not the only one because I see the same ones on many other sites.

Also for your personal culture as well as for the readers of this site who will need it, I would like to advise you to watch this next video really very interesting and very complete
available on Youtube ->

Here are also 2 tables to complete my intervention ->

My little idea 'to explore' for our designers of the AMPAPA D1 in order to improve the ventilation of the case while retaining the good idea for accessibility to OPAMPS ->

And finally use this type of potentiometer for volume attenuation rather than chips (NJW1194 or NJW1195) which significantly degrade the final SINAD of the amplifier ->

NB: Amplifiers such as those offered in this format CANNOT work correctly at high powers without a very large number of precautions but also within acceptable levels of distortion coming from a so-called 'Hi-Fi' device: I know what I'm talking about because I have designed several of them and I have experience of the good and bad things to do with this type of device...

Happy reading everyone

 

[DrLex]

Wait! You have Elton John and Eric Clapton filed alphabetically under ‘E’ and not ‘J’ and ‘C’? You…… monster!!!

 

[ICIETDIYEUR]

Hi.

Sorry but I'm just a 'poor Frenchman' as I said above and I have the habit of addressing someone by their first name first followed by their last name besides don't we say 'first name' and 'last name' in English-speaking countries?
There must be a reason for this...
...this is why I make my classification like this.
That said, I do not see any connection with your intervention concerning the subject in question here: the AMPAPA D1 amplifier.

 

[john61ct]

It was a joke.

But yes the surname is usually used for index order, as in the old "phone book"