Topping PA3s Review (Desktop Amplifier)

[levanter]

If it is the same as I observed, it;s not anything close to a 50/60Hz. It's more like a broadband white noise centered around the 2.5k region. Only shows up on sensitive drivers (in my case, 12" waveguides).

Infineon calls it out in their design whitepapers and I mentioned it in the 2nd paragraph here.

Topping PA3s Review (Desktop Amplifier)

Does anyone have a channel misbalance under 1/4 volume. It seems that right channel is very slow to come up to the match volume of the left channel. I worked around it by lowering the volume of the DAC but i was wondering if there is a way to fix this permanently? Another thing i noticed with...

www.audiosciencereview.com

Signal needs an analog ground ref near the chip or else stray capacitance causes injected noise. Topping uses the mode without compensation in SE, so the chip does not sample AGND. Just acts as a transistor amp like proper old school design instead of using basic on-die filtering and signal conditioning.

ETA: IIRC I want to say that there was also an op-amp in an inverting output mode in SE operation which took signal after input. So the poor pot can strike twice there? I admit, been a hot minute since i was really inside the amps. Been sitting next to my speakers just... doing amp things.

Long story short, the board deviates a pretty havily from the infineon whitepapers and infineon;s own eval options so there were def cuts to hit a price point and keep a heavy chunk of profit for intl sales.

That said, the infineon eval kits are not worth the price diff because (atr least back when i was researching) I want to say they were 25%+ more and no PSU and no case?

Yes! It is exactly what you described. White noisish hiss in mid-high frequency region.
I don't think it's anything external because it comes even without any interconnect cables connected. A little bit of noise like that is common among most AB class I've experienced, but not really so for decent D class amps.

I thought it was weird that it only comes when unbalanced input is selected. On the contrary my recently bought Topping B100 monoblocks are very silent from both inputs.
Thanks guys for your valuable opinions.

 

[Skyra]

As it is well known - and was mentioned before- , this amp is a particularly good source of, and also very sensitive to high frequency noise. This is the backlash of the filterless design. The thing works as an effective low frequency RF noise transmitter, and in unfortunate cases the radiated noise is being picked ud and fed back by the connecting cables to the input of the amp.
Mine had exactly the same "white noise-ish" problem, but after inserting a 6.8uH ferrite encased filter coil (Coilcraft) to each speaker line (as close as possible to the case of the amp) it disappeared completely. And no, the coils won't have any negative effect on its sound (at least in my amp I didn't detect any).

 

[racingpht]

Yes, this amp CAN be pretty noisy. But it's tricky. Sometimes it's a fine AMP.

In one case, it seems happen when the load is less inductive. I drove a full-range drive with very flat impedance(motor with shorting ring), it makes HF noise when the driver is playing pure bass note(I put a minidsp LP filter), but stay quiet when there's no signal. It almost sounds like the driver's fault and only when you switch AMP you know it's from the AMP. After putting an air-coiled inductor paralleled with a resistor attenuating HF energy, the AMP works fine. It almost seems the AMP expects relatively strong inductive load.

In another case, again with active speaker, when I add some ferrite core filter near the woofer section(but not the amp) with ~50cm cable I notice the AMP is picking up strange noise again.

I believe last but not least, when I use PA3S to drive a HF section(capacitor protected tweeter) , the AMP is not happy and make noise of you. I almost blaming my MiniDSP2x4 but finally found it's the AMP.

It's such an annoying little amp. Sometimes it did work fine. I hope I'll try some ferrite coil mod and hope it will work. Hopefully something that fits inside the case.

This is the perfect example that some AMP might be built for benchmarks rather than real-world use cases, probably with inexperienced designer.. To be fair, I have many other topping AMPs and PA5II/PA7 all works trouble-free.

 

[LTig]

Yes, this amp CAN be pretty noisy. But it's tricky. Sometimes it's a fine AMP.

In one case, it seems happen when the load is less inductive. I drove a full-range drive with very flat impedance(motor with shorting ring), it makes HF noise when the driver is playing pure bass note(I put a minidsp LP filter), but stay quiet when there's no signal. It almost sounds like the driver's fault and only when you switch AMP you know it's from the AMP. After putting an air-coiled inductor paralleled with a resistor attenuating HF energy, the AMP works fine. It almost seems the AMP expects relatively strong inductive load.

In another case, again with active speaker, when I add some ferrite core filter near the woofer section(but not the amp) with ~50cm cable I notice the AMP is picking up strange noise again.

I believe last but not least, when I use PA3S to drive a HF section(capacitor protected tweeter) , the AMP is not happy and make noise of you. I almost blaming my MiniDSP2x4 but finally found it's the AMP.

It's such an annoying little amp. Sometimes it did work fine. I hope I'll try some ferrite coil mod and hope it will work. Hopefully something that fits inside the case.

This is the perfect example that some AMP might be built for benchmarks rather than real-world use cases, probably with inexperienced designer.. To be fair, I have many other topping AMPs and PA5II/PA7 all works trouble-free.

Seems the amp is unstable with your speakers. I would not use it any longer, who knows when it starts to oscillate and kills the speakers...

 

[Solveit]

Differential input is always best with this MA12070 chip.
RCA is way to noisy with EMI pick up.

 

[racingpht]

As it is well known - and was mentioned before- , this amp is a particularly good source of, and also very sensitive to high frequency noise. This is the backlash of the filterless design. The thing works as an effective low frequency RF noise transmitter, and in unfortunate cases the radiated noise is being picked ud and fed back by the connecting cables to the input of the amp.
Mine had exactly the same "white noise-ish" problem, but after inserting a 6.8uH ferrite encased filter coil (Coilcraft) to each speaker line (as close as possible to the case of the amp) it disappeared completely. And no, the coils won't have any negative effect on its sound (at least in my amp I didn't detect any).

Unfortunately I tried both first/second order pass filter but they make no difference to me. It was a 12uH + 0.47uf + 12uH filter in the picture.

Seems the amp is unstable with your speakers. I would not use it any longer, who knows when it starts to oscillate and kills the speakers...

Thanks for your input. So far my issues are all noise-related and I have no evidence to say it's unstable..

 

[racingpht]

For anyone who's curious, I recorded some strange noise made by my PA3s. In the video I was playing a 60hz sine wave, the amp generates some noise at certain volume. If you pay attention you will notice that triggering the noise on/off are at slightly different volumes, as if the amp is transitioning into and out of different "states" with a threshold of a few dbs.

Very funny, I wonder if that's the "smart" features mentioned in MA12070 white paper.. such as "The power management algorithm dynamically adjusts switching frequency and modulation to optimize power loss and EMI across the output power range."

But I never noticed the same problem on my MX3s which also uses MA12070.

BTW: The extra noise are in -80db range compared to the 60hz signal. Regular measurements might say it's inaudible. This is a nice example that something might measures fine but you can easily hear something wrong Speaker driver distortion here is -20db but still sounds perfectly fine which is also kind of shocking.

 

[racingpht]

For anyone who's curious, I recorded some strange noise made by my PA3s. In the video I was playing a 60hz sine wave, the amp generates some noise at certain volume. If you pay attention you will notice that triggering the noise on/off are at slightly different volumes, as if the amp is transitioning into and out of different "states" with a threshold of a few dbs.

Very funny, I wonder if that's the "smart" features mentioned in MA12070 white paper.. such as "The power management algorithm dynamically adjusts switching frequency and modulation to optimize power loss and EMI across the output power range."

But I never noticed the same problem on my MX3s which also uses MA12070.

BTW: The extra noise are in -80db range compared to the 60hz signal. Regular measurements might say it's inaudible. This is a nice example that something might measures fine but you can easily hear something wrong Speaker driver distortion here is -20db but still sounds perfectly fine which is also kind of shocking.

Since I can't edit my post anymore, I want to add a bit more information(probably as someone else already mentioned):

The strange noise is gone when driving with XLR input in the same test environment.

So this amp has a pretty noisy RCA input. The filterless design seems not the one to be blamed. You should always use XLR input with this amp. minidsp2x4 or 2x4HD is not the a good pair for this amp but on PA5 everything seems fine.