Topping PA5 II Stereo Amplifier Review

[VVR]

Not sure if it was on ASR or elsewhere, but I read some post saying that the heat generating devices were now out of the potted assembly. So, it shouldn't be any issue with reliability in PA5 II. I personally never turn mine off.
I don't see where the problem with potted devices is. All ICs, high power transistors, and regulators are just that anyway. Once the heat management is correct things work. There may be some Topping proprietary solutions there, but leaving unpotted some miniature SMT components won't make the module more serviceable. I can tell from experience that replacing components smaller than 0603 is not fun... If repair is required it will be direct module replacement. Nobody cares to troubleshoot and repair on component level circuits that cost a couple of bucks to replace.

 

[grogi.giant]

Not sure if it was on ASR or elsewhere, but I read some post saying that the heat generating devices were now out of the potted assembly. So, it shouldn't be any issue with reliability in PA5 II. I personally never turn mine off.

Yes, it was here.

but leaving unpotted some miniature SMT components won't make the module more serviceable. I can tell from experience that replacing components smaller than 0603 is not fun... If repair is required it will be direct module replacement. Nobody cares to troubleshoot and repair on component level circuits that cost a couple of bucks to replace.

The problem is you cannot really buy those modules from Topping.

Replacement modules for the original PA5 are available, because some ridiculously insane people reverse engineered them. Topping decisions and actions made the servicing harder, not easier.

 

[antcollinet]

Not sure if it was on ASR or elsewhere, but I read some post saying that the heat generating devices were now out of the potted assembly. So, it shouldn't be any issue with reliability in PA5 II. I personally never turn mine off.
I don't see where the problem with potted devices is. All ICs, high power transistors, and regulators are just that anyway. Once the heat management is correct things work. There may be some Topping proprietary solutions there, but leaving unpotted some miniature SMT components won't make the module more serviceable. I can tell from experience that replacing components smaller than 0603 is not fun... If repair is required it will be direct module replacement. Nobody cares to troubleshoot and repair on component level circuits that cost a couple of bucks to replace.

It is not about heat damaging the components in the modules. It is the mechanical stresses caused by that heat physically breaking them, or breaking the solder joint to the PCB. Potting can be used but it requires careful consideration of thermal differential expansion - and lots of validation to ensure it is a viable design. Things that topping apparently failed to do.

And IC's, Transistors, regulators etc are not potted. There is either physical space around the silicon where bond out wires connect to the lead-frame, or if additional insulation is needed (as in high voltage components) it tends to be a soft jelly like compound, rather than a hard resin.

 

[VVR]

And IC's, Transistors, regulators etc are not potted. There is either physical space around the silicon where bond out wires connect to the lead-frame, or if additional insulation is needed (as in high voltage components) it tends to be a soft jelly like compound, rather than a hard resin.

Correct, not potted in the sense of poured epoxy. It's soft silicones or space around, no doubt. But do we know how Topping did it? Maybe there is some thought behind it. Anyway, multiple components live under same roof and cope with heat and often even profit point of view thermal feedback. Their first design was poor, but what's the reason for complaining in the case with PA5 II? Are there many known reported failures? Not a rhetorical question, I really want to know.
Too bad if they don't offer replacement modules! Maybe they will. I don't see why not. It cost them close to nothing.
From my repair shop experience I can tell that if a repair will take 2 hours, say at 100$/h shop rates, it will be recommended to the customer to just buy a new 250$ unit. Actually even 1h shop time is probably not worth the trouble on a 250$ unit. I wonder if Topping didn't come to similar conclusion. This is not an expensive luxury item. I doubt that they care to serve the occasional hobbyist who decided to fix their amp at home after the warranty has expired. Under warranty I really hope that they would replace the whole thing with a new one. Their production cost is probably 30$ for PA5, why bother with repairs.

 

[Lukelin]

what's ic? TPA3225?

 

[antcollinet]

But do we know how Topping did it?

Yes, we do. One of our members scraped away the potting and reverse engineered the circuit. So we know exactly what the construction was.

Maybe there is some thought behind it.

Not enough to prevent the failures.

Simple fact is they messed up. They did it for no benefit other than IP Protection. The IP protection is useless, as one person working on their own was able to circumvent it in a few hours. And they created what appears to be an absolutely horrendous failure rate in the first PA5.

And in the process have probably put some people off topping for life. I was a potential future purchaser (when they bring out a higher power class D amp with good numbers). But not now.

 

[Toku]

what's ic? TPA3225?

I think TPA3225 is the mistake of TPA3255. I see the TPA3255 amplifier being displayed incorrectly as TPA3225 on Aliexpress.

 

[VVR]

Yes, we do. One of our members scraped away the potting and reverse engineered the circuit. So we know exactly what the construction was.

On the PA5... Has anyone cared to do it on the working II version? I guess no, because it's working
BTW, for IP protection I would wipe all markings on the components, not pot to hide them. Without the correct opamp and the exactly same pcb layout good luck to get the same performance. Although, audio is forgiving in general.

I was one of the people who were reluctant to put money in Chinese products and quality. But Amir's tests changed my mind. And ASR in general. I had E50 first and I really liked it. Then, after the obvious issue with PA5 was fixed, I decided to give it a try and I'm happy so far with PA5 II. I understand the sentiment though.

 

[Julf]

I was one of the people who were reluctant to put money in Chinese products and quality.

You mean stuff like iphones?

 

[VVR]

You mean stuff like iphones?

... All comes from there nowadays, isn't it? But It's different in the the quality management, parts control etc. At least it was until recently. Now, not so sure. Chinese learn and adapt quickly.
Besides the high volume consumer manufacturing and some components, not everything is made in China. In my current field we have nothing manufactured there. Except the power adaptors and some bare pcb. I guess at some level of know how and IP it's not reasonable to do it cheaply abroad.

 

[TurtlePaul]

Because the PFFB design publicly available: https://www.ti.com/lit/an/slaa788a/slaa788a.pdf. In this context pirating means figuring out what Topping did so that it can be fixed.

The key to the Topping amps isn’t just PFFB. It is almost certainly a composite amplifier with a higher precision opamp responsible for the global feedback loop. Just implementing PFFB with a TPA chip won’t get you these numbers.

 

[Toku]

The key to the Topping amps isn’t just PFFB. It is almost certainly a composite amplifier with a higher precision opamp responsible for the global feedback loop. Just implementing PFFB with a TPA chip won’t get you these numbers.

That's right!
I also think that Topping's unique feedback loop technology is incorporated.

 

[ITJ]

I am on the fence about this Amp. I am in the market for a small 50W amplifier, but previous reliability issues due to heat is putting me off.
Not sure how well it will do with a nominal 6ohm full range speaker. (tweeter, midrange 2x bass)

 

[Julf]

That's right!
I also think that Topping's unique feedback loop technology is incorporated.

I am always curious about "unique feedback loop technology" - there aren't that many ways to do feedback loops, but please tell us more!

 

[Toku]

I am always curious about "unique feedback loop technology" - there aren't that many ways to do feedback loops, but please tell us more!

I'm sure you already know what a PFFB is. PFFB stands for Post Filter Feedback.
Typically, Class D amplifiers are equipped with a low-pass filter on the speaker output, which removes noise and harmonics from the PWM (Pulse Width Modulation) speaker output signal, resulting in a clean audio signal that is sent to the speaker. It will be sent.
Previously, due to a patent from an EU company, the feedback signal of a class D amplifier was extracted before it passed through the output filter. That patent has now expired, but Topping has now extracted the feedback signal after it has passed through the output filter, and the output filter has also been incorporated into the feedback loop, greatly improving the number characteristics.
The characteristics of the output filter of a class D amplifier vary greatly depending on the impedance of the speaker. As a result, when you connect a typical 4-8Ω speaker, the amplifier's overall frequency response fluctuates significantly around 20KHz. However, by using PFFB, this fluctuation is eliminated and a flat frequency characteristic is obtained.

I think Topping adopted PFFB and at the same time devised its own feedback loop. The reason is that the excellent THD+N characteristics of PA5 ll cannot be achieved simply by using PFFB. I don't know what the circuit is like. I imagine that it probably applies feedback to multiple amplification stages. Topping has modularized this part of the circuit and housed it in a small case to prevent it from being imitated by other companies. Furthermore, the inside of this case is filled with resin and has a completely solid structure.

 

[MAB]

I am always curious about "unique feedback loop technology" - there aren't that many ways to do feedback loops, but please tell us more!

Here are the details in case you hadn't seen:

Topping PA5 fix - D01 Module Replacement for everyone

Ive been working on the PA5 for a while now after mine died suddently, like many others https://www.audiosciencereview.com/forum/index.php?threads/poll-for-topping-pa5-owners-only-please.33293/ Most of the PA5 units develop a crackling, hissing or just go a little silent or completely dead on...

www.audiosciencereview.com

 

[howard416]

I'm sure you already know what a PFFB is. PFFB stands for Post Filter Feedback.
Typically, Class D amplifiers are equipped with a low-pass filter on the speaker output, which removes noise and harmonics from the PWM (Pulse Width Modulation) speaker output signal, resulting in a clean audio signal that is sent to the speaker. It will be sent.
Previously, due to a patent from an EU company, the feedback signal of a class D amplifier was extracted before it passed through the output filter. That patent has now expired, but Topping has now extracted the feedback signal after it has passed through the output filter, and the output filter has also been incorporated into the feedback loop, greatly improving the number characteristics.
The characteristics of the output filter of a class D amplifier vary greatly depending on the impedance of the speaker. As a result, when you connect a typical 4-8Ω speaker, the amplifier's overall frequency response fluctuates significantly around 20KHz. However, by using PFFB, this fluctuation is eliminated and a flat frequency characteristic is obtained.

I think Topping adopted PFFB and at the same time devised its own feedback loop. The reason is that the excellent THD+N characteristics of PA5 ll cannot be achieved simply by using PFFB. I don't know what the circuit is like. I imagine that it probably applies feedback to multiple amplification stages. Topping has modularized this part of the circuit and housed it in a small case to prevent it from being imitated by other companies. Furthermore, the inside of this case is filled with resin and has a completely solid structure.

Not having gone in depth on the PA5 magic, I assume it's due to using a crap-ton of PFFB and then restoring the lost gain with an ultra-low-noise preamp stage.

 

[Julf]

I'm sure you already know what a PFFB is. PFFB stands for Post Filter Feedback.

Yes, as pioneered by Bruno Putzeys at Philips, and as seen in the Hypex and Purifi amps, right? So hardly "unique", I would say.

 

[Lukelin]

I think TPA3225 is the mistake of TPA3255. I see the TPA3255 amplifier being displayed incorrectly as TPA3225 on Aliexpress.

yep, I mean TPA3255. same as PA7?

 

[grogi.giant]

There are TPA3220, 3221, 3245 and 3244 thought. In addition to 3250, 3251 and 3255.