Topping PA5 Review (Amplifier)

[sarumbear]

The pinning isn't allowed in the EU due to IEC 60320, despite it features de C E-sign which is curious enough, we won't see that in EU, only for the US allowed.

What do you mean by pinning?

 

[Walter]

Sorry, if it's a stupid question, but why does it give you highest SNR in this case? Could not the pre also be clipping if you (as in this case) drive it to it's max...?

Usually, the amp has the most noise, so you want to keep the gain on the amp as low as possible. The PA5 is so quiet that this will not be much of an issue, but it is still a good technique because it let's you utilize nearly the full range of the remote controlled volume of the preamp.

 

[Robbo99999]

Thank you so much Amir for the review. This is a very important project for me and I'm so glad the hard work paid off.

Any plan for bigger one

In the work.

Well I'm not a buyer in this space, but this sounds very exciting!

 

[Rottmannash]

The PA5 is a great, class-leading achievement. But I won't be selling my VTV Purifi, (which is stereo with VTV buffers). The PA5's SINAD is great but not as stunningly low as the Purifi (with the Eval-1 buffer anyway). Personally I'd be happier not see the higher-order spike in the PA5, albeit they are likely inaudible.

Probably the PA5 actually has enough power for me, but don't see the extra juice as doing any harm. In my case the money is spent so the price differential is irrelevant.

Topping sees to like the small form factor which might explain the external power supply, (though maybe it reduces noise too). I would like them to do some thing closer standard format, say even 17"/430mm wide, with internal supply.

View attachment 168611

And the power output difference...

 

[Billy Budapest]

In another thread, I had hypothesized that Topping was not focusing its considerable engineering chops on speaker amplifiers. I stand corrected!

 

[vkvedam]

Ahh, why did they omit the remote

 

[pjug]

I was looking at the specs for the Hypex NC222MP and I have to think it was at least one of the targets here. The PA5 slightly beats it in power and significantly in noise and distortion, losing only in the less linear THD to frequency, in a cheaper and smaller package with a volume knob--although I'm not sure how much smaller it really is if you include the power supply. Well done, JohnYang.

Do you mean NC122MP? I think this is a great comparison and to me they do look very similar. Edit: scratch question I think answered later in thread.

@JohnYang1997 Nice work. Do you expect you will be bringing the power supply inside the box as you go up in power for future devices?

 

[Tangband]

I was able to achieve 0.00074% THD on TPA3255 at 1W using standard factory recommendations for PFFB. I would imagine that with some more optimizations, one can achieve 0.000524% at 5W. The distortion vs frequency plot measured here is a fingerprint that matches the TPA32XX with PFFB. Look at Fig 70 in the PFFB white paper from TI. 0.0015% THD spec from Topping is consistent with this plot.
View attachment 168622

Yes, It has already been shown and tested at audiosciencereview that the tpa3255 can have really good performance.
Read here:

JL ELectronics Sylph-D200 Amplifier Module Review

This is a review and detailed measurements of the JL Electronic's (JLE) Sylph-D200 stereo amplifier module built on top of TI's TPA3255 amplifier IC. Depending on the configuration you pick, the module costs US $130 to $170. You can read more about it here...

www.audiosciencereview.com

 

[Walter]

Do you mean NC122MP?

Yes, thanks for catching that. I've corrected the typo.

 

[Tangband]

Comparison between topping pa5 and JLE , JLE using tpa 3255 with PFFB

 

[daniboun]

I was able to achieve 0.00074% THD on TPA3255 at 1W using standard factory recommendations for PFFB. I would imagine that with some more optimizations, one can achieve 0.000524% at 5W. The distortion vs frequency plot measured here is a fingerprint that matches the TPA32XX with PFFB. Look at Fig 70 in the PFFB white paper from TI. 0.0015% THD spec from Topping is consistent with this plot.
View attachment 168622

Hi,

We could make a comparison by analogy ... Allo has already shared the measurements of its future TPA325X. The numbers were shared by the manufacturer (Allo's founder), so far it remains a hypothesis ....

I have already revealed some info here but I think Allo has to worry if the results are not kept.
What I can tell you is that the new Allo; to achieve the announced measures has shown fine innovations in the implementation of the TPA325X such as : new OP amps / Air core inductors / new ultra low noise PSU / Fully balanced etc

Allo TPA3255.

THD+N 5W , 1Khz 0.00014
THD+N 100W 20-20Khz 0.0097
THD+N 10W 20-20Khz 0.0091
SNR : about 117DB


I can't wait to see the final match)

 

[PhnomPenhPaul]

Thank you Amir for the review. Well done @JohnYang1997, you've excelled yourself with this one! I hope you sell a boat load of these [figuratively and literally!] although I'll wait to see what else [as per your post # 253 above] you have in store for us [hopefully with a combined power / input switch as per the PA3 to match the the D90SE and Pre90 rather than the toggle switches of the A90].

 

[Tangband]

The JLE amplifier with tpa 3255 and PFFB .
The criticism PMA has against the chip with load dependent issues in the treble dissapears with PFFB.

 

[pma]

I was able to achieve 0.00074% THD on TPA3255 at 1W using standard factory recommendations for PFFB.

Not only you (see below), btw would you show YOUR and not the datasheet plot?
Not difficult to get it at 1W, but to keep it at high power. And how about low noise? You get it with TPA? I do not think so.

TPA 3255 noise from datasheet

 

[Tatteredmidnight]

Help me out here, but I’m a bit perplexed by the max input and gain choices. The suggested DAC is the E50, but it’s 4v nominal for TRS output, which would clip the input on this amp. 2.6v is over what we usually see for RCA and under what we see for balanced. Would it not be possible to have user-selectable gain like a HP amp for use with a balanced DAC, so one could take advantage of the full volume range provided by the source. Does the volume knob actually attenuate the input signal pre-buffer to accomplish this?

 

[Billy Budapest]

Exceptionnal.
Thank a lot Amir.

Was there any audible hiss in the speaker? It is a common issue with those products. eg. Aiyima A07 produces hiss but SMSL AD18 not.

The extremely low noise measurements indicate this should not be an issue.

 

[pma]

which would clip the input on this amp. 2.6v is over what we usually see for RCA

Why? If the pot is at the input then you can have any input voltage. Ask @JohnYang1997 to clarify. However he is not telling much about design (contrary to Bruno Putzeys e.g.).

 

[Tatteredmidnight]

Why? If the pot is at the input then you can have any input voltage. Ask @JohnYang1997 to clarify. However he is not telling much about design (contrary to Bruno Putzeys e.g.).

This is what I’m trying to understand, I believe he volume control could be implemented in a number of ways and I just don’t know enough to make an educated inference. The rest of my question touches on this but maybe it wasn’t clear enough.

 

[tw99]

@pma Do you have your own theory about which chip amp they're using then ? Aside from your scepticism about the achievable performance with the TPA device, a lot of the other things seem to match (switching frequency, graph shapes..) I'm sure we can agree that they haven't designed their own chip as that would be way beyond their capabilities as a company ?

 

[Tks]

... where do we go from here?

More power, better IO using XLR perhaps and such (gonna need to be bigger obviously)

Aside from that, just take Benchmarks current half-decade+ old crown from it, as it still reigns on paper.

@JohnYang1997 Congratz btw

Was this the Benchmark competitor you teased what seemed to be all those years ago?