Review and Measurements of Purifi 1ET400A Amplifier

[Audiocrusader]

I assumed that the same requirements in traditional amplification would also be considered with class D. Well, it turns out that it has not been so, hence my surprise and anger with the 1200AS2.

Reading through this thread, Bruno clearly stated that any phase shift that turns up in the measurements is only a bi-product of the errors produced by the measurement gear when preforming those tests. They aren't actually there in real life from the gear, and there's no audible consequences. So what makes you think that someone who has absolutely no hands on experience listening to class d amps, or audio measurement gear whatsoever, should have the confidence to override the opinion of the greatest class D pioneer in the industry? You should really be ashamed of yourself for attempting to mislead so many novices to the hobby with your senseless rhetoric across so many forums. You're right up there with George.

 

[maty]

I know what he said but I do not share it from day one. I have been silent these months waiting for others to realize but they have been silent. The fallacy is so obvious... I am sure he does not even believe it.

 

[Audiocrusader]

I know what he said but I do not share it from day one. I have been silent these months waiting for others to realize but they have been silent. The fallacy is so obvious... I am sure he does not even believe it.

2 great options for you in order to obtain the confidence to continue posting about class D amps:

Option 1:

1:Take a break from the forums
2: Purchase some class D amps of all types
3:Listen to them with all your best music for many months.
4: Share your subjective opinion on subjectivist forums

Option 2

1: Take a break from the forums
2: Purchase some class D amps of all types
3: Purchase or rent an Oscilloscope, high end audio analyzer, various test probes, active and passive loads
4: Take a 3-6 month course on how to use it all to the best of it's capability
5: Spend another 3-6 months testing and evaluating these class D amps
6: come here and share your findings.

 

[maty]

Option 3

Do not tell me what I have to do, especially if you do not know my knowledge.

 

[Audiocrusader]

Option 3

Do not tell me what I have to do, especially if you do not know my knowledge.

You can choose to continue to post. but don't expect any credibility or respect. If you want to continue to look like a fool it's your choice.

 

[maty]

You can choose to continue to post. but don't expect any credibility or respect. If you want to continue to look like a fool it's your choice.

It is inevitable, in the end personal disqualifications always appear in the open Internet.

- The End -

 

[renevoorburg]

Btw, did you notice that https://www.purifi-audio.com/ now has a shop area? It only shows a woofer, and the shop doesn't seem to be working yet.

 

[g29]

Btw, did you notice that https://www.purifi-audio.com/ now has a shop area? It only shows a woofer, and the shop doesn't seem to be working yet.

October has arrived !!!

 

[KSTR]

Extracting the phase is easy, an amp is a minimum phase system. Phase response is directly linked to magnitude response and vice versa.
I'd simply model a filter function in LTspice so that the mag response follows what is given in the corresponding plot plus that the 10kHz step (square) response looks the same. Phase reponse of this model will be close to the real thing.
Anyway, as long as channels are matched, phase response of an amp doesn't matter as it is totally neglegible compared to phase responses of sources (minimum phase DAC reconstruction filters, anyone) and notably, speakers.

For the 10kHz square shape sans AP's filter, this can easily constructed as well. Amir's plots shows the switching frequency is 20dB down in level for the signal output voltage corresponding to 5W@4R, which is 6Vpk (~4Vrms). Hence the 500kHz component is about 400mVrms and rather sinusodial, a very typical value.

Now this is an investigation that should be super-clear to anyone with at least some half-baked electronic engineering knowledge... @georgehifi, have you been just too lazy or are you actually uncapable to do this?

 

[Joachim Herbert]

Now that's a euphemism if I've ever heard one. Personally I'd much rather drive a Porsche at any age than stare at a piece of audio gear I've over paid for and can't hear anyways. And likely has the effect of scaring chicks away rather than attracting them like the Porsche would.

Its potential performace aka measurements vs. speed limits.

Driving the Porsche is fun no matter what.

Never tried to attract the other sex with my audio.

Never had to buy a Porsche for that purpose.

 

[Audiocrusader]

Now this is an investigation that should be super-clear to anyone with at least some half-baked electronic engineering knowledge... @georgehifi, have you been just too lazy or are you actually uncapable to do this?

He is simply another forum crusader with absolutely no knowledge of how audio gear works. He made some investments in GAN technology for class D amps and is hoping for a fruitful ROI. So when Bruno said that they found a more effective way of dealing with deadtime using conventional Mosfets, he decided to ramp up his crusade into high gear. But the sad thing is it's just so weak. Because only like 0.1% of the real market reads these forums anyways. He can crusade till he's blue in the face and never make it anywhere.

 

[Audiocrusader]

- The End -

 

[ti33er]

Argh this is turning into an argumentative rambling on, frustration thread ...the one on DIY for the nc500 for this reason got shut down as a result

Bruno is not going to read all of this and not I doubt comment here also :-(

Think we need a moderator to clear out all the irrelevant banter (please)

 

[ahofer]

So, what are people suggesting are the audible effects of these alleged phase and switching frequency measurements? (Stipulating, for the moment, contra Bruno, that the phase anomaly is not a measurement artifact). How do my speaker wires and speaker react, in the audible band, to the high frequency signal presence? I do believe we’ve covered that in a few other places and concluded there should be no audible artifacts, but perhaps we missed something.

And, of course, inclusion of Class D options in some good double blind tests could show us for sure. As I’ve said elsewhere, I would be happy to join funding for a bet that NCs and this new module cannot be distinguished from linear class AB designs in blind testing.

 

[March Audio]

I know what he said but I do not share it from day one. I have been silent these months waiting for others to realize but they have been silent.

The fallacy is so obvious... I am sure he does not even believe it.

You are wrong and Bruno is correct. Others haven't realized your point of view because it is incorrect. Brunos point re phase is actually an incredibly simple technical concept and quite obvious when you think about it.

Your comments there essentially question Brunos honesty along with his technical ability. Considering your continued penchent for spamming multiple threads with uninformed nonsense despite multiple warnings, I am of the opinion you should be given a permanent holiday from the forum.

@Thomas savage ?

 

[March Audio]

Reading through this thread, Bruno clearly stated that any phase shift that turns up in the measurements is only a bi-product of the errors produced by the measurement gear when preforming those tests. They aren't actually there in real life from the gear, and there's no audible consequences. So what makes you think that someone who has absolutely no hands on experience listening to class d amps, or audio measurement gear whatsoever, should have the confidence to override the opinion of the greatest class D pioneer in the industry? You should really be ashamed of yourself for attempting to mislead so many novices to the hobby with your senseless rhetoric across so many forums. You're right up there with George.

Sorry to pick but it's not a measurement error per se, but if the phase shift is linear it just produces a fixed time delay across the audible frequency range. This is obviously of no concern.

However you are absolutely correct about uninformed people derailing threads with garbage.

 

[ahofer]

Well, I certainly hope I haven't contributed to any derailing, either with questions on measurements or the points on BVI persons,and I never had the intention of being "politically correct". I apologize if it came off that way.

However (sorry?), since Class D seems like an incredible leap forward in efficiency, convenience, and power delivery, I'm very interested in chasing down the remaining objections to it to determine if there is any there there. I'm not qualified to opine on Maty's/George's potential/theoretical objections, but I still think it would be useful to know whether any of the artifacts of switching *possibly* are make their way into audible artifacts (through some transmission/translation mechanism from the non-audible spectrum). My priors (and my own subjective, and high frequency impaired comparison) say that they do not, but chasing down the details seems worthwhile, if it can be done scientifically.

We already have more than enough data to show that relatively linear class A/AB (including some not-so-linear tube designs like Futterman and VTL) simply aren't audibly distinguishable from one another in blind, level-matched tests. It would be good to get some Class D into the body of evidence. If my priors are right, then we can put it into the "but thousands of audiophiles hear it subjectively" non-evidence slop bucket. If not, I'm sure Bruno or someone else has a new design coming out....

 

[March Audio]

Well, I certainly hope I haven't contributed to any derailing, either with questions on measurements or the points on BVI persons,and I never had the intention of being "politically correct". I apologize if it came off that way.

However (sorry?), since Class D seems like an incredible leap forward in efficiency, convenience, and power delivery, I'm very interested in chasing down the remaining objections to it to determine if there is any there there. I'm not qualified to opine on Maty's/George's potential/theoretical objections, but I still think it would be useful to know whether any of the artifacts of switching *possibly* are make their way into audible artifacts (through some transmission/translation mechanism from the non-audible spectrum). My priors (and my own subjective, and high frequency impaired comparison) say that they do not, but chasing down the details seems worthwhile, if it can be done scientifically.

We already have more than enough data to show that relatively linear class A/AB (including some not-so-linear tube designs like Futterman and VTL) simply aren't audibly distinguishable from one another in blind, level-matched tests. It would be good to get some Class D into the body of evidence. If my priors are right, then we can put it into the "but thousands of audiophiles hear it subjectively" non-evidence slop bucket. If not, I'm sure Bruno or someone else has a new design coming out....

These are all perfectly reasonable questions. There is a clear differentiation between someone wishing to consider further information and have a discussion regarding a subject and some of the ill-informed argument causing dogma that destroys threads.

With the latter it doesn't matter what information is presented, it's a waste of time. The mind is made up and won't be changed.

 

[March Audio]

So, what are people suggesting are the audible effects of these alleged phase and switching frequency measurements? (Stipulating, for the moment, contra Bruno, that the phase anomaly is not a measurement artifact). How do my speaker wires and speaker react, in the audible band, to the high frequency signal presence? I do believe we’ve covered that in a few other places and concluded there should be no audible artifacts, but perhaps we missed something.

And, of course, inclusion of Class D options in some good double blind tests could show us for sure. As I’ve said elsewhere, I would be happy to join funding for a bet that NCs and this new module cannot be distinguished from linear class AB designs in blind testing.

The phase shift with the Purifi module is insignificant, less than 1 deg at 20kHz. Also If phase was a significantly audible problem we would find traditional passive speakers and their significant phase anomalies unlistenable.

How do your speakers and wires react? They do nothing. Well not quite true, the tweeter will dissipate maybe a few mW in heat, but nothing else.

 

[kazsud]

Just got the new Purified Nords. I did not get a chance to listen (and with my current work load it may take a while ...) but I did manage to open them up and take some pictures.

What is most interesting, is that the Purifi modules show some 20 deg celsius lower working temps than the Ncore ones. They still run hot at idle, but not just as hot.

Here are two amps side to side:

Purifi top
NCore NC 500 bottom

View attachment 34873

This pic doesn't is not accurate enough, but the NCore modules get as hot as 68'C (154 degrees Fahrenheit). By far the hottest component is the coil:

View attachment 34874

Purifi on the other hand maxes out at 50'C (all temps at idle with the tops open), with the hottest components beeing that vertical daughterboard an a chip next to it.

PSU transformer, Sparkos discrete voltage regulators on the input board also reach 50'C.

View attachment 34875

Purified Nord inside:

View attachment 34876

Purifi board (here you can see the vertical doughterboard that gets so hot and the long chip next to it; the coil on the other hand is warm, but not hot to the touch):

View attachment 34877

What speakers will you be pairing them with?