Purifi 1ET400A new range of class D amp modules

[March Audio]

39 minutes ago, ghost4man said:

George, 

 

I don't quite get what Alan is saying about there being no phase shift. If I'm wrong or taking things out of context then please advise but I can't see the accuracy in stating that there is no phase shift. This is what I'm reading as coming from Bruno

 

Otherwise the linear phase shift corresponding to the distance between the speaker and the mic completely clouds the picture. In the case of the 1ET400 module it’s just under 1 degree at 20kHz

 

 

 

If being a little pedantic  , yes it was a factually incorrect statement.  I should have said "less than 1 degree".  My opinion is that that is of no audible consequence whatsoever, which is what I should have said.

 

You may feel otherwise, but please explain in technical detail why you think under 1 degree of phase shift at 20kHz is an audible problem.

 

Then explain why a speaker which has phase all over the place, or that valve amp shown earlier, somehow is something to be ignored or isnt a problem.

Edited September 8, 2019 by March Audio

[aussievintage]

42 minutes ago, March Audio said:

To take some rough numbers, it if you have a circuit that has a 0.2 degree phase shift at 200Hz, 2 degrees at 2kHz and 20 degrees at 20kHz, that’s the same as saying it has “0.001 degree per Hertz” phase shift. That’s another way of saying that the whole signal is simply delayed by 2.8 microseconds. If you plot phase shift on a linear frequency scale that’s immediately obvious because you get a straight line. Of course a simple delay doesn’t change the sound. It’s literally the same as starting your music a few microseconds later.

That makes sense.   Here  https://lpsa.swarthmore.edu/BackGround/TimeDelay/TimeDelay.html   the provide some nice example plots

 

 

and log plots

 

[ghost4man]

1 hour ago, March Audio said:

 

If being a little pedantic  , yes it was a factually incorrect statement.  I should have said "less than 1 degree".  My opinion is that that is of no audible consequence whatsoever, which is what I should have said.

 

You may feel otherwise, but please explain in technical detail why you think under 1 degree of phase shift at 20kHz is an audible problem.

 

Then explain why a speaker which has phase all over the place, or that valve amp shown earlier, somehow is something to be ignored or isnt a problem.

Alan, 

 

You are misrepresentating my comment by virtue of a straw man fallacy which I know you would be familiar with as you are not stupid. 

 

I did not say that it would. You asserted that there was NO phase shift. However small it is that was factually incorrect. You've acknowledged that. 

 

Does or would that make an audible difference is something that I could not answer. 

 

My gut at the moment tells me that the implementation of GaN technologies must surely be heading in the right direction given that Mola and Technics have invested so heavily in their products to produce what would arguably be considered hi end gear for Class D at least. 

 

I think what would be of value is a direct comparison with what you have on the bench and say a Mola or Technics. 

[georgehifi]

1 hour ago, March Audio said:

You may feel otherwise, but please explain in technical detail why you think under 1 degree of phase shift at 20kHz is an audible problem.

It's not 1 degree, even Bruno said it's 20 degrees at 20k, Jeeze!! And I doubt that, I say its even more if it's a 500khz switching frequency your trying to eliminate! The ICE Power is 70 degrees phase shift at 10khz!!!!!!!!

 

 

 

 

If not and you set the filter higher  to get lower phase shift figures your then letting through way too much to the speaker, I've seen Wilson Watt 8's  with blued tweeter voice coils because too much of the 700khz  switching frequency was let through to the speakers (you don't hear it, it just cooks the tweeter voice coils in silence)  

I'll say it again, this is why GaN Tech is so good because you can have 1.5mhz switching frequency, and the output filter can also be higher and filter it all out and not have any phase shift at 20khz.

 

End of story again, as it seems  with over 80 posts from you on this thread and nothing else, it's just as I said on the 4 page. "this thread is taking a certain advert feel, which I don't think is right, so I won't be fueling the fire anymore."

Edited September 8, 2019 by georgehifi

[aussievintage]

2 minutes ago, georgehifi said:

Bruno said it's 20 degrees at 20k,

wasn't that just an example, where he said if you have a circuit that has a 0.2 degree phase shift at 200Hz, 2 degrees at 2kHz and 20 degrees at 20kHz,

 

 

 

 

[Wobba]

12 minutes ago, ghost4man said:

Alan, 

 

You are misrepresentating my comment by virtue of a straw man fallacy which I know you would be familiar with as you are not stupid. 

 

I did not say that it would. You asserted that there was NO phase shift. However small it is that was factually incorrect. You've acknowledged that. 

 

Does or would that make an audible difference is something that I could not answer. 

 

My gut at the moment tells me that the implementation of GaN technologies must surely be heading in the right direction given that Mola and Technics have invested so heavily in their products to produce what would arguably be considered hi end gear for Class D at least. 

 

I think what would be of value is a direct comparison with what you have on the bench and say a Mola or Technics. 

By Mola, I assume you’re referring to Mola Mola Kaluga amps? If so, can you please provide a reference/link to show that they are using GaN’s. 

Edited September 8, 2019 by Wobba

[aussievintage]

3 minutes ago, aussievintage said:

wasn't that just an example, where he said if you have a circuit that has a 0.2 degree phase shift at 200Hz, 2 degrees at 2kHz and 20 degrees at 20kHz,

 

Yeah he definitely said  "In the case of the 1ET400 module it’s just under 1 degree at 20kHz"

[georgehifi]

22 minutes ago, aussievintage said:

Yeah he definitely said  "In the case of the 1ET400 module it’s just under 1 degree at 20kHz"

Impossible then the output filter would have to be far higher order and would burn out very quickly with this kind of power, that why the A.P. test filter is high order passive but can only take milliwatts when test are done using it.

 

Mark Levinson with their $80k No. 53 Class-D monoblocks tried to series up two 2nd order filters ones to get a steeper slope to null out the 600khz switching frequency but it sounded rubbish.

Yes that's just one of the monoblocks and those 4 chokes are part of the output filters

 

  

 

They actually managed to kill off most of the switching noise as can be seen by the 10khz square wave without AP filter no switching noise ringing across the top of it, just one small over shoot.

 

 

 

Fig.2 Mark Levinson No.53 Reference, small-signal 10kHz squarewave into 8 ohms (no AP low-pass filter).

 

 

Edited September 8, 2019 by georgehifi

[andyr]

49 minutes ago, Wobba said:

By Mola, I assume you’re referring to Mola Mola Kaluga amps? If so, can you please provide a reference/link to show that they are using GaN’s. 

 

Are you a slacker, W?    Why not get Dr Google to look it up, yourself?

 

Andy

 

[Wobba]

I did  And couldn’t find any reference  to Mola Mola Kaluga amps using GaN tech. They use the same Hypex NC1200 & SMPS1200 modules as March Audio P701 amps.

[kukynas]

@georgehifi first of all, did you read entire thread from Amir and consequent 22 pages of discussion with further measurements including Q&A with Bruno P. I attached into my first post? coz if not then I would highly recommend it before we discuss any technicalities further, I believe it would save time and arguments... and I would recommend it to everyone reading this thread...

 

second re your repeating point about fried tweeters, if you've read any of Bruno's comments regarding this topic you know by now that in good class D design even without GaN there is no way to cook tweeters with amp switching frequency coz there's not enough content and energy to do so unless there's a fault elsewhere, I once experienced what might be similar effect by listening to faulty DAC (high amount of oscillation leaking into audio band confirmed later by the scope) connected to the class D amp, I couldn't hear it but I could feel it by fatigue sound, since than after fixing the DAC zero issues, would I blame amp for it? not saying it was your case but without knowing what was happening under the scope is hard to judge, and even than we don't know if that specific amp wasn't faulty ...     

[aussievintage]

1 hour ago, georgehifi said:

Impossible then the output filter would have to be far higher order and would burn out very quickly with this kind of power, that why the A.P. test filter is high order passive but can only take milliwatts when test are done using it.

 

Nevertheless, it IS what he said, and that's all I was saying.

[andyr]

1 hour ago, Wobba said:

I did  And couldn’t find any reference  to Mola Mola Kaluga amps using GaN tech. They use the same Hypex NC1200 & SMPS1200 modules as March Audio P701 amps.

 

Your post prompted me to do the same, W - and, yes, all any review says is that the Kaluga uses Hypex NC1200 modules.    I doubt whether Bruno has replaced the output transistors n this module ... so it would seem to be a myth that the Kaluga uses GaN technology. 

 

Andy

 

[Ittaku]

17 minutes ago, andyr said:

Your post prompted me to do the same, W - and, yes, all any review says is that the Kaluga uses Hypex NC1200 modules.    I doubt whether Bruno has replaced the output transistors n this module ... so it would seem to be a myth that the Kaluga uses GaN technology. 

I don't recall anyone saying it did? I brought it up a while ago as an example of an expensive class D amp only.

[March Audio]

On 08/09/2019 at 4:53 PM, georgehifi said:

It's not 1 degree, even Bruno said it's 20 degrees at 20k, Jeeze!! And I doubt that, I say its even more if it's a 500khz switching frequency your trying to eliminate! The ICE Power is 70 degrees phase shift at 10khz!!!!!!!!

 

 

 

 

If not and you set the filter higher  to get lower phase shift figures your then letting through way too much to the speaker, I've seen Wilson Watt 8's  with blued tweeter voice coils because too much of the 700khz  switching frequency was let through to the speakers (you don't hear it, it just cooks the tweeter voice coils in silence)  

I'll say it again, this is why GaN Tech is so good because you can have 1.5mhz switching frequency, and the output filter can also be higher and filter it all out and not have any phase shift at 20khz.

 

End of story again, as it seems  with over 80 posts from you on this thread and nothing else, it's just as I said on the 4 page. "this thread is taking a certain advert feel, which I don't think is right, so I won't be fueling the fire anymore."

 

Gearge PLEASE read the text again.  Im not going to argue with you if you havent read and understood the text.  Bruno did not say that.

 

Otherwise if you feel this is an advert instead of actually discussing the technicalities of class D generally, please complain to the mods.  BTW I have already discussed your previous comment of that nature with them.  No issue was perceived.

Edited September 9, 2019 by March Audio

[March Audio]

3 hours ago, aussievintage said:

wasn't that just an example, where he said if you have a circuit that has a 0.2 degree phase shift at 200Hz, 2 degrees at 2kHz and 20 degrees at 20kHz,

 

 

 

 

Precisely

[March Audio]

5 hours ago, ghost4man said:

Alan, 

 

You are misrepresentating my comment by virtue of a straw man fallacy which I know you would be familiar with as you are not stupid. 

 

I did not say that it would. You asserted that there was NO phase shift. However small it is that was factually incorrect. You've acknowledged that. 

 

Does or would that make an audible difference is something that I could not answer. 

 

My gut at the moment tells me that the implementation of GaN technologies must surely be heading in the right direction given that Mola and Technics have invested so heavily in their products to produce what would arguably be considered hi end gear for Class D at least. 

 

I think what would be of value is a direct comparison with what you have on the bench and say a Mola or Technics. 

 

No Im not.  There is no straw man, I just asked you a question.  Yes I used lazy terminology, thats all.

 

Im not sure why there is all this noise about GaN.  It offers some advantages and also some disadvantages.

 

Mola Mola Kaluga does not use GaN.  Its a modified Hypex NC1200.  

 

Edited September 8, 2019 by March Audio

[March Audio]

4 hours ago, georgehifi said:

Impossible then the output filter would have to be far higher order and would burn out very quickly with this kind of power, that why the A.P. test filter is high order passive but can only take milliwatts when test are done using it.

 

Mark Levinson with their $80k No. 53 Class-D monoblocks tried to series up two 2nd order filters ones to get a steeper slope to null out the 600khz switching frequency but it sounded rubbish.

Yes that's just one of the monoblocks and those 4 chokes are part of the output filters

 

  

 

They actually managed to kill off most of the switching noise as can be seen by the 10khz square wave without AP filter no switching noise ringing across the top of it, just one small over shoot.

 

 

 

Fig.2 Mark Levinson No.53 Reference, small-signal 10kHz squarewave into 8 ohms (no AP low-pass filter).

 

 

 

George, with the greatest of respect you are technically out of your depth here.

 

Re ringing and square waves look up Gibbs effect.

 

I have already shown you a square wave from a Hypex NC400.

 

There is no requirement to completely filter out the switching frequency BTW.  A tweeter does not respond at 450 or 500kHz, its an electrical and mechanical filter.  As such it cannot inter-modulate anything down into the audio frequency range.  The signal that is left behind is low level.  The tweeters inductive reactance mean instead of being say an 8 ohm load at audio frequencies it is probably more like a 50+ ohm load at 500kHz.  The power dissipated is a therefor only a few mW.

 

I can run through the calculations if you like but perhaps you could yourself and you will see why its a non issue.  Look up the inductance of a typical tweeter, calculate its inductive reactance at 500kHz and add R.  Calculate the current and power considering the residula switching signal is a feew hundred mv.  Oh and thats for a directly connected tweeter without an intervening crossover which would filter it even more.

 

NC1200 Switching waveform

 

 

 

Below is a general paper not related to class D about ultrasonic intermodulation in speakers plus some of Brunos commentary.

 

http://www.aes.org/e-lib/browse.cfm?elib=20459

 

 

pjug said:

In Benchmark's literature, they state:

“The important difference between the AHB2 and class-D amplifiers is that the AHB2 does not produce switching noise. Class-D amplifiers are measured with brick-wall AES17 filters that ignore the switching noise above 20 kHz or 40 kHz. Tweeters can demodulate this ultrasonic noise and fold it into the audible band.”

That’s a curiously shaky argument to put forward in favour of a fine amplifier like the AHB2. It doesn’t need to make any excuses. I imagine that it was just a throwaway comment, otherwise it’d qualify as Spreading Uncertainty and Doubt.

I am fully aware that literature exists testing tweeters at blistering levels with signals close to the audio band, but it’s quite a stretch to extrapolate from there to a small 500kHz residual and side bands. Keep in mind that the electrical impedance of a speaker goes up with frequency so the residual (heavily attenuated by the LC filter) amounts to an absolutely tiny current. But hey, should anybody really want to make a point of this claim, the onus is of course on them to provide data showing that the switching residual of a good class D amp adds distortion to the acoustical output that isn’t there when the switching residual isn’t there. If such data were around, it’d probably be online already.

That’s not a very informative answer, so let’s see if we can take the argument anywhere useful. This is about demodulating the sidebands of the switching residual downward. The word “noise” is potentially misleading here, it’s not noise as in “hiss” but as in “signal I didn’t pay for”. The switching residual is actually a highly predictable signal.

Anyhow. Demodulating requires an even order distortion mechanism. One which, as several of you noted, would have to be present even when the voice coil isn’t moving anymore. For the sake of an argument, let’s just imagine that there is something in the tweeter that is able to turn not just the current but also the square of the current into a force. If you designed a tweeter specifically to be that bad, you could actually do that, but it wouldn’t do much else besides: the exact same nonlinearity would also produce absolutely massive amounts of distortion in the baseband in the regular way.

The hypothesised additional distortion would be the intermodulation between the carrier and its sidebands, and between the sidebands amongst themselves. Spectrally they would therefore appear exactly where regular distortion components are already present, except much lower. These regular distortion products are caused by the same nonlinearity but with a much larger input signal (the actual audio signal instead of the filtered residual). The hypothesised fold-down distortion would be much smaller than regular distortion and indistinguishable from it because it’s in the exact same place.

If we put in a better tweeter, regular and fold down distortion (if any) would fall in tandem. There’s simply no scenario in which fold-down distortion could appear separate from its associated regular base-band distortion and/or with remotely comparable magnitude.

From that thought experiment we can conclude it’s rather unlikely that any measurable or audible mechanism exists that can be traced back to the carrier residual.

As an aside I find the “demodulation” argument also a bit of an own goal when proposed in favour of high-res audio. What they’re saying there is that maybe a bad tweeter might manage to produce distortion components inside the audio band which wouldn’t be there if either the tweeter were better or the bandwidth of the audio signal were limited to 20kHz. That sounds more like an argument against mindlessly doubling the sampling rate every 3 years and inventing a new format every decade.

 

 

Edited September 8, 2019 by March Audio

[georgehifi]

6 hours ago, March Audio said:

George, with the greatest of respect you are technically out of your depth here.

No sorry it is you, as you have no grasp of what kind of output filter it takes on a Class-D to rid the switching frequency noise, with a 500khz switching frequency.  Time will show, this ad will run it's course.

Edited September 8, 2019 by georgehifi

[ghost4man]

7 hours ago, March Audio said:

 

No Im not.  There is no straw man, I just asked you a question.  Yes I used lazy terminology, thats all.

 

Im not sure why there is all this noise about GaN.  It offers some advantages and also some disadvantages.

 

Mola Mola Kaluga does not use GaN.  Its a modified Hypex NC1200.  

 

Apologies you were correct the Mola dies not. 

 

I have confused it for the Merrill which does. 

 

This is worth the read and in particular focuses on the Merrill 118 monoblocs and gets into how the technology works. Similarly priced to the Technics. 

 

http://www.enjoythemusic.com/superioraudio/equipment/1018/Merrill_Audio_Element_118_Monoblock_Amplifiers_Review.htm

[georgehifi]

10 hours ago, andyr said:

12 hours ago, Wobba said:

I did  And couldn’t find any reference  to Mola Mola Kaluga amps using GaN tech. They use the same Hypex NC1200 & SMPS1200 modules as March Audio P701 amps.

 

Your post prompted me to do the same, W - and, yes, all any review says is that the Kaluga uses Hypex NC1200 modules.

I don't believe they use GaN either, I couldn't find anything on it.

The only ones so far I know are:

Technics SE-R1 and the magic SU-G30 integrated wifi streamer amp

Merrill Element 114, 116 and 118

AGD Production GaN tube Stereo and Monos

Orchard Audio BOSC monoblocs

 

The value one is the SU-G30 integrated with GaN Technology, that still gets you far better "Dead Time" figures (than others spruik'd here), but not all what GaN can do.

 

Technics in the SE-R1 the only one using the 1.5mHz switching frequency, which with the output filter gets you "nearly" zero phase shift at 20khz, as you can see below in the GaN fet is far more square like the original 1k square wave, it's sides are not falling inwards like the old tech Class-D that shows they have phase shift.

 

 

Cheers George

 

Edited September 8, 2019 by georgehifi

[ghost4man]

1 hour ago, georgehifi said:

I don't believe they use GaN either, I couldn't find anything on it.

The only ones so far I know are:

Technics SE-R1 and the magic SU-G30 integrated wifi streamer amp

Merrill Element 114, 116 and 118

AGD Production GaN tube Stereo and Monos

The only value one is the SU-G30 integrated with GaN Technology, that still gets you far better "Dead Time" figures (than others spruik'd here), but not all what GaN can do.

 

Technics in the SE-R1 the only one using the 1.5mHz switching frequency, which with the output filter gets you "nearly" zero phase shift at 20khz.

 

Cheers George

 

George, 

 

Don't forget Orchard Audio BOSC monoblocs also utilise GaN technology and at a much cheaper price. I can't comment on quality though but this isn't about quality but rather implementation. 

[georgehifi]

2 minutes ago, ghost4man said:

Don't forget Orchard Audio BOSC monoblocs also utilise GaN technology and at a much cheaper price.

Oh yes forgot about them, just a bit sus on what the powersupply's like. it'll be smp, which I'm not keen on anyway, but these must be minuscule.

 

Cheers George

[March Audio]

4 hours ago, georgehifi said:

No sorry it is you, as you have no grasp of what kind of output filter it takes on a Class-D to rid the switching frequency noise, with a 500khz switching frequency.  Time will show, this ad will run it's course.

 

George the reality is what it is.  Bruno and Purifi have done what you claim is impossible. 

 

No need to wait, its already here

 

As mentioned there is no requirement to completely filter the switching frequency and as I have already shown you its quite capable of reproducing square waves. 

 

I'm away from the office until Thursday but will measure the phase shift of the Purifi. 

Edited September 9, 2019 by March Audio

[March Audio]

14 hours ago, georgehifi said:

Oh yes forgot about them, just a bit sus on what the powersupply's like. it'll be smp, which I'm not keen on anyway, but these must be minuscule.

 

Cheers George

 

Nothing wrong with properly designed smps power supplies. You seem to have a lot of preconceptions. 

 

If smps was a noisy issue how does the Purifi acheive performance like this? 

 

Edited September 9, 2019 by March Audio