• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

NC252MP (class D) vs. A250W4R (classAB) burst measurements into 4ohm//2.2uF load

D

Deleted member 48726

Guest
Both ECUs and amplifiers use PWMed FETs in the audio band, so yes, they are quite comparable. Moreover many Ncore implementations use automotive FETs.

I was replying to

When you look outside the very small audio world, it's not so natural.
No, I don't think they are comparable. In diesel tech you need to lift a needle pilot bore via. an electromagnet or piezo element. Are we going to discuss how high pressure common rail nozzles are comparing as load opposed to dynamic drivers or where are we headed with this? Let's end these car analogies.
--
But we are looking at the audio world.
 
  • Like
Reactions: pma

boXem

Major Contributor
Audio Company
Joined
Jun 19, 2019
Messages
2,018
Likes
4,901
Location
Europe
No, I don't think they are comparable. In diesel tech you need to lift a needle pilot bore via. an electromagnet or piezo element. Are we going to discuss how high pressure common rail nozzles are comparing as load opposed to dynamic drivers or where are we headed with this? Let's end these car analogies.
--
But we are looking at the audio world.
You are right, who am I to try to get you out of your small world. I will leave you there, have a good night.
 

IAtaman

Major Contributor
Forum Donor
Joined
Mar 29, 2021
Messages
2,397
Likes
4,128
Even in audio world, Class D is not new. When I was a teen, trying to build my own amps, they were widely available. Quick google search reveals they have been around as an option commercially since 1978 Sony release of TA-N88. That is 45 years.

I have to admit, I was also quite intrigued by the findings initially. But the more you think about it, more you realize the absurdity of being surprised by a switching amplifier not behaving nicely when connected directly to a capacitor. I guess we are getting lost in the details and missing the big picture here. In any case, I personally think these measurements have no implication for Class D as an audio amplifier topology, but might have some very limited implications for N252 implementation.

Regarding the wider question whether amps effect the tonality of the music due to their changing FR in connection with changing load impedance, I think it might possible but the fact that such edge case examples are selected to demonstrate increase in distortion with varying load does not really increase my confidence in this thought.
 
D

Deleted member 48726

Guest
You are right, who am I to try to get you out of your small world. I will leave you there, have a good night.
At times it seems small, at times it seems big. As your analogy I don't get this post either.
I may be too small minded.
 
D

Deleted member 48726

Guest
Even in audio world, Class D is not new. When I was a teen, trying to build my own amps, they were widely available. Quick google search reveals they have been around as an option commercially since 1978 Sony release of TA-N88. That is 45 years.

I have to admit, I was also quite intrigued by the findings initially. But the more you think about it, more you realize the absurdity of being surprised by a switching amplifier not behaving nicely when connected directly to a capacitor. I guess we are getting lost in the details and missing the big picture here. In any case, I personally think these measurements have no implication for Class D as an audio amplifier topology, but might have some very limited implications for N252 implementation.

Regarding the wider question whether amps effect the tonality of the music due to their changing FR in connection with changing load impedance, I think it might possible but the fact that such edge case examples are selected to demonstrate increase in distortion with varying load does not really increase my confidence in this thought.
I agree. But feel like there's proof that's missing. If just one more step can be made..
 

PeterOo

Member
Joined
Jun 6, 2021
Messages
91
Likes
64
Location
Zutphen, The Netherlands
Oh no. You need to backtrack the thread a bit. Equivalent capacitance as seen by an amplifier can be much much higher than 2.2 uF on even an ordinary dynamic driver bookshelf speaker.
Can you explain to me what you mean with “equivalent capacitance” in this context? I can understand it with only Cs in series or parallel, even with R // C, but what does it mean in a more complex network that represents the complete behaviour of a loudspeaker. Such a network will have multiple Ls, Rs a d Cs to represent impedance and phase over a certain bandwidth.

if the equivalent capacitance is a C in an R//C network that only represents phase and impedance at one frequency (I get the impression this is how it is used in this thread, but please correct me if I am wrong), then the interaction with the Ncore amp might be different for that simplified network then for the complex network, right?
The simple network might have some theoretical importance, but is it relevant for the actual use of the amp?
 

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,207
Location
Northern Virginia, USA
Whatever the meaning of these tests, have they been performed at reasonable levels also? The ones I saw (and I may have missed some) were all run at enormous output power. I don't know about others, but I listen at about 0.5W average to the cones.
 
D

Deleted member 48726

Guest
Can you explain to me what you mean with “equivalent capacitance” in this context? I can understand it with only Cs in series or parallel, even with R // C, but what does it mean in a more complex network that represents the complete behaviour of a loudspeaker. Such a network will have multiple Ls, Rs a d Cs to represent impedance and phase over a certain bandwidth.

if the equivalent capacitance is a C in an R//C network that only represents phase and impedance at one frequency (I get the impression this is how it is used in this thread, but please correct me if I am wrong), then the interaction with the Ncore amp might be different for that simplified network then for the complex network, right?
The simple network might have some theoretical importance, but is it relevant for the actual use of the amp?
An equivalent circuit is just that. An equivalent to what you connect your amp to. That means you have your speakers crossover, wiring, drivers and enclosure boiled down to one single circuit that behaves just like this speaker. Well, without it's non linearities like distortion and compression.
 

sonder

Member
Joined
Mar 6, 2023
Messages
92
Likes
75
@pma, although your experiments are very interesting from a technical perspective, it's clear from different comments they also cause confusion. That's a shame for a forum that aims to give insight. I think it would help if you could add your post with the conclusion about resistive and complex loads to the opening post?

It would, however I must clarify that topics like this also lead the willing to explore much further, within ~48 hours of asking that, I had reached this level of understanding and in trawling through lots of historical posts and learning, many of @pmas posts were both useful and insightful, remaining grounded throughout (along with many other's here of course). @pma, I did appreciate your follow up post greatly, showing performance under more normal conditions, I'd already pulled the trigger but it was reaffirming to read.

I was in the wrong, I hadn't read the full thread thoroughly, I didn't ask questions where I was unsure, I simply said I don't understand so this must be noise can somebody tell if A will work with B pls.

I guess the only note I'd say, is that whilst many of your journey's are long, and the discourse here long running with lots of historical context, be aware that often people are hitting something you write as the first thing they see, with none of that knowledge or background, so providing a bit of context or quantification would help, even if that's just a review later and adding in a couple of edits (often to the richest comment, rather than the op, landing on p3 or p12 of a thread is normal).

However, the same can be said for most things on here, for example if a cheap integrated amp is reviewed, showing it's performance and distortion/imbalance etc at different "volume" levels, or pointing to some posts which explain and show how performance varies under different conditions (rca vs xlr, different gains, different power connectors, would help.

Those speakers are more sensitive than average. Put foam corks in port's to cut the port resonance, buy a subwoofer and high/low pass them at 80~100 Hz instead. If subwoofer is acceptable for you of course and better siled enclosure one.

Thanks for the input, honestly I can see myself working right through my stack again swapping things about, new speakers are not off the cards even though I've only had these for 2 weeks. I'll get everything between digital and speaker cables optimal, then look at speakers again, if mrs sonder doesn't kill me first.

I don't want to share details here, as taken it off topic too far already.
 
Last edited:

Davide

Senior Member
Joined
Jul 6, 2020
Messages
468
Likes
175
Location
Milan, Italy
This is an interesting question. I haven't looked at class AB amps in some time but I really can't seem to find something that does:

100 W x 2 @ 8 ohm
200 W x 2 @ 4 ohm
400 W x 1 @ 8 ohm BTL

Preferably with gain <= 26 dB and 5W in to 4 ohm SNR > 90 dB. Anybody know of such a product (regardless of price)?

Michael
I only found the Neurochrome modules but we're talking about very different money.
Ditto in terms of finished products, there is the Rotel RB 1552 Mk2 but we are almost at 1000 euros, and in any case it does not have the same performance or many protection circuits.
 

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,513
Likes
3,364
Location
Detroit, MI
I only found the Neurochrome modules but we're talking about very different money.
Ditto in terms of finished products, there is the Rotel RB 1552 Mk2 but we are almost at 1000 euros, and in any case it does not have the same performance or many protection circuits.

After thinking about this some more the Benchmark AHB2 pretty much ticks all my boxes (again at a very different price point than Hype).

It doesn't seem like many of these class AB amplifiers support bridging. For example Neurochrome 686 is already internally bridged and only does 260 W in to 8 ohm. Rotels and current ATIs don't seem to support bridging either. Although I wonder if something like the ATI 1820 series which is not internally bridged could be manually bridged by inverting one input and taking an output from the positive terminals of two channels.

As an aside I find the criticism of @pma in very poor taste. Running AND documenting these tests takes a lot of effort and is a huge contribution to the community. You can decide for yourself if you think the results will be an issue for your specific setup but it is much better to have the data than not. I see this as someone trying to advance the understanding of amplifier performance, not someone with an axe to grind. He has not claimed this is an issue with all class D amplifiers and has even done tests to show the AIYIMA A07 does not have the same issue as the Hypex.

Michael
 

tmtomh

Major Contributor
Forum Donor
Joined
Aug 14, 2018
Messages
2,761
Likes
8,116
I agree. But feel like there's proof that's missing. If just one more step can be made..

A statement that will be repeated by the Class D FUD folks after every “just one more step” is made.

(Not necessarily referring to you specifically)
 

damonhill

Member
Joined
Apr 1, 2023
Messages
25
Likes
47
I am starting to wonder if you do not understand the advises on purpose. "the transformer has sufficient leakage inductance and series resistance"

@pma test is not representative of a real world scenario. No speaker imposes a purely capacitive load in parallel with a resistive one. There is always additional series resistance and inductance from various sources.

Also, impedance wise, he is only considering what is going on at low frequencies.

A 2.2uF cap directly across the amp output of a switching amp will impose a very low impedance at the switching frequency.

At 450kHz this will be in the creation of 0.16 ohms. No wonder the amp exhibited some odd behaviours.

Add speaker cable resistance, inductance plus that of a speaker system and problem goes away.

This thread is a great example of a little knowledge being dangerous.
 

tmtomh

Major Contributor
Forum Donor
Joined
Aug 14, 2018
Messages
2,761
Likes
8,116
@pma test is not representative of a real world scenario. No speaker imposes a purely capacitive load in parallel with a resistive one. There is always additional series resistance and inductance from various sources.

Also, impedance wise, he is only considering what is going on at low frequencies.

A 2.2uF cap directly across the amp output of a switching amp will impose a very low impedance at the switching frequency.

At 450kHz this will be in the creation of 0.16 ohms. No wonder the amp exhibited some odd behaviours.

Add speaker cable resistance, inductance plus that of a speaker system and problem goes away.

This thread is a great example of a little knowledge being dangerous.

I agree, but sadly I’d say @pma has a lot more than “a little” knowledge. He’s just so intent on pursuing a single-minded argument about Class D that he doesn’t care what key information he has to ignore or how many parameters he has to strip away in order to come up with the graphs he posts.

Designing tests that are not representative of a real world scenario, and then claiming those tests have important real world implications, is unfortunately not a result of a lack of knowledge; it’s his entire purpose. And he doesn’t seem to care how misleading his posts end up being as a result.
 

ahofer

Master Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
5,037
Likes
9,108
Location
New York City
@pma test is not representative of a real world scenario. No speaker imposes a purely capacitive load in parallel with a resistive one. There is always additional series resistance and inductance from various sources.

Also, impedance wise, he is only considering what is going on at low frequencies.

A 2.2uF cap directly across the amp output of a switching amp will impose a very low impedance at the switching frequency.

At 450kHz this will be in the creation of 0.16 ohms. No wonder the amp exhibited some odd behaviours.

Add speaker cable resistance, inductance plus that of a speaker system and problem goes away.

This thread is a great example of a little knowledge being dangerous.
Is it not possible to measure speaker/cable capacitance in real world conditions? Can't this be settled (for us non-EEs)?
 

Matias

Master Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
5,072
Likes
10,923
Location
São Paulo, Brazil
This is never going to be settled as long as one is on an irrational crusade against class D, for whatever reason.
 

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,513
Likes
3,364
Location
Detroit, MI
It sounds like there are a few in this thread with Hypex amplifiers and electrostatics. A very useful datapoint would be acoustic measurements of 1 kHz and 5 kHz test tones at a reasonable power level (10 W) to see if you can see this issue in the spectrum. This would help put this to bed, especially if the same measurements were run with a clean class AB amplifier powering the electrostatics for comparison.

My guess is that you won't be able to see it above base room noise / speaker distortion but would love to see the data.

Michael
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,952
Likes
3,569
As an aside I find the criticism of @pma in very poor taste. Running AND documenting these tests takes a lot of effort and is a huge contribution to the community.

Tests like this, without context and clarification, is all an audiophile needs as proof that class D amps are crap after all. Subjectivist are rolling over the floor laughing, as they see the experts fighting over graphs.
 

IAtaman

Major Contributor
Forum Donor
Joined
Mar 29, 2021
Messages
2,397
Likes
4,128
This is never going to be settled as long as one is on an irrational crusade against class D, for whatever reason.
It is never a good idea to pass opinion on someone else's intention but I have to say I don't think there is a crusade against Class D per se. My probably wrong theory is that @pma disagrees with the way amps are measured here on ASR, maybe believes a few key tests are missing to uncover differences in performance, maybe thinks more standardization is required, and is using this example to make a point. To what success, hard to say.
 
D

Deleted member 48726

Guest
A statement that will be repeated by the Class D FUD folks after every “just one more step” is made.

(Not necessarily referring to you specifically)
"That one step" = Test circuits based on real world measurements of real world loudspeakers.
Fringe cases are what they are.
 
Top Bottom