• 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!

Hypex Nilai500DIY Amplifier Review

Rate this amplifier:

  • 1. Poor (headless panther)

    Votes: 4 1.1%
  • 2. Not terrible (postman panther)

    Votes: 13 3.5%
  • 3. Fine (happy panther)

    Votes: 74 19.9%
  • 4. Great (golfing panther)

    Votes: 281 75.5%

  • Total voters
    372

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
Well, I call that progress. We went from: it’s certainly audible, to: we’ll need to check. I can live with that :)

But I do think it’s far better than anecdotes. Logic can help you understand a problem, filter out evidence, help you understand and interpret it
If I ever wrote/said "certainly audible", sorry. Don't remember doing that and too lazy to re-check. AFAIR, I only argued that you can't say "not audible" for sure.

And btw, this missing fundamental "debate" is just a tangent. The original Q was about the audibility of -75dB THD at 15kHz. The two questions are somewhat close/related but still, the missingF is a matter of pitch-perception and the other one is timbre-perception.

And before someone jumps, I also haven't said that -75dB THD is surely audible. Just that I do not consider it ok .. or 'safe' .. or acceptable.
Nobody can say "-75db is audible", at least not in a generic way. But same as sure, nobody can say "not audible" or "no concern".
And that's not only true about -75dB, but pretty much any SINAD number between ~50 and 1xx.

Personally, I prefer 106+dB as a sort of 'safe' number. As in: 10 dB below the 96dB range of a CD. IIRC, that 'logic' comes from an old post by amir. Or I might just have 'invented' it :)
 
Last edited:

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
..
Your light analogy is also wrong. The presence of infra or ultra electromagnetic radiation (infra red or ultraviolet) doesn't alter the perception of visible light.
Generic "infrared" was a pretty poor choice of words. Of course, not talking about _any_ infrared. We were talking about 30kHz, pretty close to the audible spectrum. The light equivalent would be a near-infrared, say a faint laser pointer. Not visible if you point it to the sky but you'll see its effect if pointed to a bright-yellow paper close by.
It's also just an analogy, doesn't have to be 100% perfect.

Anyway, please point me to a paper that clearly states: missingF is not audible if its HDs are not audible. Or do the experiment and I'll simply accept the results. In the meantime, if you don't mind, I'll keep being skeptical ... or 'undecided' if you prefer.
 

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
Talking about proof, is there one showing that the increase of THD+N is due to THD and not to N?
I'd surely want to know that too. I simply assumed HD but I could easily be wrong and the whole 'discussion' would be (at least) offtopic if it was a case of N.
Would make the hypex/purifi classD amps even worse, though
 
Last edited:

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
I would love to be able to hear anything at 15 kHz. :)
Luckily (or not) I can still hear a somewhat fainter 15kHz and an even fainter 16kHz. May hear a 17kHz too if I push the volume but I don't really want to try.
Quite a wonder after all the bum-bum concerts & clubs I visited :)
I think that you backup your claim here, and, btw, your Wikipedia link does not support your claim.
No it does not. Not 100%. It's not the argument, just an argument. A '+1' if you wish. That's why I keep asking for a conclusive test.
 
Last edited:

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
Luckily (or not) I can still hear a somewhat fainter 15kHz and an even fainter 16kHz. May hear a 17kHz too if I push the volume but I don't really want to try.
Quite a wonder after all the bum-bum concerts & clubs I visited :)

No it does not. Not 100%. It's not the argument, just an argument. A '+1' if you wish. That's why I keep asking for a conclusive test.
Knock yourself out.

 

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
That was my first thought too. Trouble is, I do not have any supertweeters. Or any transducer that can surely reproduce 25+kHz.
But maybe someone does. Or maybe there are some HPs known to work to 50+ kHz?!

And another missing piece of info: how many HDs do you need to hear/test a missingF? Is 3 of them enough?
 

HarmonicTHD

Major Contributor
Forum Donor
Joined
Mar 18, 2022
Messages
3,326
Likes
4,829
That was my first thought too. Trouble is, I do not have any supertweeters. Or any transducer that can surely reproduce 25+kHz.
But maybe someone does. Or maybe there are some HPs known to work to 50+ kHz?!

And another missing piece of info: how many HDs do you need to hear/test a missingF? Is 3 of them enough?
Aren’t we talking about the audibility with “regular” audio equipment here, which you and I have in their listening rooms? Who has transducers in their speakers for 25+kHz? (Cats? dogs? bats? ;-))
 

antcollinet

Master Contributor
Joined
Sep 4, 2021
Messages
7,414
Likes
12,296
Location
UK/Cheshire
Generic "infrared" was a pretty poor choice of words. Of course, not talking about _any_ infrared. We were talking about 30kHz, pretty close to the audible spectrum. The light equivalent would be a near-infrared, say a faint laser pointer. Not visible if you point it to the sky but you'll see its effect if pointed to a bright-yellow paper close by.
It's also just an analogy, doesn't have to be 100% perfect.

If its a laser pointer then it is visible light, not infrared. It’d be pretty useless as a pointer if it wasn’t.
 

JSmith

Master Contributor
Joined
Feb 8, 2021
Messages
5,153
Likes
13,219
Location
Algol Perseus
And if anyone complains about harsh/hard highs they get quite a rebuke.
It could be viewed (although I hope it's not the case) that you're being a bit disingenuous here as you state this in your first post in this thread, then continue with the current discussion as though the two are correlated somehow. Was that your intention?


JSmith
 

NTK

Major Contributor
Forum Donor
Joined
Aug 11, 2019
Messages
2,660
Likes
5,820
Location
US East
Interesting one. So, does the paper clearly state that the missing fundamental is inaudible when its HDs are inaudible? Is there a clear test for that?

Here's an experiment that points to the contrary: i.e. the missingF is audible when its HDs are not:
"experiments subsequently showed that when a noise was added that would have masked these distortions..., listeners still heard a pitch corresponding to the missing fundamental, as reported by J. C. R. Licklider in 1954.[4]"
Not exactly 100% proof for this case but might be close enough. Or not.

In any case, this 15kHz missingF question simply requires an experiment. That would save us 100+ posts full of assumptions and "logic"
Since the Δf between each harmonic is the signal frequency f (e.g. f_2 = 2 × f, f_3 = 3 × f, ..., etc.), your "missing fundamental" would be f_0 = f - Δf = 0, which is an inaudible DC offset.
 

Julf

Major Contributor
Forum Donor
Joined
Mar 1, 2016
Messages
3,006
Likes
3,998
Location
Amsterdam, The Netherlands
Interesting one. So, does the paper clearly state that the missing fundamental is inaudible when its HDs are inaudible? Is there a clear test for that?
Yes. Play a 30 kHz tone and see if you can hear a 15 kHz tone. It has been done many, many times.
Here's an experiment that points to the contrary: i.e. the missingF is audible when its HDs are not:
"experiments subsequently showed that when a noise was added that would have masked these distortions..., listeners still heard a pitch corresponding to the missing fundamental, as reported by J. C. R. Licklider in 1954.[4]"
Can you point me to what part of Licks paper supposedly supports your speculation.

By the way, he does echo my point about the shape of the waveform on the scope having nothing to do with how it sounds:

Screenshot from 2023-02-09 08-00-36.png

In any case, this 15kHz missingF question simply requires an experiment. That would save us 100+ posts full of assumptions and "logic"
Not posting uninformed speculation would be a start. But go ahead, do the experiment, and report back on the reults.
 

Julf

Major Contributor
Forum Donor
Joined
Mar 1, 2016
Messages
3,006
Likes
3,998
Location
Amsterdam, The Netherlands
That was my first thought too. Trouble is, I do not have any supertweeters. Or any transducer that can surely reproduce 25+kHz.
But maybe someone does. Or maybe there are some HPs known to work to 50+ kHz?!
Less than $8. Piezo Horn Speaker Tweeter
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,938
Likes
3,526
Even if yours ears can't hear 15kHz there is still bone conduction.
"The BC-AC sound field sensitivity is proposed, by combining the present result with others, as frequency independent at 50 to 60 dB at frequencies up to 900 Hz. At higher frequencies, it is frequency dependent with minima of 40 to 50 dB at 2 and 8 kHz, and a maximum of 50 to 60 dB at 4 kHz". (Examination of bone-conducted transmission from sound field excitation measured by thresholds, ear-canal sound pressure, and skull vibrations - 2007 Sabine Reinfeldt).

In short, bone conduction is about 50dB less sensitive than air conduction. So signals that are at an inaudible air conduction level will for sure not be audible via bone conduction. This insensitivity is why bone conduction transducers need to be attached to the head.
 

Rob Fens

Member
Joined
Feb 8, 2023
Messages
39
Likes
52
Location
Haarlem Holland
It is good to see there is a lot of knowledge on this forum, but i don't see the link anymore between the current state of the thread and the original subject, the Nilai.
The thread started to derail when some members missed the correction by Amirm in reply 183, in which he explained that there was no issue afterall at 10 and 15kHz and now we are 110 replies further. The thread turned into a discussion about distortions and psycho-acoustics. (FYI, the "problem" apeared to be caused by an open amp case during the measurements). As far as i understand the measured values, the Nilai, Purifi and Benchmark are on the same Top level, with minor differences.
 

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
It could be viewed (although I hope it's not the case) that you're being a bit disingenuous here as you state this in your first post in this thread, then continue with the current discussion as though the two are correlated somehow. Was that your intention?


JSmith
I'd say my first post was pretty clear: based on the THD vs. Freq measurements, I think the classD amps are improperly called TOTL or top10 or "amplification was solved". Not sure if that can be called 'intention', it's just an opinion. Some may share it, some may not. Fine with me, there seem to be good arguments for both sides.

A tangent argument arose about the missingF being audible for 15kHz. I'd like to see proof for that. A clear direct test, not just logic/arguments.
I offered a €50/€100 ASR donation for that test. Which I'll pay in case my assumption is wrong. Haven't seen any volunteers to take the other side of that 'bet'.

Not sure what other suspicious 'intentions' are you suspecting but I don't really have anything more than stated above.

More generally, this whole 'discussion' went way out of proportion. I share the blame there but it's not like I did it alone...
 

Julf

Major Contributor
Forum Donor
Joined
Mar 1, 2016
Messages
3,006
Likes
3,998
Location
Amsterdam, The Netherlands
A tangent argument arose about the missingF being audible for 15kHz. I'd like to see proof for that. A clear direct test, not just logic/arguments.
I offered a €50/€100 ASR donation for that test. Which I'll pay in case my assumption is wrong. Haven't seen any volunteers to take the other side of that 'bet'.
Can you point me to that bet? Happy to take it.
 

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
How so? That would mean that noise is increasing above 30 kHz. This is noise shaping, something commonly used by DACs but also DSD.
Yes, it might be some of the noise below. I can't really say for sure from this graph but I can't see any N that will translate into 75 SINAD. Not until 500kHz. It's one of the reasons why I went with the HD assumption. And still think it's valid.

Anyway, you should be able to use one of your amps for a (more) conclusive measurement

noise.png
 
Last edited:
Top Bottom