Hypex Nilai500DIY Amplifier Review

[Sokel]

Is that something that was ever looked at?

Not that I know of.
@boXem also raises this question often and I think it's more than interesting.

I even try a test procedure myself (don't shout,I'm a newbie in measurements) :

BUCKEYEAMPS Hypex NCx500 Amplifier 2channel Review

That is what AP generates. So Ap generates 5.8 V out, DUT(amps)'s max input voltage is 5.8 V. Last time the value was actually 700W point by my experiment. This time theoretical(by gain) value suggestions from the company. Points are decreased 150 -> 100, Shorter duration of stress. Filter...

www.audiosciencereview.com

 

[Jero]

Well then noise in such an amp is mostly residual electric noise. That’s why the THD+N curve first goes down and it is a nice line in those logarithmic plots, because the S/N ratio is worse at lower output. Then distortion starts to pick up and at some point start to increase more and more steeply. It may have a new local minimum or two before shooting up because of the order of the feedback loop: second order may have one extra minimum, third order like Eigentakt or Nilai could even show a second one, in theory, but that additional (subtractive) resonance may be well after the clipping point. This is by far the most likely explanation.

However, it could also be a strange increase of the noise, and in theory one could observe some subharmonics of the switching frequency, but they are still quite far from the measured frequency.

In Munich Putzeys this year told me that the Eigentakt does not have that type of curve at most frequencies, that the distortion factors just jump up only at the last moment at all frequencies, so residual electric noise dominates almost always, and of course he made sure I understood that this is because he did a superior work, and the others can only catch up with trial and error while he uses mathematics We know him. I actually find him a likeable person, I do not like people that hide the fact that they are aware of their intelligence - I know I have myself a “stellar” IQ and this is for me just a random gift I have been given with which the moral imperative is that I have to do good stuff with it, contributing to a make this world a better place. So it is just responsibility, definitely not superiority, and for the most I am even failing.

In Munich Putzeys this year told me that the Eigentakt does not have that type of curve at most frequencies, that the distortion factors just jump up only at the last moment at all frequencies, so residual electric noise dominates almost always.

I would like to see a multi frequenty measurement of Eigentakt compared to Nilai, see if indeed Eigentakt doesnt have that 10-15 kHz "bump".
Does anyone has the means to do this besides Amir who is busy as it is already

I might be nit picking here because i have a ****** small Denon poweramp to drive my speakers now, and i hear a bit of grain in the highs. Probably because this amp cant handle the powerhungry speakers.

Im about to push the button on the Nilai500 mono's but the same moneys i can get a used Benchmark AHB2 so i dont know what to do really.
.

 

[boXem]

True, but that would mean there is noise modulation at play here. Is that something that was ever looked at?

Noise modulation would be in band noise. I am more thinking of noise shaping, like with DACs. The principle is the exact same: lots of feedback in a given bandwidth decrease the noise level in this bandwidth while increasing the noise level outside the bandwidth. DACs have the comfort to run in the MHz range, so the high feedback bandwidth is quite large while class D amps are much more limited in frequency, causing the noise shaping bandwidth to be limited to audible range.

 

[Sokel]

Noise modulation would be in band noise. I am more thinking of noise shaping, like with DACs. The principle is the exact same: lots of feedback in a given bandwidth decrease the noise level in this bandwidth while increasing the noise level outside the bandwidth. DACs have the comfort to run in the MHz range, so the high feedback bandwidth is quite large while class D amps are much more limited in frequency, causing the noise shaping bandwidth to be limited to audible range.

At NCx500 is distortion at Apollon Audio implementation:

Apollon Audio Class-D Amp Build Quality

Of course. Here are the measurements of the dual mono and with different op amps (Sonic Imagery 994, OPA1656, LM4562). Any chance you have some measurements of the OPA1612 compared to the Sparkos SS2590 for the regular dual mono? I was also curious how many of the SS2590 opamps are...

www.audiosciencereview.com

That's one on the list,others will hopefully follow.

 

[Rob Fens]

I may be missing something, but Amir explained in page 10 the cause of rising distortions at higher frequencies, the test was done with the lid open.
With the lid closed it is gone. Again i might be mixing things up and be mistaken...

Regards,
Rob.

 

[Matias]

I have only 1 question about the Nilai500. That 10 and 15kHz bump in the graph bothers me really. Is this audible? like on very revealing speakers as ESL or Planar midrange?

FAQ #7.

Class D FAQ

This topic is dedicated to Frequently Asked Questions about class D amplifiers. If you have been directed here the reason is that it is easier to concentrate the frequent answers here instead of repeating them in several topics. Q1: Class D power ratings cannot be true. How can these small and...

www.audiosciencereview.com

 

[voodooless]

I may be missing something, but Amir explained in page 10 the cause of rising distortions at higher frequencies, the test was done with the lid open.
With the lid closed it is gone. Again i might be mixing things up and be mistaken...

The graphs in the review were already updated after this mishap. So whatever we see there is with the lid closed.

 

[Rob Fens]

The graphs in the review were already updated after this mishap. So whatever we see there is with the lid closed.

Thank you for the swift and clear answer.

 

[Smoey22]

The audioXpress website just released their Nilai review, this article was originally published in the audioXpress magazine, July 2023.

Fresh From the Bench: Hypex Nilai500DIY 500W Mono Power Amplifier Kit - A New DIY Benchmark

The Hypex Nilai500DIY 500W Mono Power Amplifier Kit is new DIY benchmark from the company's dedicated DIYClassD brand. The kit is specified to deliver 500W with 1% THD at 4 ohms, improves output noise and idle losses significantly, and comes with a discrete buffer stage and voltage regulators...

audioxpress.com

 

[maty]

Can anyone indicate which opamp the Hypex Nilai uses? Maybe OPA1656?

Apollon Audio Class-D Amp Build Quality

Of course. Here are the measurements of the dual mono and with different op amps (Sonic Imagery 994, OPA1656, LM4562). Any chance you have some measurements of the OPA1612 compared to the Sparkos SS2590 for the regular dual mono? I was also curious how many of the SS2590 opamps are...

www.audiosciencereview.com

 

[boXem]

Can anyone indicate which opamp the Hypex Nilai uses? Maybe OPA1656?

Apollon Audio Class-D Amp Build Quality

Of course. Here are the measurements of the dual mono and with different op amps (Sonic Imagery 994, OPA1656, LM4562). Any chance you have some measurements of the OPA1612 compared to the Sparkos SS2590 for the regular dual mono? I was also curious how many of the SS2590 opamps are...

www.audiosciencereview.com

View attachment 310273

Discrete opamps, like NC400 DIY.

 

[Rick Sykora]

The audioXpress website just released their Nilai review, this article was originally published in the audioXpress magazine, July 2023.

Fresh From the Bench: Hypex Nilai500DIY 500W Mono Power Amplifier Kit - A New DIY Benchmark

The Hypex Nilai500DIY 500W Mono Power Amplifier Kit is new DIY benchmark from the company's dedicated DIYClassD brand. The kit is specified to deliver 500W with 1% THD at 4 ohms, improves output noise and idle losses significantly, and comes with a discrete buffer stage and voltage regulators...

audioxpress.com

Nice effort from @SIY to add to Amir’s review.

See he got caught by the tiny front power switch too!

 

[FERCSA]

Nice review indeed. Imho the tiny switch is a cheapo but clever solution. Most cases it not used at all since your preamp do the power switching for you via the trigger link.

 

[Reddoc]

Funny how people who BUILD (assemble) their amps, and this includes inserting the switch into the faceplate, don’t realise that there is a switch in the faceplate!

 

[SIY]

Funny how people who BUILD (assemble) their amps, and this includes inserting the switch into the faceplate, don’t realise that there is a switch in the faceplate!

The switch is part of a circuit board which also holds the LED. There's no actual "insert the switch."

Still, my oversight.

 

[Reddoc]

The switch is part of a circuit board which also holds the LED. There's no actual "insert the switch."

Still, my oversight.

Yes, it is a TINY switch attached to the LED board

 

[Rick Sykora]

The switch is part of a circuit board which also holds the LED. There's no actual "insert the switch."

Agree, for that matter, will add it is a momentary switch as well. If the amp loses power, you need to press it again to power up. Given Hypex’s otherwise solid choices, this is a small but annoying blemish on the design.

 

[David_M]

How exactly that amazing D-class SINAD is beinf stated at 112 dB if it turns into 80 dB pumpkin @10 kHz+?
Seems a bit tricky and yes, not this amp alone, not Hypex alone - almost all of them.
And yes again, 80 dB is okay too.
Question is if it's correct to mark such amp as a 110 dB+

SINAD is spec'd at one frequency only, i.e. 1KHz. A proper series of swept graphs would be SINAD vs Frequency for set power levels, say from 1W, 5W, 50W, 100W etc. for each impedance load.

 

[Stefano86TV]

Hello everyone, I'm writing from Italy and I need some advice. In early May 2023 I received my Hypex Nilai mono kits. I immediately noticed that some capacitors were bent and/or detached, so I sent them both back for service. They came back, and I assembled them. I immediately noticed that they were hotter than I thought. However, I have read in various forums that it is tolerable. I was listening to music the other day when I suddenly heard a loud "bump" coming from the left speaker. Then the speaker stopped playing. I immediately checked that it was intact and at first glance it appears to be. Then I looked at the amplifier that controls it and realized that it was off (even the LED). I tried turning it back on but it wouldn't turn on. I put my hand on it and it wasn't even warm. After a few hours I tried to turn it back on and saw the LED light up, but I no longer trusted it to ring. I opened it up and everything seems fine (no burning, no strange smell, no loose connectors, no swollen capacitors). At this point, how would you behave?
As for the speaker, if with another amplifier it seems to sound normal, do I have anything to worry about?

 

[David_M]

Your first and unfortunately greatest mistake was to accept the kit once you realized it had a problem. You should have returned it and requested a new one, having noted the original kit serial number. Probably you have a bad capacitor, since you've already experienced bent or detached capacitors. Return the kit for repair asap or get your money back and buy it from another vendor, if you still insist on using Nilai?