Hypex NCx500 Class D Amplifier Review

[restorer-john]

I might be getting something wrong with the graphs when I take them in comparison. These Hypex module seems to archive a slightly better SINAD than the Purifi module with a similar gain (around 27 db). The problem some of you are pointing out is the rising distortion that comes with rising power output. Am I reading it right?

Yes, but with a caveat. The Purifi seems the better amp to me.

Edit: Amit got there first.

 

[Vacceo]

Thanks to both John and Amir!

 

[JSmith]

The Purifi seems the better amp to me.

I suppose some may prefer the lower distortion at lower power of the NCore compared to lower distortion at higher power of the Purifi though. Of course this will vary from person to person based on their use case, speaker sensitivity etc. as to which trade off suits.


JSmith

 

[pma]

Marantz PM-10 Comparison
(bridged NC500OEM with Marantz buffer+preamp)

Hi Alan, I have also made some measurements in BTL mode, on another Hypex module, NC252MP. The new thread is here:

Meaurements of Hypex NC252MP in bridge-tied load (BTL)

Following the discussion in the NCx500 review thread, I was curious how my PMA-NC252MP amplifier based on NC252MP module would perform in a bridged configuration. Original Hypex datasheet allows for this configuration, as we can see in the following part from the datasheet: It declares...

www.audiosciencereview.com

 

[PeteL]

No, I sent the Buckeye back already.

Would you still agree that there is a real probability that the main reason for giving a not recommended rating for the Buckeye Purifi amp could have been an instrumentation flaw tough?

 

[sernikus]

Yes, but with a caveat. The Purifi seems the better amp to me.

Edit: Amit got there first.

But...we have to consider a very important thing : the buffer in the Ncx500 is really very good sonically.... believe me...

 

[pma]

Would you still agree that there is a real probability that the main reason for giving a not recommended rating for the Buckeye Purifi amp could have been an instrumentation flaw tough?

It may happen, I have a similar experience as well and it happened to @John Atkinson from Stereophile as well.

My cure to this issue is as follows: as I use massive binding posts at the amp output, I tightly screw wires from the load below them. The measuring signal (voltage, negligible current) I take by banana plugs inserted to the binding posts. So, the additional contact between load wire and the binding post is not taken into measurements. It works reliably.

 

[Rick Sykora]

No, I sent the Buckeye back already.

Would you still agree that there is a real probability that the main reason for giving a not recommended rating for the Buckeye Purifi amp could have been an instrumentation flaw tough?

Unlikely, despite similarities, I happen to have the Buckeye and the NCx500 on my test bench at the same time. Here is the comparison…

While not the exact same Buckeye unit that Amir tested, as was noted in that review thread, my previous Buckeye posted measurements followed Amir’s results. I mentioned this to Amir as part of post review discussion of the NCx500. Since I could not reproduce the distortion rise, our alternative was to ship his unit to Hypex. Before we considered this, he decided to recheck his results. This is when he found his wiring issue. I have encountered a similar issue when the screws to the amp output on my dummy load got loose.

P.S. Bottom line, despite very comparable distortion rise, the root cause is not the same.

 

[Rottmannash]

Would you still agree that there is a real probability that the main reason for giving a not recommended rating for the Buckeye Purifi amp could have been an instrumentation flaw tough?

Dylan also measured a similar level of distortion.

 

[pma]

Dylan also measured a similar level of distortion.

And I measure lower distortion at high power even in the bridged mode, which in fact stresses the amp sections by 2 ohm load.

 

[Rick Sykora]

But...we have to consider a very important thing : the buffer in the Ncx500 is really very good sonically.... believe me...

Now that the performance measurements are clearly more comparable too, expect the discussion will shift towards other key differentiators like features and price. This is where the OEMs add their value. We have a number of good examples of OEM differentiation on ASR already. Will be interesting to see what the future holds!

 

[NikolasA]

How does it compare to Nilai? Hope we see the details soon

 

[Matias]

How does it compare to Nilai? Hope we see the details soon

 

[Vacceo]

Now that the performance measurements are clearly more comparable too, expect the discussion will shift towards other key differentiators like features and price. This is where the OEMs add their value. We have a number of good examples of OEM differentiation on ASR already. Will be interesting to see what the future holds!

There is an element I'd like to know for comparison: efficiency. Cooler-running electronics is always good news.

 

[Rick Sykora]

There is an element I'd like to know for comparison: efficiency. Cooler-running electronics is always good news.

Not sure how you measure the amp overall (as it would have to have known set of conditions), but the NCx500 module data sheet claims 93% efficiency.

 

[pma]

Efficiency depends on output power and load, both can be found in the datasheet. It is not a single number. Efficiency gets worse with lower power. My NC252MP eats 17W at idle.

 

[HarmonicTHD]

Would you still agree that there is a real probability that the main reason for giving a not recommended rating for the Buckeye Purifi amp could have been an instrumentation flaw tough?

No I asked @Buckeye Amps this question as I also was unclear and he clarified it here:

Post in thread 'Buckeye 3 Channel Purifi Amp Review'
https://audiosciencereview.com/foru...-channel-purifi-amp-review.40293/post-1451808

BTW. Very commendable @Buckeye Amps how you are dealing with this. Respect.

 

[Vacceo]

Not sure how you measure the amp overall (as it would have to have known set of conditions), but the NCx500 module data sheet claims 93% efficiency.

I guess measuring heat after a certain amount of time with a certain amount of power; I honestly don't know either.

Power draw at idle is another option, as PMA has pointed out.

You have tried several modules and you know about the need of heatsinks, so even if that is not particularly accurate, I'd trust your word.

All in all, this gives an indication about how well you can stack components on top of each other or if the amp will need a fan or not.

 

[Piere]

It may happen, I have a similar experience as well and it happened to @John Atkinson from Stereophile as well.

My cure to this issue is as follows: as I use massive binding posts at the amp output, I tightly screw wires from the load below them. The measuring signal (voltage, negligible current) I take by banana plugs inserted to the binding posts. So, the additional contact between load wire and the binding post is not taken into measurements. It works reliably.

Really something to take care of! Surely at these low levels of distortion. When I once measured a low sitortion CA amp I noticed distortion harmonics at very low level (below -120 dB) that looked like a train of harmonics going beyond 20 kHz at the spectrum view. Very similar to the typical cross-over distortion profile of a standard class-AB BJT amp. The dummy load was connected to the amp's binding post by screwing the bare copper wire directly underneath the binding posts. To my amazement the distortion disappeared when I tightened te binding post a lot more firmly. What was going on here? Well the thin oxide of the bare copper together with the gold plating of the binding posts acted as small Schottky rectifier diodes! By tightening the posts more firmly I went trough the copper oxide destroying the small diodes. Lesson learned: Stay away from pure copper regarding loudspeaker interconnect transitions!

I do not say it will be the case here, but these unexpected mechanisms can play havoc!

 

[HarmonicTHD]

I guess measuring heat after a certain amount of time with a certain amount of power; I honestly don't know either.

Power draw at idle is another option, as PMA has pointed out.

You have tried several modules and you know about the need of heatsinks, so even if that is not particularly accurate, I'd trust your word.

All in all, this gives an indication about how well you can stack components on top of each other or if the amp will need a fan or not.

Nah. Just measure how much power you put in and how much you get out (the difference is pretty much the heat). As it was said before at idle it draws 17W already.