Review and Measurements of Hypex NC400 DIY Amp

[rajapruk]

I read somewhere that the large heatsinks of the Benchmark AHB2 is only needed for passing tests. Absolutely not needed for listening to any kind of music.
That is kind of crazy in a way, making the product more expensive for no REAL reason.

 

[vitalii427]

I don't want you all to be disappointed with your life after you see mine.

We can stand it

 

[svart-hvitt]

ON STANDARDS:

Double standards are twice as good.

Three standards are thrice as good.

After counting on all your fingers, use your toes.

 

[March Audio]

I read somewhere that the large heatsinks of the Benchmark AHB2 is only needed for passing tests. Absolutely not needed for listening to any kind of music.
That is kind of crazy in a way, making the product more expensive for no REAL reason.

And this is what manufacturers have debated with the FTC for some time.

 

[Thomas savage]

Well amirs generally focused on consumer relevant testing , getting results that relate to what the product might do or how it might behave at your house. We could test some amps with the full constant power standard but then others ( when the owners object ) in another way... sounds confusing.

We need a standard than can be applied to all amps that amir receives and observes conditions of practical usage . That’s for what we do here, I’m not saying that should be a standard in general. Personally when I buy a amp id want to know it’s been tested to destruction and be given it’s true specs as a amplifier.

 

[rajapruk]

And this is what manufacturers have debated with the FTC for some time.

But then you can not compare them to the precious 50 year old vintage amps. They are too heavy to move to an measuring system again!

 

[Burning Sounds]

This will be one of four amps for a 4 way active system with DSP and a certain amount of FIR corrections.

This is where these modules (and the Icepower equivalents) come into there own IMO. It is relatively cheap to build a multichannel amp that delivers what active speakers require without any loss in SQ. I have 4 Hypex NC502MP modules (Apollon Audio NCMP8350) driving my speakers - I do wish they would spec their own build and not simply use the headline figures from Hypex, though. I applaud March Audio for planning to do this.

However, when I bought this amp I was perfectly aware that the NC502MP modules are 100 watt continuous with peaks of about 200 watts @0.002% into any load from 2-8 ohm. It's all there on their website, including the fact that continuous power is about one fifth of peak power. I don't think they have tried to hide anything. It will be interesting to see how the reliability fares compared to a couple of my older class AB multichannel amps.

I would love to see some of the quite comprehensive testing that Paul Miller used to do for HiFi News carried out on the class D amps. When I first got into active speakers about 10 years ago I was somewhat wary of the multichannel amps available and the general consensus on the forums was that they were really only suitable for home theatre setups. Miller's testing showed that some of these less expensive amps (ie not Bryston etc) were very competent and saved me a ton of money.

 

[Juhazi]

I guess that I'm too old for this, or just a wimp... I have been using ICEpower 125ASX for woofer and dipole low-mid for five years now without knowing that they are insufficient. It is easy to make drivers clonck with these, but I never listen that loud at home.

I have now listened for 2 weeks a pair of Fusion FA123 with it's 125W rated stereo module driving woofer and mid in other speakers -, same scenario. Both are passively cooled and in way too small boxes. Temp of FA123 backplate stays 44¤C and ICE box is 30¤, 230VAC line voltage.

 

[JustIntonation]

And this is what manufacturers have debated with the FTC for some time.

Isn't this possibly different in the case of active speakers?
For instance, if we use an amp to drive only the bass woofer / subwoofer. With for instance organ music or some drum&bass / jungle music with a continuesly sounding bass, or for making music. In think in these situations we can see a prolonged loud bass pushing the woofer to its linear xmax and the regular dynamic range of music doesn't apply?
I think it makes sense to select amp power for long term / FTC rating to push the woofer to linear xmax in such an application?

 

[Armand]

I have done som tests on a prototype of a stereo amplifier built around Hypex NC500 modules that will be released in a few months. This amp is bridgable and can deliver 1000W continous for hours without any component reaching more than 75 degrees celsius. Below is a graph showing the temperature during 1 hour at 1000W into 8 ohms. In the beginning of the graph the power was increased from 800W.

At 1000W the THD+N is about 0,01% but I am trying to reduce this in the final version. In short periods it can deliver 1800W but the temperature get out of control after a few seconds.
The problem with Hypex's modules is heat in the output coil and the only solution is fan cooling to keep the coil and surrounding capacitors cool. With "normal" use the fans does not need to start. At about 300W continous they will start up in a vurtual silent mode that is silent unless you put your ear into the fan tunnel.. For all practical purposes the amp will be silent with music-power up to 1000W.
Stressing these modules to the maximum without active cooling raises the temperature of the coil and capacitors to 120 degrees and it will shut down. I suspect that many Hypex based amplifiers without active cooling will fail prematurely due to faulty capacitors if used hard.

 

[garbulky]

As I noted, I am hesitant to test loaner equipment to such limits (thermal fault).

I dont blame you Amir. It is risky with loaner equipment

 

[Armand]

With regards to the discussion of Hypex's naming of their products VS rated power I don't Hypex does anything wrong. It is up the manufacturer to rate the products maximum power.
Why not make an amplifier with NC400 modules and a 600W (150W continous) power supply if you know you will never use this amplifier in a DJ club but at home where you need peak power just for transients.
Peronally I think this continous power rating is outdated.

 

[DonH56]

I do not think continuous power is outdated, but we need a good metric for comparison of burst power.

The continuous power rating is an FTC standard. It provides a basis for comparing among different products over many years so I would not lightly dismiss it. Note it is not John's standard and not some made-up arbitrary thing; it is what has been in place from the FTC for decades. Like John, I don't have a particular problem with how Hypex rates their modules, but agree their (and more so other manufacturer's using their modules) marketing material should make it clear that the 400 W (or whatever) is peak power and under what conditions. I went through the whole 1970's (and well into the 80's) when every manufacturer defined a different "peak maximum dynamic music power" to inflate their power ratings. (I have before commented upon a Sound Design cheapie all-in-one that claimed "100 W peak dynamic music power" from a pair of transistors supporting ~1 W max.) I liked (and still do) the idea of a continuous plus burst rating so you could reliably assess continuous power and headroom among amplifiers.

AFAIK the FTC never ruled on a burst/headroom metric, just (proposed) relaxing the preconditioning requirement and left the max power at 5 minutes sustained. It is pretty tough coming up with a metric since peak-to-average ratios differ so much. Most songs are what, 3 minutes? Orchestral works much longer but may only have short peak sections. It'd be interesting to see an analysis over a large variety and quantity of music to see what a reasonable peak rating should be. This has been done many times over the years but AFAIK the only "standard" is the IHF burst test (?) Not my field so I have not kept up. And I have not been able to find the multi-volume IHF set used to get my audio consultant's license way back when...

How large a burst over average and for how long would be the question to me.

 

[Deleted member 65]

I do not think continuous power is outdated, but we need a good metric for comparison of burst power.

The continuous power rating is an FTC standard. It provides a basis for comparing among different products over many years so I would not lightly dismiss it. Note it is not John's standard and not some made-up arbitrary thing; it is what has been in place from the FTC for decades. Like John, I don't have a particular problem with how Hypex rates their modules, but agree their (and more so other manufacturer's using their modules) marketing material should make it clear that the 400 W (or whatever) is peak power and under what conditions. I went through the whole 1970's (and well into the 80's) when every manufacturer defined a different "peak maximum dynamic music power" to inflate their power ratings. (I have before commented upon a Sound Design cheapie all-in-one that claimed "100 W peak dynamic music power" from a pair of transistors supporting ~1 W max.) I liked (and still do) the idea of a continuous plus burst rating so you could reliably assess continuous power and headroom among amplifiers.

AFAIK the FTC never ruled on a burst/headroom metric, just (proposed) relaxing the preconditioning requirement and left the max power at 5 minutes sustained. It is pretty tough coming up with a metric since peak-to-average ratios differ so much. Most songs are what, 3 minutes? Orchestral works much longer but may only have short peak sections. It'd be interesting to see an analysis over a large variety and quantity of music to see what a reasonable peak rating should be. This has been done many times over the years but AFAIK the only "standard" is the IHF burst test (?) Not my field so I have not kept up. And I have not been able to find the multi-volume IHF set used to get my audio consultant's license way back when...

How large a burst over average and for how long would be the question to me.

It's been a great learning experience for me reading your always insightful posts @DonH56 !

 

[DonH56]

It's been a great learning experience for me reading your always insightful posts @DonH56 !

Thanks! But do remember this (audio) is not my day job and much of my experience is horribly out of date. That is why I rarely comment upon current products, just have not been getting out that much. And am satisfied with what I have for the most part so no desire to spend limited free time window-shopping as it were. Plus sales folk these days seem less tolerant of people just coming in to try out a few things (but that may be my old memories speaking, not reality, since the high-end stores I worked were very tolerant of people spending time to listen and compare products over several sessions -- may well have been the exceptions even back then).

 

[garbulky]

Well there is a difference between 400 and 40 watts. And one big differentiator is the cost to build these units and the materials it takes. Though my listening levels are about 1 to 2 watts, you could use 40 watts of power pretty consistently in home theater though perhaps not constantly, very often throughout a movie.
Emotiva had on occassion tried to use this game of advertising implying a higher power than what one thinks.
They used to have a UPA-5 which was this ginormous thing with 5 independent channels amps and heat sinks and a large torroidal 600VA transformer with 90,000 mf cap selling at a paltry $585 free shipping.

It was rated 125X5 and 185X5 weight 58 pounds. Now this was an all channels rating which was their standard rating they used for all amps.

Their current replacement amp is the A-500 a 5 channel amp
which looks like this:

Anybody looking will see that it has one single amp blade. Much smaller heat sink. Two Fans. An unspecified torroid (about 300VA) and unspecified capacitance probably close to 30,000 mf or less. It was rated on their gallery page as
5 channel amp 110 watts per channel. (And 195 watts @ 4 ohms).
Weight 26 pounds
Sells for $500
Now anybody looking at these two amps on the gallery page comparing the amps would see this as being a replacement amp with similar ratings. But it's not even close.

Looking closer at the specs one sees that it's only 80 watts per channel all channels driven @ 8 ohms and NO 4 ohm rating given (not an accident). After protests by me and some other members they changed the wording and was more clear about exactly what this amp is and isn't capable of.
A better way to honestly rate it the way they used to with all channels driven continuous would have been 80 wattsX5 8 ohms amp which doesn't look nearly as good as their older stuff.


So it's not a trivial thing to rate amps in specific ways. It takes a lot for amps to be able to provide continous power and all channels driven power and they are the differentiators between them and significantly lesser amp. An amp that can provide 1000 watts continuos power like my XPA-1 gen 2 amp and an amp that does 1000 watts "burst" but can't manage over 60 watts of power continuous is a big difference. It takes money materials, and build (that you are paying for) to have high continuous watt ratings.

The Ncore wouldn't look as good if it was advertised as a 2X40 watt amp for $2000 despite its stellar distortion measurements. But that's what it is.

 

[jmmaher]

I have no measuring equipment.
I do have a stereo amplifier using NCore DIY modules with their 1200 watt power supply. I have never heard any compression or clipping using there with normal speakers (87) even when testing them at loud volumes. I have had two power supplies burn out in 3 years. Amps are on 24 hours per day. Both times the power supplies were replaced or repaired at no cost (other than shipping) by Hypex. Obviously a heat issue but no observable problems in real world use (sound quality - - not heat issues)

 

[gtb75]

I was browsing the member area and stumbled across this: https://www.audiosciencereview.com/...pex-ncore-amp-cases-and-might-make-more.5969/

Looks like he has an archived board on AudioCiricle: https://www.audiocircle.com/index.php?board=129.0

Definitely beautiful machine work!

 

[Ron Texas]

I read somewhere that the large heatsinks of the Benchmark AHB2 is only needed for passing tests. Absolutely not needed for listening to any kind of music.
That is kind of crazy in a way, making the product more expensive for no REAL reason.

Yeah, but it's pretty and looks powerful.

 

[amirm]

The problem I remember that caused FTC regulation was momentary power specs. These were ratings of producing such for milliseconds. Now it is stated as peak, or peak music power. That is not what I am measuring. I am measuring continuous power for a minute or so.

What we have here is a middle ground so I am still hesitant to say it is a useless approach and we should only measure them under strict FTC guidelines.