1. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

KJF audio Introduce the MA-01 flexible amplifier

Discussion in 'audio' started by orangeart, Dec 12, 2018.

  1. orangeart

    orangeart KJF Audio Ltd.

    Following in a long line of products launched and supported by members of the PFM family here is my next venture.

    After months of hard work I am ready to release my new amplifier- the MA-01. This is your chance to get one for a discount as the initial production run is being funded through a Kickstarter campaign. Kickstarter have become the leading website for launching new and exciting products and attract investment from across the world and bring products to highly expectant markets with a wide user base.

    We are well used to supplying very high performing speakers at bargain prices and wanted to design an amplifier that was flexible and truly crossed audio boundaries. Flexible enough that each build could be a custom build while making construction a standard process. In this way the customer could buy an amp that matched their needs without me having to do a ground up custom build for each customer. This locks in the cost saving for the customer associated with production line techniques while maintaining a flexible design. So flexible it can even be reconfigured should the customer decide at a later date.

    The design utilises the Hypex N-core modules which appear in some of the most sought-after high-end amplifiers, so absolute sound quality is beyond doubt. They come in a range of power output from 100 Watt to 500 Watt so power is available too. The amplifier itself can be configured with up to 8 channels so a customer can have a 2 channel 125 Watt mono-blocked amplifier for their top flight audiophile speakers or 2x 500 Watt, 2x 250 Watt and 2x 125 Watt for their active speakers, in fact there are many different configurations possible.

    The amplifier will be equally at home in a home environment or a portable gig rack for small gigs with the optional rackmount front panel. With its intelligent active cooling even the hardest of loads at the highest volumes won’t phase it.

    The package is topped off with a simple but intelligent user interface communicated to the user via the ring light at the front. Clip indication, over temperate indication, power state, DC detection and module failure are all indicted and the amplifier will auto shut down when protection is needed and indicate to the user why.

    To help make the product a reality take a look at the full specifications and introductory video over at the Kickstarter page.


    Mintsauce likes this.
  2. Ian M

    Ian M pfm Member

    Seems a good deal but I know nothing of these N-Core amps. How would the sound compare to an Avondale CAP6/NCC220 amp. Eagerly awaiting to hear what others think of this project.
  3. orangeart

    orangeart KJF Audio Ltd.

    @Ian M

    I spent years using Naim amps and doing all the usual mods, lovely they are too. Later I built NCC200 amps with Les' excellent boards and lastly moved onto the HackerNap boards that are similar to Les' voyager setup with a slightly different decoupling and earthing setup. That last amp was absokitabs brilliant, MUCH better than a NAP250 and a credit to what it possible when like minded folk come together to deliver something without commercial comstraints.

    I moved into designing active speakers that needed 6 channels, the HackerNap build was so huge I just couldn't face doing 4 more channels!

    At that point I jumped ship to the hypex modules that at that time were the UCD modules. They were simply brilliant, they just got out of the way of the music, seeming to offer no auditory footprint of their own. Just what I needed to develop speakers.

    The UCD amp gave way to the Ncore which has garnered a huge following and has appeared in some really high end amplifiers from top marquees, the modules have continued to develop so that they can work as part of a really well integrated and intelligent design which is what we have here.

    Sound wise they are quite different to the Naim offering in that they control bass much better and have a much lower noise floor. Some folk might find that the extra detail on offer is a bit much for them, for folk that want the proverbial strsight wire with gain that are load agnostic these modules are as good as it gets IMHO.

    Being class-D with properly designed SMPS they are also very efficient meaning you use much less electricity for added kindness to the planet which is important to me.

    YNWOAN 100% Analogue

    By ‘modular construction’ it sounds like it is constructed from modules, not that the customer can reconfigure it themselves. I can easily see the benefit to the manufacturer of such a modular construction but the customer benefit is less clear.

    Regarding the N-Core amps; I’ve heard a number of amps based on them and wound say that they actually have quite an obvious sonic character.
  5. Purité Audio

    Purité Audio Trade: Purite Audio

    No, I would say they have no sonic character at all.
    Just really fuss free, transparent amplification.
  6. Ian M

    Ian M pfm Member

    Thanks Stefan. I’m currently running Frugal horn Mk3s with my Avondale amp. This goes plenty loud enough although I wouldn’t want any less power. Would the Dual mono 250w per channel be overkill. I’m presuming dual mono will be better sound quality?

  7. orangeart

    orangeart KJF Audio Ltd.

    Hey Ynwoan, take a look over at the kick-starter page, all is explained there. Yes the amplifier (as a whole) is made up of amp modules from Hypex and a microprocessor control board for house keeping duties designed by me. It's not able to be reconfigured at home, it would need to be returned to base for that. The beauty of the modular construction form a customer point of view is that, firstly they can get the amp they really need in the first place, without the niggling doubt that when their needs change, they will have to buy another amplifier. Simply return it to base and add extra channels or higher powered modules. Secondly they benefit from lower pricing, it is no different to build a 6 channel amp from a construction point of view as the case is the same with the same drill holes pre drilled, the control board is the same and so is the software controlling the boards, therefore a mass production technique can be applied to what is essentially a custom amplifier.

    It always really interests me how we all view sonic character differently, I don't think they really add or take anything away and that seems to be the general consensus with these amps when a review of a product containing them crops up, however, like all things HiFi your mileage may vary. I'm not the kind of person that goes around saying things like, 'it's the best amp ever', or 'amps by XYZ are crap in comparison' as I know that is futile. We all like what we like or what we have got used to, so no point arguing differently or denigrating the competition. Like all of my other speaker products I just want to bring something that is as good as I can do it at sensible prices. Of course not everyone will like those products, that is a given I think. My main concern is that I bring something to the party that wan't there before without the customer having to sell their children to do so :)
  8. orangeart

    orangeart KJF Audio Ltd.

    I bet those FH Mk3 sound great with the Avondale amp. I've not got any Naim type stuff around here at the moment which is a shame.

    Overkill, what's that!! HAHAHA I run mine with 500W per channel, although be careful, this happens if you let your children and brother in law loose.



    Seriously any size amp is just fine but you do obviously need to be careful with the loud pedal. I'd be more inclined to go with dual 125W as there is a little less chance of the above happening.
  9. Shadders

    Shadders pfm Member

    I would have to disagree.
    A Class A, A/B amplifier is as accurate as you can get. The signal path through the amplifier can be traced - they are analogue amplifiers and the output is a continuous function (smooth - no discontinuities).

    For class D, this is an approximation. The output voltage is always an approximation of the input signal in terms of precision. I examined the Hypex N-Core NC400 datasheet. The specification for THD in the bandwidth 20Hz to 20kHz is stated to be no more than 0.002%. This is excellent performance and better than most commercially available class A/B amplifiers. So, if class D is an approximation, then how can it exceed the performance of a class A/B amplifier whose output is a continuous function ???

    The datasheet for the N-Core NC-400 has a note for the THD specification which reads :
    "Note 1: At higher audio frequencies there are not enough harmonics left in the audio band to make a meaningful THD measurement. High frequency distortion is therefore determined using a 18.5kHz+19.5kHz 1:1 two-tone IMD test."

    What this is saying is, that at 10kHz, the harmonics are outside the audio band (greater than 20kHz). So this means that the THD specification (0.002%) is based only on the measurement bandwidth which i would assume has been set to 22kHz (as seen in other measurements for other amplifiers using the Audio Precision (AP) THD analyser). If you examine the NC400 datasheet, section 6 Typical Performance Graphs, the 4th graph shows the THD reducing as it passes 10kHz. So, one would think the amplifier gets better (lower THD) as it increases in frequency.

    The AP THD measurements i have seen usually use 30kHz or 80kHz measurement bandwidth. This provides a more accurate representation - levels the playing field.

    If you examine the Elektor design for a class D :

    Plot G or plot I, it shows both the 22kHz and 80kHz measurement bandwidth curves. As can be seen, at 20kHz with the 80kHz measurement bandwidth, the THD exceeds 0.1%. Many class A/B amplifiers will be at least 10x less than this. The curves for 22kHz measurement bandwidth are similar in shape to the Hypex NCore NC400

    After all this text - my point would be that class D is not transparent since it has a worse performance than a good class A/B amplifier. It will have a sonic signature - just different to a class A/B.

  10. YNWOAN

    YNWOAN 100% Analogue

    Yes, I watched the KS video.
  11. Purité Audio

    Purité Audio Trade: Purite Audio

    Shadders I am sitting in front of Ncores, Classic A/B and the Benchmark hybrid, once level matched you cannot tell them apart.
    I have had poor class D we had some SPEC mps that just weren’t right and that was reflected in their Stereophile measurements, but every NCore measurement I have seen places them in terms of measurement as the very finest measuring available.
  12. JensenHealey

    JensenHealey pfm Member

    If anyone thinks they can tell if an amp has 0.1% distortion at 20Khz, then seriously, go get your ears tested. It has to be simply nonsense. You may be hearing something else, but distortion it ain't
  13. Shadders

    Shadders pfm Member

    Hi Keith,
    If you are listening at levels where the THD is so low that the differences are not heard, then OK. As per the other thread, it is the distortion characteristics that gives an amplifier its specific sound as proved by Bob Carver.

    My point was, that class D is not transparent based on their inferior high frequency performance. I searched on "hypex ncore stereophile" and the top response was :


    These are $12k per pair of monoblocks. The measurements show that at 20kHz for all impedances that the THD is 0.08% (50watts into 8ohms).

    My design for a class A/B amplifier which is essentially Bob Cordell's book design (Fig 3.14) with only 2 pairs of transistors for the output stage, for 500watts peak into 4ohms, under simulation, the THD is 0.007%. This is 10x better at a much higher power, and i am a novice designer. If i match the stereophile 20volts RMS into 4ohms, then the THD is 0.0038% - which is 20x better than the Theta Prometheus. My measurement bandwidth is 800kHz, reducing to 80kHz, the THD is 0.0026%.

    As such, the NCore module used in the Theta Prometheus $6k monoblocks are really, not that good when compared to a class A/B amplifier.

    I was only disagreeing with the description of transparent, since their THD profile indicates a worse performance than a good class A/B amplifier.

  14. Ian M

    Ian M pfm Member

    Stefan, I can’t seem to see a pledge for a dual mono 125? And yes they sound very good!
  15. Purité Audio

    Purité Audio Trade: Purite Audio

    I forwarded your post to Jan at Hypex, he is always busy but if he does reply I will post it.
    Ncores areas used in two out ot three of the most transparent loudspeakers I have heard, Grimm LS1s that was the first time I heard them.

  16. Shadders

    Shadders pfm Member

    Hi Keith,
    If it was active speakers then maybe you were hearing the benefit as such ???

    Either way, a good class A/B amplifier will always have a better performance than a class D for high frequency response (10kHz to 20kHz) given the current switching speeds for class D.

  17. Ian M

    Ian M pfm Member

    Can someone tell me the output of the Avondale NCC220 boards and will the equivalent power rating of N-Core amps have the same loudness? Thanks
  18. orangeart

    orangeart KJF Audio Ltd.

    Hi Ian, there aren't Mono 125 W modules available. To be honest though the integrated PSU is mono after the switching part of the circuit with separate regulation for each channel. The audio circuits for both channels share a common ground plane but no other shared components. If you want Mono you will have to go for the 250W
  19. orangeart

    orangeart KJF Audio Ltd.

    I don't think you can really compare those results, it's a different amplifier, it might be class-D but as we all know just because it's the same topology, doesn't mean it's the same. It is certainly true that that to remove the switching element of the output, filtering is required of that signal to return it to the smooth analogous output you described (remember class-D doesn't mean digital) . The output filtering has been the Achilles heel of older Class D designs and some from other manufactures, that is because the filters interact with the load (speakers and cables) and although that is true also of some other topologies, early class-D designs did suffer with the high frequencies being shifted one way or the other. The amp you have linked to above is one of these and they are testing for the THD distortion introduced using different filter elements and feed back elements. The N-core is a different beast it doesn't do it's output filtering in the same way so less load interaction meaning a load agnostic HF output is achieved.

    The THD is measured using the accepted standard for devices that include a switching element which is min 80Khz in the case of the N-core. The AES17 standard which I believe calls for the signal to be filtered at 40Khz pre testing so a pretty level playing field :)
  20. Shadders

    Shadders pfm Member

    Yes - i understand that the Elektor amplifier can be set up with different feedback options, and hence for different loads.

    My point was, the specifications "seem" to use a measurement bandwidth of 22kHz, and the shape/form/profile of the specification THD is similar to the Elektor amplifier which is also measured with 22kHz measurement bandwidth. Of course, the Hypex NC400 has a better performance than the Elektor amplifier.

    If you then examine the other link to Stereophile - the Theta Prometheus amplifier has been tested with the measurement bandwidth of 80kHz as per class A/B amplifiers, and it has a much worse performance at higher frequencies than a good class A/B amplifier.

    This is all i was saying, and that the description "transparent" is perhaps not the most appropriate description given the high frequency performance of class D amplifiers - they will have a "sound" as do class A/B amplifiers.

    For AES17 - seems to be a standard for testing digital audio equipment, not amplifiers. It refers to 40dB stop band, and 20kHz bandwidth, but not 40kHz bandwidth.

    Cereal Killer likes this.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice