• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Audible noise/hiss from Hypex NC502MP based amp (KJF MA-01), is it to be expected?

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
Hi. I've recently received a KJF MA-01 amplifier configured with 3x Hypex NC502MP modules.

When testing it with my speakers (Ascendo Immersive Audio CCRM12-P), which are specified as fairly sensitive at 99 dB/W, I can hear a hiss from the speaker.

My room is a dedicated mancave/tiny home theater with decent sound isolation. The background noise is fairly quiet at about 29 dBA or lower (don't know if my Umik-2 measurement microphone is the limiting factor there).

My Denon AVR-X4200W also has some hiss, but it is only audible from about 1 m, while channel 2 on the MA-01 is clearly audible from my listening position about 2.5 m away from the speaker. Channel 1 on the MA-01 is pretty good though, with the hiss only audible from fairly close (say 0.5 m or so).

See attached pictures of my measurements (both individual unsmoothed and all with psychoacoustic) and corresponding mic setup.

When testing the MA-01 the input pins (XLR pin 2 and 3) were shorted using my highly advanced customized XLR cable (see picture below).

I've read other threads about higher noise on one channel of the NC502MP, likely because of input pins of one channel being very close to some noisy components. I guess my question is if it's possible that you just have to expect one channel to be noisier than a Denon AVR from 2015, or if it could be a problem with this particular amp?

all-psychoacoustic.png
background.pngdenon_avr_x4200w.pngkjf_ma-01_ch1.pngkjf_ma-01_ch2.png
kjf_ma-01_back.jpg
mic_setup.jpg
 

AdamG247

Strive not for validation, but rather to be valued
Moderator
Forum Donor
Joined
Jan 3, 2021
Messages
2,326
Likes
5,639
Can you post a video so we can hear the hiss ? Could help in trouble shooting. Thanks
 

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
3,010
Likes
6,043
Location
Prague
Maybe a record made with Audacity or something similar as a wav file. It can be analyzed, then.
 
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
Thank you, I've recorded all the variants (including background noise) with the mic in the same position using audacity, boosted +20 dB and exported as mp3. Since mp3 isn't an allowed extension I've zipped them.

Edit: I recorded them before seeing you suggestion to save as wav @pma, is mp3 ok?
 

Attachments

  • hiss_recordings.zip
    751.3 KB · Views: 48
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
Interestingly, my ~20 year old NAD 214 amp is the best by far, I can only slightly hear some hiss with my ear right up to the speaker...
nad_214.png
 
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
Okay so this has been bugging me alot so I decided to try something on a whim.

I decided to try and shield the signal ribbon cable. I happened to have some 1.5 mm steel fittings plates here, so I wrapped them in electricians tape (to avoid shorting something or shocking myself).
tools.jpg


Opened up the amp and found the signal cable:
open.jpg

Here:
before.jpg

Put the plates in place:
after1.jpg

after2.jpg


And to my huge relief, the speaker is dead quiet, I can't hear anything even with my ear right at the speaker! So now it's better than any of my other amps, which frankly is what I expected from such an expensive and well known amp module. Just to make absolutely sure I wasn't fooling myself, I carefully removed the plates, and the hiss was back.

I'll be contacting KJF Audio to hear what he has to say. I think it's basically a design defect to run the ribbons along the power supplies like this. I guess it should either be shielded, or run a different (longer) path well away from the PSU's.

Any suggestions on how I can fix my amp in a more permanent manner? I could always manage to permanently attach some similar metal plates somehow, but perhaps a better approach would be to shield the ribbon using some copper tape, connect that to ground, and wrap it all in electrical tape again? Or is it possible to buy compatible ribbon cable with a shield already on it?
 

sq225917

Addicted to Fun and Learning
Joined
Jun 23, 2019
Messages
868
Likes
903
You can get the same idc wire twisted inside a foil screen with an exteruded outer. It's a bugger to route though.
 
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
You can get the same idc wire twisted inside a foil screen with an exteruded outer. It's a bugger to route though.
The amp isn't very crowded so I doubt routing would be a huge issue. Any hint how/where to get that kind of cable? My knowledge on the subject of ribbon cables is absolutely zero :p
 
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
On a related note, I wonder what the results would have been if Amir had run his MA-01 Review on channels 5 and 6 instead of 1 and 2. In the teardown it looks like channels 1-4 avoid running the ribbons along PSU's, while 5 and 6 do not.
 

sq225917

Addicted to Fun and Learning
Joined
Jun 23, 2019
Messages
868
Likes
903
Qbd, I bought mine from Farnell, mouser sell it too.


Search shielded idc cable, shielded ribbon cable
 

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
3,010
Likes
6,043
Location
Prague
Thank you, I've recorded all the variants (including background noise) with the mic in the same position using audacity, boosted +20 dB and exported as mp3. Since mp3 isn't an allowed extension I've zipped them.

Edit: I recorded them before seeing you suggestion to save as wav @pma, is mp3 ok?

The results are poor. Briefly viewed and listened, it looks like cable or grounding problem or both. Detailed drawing of all components, interconnections and cable types would be needed, and also a note which of the components is class I (signal gnd connected to PE, 3-pin mains plugs/sockets).
 

sq225917

Addicted to Fun and Learning
Joined
Jun 23, 2019
Messages
868
Likes
903
My money is on it being proximity to inductors
 
  • Like
Reactions: Qbd

audioBliss

Active Member
Joined
May 7, 2019
Messages
258
Likes
282
Location
Sweden
I had a similar issue with my NC502MP modules. Here is the TLDR of that thread https://www.audiosciencereview.com/...llon-ncmp6350-hypex-nc502mp.15971/post-534976 (Maybe it's the thread you found?)

I did a lot of testing on my modules - one of the channels is just noisier but completely goes away when using a balanced pre-amp like an AV processor with XLR outputs.

I'm quite sure that this noise is on all of these modules and they count on the customer always using XLRs. I'm not sure why it's not talked about more because most people don't run a balanced preamp. For 2-channel maybe but for multichannel setups it's rare due to costs.

If it's of any help I did not have this issue with my Ncore NC400 which have a different design. They are a lot more expensive though..
https://www.audiosciencereview.com/...llon-ncmp6350-hypex-nc502mp.15971/post-528095
 
Last edited:
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
I had a similar issue with my NC502MP modules. Here is the TLDR of that thread https://www.audiosciencereview.com/...llon-ncmp6350-hypex-nc502mp.15971/post-534976 (Maybe it's the thread you found?)

I did a lot of testing on my modules - one of the channels is just noisier but completely goes away when using a balanced pre-amp like an AV processor with XLR outputs.

That’s the thread I had read yeah. It may well be a similar cause. My amp is extremely noisy with nothing connected, but the noise pickup is probably just so bad that even with pins 2 and 3 shorted it’s still audible (but much less than when unconnected).


The results are poor. Briefly viewed and listened, it looks like cable or grounding problem or both. Detailed drawing of all components, interconnections and cable types would be needed, and also a note which of the components is class I (signal gnd connected to PE, 3-pin mains plugs/sockets).

Since the metal plates between the ribbons and PSUs fixed it, I’m guessing it’s picked up noise from the PSUs. Luckily seems relatively easy to fix at least!
 

sq225917

Addicted to Fun and Learning
Joined
Jun 23, 2019
Messages
868
Likes
903
@audioBliss, but these are power amps, the last item in the chain, it doesn't matter if the inputs are balanced or SE coming from the pre-amp unless you're running mega long cables. The noise is internal to the power amps irrespective of source.
 
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
I thought I should provide a quick update here.

I did try shielding the signal ribbon cable, but it actually had essentially no effect on the noise with pins 2 and 3 shorted. It did drastically reduce the noise with the input pins open (which is really bad out of the box), but that’s not really a problem so the shielding isn't very interesting. I tried leaving the shield ungrounded, or connected to the amp ground (the Wago-like ground clamps on the amp PCB).

So instead of trying to manually attach heavy steel plates inside the amp, I loosened the ribbons and tried rerouting them. For one of the modules there is no way to avoid running past a PSU, but obviously not all components are as noisy.

Here’s the original (before mods) picture:
original.jpg


And here is the rerouted setup. Note that only the widest ribbons are signal cables, so the narrower cables are left in their original route:
new.jpg


This routing is basically completely silent. I'm guessing the culprits are the inductors with some green stuff visible on their tops (sorry I've no idea what they are for or their proper names), but I am by no means sure about that!

Anyway this solves my problem, so the amp will be put in service as planned like this. I don't much want to have to do this again though, so for additional amps I'm probably going to look at Apollon Audio. Based on the pictures I've seen of their amps, the boards are oriented with the PSUs towards the front of the case, so no signal or output cables have to cross by the PSUs. From what I can tell the only thing I'd miss out on compared to this KJF amp is active temperature-dependent cooling and clipping indicators.
 
OP
Q

Qbd

Member
Joined
Aug 27, 2021
Messages
56
Likes
46
@audioBliss, but these are power amps, the last item in the chain, it doesn't matter if the inputs are balanced or SE coming from the pre-amp unless you're running mega long cables. The noise is internal to the power amps irrespective of source.
I've not been able to get this amp to be noise-free with single-ended input from my AVR, while another (RCA-only, ungrounded amp) works fine. It is dead silent with balanced input though (only tested with my Topping EX5). My general impression is that the hypex modules are a crapshoot with unbalanced sources, at least if you put it in a rack with various equipment and/or the source is e.g. an AVR with multiple HDMI sources. Perhaps if someone made an ungrounded double insulated version it would work better.

I'll be switching to a processor with XLR outs to hopefully never have such problems again.
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
2,183
Likes
2,069
Location
.de
I've not been able to get this amp to be noise-free with single-ended input from my AVR, while another (RCA-only, ungrounded amp) works fine. It is dead silent with balanced input though (only tested with my Topping EX5).
Now that part is sort of expected. You would need the right kind of adapter cable, possibly even a custom job that takes the AVR's output impedance into account which I bet is several hundred ohms. From Douglas Self - Balanced Line Technology:

balfig2.gif

If the AVR is already grounded somewhere else (which it usually is), you may also be getting issues with ground currents:

balfig3.gif

In this case it may be beneficial to leave the shield unconnected at point B. Assuming the amplifier is AES48 compliant, the culprit is most likely to be inside the AVR. (I can still remember very well when Denon AVRs were majorly affected by HDMI hiccups caused by ground loops. That obviously shouldn't be happening with differential signalling, so I bet it was something similar to a Pin 1 Problem.)
 
Top Bottom