• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Reliability of Hypex modules - Early failure of NC252MP module

OP
S

suttondesign

Addicted to Fun and Learning
Forum Donor
Joined
Feb 4, 2019
Messages
732
Likes
1,310
Location
Bellingham, WA
To have an amp module fail less than 2 years out is disheartening. I don't worry about the warranty issue, and in fact I don't mind installing a replacement module myself instead of shipping the whole amp back to England for Nord to do it, but this does worry me about these convenient, high-performing class-D modules generally.

ps

Update: After I sent Nord an explanation how I isolated the problem to one 2-channel module, Nord is shipping me a new module to install since it's such a straightforward swap-out.
 

AdamG

Proving your point makes it “Science”.
Moderator
Forum Donor
Joined
Jan 3, 2021
Messages
4,636
Likes
14,918
Location
Reality
Update: After I sent Nord an explanation how I isolated the problem to one 2-channel module, Nord is shipping me a new module to install since it's such a straightforward swap-out.
That’s great to hear and reflects well on Nord’s Customer Support. Nice!
 

retro

Active Member
Joined
Sep 22, 2019
Messages
164
Likes
296
As for older NAD amps, I had a couple of 2600A's which sounded fine to my ears back then. However, both failed after about three years. If I remember correctly, it was the output relays, a common problem with these. Cheap to replace, no problem.

But when I switched both to an Electrocompaniet AW250DMB, I realized the NAD's weren't the best sounding amps in the world. Ok for the price I guess, but the EC opened up a new world of music for me. Yes, totally subjective. Will I get banned now..:cool:
Now using a couple of Kevin Halvorsen amps...
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,280
Location
Gold Coast, Queensland, Australia
A failure is one thing. What caused the failure is the interesting part. If Nord don't want the dead module back, someone should do an autopsy on it, find the culprit (component/s) and generally investigate. Otherwise nobody learns anything.

I'll volunteer if postage to Australia is reasonable.
 

dfuller

Major Contributor
Joined
Apr 26, 2020
Messages
3,337
Likes
5,053
A failure is one thing. What caused the failure is the interesting part. If Nord don't want the dead module back, someone should do an autopsy on it, find the culprit (component/s) and generally investigate. Otherwise nobody learns anything.

I'll volunteer if postage to Australia is reasonable.
I'd like to see this too. My bet is a power transistor shorted and it went into protection, but I'm always curious.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,280
Location
Gold Coast, Queensland, Australia
I'd like to see this too. My bet is a power transistor shorted and it went into protection, but I'm always curious.

They use MOSFETs originally designed for diesel fuel injectors. They are pretty tough. My bet is on something else as a shorted/failed output device would present DC and the SMPS would shut down, taking the other channel (shared power supply) offline also. But one channel is still functioning. I'd like to see for myself.
 
Last edited:

dfuller

Major Contributor
Joined
Apr 26, 2020
Messages
3,337
Likes
5,053
They use MOSFETs originally designed for diesel fuel injectors. They are pretty tough. My bet is on something else as a shorted/failed output device would present DC and the SMPS would shut down, taking the other channel (shared power supply) offline also. But one channel is still functioning. I'd like to see for myself.
Huh, didn't know that. In that case, I'll defer to your thoughts.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,280
Location
Gold Coast, Queensland, Australia
Huh, didn't know that.

These are used in the NC-400OEM (from the schematic):

1620783687830.png


1620783794056.png
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,830
... don't use the best quality capacitors and nor do Purifi. One amp supplier offers to change these for higher quality caps that have a longer rating.

2 years is much too short a time though, even if it was on 24/7. Even though you say it doesn't get more than warm, the early failure suggests the caps were running close to the rated temp and so failed prematurely.

Have you contacted Nord yet about this failure?



JSmith


Caps normally degrade, not fail outright except for random failures. You can't make the guess you are making. Almost no supporting information.
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,830
About 35 years.;)

Found this detailed answer from @restorer-john about the difference between Class D amp's an the older Amp's.


What‘s the difference between NAD 2200 and NAD M22/M22.2 35 years.:)

Topology is completely different. The M22 is Hypex OEM NC400 Class D. The 2200 is a commutating rail through-hole dinosaur from the mid 1980s, but, as you can see, not only is it still working, it's repairable and not too bad looking for a NAD, and it still performs very well.

There's no way on earth the SMPS powered M22 will last 35 years. It's chock full of microcontrollers and SMD lead-free componentry. But, it is new and some people want that. I get it.


I expect this stuff on Facebook forums, not here. Not that "bad" microcontrollers! Not those awful PICS! .. Not that awful SMD and lead-free (just like everything else these days).
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,830
NAD replace them with United-Chemi-Con which is fine and Apollon Audio also offer a cap replacement on both NCore and Purifi amp modules. The standard ones used are supposedly rated at 5000 hours (at max temp) and the replacements are 20000 hours.

https://www.audiosciencereview.com/...28-7-channel-purifi-amplifier-teardown.15981/


https://www.audiosciencereview.com/...integrated-amplifier.20426/page-8#post-693708



JSmith

That 5,000 hours at maximum temperature and maximum ripple current. If you are running an average of 30C under max, and averaging a fraction of max ripple current you are more like 40-50,000 hours continuous, then again, it could be even cooler and last longer. That's 80% of initial capacitor value. It will still work.
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,830
They use MOSFETs originally designed for diesel fuel injectors. They are pretty tough. My bet is on something else as a shorted/failed output device would present DC and the SMPS would shut down, taking the other channel (shared power supply) offline also. But one channel is still functioning. I'd like to see for myself.

That's just some crap on a data sheet. It may have been on the data sheet, but that does not imply there is anything overly robust about them. They meet automotive quality requirements but so does a lot of stuff. The benefit for the injector would be the relatively high voltage that reduces snubbing requirements. The avalanche energy, which would be one of the robustness parameters, is about what one would expect at this voltage/current/switching performance.
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,830
Sure, but these tend to happen fairly quickly after first usage... not 2 years.

OP of that post has not advised Nord's response as yet.

JSmith

No. That is why they are called random failures, not infant mortality. Random failures happen, They happen all the time. When you build enough of anything and you have enough total operating hours across all units, then there will be random failures.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,280
Location
Gold Coast, Queensland, Australia
I expect this stuff on Facebook forums, not here. Not that "bad" microcontrollers! Not those awful PICS! .. Not that awful SMD and lead-free (just like everything else these days).

That's just some crap on a data sheet. It may have been on the data sheet, but that does not imply there is anything overly robust about them. They meet automotive quality requirements but so does a lot of stuff. The benefit for the injector would be the relatively high voltage that reduces snubbing requirements. The avalanche energy, which would be one of the robustness parameters, is about what one would expect at this voltage/current/switching performance.

No. That is why they are called random failures, not infant mortality. Random failures happen, They happen all the time. When you build enough of anything and you have enough total operating hours across all units, then there will be random failures.

You appear to be itching for an argument about something, anything, with anybody, regarding a failure on a product you haven't even inspected. :facepalm:

I sure can't be bothered with the attitude.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,280
Location
Gold Coast, Queensland, Australia
I'll leave this discussion with you now as I have no idea wtf you're talking about with this comment.

Early life failure rate is often called "infant mortality" by the less emotionally intelligent. Not a great idea on a public forum.
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,830
Early life failure rate is often called "infant mortality" by the less emotionally intelligent. Not a great idea on a public forum.

Or people who actually work in industry designing products and bringing them to market, who have to do reliability studies, etc. and who came from an era, not too long ago, where we didn't need safe spaces to get through the day nor be PC about everything. Ask any electronics product development engineer what infant mortality is w.r.t. electronics and they will know exactly what you mean.
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,830
You appear to be itching for an argument about something, anything, with anybody, regarding a failure on a product you haven't even inspected. :facepalm:

I sure can't be bothered with the attitude.

No, I am itching for people not to post wrong and misleading information on a technical site. There is nothing wrong with microcontrollers, and they don't lead to early failures, especially PICs. SMD is more reliable than through-hole. Almost everything is lead free today. There is nothing particularly robust about the FETs used and you can't ascertain they were specifically designed for diesel injectors from the data sheet, and random failures are a reality of any electronics product.
 
Top Bottom