Apollon Hypex NC2K Amplifier Teardown

[March Audio]

So we can conclude brand capacitors are better than cheap chinese capacitors? and a $2000 amplifier should not try to save $10 on capacitors.

YOU can conclude whatever YOU like

 

[restorer-john]

I'd have no problem paying $10 extra to have high quality Japanese capacitors in my amplifier instead of questionable Chinese caps, especially when we are talking a $2000 amplifier then it's not even a questions, everyone would go with that option.

High quality capacitors are expensive. The differences between these Hypex third tier caps and premium capacitors is significant.

Do you think for one second respected HiFi companies compromised on their top of the line models when it came to PSU capacitors? No, they didn't. They worked hand in hand with Nichicon, Elna, Matsushita, Rubycon, Nippon Chemicon and had capacitors custom wound (geometry/dielectric etching patterns etc), custom values, sizes, electrolyte compounds, lead out material upgrades etc.

And it paid off. Here's a set of four Nippon Chemicon 22,000uF 100V capacitors from a TOTL 1978 Pioneer SX-1980. Each one weighs 1.5lbs. They are bigger than a beer can (375ml coke can for scale).

The biggest receiver they ever built and the heaviest, at 34kg. And the most expensive. These capacitors are now 42 years old and test perfectly. Date code week 21 1978. All are around 10-20% above their capacitance rating (unlike modern cheap caps where they are more oftern than not 20% below spec. ESR is low enough that I cannot measure a difference between them and the brand new ones I put in the unit. The owner wanted them replaced along with a full rebuild so the receiver would last another 40 years. Will the United Chemicons I used last as long as these originals? I don't think so, but what do you do?

Hypex use cheap, non premium capacitors in their TOTL power supply and TOTL power amplifier. That is a fact. It is that contempt for their customers I find hard to stomach. No Japanese maker would do that. I could show you the internals of a hundred Japanese amplifiers, ranging from $499 units to $6000 statement amplifiers. Not a cheap Chinese/Korean cap anywhere. No US manufacturer would use cheap capacitors in their amplifiers either. Look inside a Benchmark AHB-2 and tell me what you see. You see Panasonic, United Chemicon and Nippon Chemicon. All tier one components.

...so in your experience are reputable Japanese caps more or less likely to leak than cheap Chinese ones?

I know you directed the question at Alan, but I could dig out the wall of shame Chinese capacitors I keep in a few boxes and give you a run down, along with the Japanese capacitors I have removed which are still all in spec at 30-45 years of age. I certainly don't have a huge number, maybe the last few years worth of work as they get thrown out.

Japanese capacitors (or at least the Japanese big brands, even though they are made all over the world now) are less likely to catastrophically fail than any Chinese/Korean/Tiawanese caps you'd like to mention. No comparison whatsoever- they hold on in situations where lesser caps explode.

 

[March Audio]

With justification of having seen such components fail. You can question his experience with such components if you like but not concluding that such components don't belong in a SOTA module. This is what a consumer would expect too. I can see your defense as a manufacturer using Hypex modules but those are subject to bias.

That depends on the scale with which they fail and the distribution within various producers assuming the owners even come to know of the reason. This isn't just a Class D issue. Adcom amps for example were notorious for using bad caps in some models and batches that would leak and damage the board and the components around them. It only became known a decade or more later, not at the time people were buying them.

Poor drives from Seagate and WD are well-known but how much data do we have we of a specific computer brand failing from using them? Doesn't mean it doesn't happen.

If I was a consumer looking for something like this in this price range, why would I willingly put myself to be part of a statistic in the future? Like I said before absence of evidence isn't evidence of absence. You are using the same fallacy.


Like I said earlier, the burden of proof would then be on the Hypex module maker/integrator to show that the operating conditions are different in these modules to not suffer the same fate.


This is a rather weak "both-sides" argument.

I think the gent doth protest too much.

No, he has seen different components fail in different products with different operating conditions.

Your second sentence is only based on the incorrect assumptions made in the first sentence.

Its not a fallacy, I have direct experience of these amps not failing due to these capacitors (yes I have been using them longer than I have been manufacturing). Considering that the assertion is that these are allegedly "renowned" for failing it would be by definition be a KNOWN and documented issue. Except of course you cant provide any references to it.

You can call it bias, I just call relying on facts and science, not myth and rumour.

When you come up with some evidence that these caps in these products are fundamentally unreliable we will have something to talk about.... until then....

Also have to ask again why do you put the emphasis on "in this price range" as if its some kind of high end premium product? For the performance it offers it is not expensive.

 

[March Audio]

High quality capacitors are expensive. The differences between these Hypex third tier caps and premium capacitors is significant.

Hypex use cheap, non premium capacitors in their TOTL power supply and TOTL power amplifier. That is a fact. It is that contempt for their customers

I know you directed the question at Alan, but I could dig out the wall of shame Chinese capacitors I keep in a few boxes and give you a run down, along with the Japanese capacitors I have removed which are still all in spec at 30-45 years of age. I certainly don't have a huge number, maybe the last few years worth of work as they get thrown out.

Japanese capacitors (or at least the Japanese big brands, even though they are made all over the world now) are less likely to catastrophically fail than any Chinese/Korean/Tiawanese caps you'd like to mention. No comparison whatsoever- they hold on in situations where lesser caps explode.

John thats total nonsense. Your attitude is purely down to your penchant for 1970s electronics and nothing to do with the reality. You know nothing of the usage of the components that are OK after 30 years, could have been barely used, so you cant draw the conclusions you are.

Come on John, still waiting for an example of exploded Hypex caps

 

[Vasr]

I don't have any vested interest here since I am not in the target market for this kind of amp.

But for anyone that does DIY PC builds, PSUs without Japanese capacitors is a non-starter except for the very low-end < $50 PSUs in budget builds. So much so that even Chinese brands and ODMs that build PSUs advertise use of Japanese caps in their PSUs. And this is in a $100 PSU for a typical $1000+ build! So, I don't get why this isn't an expected thing in high-end audio when people are paying much more. Even just as a precautionary measure. The price difference isn't a deal breaker.

The problems with capacitors in PSUs are well documented and tiers well-established. For example:

https://www.tomshardware.com/reviews/power-supplies-101,4193-5.html
The above link contains the brands arranged in tiers.

While PSUs undergo a lot more temperature stresses than perhaps in this audio use, the top tier ones are also preferred for better QC in their production within their operating range.

What I expect will happen is that one of the Hypex integrators is going to put out modules with Tier One capacitors (unless Hypex arrangement forbids it) as standard and make that a big selling point (as the amount of TLC that goes into their products). All the competitors are then going to jump in with their marketing brochures screaming how they care about the longevity of the product and use only the best components.

 

[Haskil]

In French : il est instructif de constater que Restorer John ET March Audio ont raison tous les deux !
D'un côté, une attitude logique qui consiste à demander à un fabricant d'utiliser les meilleurs composants dans ses circuits d'appareils de haut de gamme.
De l'autre, une attitude pragmatique qui consiste à dire qu'il faut adapter la qualité des composants à l'endroit du circuit et leur utilisation en conditions réelles.
La première attitude fait penser à ses amateurs de tweaks qui par principe changent tous les condensateurs des appareils qu'ils achètent pour améliorer des performances qui étaient déjà au dessus du seuil d'audibilité des défauts éventuels...
La seconde attitude fait quand même penser que trop de fabricants diminuent la qualité des composants employés année après année avec le danger de finir par en employer qui posent des problèmes réels. On voit ça tous les jours avec les appareils - pas hifi - que nous employons : durée de vie limitée, réparations plus coûteuses qu'acheter un produit neuf...

La morale de tout ça ?
Des composants industriels de grade élevé donnent plus de chance au produit de conserver longtemps ses performances mesurées quand il est neuf, c'est incontestable.
Des composants de qualité moyenne risquent davantage de dériver voire de tombe en panne...

Mais bon, j'envoie ce message à l'aide d'un modem ADSL qui est brûlant et n'a pas subi une seule panne depuis des années qu'il est allumé 24 heures sur 24 bien qu'il chauffe beaucoup trop tout comme le décodeur de télévision par satellite tout aussi brûlant... Et je ne crois pas qu'ils contiennent des condensateurs autres que chinois bas de gamme... Ils doivent donc être choisis d'une façon qui fait qu'ils supportent d'être sous tension 24 sur 24 dans une petite boite fermée sans aucune aération...

 

[tomtoo]

Is it not possible that chinese cabs got better in the last years and hypex not sees a reason for the price difference any more? We should ask someone from hypex.

 

[restorer-john]

...We should ask someone from hypex...

Great idea. That's like asking Apple if their butterfly keyboard is reliable or asking an OLED TV manufacturer whether burn-in is an issue....

Or maybe go ask Louis Rossmann about Apple's quality and reliability? Which do you think is more qualified to give you the truth?

 

[tomtoo]

Great idea. That's like asking Apple if their butterfly keyboard is reliable or asking an OLED TV manufacturer whether burn-in is an issue....

Or maybe go ask Louis Rossmann about Apple's quality and reliability? Which do you think is more qualified to give you the truth?

Oh, at least they could give you insight on there decision process. I mean do we have objective data comparing the actuall used caps to higher priced caps? Do we know how chinese manufacturing an QA process has changed over years?

 

[restorer-john]

Do you think Mouser Electronics, a company owned by Berkshire Hathaway (Buffet) would sell Su'scon, Samwha or Aishi? Let's find out!

Nope.

Let's have a look at Farnell (Element 14), RS components, Digikey etc and see what brands they sell.

See any Su'Scons? Or Samwhas? Aishis? Nope.

RS:

I could go on but you get the point.

If these component suppliers, companies that supply manufacturing, research, aviation, the military, and HiFi manufacturers don't stock these crappy capacitors, why should you, the consumer get them in a TOTL amplifier module and SMPS from Hypex? Answer: You shouldn't.

 

[fredoamigo]

the concern of long-term reliability is especially important for active monitors . Genelec uses proprietary modules while D&D 8c or the latest DYNAUDIO CORE use PASCAL modules .

to think that inside these speakers there are cheap chinese capacitors ... would really cool me down ... and would not incite to a total peace of mind because here in case of breakdowns it would be a very big problem ...

 

[restorer-john]

Do we know how chinese manufacturing an QA process has changed over years?

My experience is the cheap electrolytic capacitors have a shorter life than they did 10-15 years ago. As they are making them smaller, less thermal mass, electrolyte and aluminium to disipate the internal heat.

 

[restorer-john]

to think that inside these speakers there are cheap chinese capacitors ... would really cool me down ... and would not incite to a total peace of mind because here in case of breakdowns it would be a very big problem ...

Have look inside. If they disappoint you, tell the manufacturer. They will only listen if enough people make enough noise. Hiding behind a pretty front panel or a name badge means nothing. It's what's inside that counts, especially for the long term.

 

[McFly]

I put out the call; does anyone know anyone that has had a capacitor failure in a hypex or similar class D module, in the 10 to 15 years they been out, that can prove it; thus far?

 

[Willem]

Gitzo tripod

They are great. We have a small carbon one for light weight, and it is superb (light and stable). We use a Leitz balhead, but I am still waiting for the first ultralight fluid video panhead that can also do vertical still shots. We now have to take two heads.

 

[March Audio]

Do you think Mouser Electronics, a company owned by Berkshire Hathaway (Buffet) would sell Su'scon, Samwha or Aishi? Let's find out!

View attachment 88858
View attachment 88859

Nope.

Let's have a look at Farnell (Element 14), RS components, Digikey etc and see what brands they sell.

View attachment 88860
View attachment 88861

See any Su'Scons? Or Samwhas? Aishis? Nope.

RS:
View attachment 88862
View attachment 88863

I could go on but you get the point.

If these component suppliers, companies that supply manufacturing, research, aviation, the military, and HiFi manufacturers don't stock these crappy capacitors, why should you, the consumer get them in a TOTL amplifier module and SMPS from Hypex? Answer: You shouldn't.

John That is TOTALLY meaningless. Can you stick to facts instead of making up stories?

 

[Willem]

These amps with their high voltage capability would be just the ticket for big electrostats like my Soundlabs

They would be too powerful for my Quad 2805s though. I know, I control the volume, but still.

 

[March Audio]

I don't have any vested interest here since I am not in the target market for this kind of amp.

But for anyone that does DIY PC builds, PSUs without Japanese capacitors is a non-starter except for the very low-end < $50 PSUs in budget builds. So much so that even Chinese brands and ODMs that build PSUs advertise use of Japanese caps in their PSUs. And this is in a $100 PSU for a typical $1000+ build! So, I don't get why this isn't an expected thing in high-end audio when people are paying much more. Even just as a precautionary measure. The price difference isn't a deal breaker.

The problems with capacitors in PSUs are well documented and tiers well-established. For example:

https://www.tomshardware.com/reviews/power-supplies-101,4193-5.html
The above link contains the brands arranged in tiers.

While PSUs undergo a lot more temperature stresses than perhaps in this audio use, the top tier ones are also preferred for better QC in their production within their operating range.

What I expect will happen is that one of the Hypex integrators is going to put out modules with Tier One capacitors (unless Hypex arrangement forbids it) as standard and make that a big selling point (as the amount of TLC that goes into their products). All the competitors are then going to jump in with their marketing brochures screaming how they care about the longevity of the product and use only the best components.

You would have to purchase the rights to use the design and manufacture your own boards or come to some arrangement with Hypex. A costly exercise unless you are selling a very significant number of units.

Removing and replacing caps from existing boards will invalidate the Hypex warranty and runs the risk of actually degrading reliability due to that removal/replacement process.

All for something that has yet to be demonstrated as a problem

We all know John wants his HiFi to last 50 years but technology is moving on at a faster and faster pace. You may not like it but how many of todays hifi products will be relevant in 20 years?

 

[March Audio]

They would be too powerful for my Quad 2805s though. I know, I control the volume, but still.

Its funny, I have had a small number of enquiries regarding offering a NC2K based amp and when I ask what speakers they plan to use with it they have all been completely incapable of handling the power. Very few domestic speakers can.

 

[Apollon Audio]

I put out the call; does anyone know anyone that has had a capacitor failure in a hypex or similar class D module, in the 10 to 15 years they been out, that can prove it; thus far?

We sold more than 1000 amplifiers worldwide. We are in this business since 2015. From around 1000 Hypex based amps that we sold there were not more than 15 faulty modules. None had a bad cap issue. 90% of the faulty modules were the Hypex MP series. The fail rate of Hypex modules is very low. I’m not advocating these caps used by Hypex at all. I’m just sharing our experience with these modules. Most of these caps have a lifetime of 5.000 rated at 105°C. None of these caps on the modules run at 105°C so a far longer lifetime is to be expected.