Koshin caps... do they suck?

[weirdlookinguy]

I'm in the market for a kickass 2-channel receiver with a Phono input at or under the $300 mark. It's down to either a very nice Yamaha RX-497 or a very nice harman/kardon 3490. The Yamaha is 75wpc, the harman is 120wpc. However, I know that 75wpc is plenty and could rock a party no problem. The wattage is not a major deciding factor at the moment.

The harman/kardon is definitely better looking than the Yamaha, and looks borderline girly in person:

but what the hell, I like it and I'm a sucker for white LED's.

I'm still a month from making the purchase. But I've made some observations. All the Yamaha's seem to have Jamicon caps in them, some of the higher end ones had NCC filter caps but that was it. The HK3390, the 3490's little brother (80wpc), had all Koshin caps in it.

Does Koshin suck? Terribly? Cause if it does, then I need to take this amp off the list of possibilities immediately before I start leaning too heavily towards it.

[370forlife]

Re: Koshin caps... do they suck?

We have a harmon/kardon system at my dads, I can't remember what model and everything, but it's been working for 2 years now, and it gets LOUD. Their subs have a ton of bass, too.

As for Koshin, I've never heard of them.

[999999999]

Re: Koshin caps... do they suck?

No, Koshin doesn't suck, but can we use reasonable terms instead of just "suck"? If you want to know about general reliability of the whole amp then a web search for failures would be most productive.

[weirdlookinguy]

Re: Koshin caps... do they suck?

Ah, yes but the general reliability of anything these days is a crapshoot IMO. I didn't ask if you guys thought this amp would hold up, I asked about the quality of the caps.

In more reasonable terms then: Do Koshin caps lean more towards Rubycon/Nichicon/UCC (Great Quality), CapXon/Teapo (decent if it doesn't run too hot), or G-Luxon?

[999999999]

Re: Koshin caps... do they suck?

You are thinking in terms of capacitors in switching PSU subcircuits, that does not apply in a linear amp. All the above manufacturers make certain models of capacitors that would fail in a switching circuit, it depends on which specific model of cap (same with Koshin, except as mentioned it won't matter much with an amp unless it has a switching PSU).

So my answer is, generally Koshin would be between Rubycon/Nichicon/HCC and CapXon/Teapo, but in this use it won't make a difference, G-Luxon would do fine in an amp so long as it wasn't one of the past models with instable electrolyte which causes them to fail just sitting unused in a drawer. That is assuming the amp isn't in an extremely high ambient temp environment, that it creates heat itself isn't otherwise much of a factor.

[acstech]

Re: Koshin caps... do they suck?

Koshin is a Hong Kong / China company, so yes, they induce a negative pressure differential.

http://www.koshin.com.hk/index02.htm

I won't speculate as to their workability in your equipment.

[weirdlookinguy]

Re: Koshin caps... do they suck?

Eek. Well I'm just gonna have to trust Harman on this one, that amp is too damn nice to pass up!

[EGuevarae]

Re: Koshin caps... do they suck?

negative pressure differential..

OMG... what a nice line to say they just plain suck!
I'll just note it to use when the time is right ......

[weirdlookinguy]

Re: Koshin caps... do they suck?

Wait, so 999999999, how do you know that Koshin caps are decent?

[999999999]

Re: Koshin caps... do they suck?

Because they are suitable for the circuits they are used in.

We seem to keep having some strange concept of capacitors, that if a particular brand and model isn't suitable for one use, that it means "suck" for a totally different use.

There is only one relevant parameter, does a capacitor live up to it's rated specs. In the past, many did not. Now, it's a matter of consulting datasheets and comparing to the circuit in question.

Why is this so? Because the vast majority of circuits in products consumers enjoy, don't need the ultimate best cap possible, they just need one that won't fail within the viable lifespan of the product. The caps we like for the reasons of lowering failure rate, tend to be in switching voltage regulation circuits, where they self-heat. If it's not that kind of circuit, the standards change till we need know more about general stability, will it self destruct because it is not chemically stable regardless of any particular use.

[weirdlookinguy]

Re: Koshin caps... do they suck?

Then why do Samxon GF caps, rated at 105C, regularly fail in LCD monitors? I also had an old Trinitron monitor that had a shitty picture, and inside I found Jamicon caps with only 10-20uf left when they were rated 47-470uf. Those caps were not in a switching PSU, they were on the chassis and neck board of a CRT monitor, and they were causing a shitty picture because they had dried out after 8 years of service. Replacing them made the monitor look much better. We're not looking at how things are supposed to perform on paper here, we're looking at how they perform in real life.

I want this amp to last me 15~20 years like my dad's old Technics (Matsushita) amp has, not whatever the manufacturers believe is a "viable lifespan".

[kikkoman]

Re: Koshin caps... do they suck?

afaik they have 2yrs warranty on home audio products. not much but i guess they last longer than that.
some dealers offer extended service plans but you won't find any consumer electronic product/brand with more than 5yrs manufacturer warranty.

anyway, here's my point: if you want something to last beyond its warranty (let's say 15 years) you have to take good care of it. which in my opinion includes -if necessary- replacing caps.
if it's a good design you only have to dust it off once in a while. but not checking would be stupid.
it's the same thing as with cars. you can pick any analogy here (gaskets, spark plugs, oil filter etc bla bla).

if it's a cheap product i often void the warranty deliberately to check the weak spots (which every product has).

about those TK caps: i haven't seen many of them (the last time was probably before i was aware of bad caps) so no idea.
but i've seen amps that have been working flawlessly for 30 years with caps that wouldn't last a second in the environments we're usually talking about here (SMPS, VRM etc).
those were all analog amps (class A/B) of course.

[Wizard]

Re: Koshin caps... do they suck?

I'm guilty of breaking warranties... For good reasons too.

Cheers, Wizard

[EGuevarae]

Re: Koshin caps... do they suck?

afaik they have 2yrs warranty on home audio products. not much but i guess they last longer than that.
some dealers offer extended service plans but you won't find any consumer electronic product/brand with more than 5yrs manufacturer warranty.

anyway, here's my point: if you want something to last beyond its warranty (let's say 15 years) you have to take good care of it. which in my opinion includes -if necessary- replacing caps.
if it's a good design you only have to dust it off once in a while. but not checking would be stupid.
it's the same thing as with cars. you can pick any analogy here (gaskets, spark plugs, oil filter etc bla bla).

if it's a cheap product i often void the warranty deliberately to check the weak spots (which every product has).

about those TK caps: i haven't seen many of them (the last time was probably before i was aware of bad caps) so no idea.
but i've seen amps that have been working flawlessly for 30 years with caps that wouldn't last a second in the environments we're usually talking about here (SMPS, VRM etc).
those were all analog amps (class A/B) of course.

I think you got a point here : the applications we use or see caps are demanding ones, either because bad heat dissipation, bad design, crap components (crap being either really crap, not suited for the place & conditions they are put to work, or because bad quality or bad batch even for good companies like Nichicon or others), or abuse. There are bad caps because they are bad, and there are bad caps for a given application. I think we all refer to bad caps mostly because they are not suited for the places we have to get them swapped out (some are crappy even in other applications, but we do not care much about some of that other applications for now). That is what I got from the last piece of post.

[999999999]

Re: Koshin caps... do they suck?

I wrote switching voltage regulation circuit, not switching PSU. The difference being, you can have a high frequency, high current circuit in the equipment even if it's not part of the primary AC-DC drop-down main PSU.

In a quality AMP (besides class D) you don't generally have this, you have a transformer alone stepping down the voltage.

As for how things perform in real life vs paper, sure you found some that failed, opposed to a majority of consumer TVs or LCD monitors still working. If you want a guarantee nothing will ever fail, simply picking a different product because of it's capacitors will have variable effectiveness depending on where those caps are in which subcircuit. To put it another way, did you pick your refrigerator based on which capacitors it has too, or your microwave, your air conditioner, etc? I've replaced the capacitor in my A/C twice in the past 3 years, but in that case it runs really hot in the summer. Really hot.

On the other hand, if you insist that certain capacitors will give you longest life, someday when the ones in whatever equipment you buy fail, then swap that particular make and model of caps in. Far more than caps, the features and audio quality should be primary considerations in an amp. Plenty of amps run over 15 years without having our favorite brands of caps inside.

[EGuevarae]

Re: Koshin caps... do they suck?

I wrote switching voltage regulation circuit, not switching PSU. The difference being, you can have a high frequency, high current circuit in the equipment even if it's not part of the primary AC-DC drop-down main PSU.

In a quality AMP (besides class D) you don't generally have this, you have a transformer alone stepping down the voltage.

As for how things perform in real life vs paper, sure you found some that failed, opposed to a majority of consumer TVs or LCD monitors still working. If you want a guarantee nothing will ever fail, simply picking a different product because of it's capacitors will have variable effectiveness depending on where those caps are in which subcircuit. To put it another way, did you pick your refrigerator based on which capacitors it has too, or your microwave, your air conditioner, etc? I've replaced the capacitor in my A/C twice in the past 3 years, but in that case it runs really hot in the summer. Really hot.

On the other hand, if you insist that certain capacitors will give you longest life, someday when the ones in whatever equipment you buy fail, then swap that particular make and model of caps in. Far more than caps, the features and audio quality should be primary considerations in an amp. Plenty of amps run over 15 years without having our favorite brands of caps inside.

Again, I second that line regarding the real life vs the "paper life". Even things that have millions of dollars in custom near bullet proof components tend to fail : Apollo 13, Challenger, or even the plane f?cked up by ducks from the miracle Hudson's Aqua-Landing.
While you can avoid brands (being amps, mobos, or even cars) because you know that components with a high failure rate are present, you can't avoid complete failure even if you chose good quality components. Sure, you can narrow the chances, but there's one factor that even in paper can not be predicted : human natural way to do things not supposed to be done. Like running your LCD in a poorly ventilated area, or a dusty one without proper countermeasures, or a regular cleaning, or having a bad line condition (here in the States the AC line is smooth, but I invite you to go to Mexicali, in Mexico, on the summer, and you'll see what I mean ....) and not using any form of either surge suppression or backup power. I have seen exceptionally good computers being trashed in months because of abuse or misuse, and I guess that with any electronics, it would be the same : use them beyond their specs and even Rubycons will go to hell. I guess that in this particular case the best choice would be research for failure rates, buy a known good amp or get your father's one .....

[NxB]

Re: Koshin caps... do they suck?

Anything can fail. Buying better stuff only increases the chances of it working longer. Sometimes cheaper stuff even outlasts good ones with better components. Also, old electronics were made to last through nuclear wars while new stuff is meant to break as soon as possible so you have to buy a new one.

[EGuevarae]

Re: Koshin caps... do they suck?

Anything can fail. Buying better stuff only increases the chances of it working longer. Sometimes cheaper stuff even outlasts good ones with better components. Also, old electronics were made to last through nuclear wars while new stuff is meant to break as soon as possible so you have to buy a new one.

Second that. WE the consumers have much of the guilt here .... "I want to pay less" ..... quality is directly proportional to money paid, or in other words, you get what you pay for. All items nowadays are made to last less so when someone makes an "improved" version, there's market for it.

[PlainBill]

Re: Koshin caps... do they suck?

Second that. WE the consumers have much of the guilt here .... "I want to pay less" ..... quality is directly proportional to money paid, or in other words, you get what you pay for. All items nowadays are made to last less so when someone makes an "improved" version, there's market for it.

That is one view of the matter. Here is another perspective. I spend about 10 years working for companies that made commercial electronic equipment. In each case, the products were designed in-house, built in-house, and repaired by company technicians. Failing assemblies were examined and the cause of the failure established. Records were kept of the failures. If enough components of a particular brand failed, they would be removed from our list of acceptable suppliers. If the failure was due to an assembly or testing error, steps would be taken to prevent repeat failures. And if the failure was due to a poor design, the design was changed. On more than one occasion assemblies were pulled out of stock and reworked to incorporate a design change.

The result was the products of both companies had a very good history of reliability. One company gave a 5 year warranty on their products, the second only a two year warranty. The second had a 0.1% failure rate under warranty.

The business has changed. Take a look at the Mitsubishi TV I mentioned in another thread. It was designed in Japan, based on a 'light engine' designed by an American company. The components are assembled in Mexico, and the TV serviced by 'factory trained' technicians. (I suspect the factory training consists of a series of videos.) When they became aware of a couple of problems, they extended the warranties on those items.

Now take a look at someone who buys a Gateway or any other brand of monitor. Gateway contracts with a Chinese company to provide several cargo containers of monitors with the Gateway logo. The monitor is built in China from sub-assemblies manufactured by other Chinese companies from components manufactured in China. The only design done in the US is probably the plastic shell and the placement of the logo. By the time the last container leaves China the manufacturer has a new design based on other sub-assemblies.

The warranty is for one year, tops. If it fails under warranty Gateway will ship out a new one and scrap the old. If it fails AFTER the warranty, Gateway won't even repair it!! (I'm not singling out Gateway, they are just an example of common industry practice. HP, Dell, etc do the same thing).

The result is that there is little if any feedback on design or assembly flaws. If it lasts for a year, everybody (except the consumer) is thrilled. And to make matters worse, the CEO of Gateway is primarily interested in the earning report for the next quarter. He has a mild interest in the earnings for next year, and doesn't give a damn what they will be in 5 years. By that time Gateway will have a new CEO.

PlainBill

[EGuevarae]

Re: Koshin caps... do they suck?

That is one view of the matter. Here is another perspective. I spend about 10 years working for companies that made commercial electronic equipment. In each case, the products were designed in-house, built in-house, and repaired by company technicians. Failing assemblies were examined and the cause of the failure established. Records were kept of the failures. If enough components of a particular brand failed, they would be removed from our list of acceptable suppliers. If the failure was due to an assembly or testing error, steps would be taken to prevent repeat failures. And if the failure was due to a poor design, the design was changed. On more than one occasion assemblies were pulled out of stock and reworked to incorporate a design change.

The result was the products of both companies had a very good history of reliability. One company gave a 5 year warranty on their products, the second only a two year warranty. The second had a 0.1% failure rate under warranty.

The business has changed. Take a look at the Mitsubishi TV I mentioned in another thread. It was designed in Japan, based on a 'light engine' designed by an American company. The components are assembled in Mexico, and the TV serviced by 'factory trained' technicians. (I suspect the factory training consists of a series of videos.) When they became aware of a couple of problems, they extended the warranties on those items.

Now take a look at someone who buys a Gateway or any other brand of monitor. Gateway contracts with a Chinese company to provide several cargo containers of monitors with the Gateway logo. The monitor is built in China from sub-assemblies manufactured by other Chinese companies from components manufactured in China. The only design done in the US is probably the plastic shell and the placement of the logo. By the time the last container leaves China the manufacturer has a new design based on other sub-assemblies.

The warranty is for one year, tops. If it fails under warranty Gateway will ship out a new one and scrap the old. If it fails AFTER the warranty, Gateway won't even repair it!! (I'm not singling out Gateway, they are just an example of common industry practice. HP, Dell, etc do the same thing).

The result is that there is little if any feedback on design or assembly flaws. If it lasts for a year, everybody (except the consumer) is thrilled. And to make matters worse, the CEO of Gateway is primarily interested in the earning report for the next quarter. He has a mild interest in the earnings for next year, and doesn't give a damn what they will be in 5 years. By that time Gateway will have a new CEO.

PlainBill

That is another good (excellent, in fact) view. Nowadays almost no one does his own things: something not new, but with the endless chain of providers, in the end HP now ends up with a BenQ product that only God knows who assembled. On top of that, corruption on development countries (I know Mitsubishi horror stories from a plant in Mexico where I used to live. They closed the plant and I wonder why ....) makes the installation of cheaper, not specified components get into good designs.
And then we have the Chinese industry of faking/copying/counterfeiting....