Big PCB, small case, a (too long?) screw and...

[Behemot]

Excuse me? They have "engineers", those should be ppl who can think in technicals terms. You telling me those ppl do not realise that a screw is actually being put into a screw worm (why it would be there in the first place??)? And that screws usually have some lenght etc.? Maybe they should establish a new bench where they put screws into chasis if they can not recognise it when they design it. I would at least expect decent manufacturer to watch their stuff and notify the customer his design may be flawed, instead of hapilly making crap…jeez.

I mean like, if I order a ton of sausages from a fab, I also have to remind them not to use road salt instead of normal table salt? So incompetence has been established to a standard today? I have often been thinking as much, but it's good I finally got a proof…

[Stefan Payne]

They have "engineers", those should be ppl who can think in technicals terms.

Yes, and?!
They might know everything about PSUs but if you want the PCB 2mm behind the screw hole, they do it, without asking because that's not their job...
Their job is to do the PSU stuff, YOUR job is to define everything and tell them how you want it!

If you want pink sleeve and pink connectors, they don't ask, they try to do what you tell them to!

You telling me those ppl do not realise that a screw is actually being put into a screw worm (why it would be there in the first place??)? And that screws usually have some lenght etc.?

1. Chinese Mentality vs. European/Western
2. Is it their job to think for you or is it your job to tell them what to do?

[Internet]

It's often not CWT's job to "remember". They work in a vacuum of PSUs sitting on benches hooked up to load testers. Not PSUs actually mounted in chassis running actual PCs. Therefore, they have to be "reminded" by the other company's engineers. If the other company doesn't have engineers, shit happens.

Just my 20 cents...

Although I agree this is ultimately the responsability of whoever is ordering the PSUs, we know most of the companies ordering those PSUs are doing absolutely no engineering, just getting a ready product and relabelling.

AFAIK, CWT has a bunch of ready-to-order platforms, from the el-cheapo to the high-end. So, it can be presumed the relabeller doesn't need to know about engineering or the size of PCB, just to know what is crap and what is good. Even a PC integrator, with no PSU engineers, could trust them to use their PSUs. And the idea of this same issue happening on branded-PC PSU isn't exactly pleasant...

Of course, CWT can also customize a platform or even create one from scratch, as long you're willing to pay. Cooler Master GM seems to be one of those, they got an already existing platform (GPK/GPN or some variant of it) but customized it with a larger PCB. I think the reason for the larger PCB was for providing space for a second cap on the primary so they could increase the hold-up time on the higher wattage models. Again, the fault is ultimately on CM (since they must supervise what the OEM is doing), but if the decision or sugestion of increasing the PCB was from CWT and CM trusted them on this, I can say CWT is being very negligent when you consider they're a company capable of designing PSU platforms and had a similar issue in the past.


BTW, a suggestion for PSU reviewers: since this has potential to create big troubles, I suggest do the screw test on PSUs where the PCB seems too big for the case.

[Jon Gerow]

If the decision was made to grow the PCB to accommodate the second primary cap (probably) then it's true that someone should then check to make sure there's no way a screw is going to dig into that PCB. But my experience with CWT is, and this is not exclusive to CWT, that nobody in that building is going to take the PSU and then mount it in a chassis for fitment.

I don't know if it's Chinese culture, the education system or what.. but this is not uncommon for Chinese factories. There's a lot of "tunnel vision" on how things are done. If everything is not clearly defined and tested, you WILL get screwed one way or another.

If you want to go to market with something "new" (even something as simple as increasing hold up time by adding a second capacitor), you better have the resources to follow that through.

I will say, it does seem that CM did have the resources to fix the hilarious photo collage they once had: http://assets.coolermaster.com/produ..._rgb_07_bg.png (For those who missed it, they changed the collage after everyone made fun of some of the content. The only copy I have of the original is barely large enough to see. It's attached.)

And it is good to see ‎Macaulay Culkin get a gig.

Now that's enough from me... I don't want to hate on CM. Just telling you how it's been from my experience.

[Behemot]

There is no need to mount that to a chasis. Not at all. Just put the bloody screw in and check the distance…but yeah, this is how it works in corporations not only in Asia lately. It's where the whole bloody System leads: They (whoever that is) will guide everything and everybody, the average Joe is expected to be mindless zombie who does only what he is told to do. Using brain is forbidden.

If this is common, it explains all the problems of SAMA platforms and somewhat proves my suspicion - they just started at reference circuit and put more and more modifications to it, it's like snowball. Find one problem, patch it. Found other one, patch it. Who knows it if actually does not even introduce new problems elsewhere. With this rate maybe few yearsf rom now the product will be good, but competition will already be elswhere…

Anyway, this with the CM GM is not only a faux-pass, this is serious problem in insulation safety, if authorities find out, CM can expect some big shit to fall on their heads.

[Internet]

If the decision was made to grow the PCB to accommodate the second primary cap (probably) then it's true that someone should then check to make sure there's no way a screw is going to dig into that PCB. But my experience with CWT is, and this is not exclusive to CWT, that nobody in that building is going to take the PSU and then mount it in a chassis for fitment.

I don't know if it's Chinese culture, the education system or what.. but this is not uncommon for Chinese factories. There's a lot of "tunnel vision" on how things are done. If everything is not clearly defined and tested, you WILL get screwed one way or another.

If you want to go to market with something "new" (even something as simple as increasing hold up time by adding a second capacitor), you better have the resources to follow that through.

I will say, it does seem that CM did have the resources to fix the hilarious photo collage they once had: http://assets.coolermaster.com/produ..._rgb_07_bg.png

And it is good to see ‎Macaulay Culkin get a gig.

Now that's enough from me... I don't want to hate on CM. Just telling you how it's been from my experience.

I understand what you're saying. But the fact (unfortunately) is most relabellers have no engineers nor at least some person who understand at least the basics of PSUs. So, my biggest worrying here is not just with CM, but if the same thing happens on a product of a company with the same level of resources of CM in terms of PSUs, or other companies with less resources which don't customize anything and just stick to the default project (or also just someone who gets PSUs directly from OEMs to use in its PCs). Because in the end of the story, the one who will be screwed will be the end consumer.

Anyway, this with the CM GM is not only a faux-pass, this is serious problem in insulation safety, if authorities find out, CM can expect some big shit to fall on their heads.

Agreed. I'm asking now if this was just an isolated case... Because I don't remember to have seen anything on other GMs...

Also I'm trying to find other potential candidates to the same issue.... Seems this Rosewill Hive could be one: http://www.jonnyguru.com/modules.php...tory5&reid=408

The case seems to fit exactly for the PCB and there are no corners (which prevents this issue).

Another one would be this Thermaltake Toughpower DPS: http://www.tweaktown.com/reviews/579...ew/index4.html

http://www.jonnyguru.com/modules.php...tory6&reid=362

[Behemot]

I think the lack of person who understands the subject is the main reason why so many units fail under standing. My testing anyway

This may indeed be isolated, maybe they put too long screws in and he has paper-thin case

But may not. You see, when a marketing guy tells me he does not understand it and that's why I got it for testing to my question why did they not put almost any information at all about the unit, incl. cable lenghts, on the box or their website…what can you expect technical wise from such companies/ppl when a marketing guy tells me he does not understand marketing? Than indeed, why a manufacturer should understand what his product should meet, right?