PCs plagued by bad capacitors

Moderator: News Team

Post Reply
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

PCs plagued by bad capacitors

Post by purrkur »

Problems with bad caps on motherboards seems to be a neverending story
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
hyperspace
Board Admin
Posts: 1395
Joined: Wed Jul 24, 2002 7:39 am
Location: Lincoln, NE USA
Contact:

Post by hyperspace »

Man, I thought we wouldn't see something like this so soon. :cry:

I haven't been to find a good local electronic repair service, so I'm going to send my two BP6s to Bad Caps.
Quantum WormHole

Image
lost in hypertime...
KliK
Posts: 350
Joined: Fri Mar 11, 2005 3:56 pm
Location: Zagreb, Croatia
Contact:

Post by KliK »

there is an old englishman saying that says: I'M NOT SO POOR, TO BUY CHEAP!!!

or, now you know why some MBO cost more than the cheap ones...but they also work!
and it's not then MBOs, it goes for everything...for example my 500$ sky jacket...it look almost new, never had any problems with it, and in fact i don't know if i would ever wore that jacket out!!! maybe if i was in alaska i would, but not in mediteranian!!! :D
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

KliK wrote:or, now you know why some MBO cost more than the cheap ones...but they also work!
I would argue that Intel designed and produced motherboards are on the expensive side of the motherboard price range and Apple charges a premium for their pretty average machines. Neither has been excluded from this problem.

I don't necessarily think that this has to do with cheap. I have heard two reasons for this problem. The first is that the electrolyte in the cap is of poor quality and the second is that there is too much fluid electrolyte in the caps. I know that even the more respected cap manufacturers have had issues.

In other words I think the problem is more complicated than being at risk when you have bought a cheap motherboard.
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
KliK
Posts: 350
Joined: Fri Mar 11, 2005 3:56 pm
Location: Zagreb, Croatia
Contact:

Post by KliK »

purrkur wrote:
KliK wrote:or, now you know why some MBO cost more than the cheap ones...but they also work!
I would argue that Intel designed and produced motherboards are on the expensive side of the motherboard price range and Apple charges a premium for their pretty average machines. Neither has been excluded from this problem.

I don't necessarily think that this has to do with cheap. I have heard two reasons for this problem. The first is that the electrolyte in the cap is of poor quality and the second is that there is too much fluid electrolyte in the caps. I know that even the more respected cap manufacturers have had issues.

In other words I think the problem is more complicated than being at risk when you have bought a cheap motherboard.

of course not, the problem is the same...but if you buy a cheap MBO, than it has small and low quality caps on it...if you buy an expensive ones, they have bigger caps, which don't have to be better quality, but they are OVERDESIGNED...and that means they can be stresses over the limit, and they won't bulge or sthg else...

so if you can't make sthg with more quality, you just put a bigger one in, which can take a load more easily...and that is the most common thing...(as a mechanical engineer i can understand it)
:D
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

KliK wrote:but they are OVERDESIGNED...and that means they can be stresses over the limit, and they won't bulge or sthg else...
Hmmm. Caps have more than one function in electronics. In power supplies and voltage regulators the larger caps usually serve as voltage stabilizers. A cap is often very much dependent on other components around it or other components are dependent on it - to fill a certain function. A cap will "charge" like a battery and then release their charge when needed. Other components (and the specifications of the cap itself) will, for example, rule how fast or slow the cap charges or discharges. This timing factor is usually of importance and is a part of the design of the cirquit that the cap is a part of. So I don't really agree with your "overdesigned" principle. Putting a larger cap in the place of a smaller won't necessarily mean that things will work better. In fact, you may screw things up royally if you are not careful.
KliK wrote:so if you can't make sthg with more quality, you just put a bigger one in, which can take a load more easily...
Like I said, it is not a matter of being able to take the load more easily. There are more factors involved than just load, and I really fail to see the load factor you are talking about. What specifics of the cap are you talking about? Are you saying that a 12 volt cap that is specified at 1000microF is better at taking "the load" more easily than a 12 volt cap rated at 100microF? Because if that is your point then I fail to see your point. Two such components have a number of parameters that are different that might screw up things like I mentioned before.

And like I also mentioned, if one of the problems for bulging/leaking caps is too much electrolyte liquid in them then they are bound to break no matter how you design your circuit. Such a problem is something I would consider to be a manufacturer production problem.
KliK wrote:and that is the most common thing...(as a mechanical engineer i can understand it)
What do I know, I am just a lowly electronics technician :)
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
purrkur
Linux Guru
Posts: 687
Joined: Fri Dec 12, 2003 5:57 pm
Location: Sweden
Contact:

Post by purrkur »

Just found this article that shows bad caps on a graphics adapter.
2x533MHz@544MHz, 2.0V
640MB PC100 memory
Realtek RTL-8139 NIC
Maxtor 6Y080L0 80GB hdd
Debian Linux stable with 2.4.8 kernel
KliK
Posts: 350
Joined: Fri Mar 11, 2005 3:56 pm
Location: Zagreb, Croatia
Contact:

Post by KliK »

purrkur wrote:
KliK wrote:but they are OVERDESIGNED...and that means they can be stresses over the limit, and they won't bulge or sthg else...
Hmmm. Caps have more than one function in electronics. In power supplies and voltage regulators the larger caps usually serve as voltage stabilizers. A cap is often very much dependent on other components around it or other components are dependent on it - to fill a certain function. A cap will "charge" like a battery and then release their charge when needed. Other components (and the specifications of the cap itself) will, for example, rule how fast or slow the cap charges or discharges. This timing factor is usually of importance and is a part of the design of the cirquit that the cap is a part of. So I don't really agree with your "overdesigned" principle. Putting a larger cap in the place of a smaller won't necessarily mean that things will work better. In fact, you may screw things up royally if you are not careful.

i agree with you about that!
but that is why you can overdesign some system to have 20% or 33% (at best) bigger caps, and it will work...in some instances even better than the original...
best example is the BP6! when i have been surfing the net, i found guy selling bigger capacitors for it...even in EC10 problem, a bigger capasitor might work! the problem is, not to put 10miliF, but you put 1500microF...and the stuff works... (got to send you more links about that!)
purrkur wrote:
KliK wrote:so if you can't make sthg with more quality, you just put a bigger one in, which can take a load more easily...
Like I said, it is not a matter of being able to take the load more easily. There are more factors involved than just load, and I really fail to see the load factor you are talking about. What specifics of the cap are you talking about? Are you saying that a 12 volt cap that is specified at 1000microF is better at taking "the load" more easily than a 12 volt cap rated at 100microF? Because if that is your point then I fail to see your point. Two such components have a number of parameters that are different that might screw up things like I mentioned before.

And like I also mentioned, if one of the problems for bulging/leaking caps is too much electrolyte liquid in them then they are bound to break no matter how you design your circuit. Such a problem is something I would consider to be a manufacturer production problem.

you miss the bigg picture...the OVERDESIGN is not to put the to big capacitor in place, but to put the safty factor inside...so in example of 100microF, you won't put the 1000miliF...
but, maybe the circut will work better with 200microF or 500microF; instead of 100microF mentioned...but that depends on the type of the circut, etc.

and about the leaking...all capacitors have more or less electrolyte liqiud inside...but when they are pushed to run OVER THE LIMIT, they will bulge...so if you put a SLIGHTLY BIGGER ONE in that place, it will be more durable, 'cause of the stress of the curcut will have on it!
for example: if on a circut 100microF cap bulges in exactly 1 year, a 200microF cap will bulge (or won't) after more than 1 year (that means 1 year and few more months!)
:D
purrkur wrote:
KliK wrote:and that is the most common thing...(as a mechanical engineer i can understand it)
What do I know, I am just a lowly electronics technician :)

good, so is my theory right?! you know better...you are the electronics guy, so can you teach me a little?!
hyperspace
Board Admin
Posts: 1395
Joined: Wed Jul 24, 2002 7:39 am
Location: Lincoln, NE USA
Contact:

Post by hyperspace »

The primary problem with the "Bad" capacitors was the electrolyte formulae. An Asian company (Chinese, I think) copied or "stole" an electrolyte formulae and sold it to some of the major capacitor companies.

More details...

Overclocking or "stressing" does not cause the problem experienced by so many. We have had no less than 3 different computers at work with bad capacitors. No overclocking.
Quantum WormHole

Image
lost in hypertime...
KliK
Posts: 350
Joined: Fri Mar 11, 2005 3:56 pm
Location: Zagreb, Croatia
Contact:

Post by KliK »

on the same page...

This is what your kit will contain to successfully rebuild your Abit BP6. The kit contains 25 capacitors, itemized as follows:
Quantity Value Voltage Diameter
12 1500uF 10v 8mm
12 1000uF 10v 8mm
1 1200uF 6.3v 8mm

The 1200uF caps is for the EC10 mod. Some swear by 1500uF's, and others swear by 470uF... I've tried them all, and have found that 1200uF or 1000uF is the most adequate. If you wish to have a different value for EC10, just let me know what value you want and I'll modify the kit accordingly.
---------------------------
This is what your kit will contain to successfully rebuild your Abit VP6. The kit contains 23 capacitors, itemized as follows:
Quantity Value Voltage Diameter
15 1500uF 6.3v or 10v 10mm
4 1000uF 16v 10mm
4 1000uF 10v 10mm


so in this case, they don't up the caps value (microF), but they up the voltage they can withstand!
hyperspace
Board Admin
Posts: 1395
Joined: Wed Jul 24, 2002 7:39 am
Location: Lincoln, NE USA
Contact:

Post by hyperspace »

KliK wrote:... so in this case, they don't up the caps value (microF), but they up the voltage they can withstand! ...
Some companies may not make a 6.3v "version". So, 10v is closest voltage to 6.3v. I don't think it is for stress, instability or overclocking.
Quantum WormHole

Image
lost in hypertime...
davd_bob
Confused
Posts: 1043
Joined: Fri Feb 13, 2004 2:30 am
Location: Houston, TX

Post by davd_bob »

Like our illustrious leader(Derek) stated, its the recipe they used for the goo in the caps that is causing this round of failures...just like when the BP6 and VP6 were being manufactured.

I think no matter how much you "overdesign" is irrelevant if the componant is bad. It seems like the stuff kinda 'spoils after time nomatter how its used but possably heat may cook it into failure at a faster rate.

BTW, for anyone out there that doesn't already know, officially NO manufacturer will support overclocking. If you want your CPU to be faster they prefer that you just buy a newer faster CPU.

(edited 11/16)My bad, The credit for showing the bad recipe should have been given to Hyperspace instead of Derek. They are both leaders in the field.
There are *almost* no bad BP6s. There are mostly bad caps.

No BP6s remaining
Athlon 2800
Sempron 2000
ViaCPU laptop with Vista.(Works great after bumping ram to 2Gig)
P-III 850@100
KliK
Posts: 350
Joined: Fri Mar 11, 2005 3:56 pm
Location: Zagreb, Croatia
Contact:

Post by KliK »

so you are saying that those are just like italian and frrench cars: NO MATHER HOW BIG IS IT, IT ALLWAYS FULL OF SHIT!!! :D
davd_bob
Confused
Posts: 1043
Joined: Fri Feb 13, 2004 2:30 am
Location: Houston, TX

Post by davd_bob »

KliK wrote:so you are saying that those are just like italian and frrench cars: NO MATHER HOW BIG IS IT, IT ALLWAYS FULL OF SHIT!!! :D
yup, that just about covers it...if your talking about bad caps and not leaders.

david
Derek
Site Admin
Posts: 2489
Joined: Tue Jul 23, 2002 3:55 pm
Location: Canada
Contact:

Post by Derek »

davd_bob wrote:Like our illustrious leader(Derek) stated, its the recipe they used for the goo in the caps that is causing this round of failures...just like when the BP6 and VP6 were being manufactured.
I can't take credit for that!
-Derek
Post Reply