Intel's Motherboards Bad Capacitors Resurface!

[Gnu_Raiz]

Looks like we will be in for another round of bad capacitors, if you
have a Pentium 4 motherboard.

http://www.digitimes.com/systems/a20041220PR204.html

<taken from link>

Mobo makers commented that equipping mobos with Taiwan-made capacitors
may lead to potential RMA issues after two years of use.
-----

Since we know that most of Intels own motherboards are made by Hon Hai,
as well as Dells, now might be a good time to reconsider AMD, Nvidia.
If you self build your systems I hope you have a three year warranty on
your MBs.

Also if your a business it might be a good time to get that extended
warranty from Dell, Gateway others if you purchased any Pentium 4 MBs,
in the last year or so.

Gnu_Raiz

 

[Tony Hill]

Looks like we will be in for another round of bad capacitors, if you
have a Pentium 4 motherboard.

http://www.digitimes.com/systems/a20041220PR204.html

<taken from link>

Mobo makers commented that equipping mobos with Taiwan-made capacitors
may lead to potential RMA issues after two years of use.

Uhh.. and AMD and nVidia based boards are not built with Taiwan-made
capacitors?!? Ohh.. I see, it's because Intel chips have somewhat
higher power consumption and the article is claiming that this will
cause more failures. Could be, though I wouldn't bet the farm on it.

In either case, my AMD/nVidia based board died due to leaky
capacitors, so this is hardly an Intel-only issue, never was.

If you self build your systems I hope you have a three year warranty on
your MBs.

I could do that, however instead I bought a board with a 1 year
warranty that only cost me ~$40 US. At that price I can afford to
replace it if it blows.

Also if your a business it might be a good time to get that extended
warranty from Dell, Gateway others if you purchased any Pentium 4 MBs,
in the last year or so.

Same goes if you bought any AMD systems because honestly, there isn't
any significant difference. The processor used means dick-all when it
comes to this sort of design flaw. If your capacitors (or voltage
regulators, or some other widgets) are not up to par, your board will
die, end of story. Don't matter one bit whether it's a P4 chip
sitting on that board or an Athlon64 chip or a Dorito.

 

[Gnu_Raiz]

I agree with you, most mother boards are made in Taiwan, with Taiwan
capacitors. If I had poor case ventilation then I might be worried, as
far as I know the only manufacture that mentioned they have started to
use capacitors made in Japan is Albatron.

One thing you cannot ignore is that the Pentium 4 puts out a lot of
watts, combine this with SLI you got a nice little thermal engine. If
the case is not vented properly and the mother board is using these
cheapo parts, then failure rates will raise.

Sure the Athlons put out some watts, but not as much as a Pentium 4,
since SLI is PCIe that leaves out a complete market segment. I have
not seen a PCIe motherboard built for socket A some 754s, 940s, 939s
sure but no A's.

I think this is more of a heat issue, poor chip design, even the BTX
form factor might not be enough to solve those problems. Even if they
use the high quality parts made in Japan do you really think that will
solve the problem? You take part A made in Taiwan, part B made in
Japan, use the same mother board I bet you still have major failures.
After all if a part is rated at x, swap for part rated at x made of
good quality, will it solve your problem if the rated x is too low to
begin with?

Most people know who read this group that heat kills, I think this was
more of a Hommer moment for the Taiwan engineers, "Doh" our
motherboards will not make it past our warranty periods.

Gnu_Raiz

 

[Yousuf Khan]

I agree with you, most mother boards are made in Taiwan, with Taiwan
capacitors. If I had poor case ventilation then I might be worried, as
far as I know the only manufacture that mentioned they have started to
use capacitors made in Japan is Albatron.

Is this Taiwan capacitor issue still a problem, or is it now an urban
myth? I know I suffered a couple of capacitor blowouts about 4 years
ago, but the last board I got about 4 years ago is now still going.

Yousuf Khan

 

[Gnu_Raiz]

According to this website its still going strong: http://badcaps.net/.

Now with the engineers admiting that in a few years they might have
major RMA's it looks like round two is about to begin.

Gnu_Raiz

 

[Tony Hill]

Is this Taiwan capacitor issue still a problem, or is it now an urban
myth? I know I suffered a couple of capacitor blowouts about 4 years
ago, but the last board I got about 4 years ago is now still going.

It has been an on-going problem, though at least now manufacturers are
aware of the potential for problems and the costs involved. I know of
some boards built within the past 12-18 months that have definitely
suffered very high failure rates due to cheap capacitors, and it's
likely that certain models of more recent boards will start showing
the exact same issue when they start to go.

Note though that I don't think it's really an issue with Taiwanese
capacitors vs. Japanese capacitors, just an issue of quality. If you
use cheap parts you can cut your manufacturing costs and continue to
compete with the other guys. If you spend more money you get better
quality but your costs go up. Engineering a commodity part (like a
motherboard) has always been the classic balancing act between the
two.


FWIW I just checked my own dead motherboard with leaky capacitors.
The ones that leaked were made by Lelon, one of the two main companies
fingered for producing the problematic components (the other being
Luxor). This board also has some perfectly fine looking capacitors
made by Rubycon, who I guess are the big name in Japanese capacitors,
and Teapo who apparently buy Japanese electrolytes instead of the
cheaper and problematic Chinese electrolyte. Of the three main types
of largish capacitors used on my system it's quite interesting to
compare them, the difference is very obvious. Where the Rubycon
capacitors (9 caps at 2200uF) and Teapo capacitors (5 of them at
1500uF) all look perfectly fine, all 18 of the Lelon caps (1000uF)
have visible signs of damage. For about half of them it's just that
the tops have bulged up (not a good sign, but these caps probably
haven't failed yet), while 3 have gunk leaking out the top and another
6 or 7 have brown goop leaking out the bottom. There's some pics of
the faulty caps here:

http://home.earthlink.net/~doniteli/index27.htm

 

[Guest]

It has been an on-going problem, though at least now manufacturers are
aware of the potential for problems and the costs involved. I know of
some boards built within the past 12-18 months that have definitely
suffered very high failure rates due to cheap capacitors, and it's
likely that certain models of more recent boards will start showing
the exact same issue when they start to go.

Note though that I don't think it's really an issue with Taiwanese
capacitors vs. Japanese capacitors, just an issue of quality. If you
use cheap parts you can cut your manufacturing costs and continue to
compete with the other guys. If you spend more money you get better
quality but your costs go up. Engineering a commodity part (like a
motherboard) has always been the classic balancing act between the
two.


FWIW I just checked my own dead motherboard with leaky capacitors.
The ones that leaked were made by Lelon, one of the two main companies
fingered for producing the problematic components (the other being
Luxor). This board also has some perfectly fine looking capacitors
made by Rubycon, who I guess are the big name in Japanese capacitors,
and Teapo who apparently buy Japanese electrolytes instead of the
cheaper and problematic Chinese electrolyte. Of the three main types
of largish capacitors used on my system it's quite interesting to
compare them, the difference is very obvious. Where the Rubycon
capacitors (9 caps at 2200uF) and Teapo capacitors (5 of them at
1500uF) all look perfectly fine, all 18 of the Lelon caps (1000uF)
have visible signs of damage. For about half of them it's just that
the tops have bulged up (not a good sign, but these caps probably
haven't failed yet), while 3 have gunk leaking out the top and another
6 or 7 have brown goop leaking out the bottom. There's some pics of
the faulty caps here:

http://home.earthlink.net/~doniteli/index27.htm

hi Tony,



I have a warranty expired MSI KT3V, MS-6712 ver:10A motherboard that is
damaged when I violently turn off the power from the wall socket because XP
failed to shut the computer down. (it stuck on the screen: "Windows Is
Shutting Down..." for about 5 minutes)



after which the computer refuses to start/boot anymore



I paid a heavy price? by turning off computer by the wall socket? the I-Cute
PSU is blown, I open it up, there is visible burn mark and it is definitely
refuses to work anymore



I replace a working PSU and the motherboard still refuses to boot/work, I
take out the MB from the casing screened it with bright torch and did not
find anything visibly wrong with it, what could have / what is the most
likely damage? Should I bother to send it in for repair?



Please advice.



Thanks.

 

[George Macdonald]

It has been an on-going problem, though at least now manufacturers are
aware of the potential for problems and the costs involved. I know of
some boards built within the past 12-18 months that have definitely
suffered very high failure rates due to cheap capacitors, and it's
likely that certain models of more recent boards will start showing
the exact same issue when they start to go.

Note though that I don't think it's really an issue with Taiwanese
capacitors vs. Japanese capacitors, just an issue of quality. If you
use cheap parts you can cut your manufacturing costs and continue to
compete with the other guys. If you spend more money you get better
quality but your costs go up. Engineering a commodity part (like a
motherboard) has always been the classic balancing act between the
two.

The cheap parts in the first round of this were made with a stolen patented
electrolyte formula though... except the purloined formula was
incomplete... reminiscent of the old tale about an early Japanese (stolen)
ship "design" which turned turtle on launch - what goes around..... I've
no idea what happened to those capacitor mfrs who were using the faulty
electrolyte... whether they are still using it?... a change of name seems
to be standard practice when you're business is ripping off others.

FWIW I just checked my own dead motherboard with leaky capacitors.
The ones that leaked were made by Lelon, one of the two main companies
fingered for producing the problematic components (the other being
Luxor). This board also has some perfectly fine looking capacitors
made by Rubycon, who I guess are the big name in Japanese capacitors,

Are you sure they're Rubycon? Take a look about mid-page here:
http://www.badcaps.net/forum/viewtopic.php?t=22 for a comparison of a real
Rubycon and the "Rulycon" rip-off - note even the vent stamp on the head is
copied.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

[Tony Hill]

I have a warranty expired MSI KT3V, MS-6712 ver:10A motherboard that is
damaged when I violently turn off the power from the wall socket because XP
failed to shut the computer down. (it stuck on the screen: "Windows Is
Shutting Down..." for about 5 minutes)

A little late now, but for future reference, hold the power button in
for 6 seconds. That will shut it down at least a tiny bit more
gracefully than just pulling the plug.

after which the computer refuses to start/boot anymore

I paid a heavy price? by turning off computer by the wall socket? the I-Cute
PSU is blown, I open it up, there is visible burn mark and it is definitely
refuses to work anymore

It does seem rather likely that the power supply blew as a result of
removing power from the system. Doing so can cause the voltage that
flows to the power supply to fluctuate greatly between 0V and 120V for
brief moments in time. In theory the capacitors should handle this
and discharge with no trouble, but sometimes it just doesn't work
right.

Of course, on the other hand it's quite possible that a component was
rather marginal to begin with, so this was just the straw that broke
the camels back, so to speak.

I replace a working PSU and the motherboard still refuses to boot/work, I
take out the MB from the casing screened it with bright torch and did not
find anything visibly wrong with it, what could have / what is the most
likely damage?

Trough to say for sure, beyond that it's likely something in the power
circuitry.

Should I bother to send it in for repair?

That's much easier to answer: No.

When my board went kaput, it was still IN warranty, and yet I decided
that it wasn't worth it for me to get it replaced because the shipping
costs for it were virtually the same as what it would cost me to buy a
replacement board (mind you, I live in Canada and would have had to
ship the board to California to get it replaced). Getting a board
repaired outside of warranty probably isn't even an option for this
board, let alone cost effective.

Replacement motherboards are cheap. I picked up a board on sale for
~$45 US and it has a newer, faster chipset, supports faster memory and
newer processors. As a result I was able to swap in both faster
memory and a faster processor when I needed another system, and I
swapped out my old processor and memory into that new (less
performance-critical) system.

FWIW the board I purchased to replace my dead MSI board was a
Chaintech 7NJL6 which www.newegg.com lists for $53. I've found it to
be a rather capable board.

 

[Tony Hill]

Are you sure they're Rubycon? Take a look about mid-page here:
http://www.badcaps.net/forum/viewtopic.php?t=22 for a comparison of a real
Rubycon and the "Rulycon" rip-off - note even the vent stamp on the head is
copied.

Yup, it's a real Rubycon, or if it's a counterfeit that it's a
sufficiently good one that I sure can't tell the difference. Either
way, those caps worked just fine, it's the Lelon ones that went all
leaky on me.

 

[cquirke (MVP Win9x)]

Ah: This is the perfect place to muse about bad caps

I'm not sure if heat is a required factor in capacitor breakdown.

Yes, I'm seeing that too - perhaps even more failures in the i845G
generation than the older i815 crowd.

The cheap parts in the first round of this were made with a stolen patented
electrolyte formula though... except the formula was incomplete...

I heard that as a rumour too, but from a reasonable source (IEEE I
think it was). At the time, I thought; we've been making electrolytic
caps for decades, surely the goo inside isn't still sekritt by now?

Then I read something abour a "low ESR" requirement, and that made
more sense. We've expected caps to cope with high frequencies and
current for a while, but the combination of these may be new. It
would be ironic if growth predictions based on Moore's Law faltered
not on the chips, but on dull bread-and-butter parts like caps.

What's interesting is that no-one has been able to maintain a list of
affected motherboard brands, for fear of litigation etc.

---------- ----- ---- --- -- - - - -

"He's such a character!"
' Yeah - CHAR(0) '

 

[George Macdonald]

Ah: This is the perfect place to muse about bad caps


I'm not sure if heat is a required factor in capacitor breakdown.


Yes, I'm seeing that too - perhaps even more failures in the i845G
generation than the older i815 crowd.


I heard that as a rumour too, but from a reasonable source (IEEE I
think it was). At the time, I thought; we've been making electrolytic
caps for decades, surely the goo inside isn't still sekritt by now?

Yes the original IEEE story was here:
http://www.spectrum.ieee.org/WEBONLY/resource/feb03/ncap.html and there's
more at http://www.niccomp.com/taiwanlowesr.htm where it's mentioned that
it was a double defection and theft - all info is coming from "well placed
sources". There are many everyday products which are basically a secret
brew - think of auto-engine lubricants where a single missing ppm additive
can be the difference between excellence and total failure - a "new"
formula there has also led to catastrophic consequences several times.

Then I read something abour a "low ESR" requirement, and that made
more sense. We've expected caps to cope with high frequencies and
current for a while, but the combination of these may be new. It
would be ironic if growth predictions based on Moore's Law faltered
not on the chips, but on dull bread-and-butter parts like caps.

From an industry-naive chemistry viewpoint, water based electrolytes *do*
seem like a bad idea with the current temps around a CPU & its VR on a
modern mbrd. I think the OP was suggesting that this 2nd round of failures
was related more to the increasing temps the capacitors were subjected to
as the mbrd substrate itself becomes a part of the heatsinking "solution".
Could be that the electrolytic's days are numbered but I've no idea what
what the ramifications might be for mbrd circuitry if designers are forced
to go to solids.

What's interesting is that no-one has been able to maintain a list of
affected motherboard brands, for fear of litigation etc.

The two prime culprits/victims would appear to be Abit and MSI - at least
their names seem to be mentioned frequently in the anecdotal reports in the
forums at places like www.badcaps.net - the "kits" he has available are
good clue I think and it doesn't take long to skim
http://www.badcaps.net/forum/. Shuttle more or less dropped out of the
U.S. market for a few years around the timeframe in question, so that might
explain the lower incidence of their name; the surprising one is
Supermicro.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

[cquirke (MVP Win9x)]

Yes the original IEEE story was here:
http://www.spectrum.ieee.org/WEBONLY/resource/feb03/ncap.html

That's the one I read, too.

more at http://www.niccomp.com/taiwanlowesr.htm where it's mentioned that
it was a double defection and theft -

Nice links from there...

http://www.low-esr.com/

....seem to confirm what I see as a common failure pattern in 815G and
845G (when built-in SVGA used); ripple interference effects on screen,
which prompts a look inside that shows bulging or leaking caps.

From an industry-naive chemistry viewpoint, water based electrolytes *do*
seem like a bad idea with the current temps around a CPU & its VR on a
modern mbrd. I think the OP was suggesting that this 2nd round of failures
was related more to the increasing temps the capacitors were subjected to

Temp is a factor, yeah, though the graphs seem happy at typical case
temps (45C is common with Prescott).

I followed a few links off Low-ESR.com and they make interesting
reading. For example, they mention liquid-centered caps use ionic
rather than electronic conduction which is slower; that implies a need
to have the electrolysis effect, and thus a polar liquid.

Could be that the electrolytic's days are numbered

The problem is getting the capacitance required. So far, combination
liquid plus solid can't get up there, though they extend the range of
the smaller purely solid capacitors.

I've no idea what the ramifications might be for mbrd circuitry if
designers are forced to go to solids.

Well, they already talk about capacitor *banks* (of electrolytics) so
to get the same capacity in solic caps is going to be ugly. They also
note that liquid caps fail open-circuit, whereas solids can combust;
that may be a safety factor, too.

The two prime culprits/victims would appear to be Abit and MSI - at least
their names seem to be mentioned frequently in the anecdotal reports in the
forums at places like www.badcaps.net

The ones I see here have been DFI, JetWay and Gigabyte - i.e. all of
the motherboards commonly used or seen from that time.

I've started using Intel motherboards (as opposed to motherboards
based on Intel chipsets) in the 865G era, but it's too early to tell
whether this will avoid the problem. So Far, So Good.

---------- ----- ---- --- -- - - - -

"He's such a character!"
' Yeah - CHAR(0) '

 

[George Macdonald]

That's the one I read, too.


Nice links from there...

http://www.low-esr.com/

...seem to confirm what I see as a common failure pattern in 815G and
845G (when built-in SVGA used); ripple interference effects on screen,
which prompts a look inside that shows bulging or leaking caps.


Temp is a factor, yeah, though the graphs seem happy at typical case
temps (45C is common with Prescott).

I got the hint that it was (extra) heat+age causing drying out of the
electrolyte more quickly.

I followed a few links off Low-ESR.com and they make interesting
reading. For example, they mention liquid-centered caps use ionic
rather than electronic conduction which is slower; that implies a need
to have the electrolysis effect, and thus a polar liquid.

Hmmm, and water is err, free.... well purify and dope it for the
purpose.

The problem is getting the capacitance required. So far, combination
liquid plus solid can't get up there, though they extend the range of
the smaller purely solid capacitors.


Well, they already talk about capacitor *banks* (of electrolytics) so
to get the same capacity in solic caps is going to be ugly. They also
note that liquid caps fail open-circuit, whereas solids can combust;
that may be a safety factor, too.



The ones I see here have been DFI, JetWay and Gigabyte - i.e. all of
the motherboards commonly used or seen from that time.

I wish there was a clearer picture of how much the defective electrolyte
was to blame globally and to what extent we're talking about low-cost
capacitor mfrs who don't have any capacitor technology but turn out cheap
junk... and to what extent the "expert" mfrs are just hitting the wall on
current technology. There's just an apparent mix of things which went
wrong, which gives an impression of pot-luck to us.

I've started using Intel motherboards (as opposed to motherboards
based on Intel chipsets) in the 865G era, but it's too early to tell
whether this will avoid the problem. So Far, So Good.

I've been led to believe that Intel desktop mbrds are coming from
same/similar PCB factories as everybody else... but if you get consistent
quality that's what really matters. I have some MSI K7 mbrds coming up for
2 years - I guess I'd better check them for capacitors with swollen
tops.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

[cquirke (MVP Win9x)]

On Fri, 31 Dec 2004 02:48:29 -0500, George Macdonald

I wish there was a clearer picture of how much the defective electrolyte
was to blame globally and to what extent we're talking about low-cost
capacitor mfrs who don't have any capacitor technology but turn out cheap
junk... and to what extent the "expert" mfrs are just hitting the wall on
current technology.

I think we may be hitting the wall.

If it were a simple matter of bad goo, then I'd expect to see the
problem affecting only around a year's worth of manufacture.

By the time mobo makers switched from 815 (Socket 370) to 845 (Socket
478) I'd have expected these brand-new mobos to have been using good
caps, and in fact by then I'd have expected the bad cap makers to have
switched to the proper goo.

But in practice, I'm commonly seeing bad 845's; if anything, they seem
to fail faster. I'm also seeing mobos with cooling ducts to protect
the caps bank nearest the CPU; if it was just a matter of buying
better caps, I'd have expected them to do that rather than add the
cost of new parts and board layout changes to fit them.

The other interesting thing, is that I seldom see this capacitor
failure pattern anywhere other than on motherboards. I've seen
bulging caps within an ATX PSU once, and none in any other component.
I see explosive cap failures in a particular series of ATX PSUs, but
this is a different failure pattern and is due to a PSU design flaw
AFAIK. IOW, the cap blows because of something else that's gone
disasterously wrong (indeed, usually all +12V components fry too).

So yes; I think there's something inherent in the role these caps are
expected to fulfil with respect to CPU power in particular, that is
causing these components to fail within expected product lifetime.

I've been led to believe that Intel desktop mbrds are coming from
same/similar PCB factories as everybody else... but if you get consistent
quality that's what really matters.

It's too early to tell. As posted in another thread, I've not been
impressed with board quality, though at least the BIOS sucks less than
earlier Intel motherboards (e.g. 440BX era).

Overall, it seems as if we've gone from mobos as seldom-fail items (as
one expects of solid-state, no-moving-parts gear) to being a high
failure rate item. This in itself is quite remarkable.

--------------- ----- ---- --- -- - - -

Tech Support: The guys who follow the
'Parade of New Products' with a shovel.

 

[Scott Alfter]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I've been led to believe that Intel desktop mbrds are coming from
same/similar PCB factories as everybody else... but if you get consistent
quality that's what really matters. I have some MSI K7 mbrds coming up for
2 years - I guess I'd better check them for capacitors with swollen
tops.

I had a couple of MSI K7N420 Pros (nForce boards with integrated graphics)
that both needed recapping a year or so ago. Each of them started getting
flaky and eventually wouldn't complete booting before it started wigging
out. Some of the 1000-uF capacitors blew out the top or bottom, some only
bulged a little bit, and a small number still looked OK. It was only the
1000-uF capacitors (with 18 on each board) that appeared to be affected. A
few dollars at DigiKey for replacements got both boards running again, and
they've run fine ever since.

I also had a K7D Master that started malfunctioning. I don't recall if it
was a capacitor problem or if it was something else, but sometimes it'd only
see one processor instead of two (it was a dual Athlon MP board). I bought
a Tyan Tiger MPX to replace it, got it fixed under warranty, and sold the
board MSI shipped back to me on eBay.

_/_
/ v \ Scott Alfter (remove the obvious to send mail)
(IIGS( http://alfter.us/ Top-posting!
\_^_/ rm -rf /bin/laden >What's the most annoying thing on Usenet?

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (Linux)

iD8DBQFB1ygDVgTKos01OwkRAt8mAJ0a4D3buHIO5rFi/09Ok0QcyR78HQCg+RL/
sO6G+0npQHO6Mzb3QzOgfOs=
=11iA
-----END PGP SIGNATURE-----

 

[keith]

On Fri, 31 Dec 2004 02:48:29 -0500, George Macdonald


I think we may be hitting the wall.

If it were a simple matter of bad goo, then I'd expect to see the
problem affecting only around a year's worth of manufacture.

I disagree. I'm on the "bad goo" side of this one.

By the time mobo makers switched from 815 (Socket 370) to 845 (Socket
478) I'd have expected these brand-new mobos to have been using good
caps, and in fact by then I'd have expected the bad cap makers to have
switched to the proper goo.

Cheap is forever, in this business. Good capacitors (or more) isn't
rocket-surgery, but it doesn't help the boottom line; PCs are cheap crap.
The cheaper, the better. If there is a way to save a buck, everyone will
soon be saving that buck.

But in practice, I'm commonly seeing bad 845's; if anything, they seem
to fail faster. I'm also seeing mobos with cooling ducts to protect the
caps bank nearest the CPU; if it was just a matter of buying better
caps, I'd have expected them to do that rather than add the cost of new
parts and board layout changes to fit them.

I wouldn't. Intel and M$ have fashioned the market so they are the only
ones who can make money. The rest cut each other's throats to supply the
cheapest garbage possible. If crap capacitors are cheaper and "work", so
be it.