Badcaps.net Forum
Go Back   Badcaps Forums > Motherboards with Bad Capacitors > Abit - Asus - Aopen
Register FAQ Calendar Search Today's Posts Mark Forums Read

 
Thread Tools Display Modes
Old 05-21-2020, 03:59 PM   #1
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Killed my Asus A7V266-E? Need advice

Hi guys, I have troubles after recapping my board. One of the 4 caps on the output side of the VRM is slightly bulged, otherwise there was no issue with the board at all. I read the other thread about polymodding this board, it was really helpful! So I went on and replaced the 4 caps with the NCC ones too. I used my spare XP1600+ to test first, as I do not want to fry my CPU, it's a nice 1700+ that can run 1.35V and very high clock too. It took me at least 3 tries to get it finally boot up into bios(I was anxious and mistaken the power LED and then power reset for the power switch wiring at first), I was really happy to see it POST into bios normally. Before I switch back the CPU, I wanted to test the 1.35V Vcore too, this time it didn't boot up, the CPU fan spins and there is no beep. I thought OK, it's too low and I switched back to the 1.7 Vcore, unfortunately it never boot up again, CPU fans spins and there is no beep.

At this point I thought the motherboard is dead. I was frustrated and sad, so the stupidity began, I remove the CPU cooler, power it on and just to see if it gets hot at all. Surely it get hot quicker than my solder iron, before I could shut it off, one of the coils of the VRM shot up smoke for about 1 second. CPU smoked first, I don't care. The VRM actually still looked fine. Later I came to my senses and power it on without the CPU, and measure the Vcore directly from the socket, it actually matches up perfectly to the different jumper settings I tried, so VRM appear to be working still?

Anyway, what bothers me is that I don't know what went wrong, I don't know why it won't boot up after the first time. I was clearing the hole with a needle, I think was a bit careless and pushed one hole too hard, that probably delaminated the board? So it was one last boot and then the board died?? The VRM seems to work fine, but it took a hit and probably will fail soon? So I shall give it up, no need to try to boot it up with an other CPU again? Please advice, thanks!

Last edited by delide; 05-21-2020 at 04:03 PM..
delide is offline   Reply With Quote
Old 05-22-2020, 03:16 AM   #2
Per Hansson
Super Moderator
 
Per Hansson's Avatar
 
Join Date: Jul 2005
City & State: ----
My Country: Sweden
Line Voltage: 230v 50Hz
I'm a: Knowledge Seeker
Posts: 4,412
Default Re: Killed my Asus A7V266-E? Need advice

Please post good clear pictures of the top and bottom of the board where you worked.
__________________
"The one who says it cannot be done should never interrupt the one who is doing it."
Per Hansson is offline   Reply With Quote
Old 05-22-2020, 11:22 AM   #3
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Thanks for your quick response, I attached some photos of those area. I don't see visible damage, the choke that smoke doesn't look different either. But now I see that one joint is still partially uncovered(bottom3 top right), should have bought much thinner solder wire. For now the VRM still seem to work though, as the VCC at the socket looked good and accurate, on my multi meter at least. I don't have a scope. Still I don't want to give up yet. What are your thoughts? Thanks!
Attached Images
File Type: jpg bottom1.JPG (716.3 KB, 14 views)
File Type: jpg bottom2.JPG (717.5 KB, 16 views)
File Type: jpg bottom3.JPG (914.9 KB, 17 views)
File Type: jpg smoked choke.JPG (715.0 KB, 21 views)
File Type: jpg top1.JPG (905.4 KB, 22 views)
File Type: jpg top2.JPG (654.4 KB, 17 views)
File Type: jpg top3.JPG (736.5 KB, 17 views)

Last edited by delide; 05-22-2020 at 11:25 AM..
delide is offline   Reply With Quote
Old 05-22-2020, 01:51 PM   #4
momaka
Badcaps Veteran
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 9,323
Default Re: Killed my Asus A7V266-E? Need advice

It doesn't look like the smoke came from the coil. Perhaps it was from the CPU and air movement (from a fan or whatever else) blew it away, making it look like it came from the coil. If it was the coil smoking, you would have seen the wire darken and also smelled it. Overheated magnet (coated copper) wire typically has a very specific and strong smell.

Also, if the VRM appears to be outputting the correct voltage and nothing is overheating in the VRM circuit, then you can give the board a try with another CPU. Just don't try it without the heatsink this time.

On that note, beware of online "techs" (particularly on YouTube and social media platforms) that maintain / give advice to performing a "boot test" with no heatsink on the GPU or CPU (or whatever else you're troubleshooting) - it's NEVER a good idea! Heatsinks are there for a reason. Admittedly, modern CPUs and GPUs don't usually go full 100% load when booted, so often times, omitting the heatsink on a modern CPU or GPU is rarely catastrophic. But if thermal throttling is reached, it probably will compromise the reliability somewhere down the road. So it's still something that should be avoided.

And as for older CPUs and GPUs, you saw what happened already - instant magic smoke. This is especially applicable to old AMD CPUs, particularly socket 462. I think every manual for every socket 462 motherboard I've read included at least one warning that running the CPU without a heatsink will cause likely CPU and possible board damage.

In your case, however, I don't think the rest of the board took damage. It appears to use a synchronous VRM for the CPU V_core. So the CPU VRM's controller IC probably prevented anything bad from happening, even if the CPU core shorted out.

With that said, before you try any valuable CPU you care about, perhaps consider getting a cheap AMD Duron or Sempron CPU that the board is compatible with, just in case. Since it appears the board did boot with the recap at some point in time, I think your recap was good - or at least for the caps around the CPU VRM. (On that note, your solder joints don't look that bad, honestly. Most look OK, and a few could just easily be smoothed out by putting a little bit of flux on them and then re-heating them with your soldering iron again.) Of course, if you haven't fully recapped the whole board (that is, most caps 470 uF and up should be change) and there are still bad cap brands left, that could also be the reason why the board is not booting up.

Before anything, though, I suggest you remove the CMOS battery to reset CMOS settings. This should rule out issues stemming from incorrect CMOS settings (like for example, resetting the CPU V_core voltage.)
Then try the board with another CPU (and heatsink) to see what happens.

Also, may we see one or two more pictures of your entire board? (In case we see something that you didn't, possibly. )

Last edited by momaka; 05-22-2020 at 01:55 PM..
momaka is offline   Reply With Quote
Old 05-22-2020, 02:01 PM   #5
ChaosLegionnaire
HC Overclocker
 
Join Date: Jul 2012
City & State: Singapore
My Country: Singapore
Line Voltage: 240VAC 50Hz
I'm a: Hobbyist Tech
Posts: 2,051
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by delide View Post
At this point I thought the motherboard is dead. I was frustrated and sad, so the stupidity began, I remove the CPU cooler, power it on and just to see if it gets hot at all. Surely it get hot quicker than my solder iron, before I could shut it off, one of the coils of the VRM shot up smoke for about 1 second. CPU smoked first, I don't care. The VRM actually still looked fine.
^this...^ never run the cpu without the heatsink! not even briefly esp. if the cpu and board isnt designed with overheat protection. i know the athlon xp series doesnt have overheat protection to prevent the cpu from being fried.

the proper way to do this is to run it still with the heatsink on the cpu but unplug the cable for the cpu fan. the heatsink should still provide some thermal mass to absorb the heat while the lack of the cpu fan should allow u to touch the heatsink to see how fast it heats up. thats the proper way to do it.

that test xp 1600+ is probably fried. try to get another cheap test cpu to see if it works or test that supposedly fried cpu in another board to see if it works.

u didnt say what psu u are using. are u using a period specific 5v heavy psu? if u are, it means the psu is also quite old and it may be failing. check the psu to make sure it hasnt failed to rule it out as the cause of the non-booting board.

also try to clear the cmos to see if it makes it boot. sometimes either during overclocking or undervolting, if the clocks are too high or the voltage is too low to allow the cpu to work properly, the cmos settings somehow become corrupt and the board cant boot with the bad cmos settings. clearing cmos fixes this problem.
Quote:
Originally Posted by delide View Post
I see that one joint is still partially uncovered(bottom3 top right), should have bought much thinner solder wire.
if u have read through the forums, the proper procedure to check if u have soldered the caps on properly would be to try to rock it from side to side perpendicularly across the line of the two legs of the cap. it should resist movement firmly. if it feels flimsy, the solder didnt go all the way into the hole. reflow the joint.

and another joint has too much solder in the bottom2.jpg picture. the left most joint.
ChaosLegionnaire is offline   Reply With Quote
Old 05-22-2020, 05:11 PM   #6
RJARRRPCGP
Badcaps Veteran
 
Join Date: Jul 2004
City & State: North Springfield, Vermont
My Country: USA
Line Voltage: 118-127V 59-63.5 Hz-> actualizo: pérdido de voltaje
I'm a: Knowledge Seeker
Posts: 5,124
Default Re: Killed my Asus A7V266-E? Need advice

The video of a T-bird frying, was with the heatsink taken off with a 3D game running, IIRC, LOL.
__________________
Asus P6T6 WS Revolution

Core i7 Extreme "Bloomfield" 965

Asus Strix GeForce GTX 970

Windows 7 SP1

SoundBlaster ZXR

Corsair TX850M PSU

Looks like I have a laptop that kills a desktop PC!

Received on November 4, 2019:
MSI GF63 Thin 9SC-649US


"There's nothing more unattractive than a chick smoking a cigarette" -Topcat

"Don't eat yellow snow!" -Salem

"did I see a chair fly? I think I did! Time for popcorn!" -ratdude747
RJARRRPCGP is online now   Reply With Quote
Old 05-22-2020, 05:26 PM   #7
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by momaka View Post
It doesn't look like the smoke came from the coil. Perhaps it was from the CPU and air movement (from a fan or whatever else) blew it away, making it look like it came from the coil. If it was the coil smoking, you would have seen the wire darken and also smelled it. Overheated magnet (coated copper) wire typically has a very specific and strong smell.

Also, if the VRM appears to be outputting the correct voltage and nothing is overheating in the VRM circuit, then you can give the board a try with another CPU. Just don't try it without the heatsink this time.

On that note, beware of online "techs" (particularly on YouTube and social media platforms) that maintain / give advice to performing a "boot test" with no heatsink on the GPU or CPU (or whatever else you're troubleshooting) - it's NEVER a good idea! Heatsinks are there for a reason. Admittedly, modern CPUs and GPUs don't usually go full 100% load when booted, so often times, omitting the heatsink on a modern CPU or GPU is rarely catastrophic. But if thermal throttling is reached, it probably will compromise the reliability somewhere down the road. So it's still something that should be avoided.

And as for older CPUs and GPUs, you saw what happened already - instant magic smoke. This is especially applicable to old AMD CPUs, particularly socket 462. I think every manual for every socket 462 motherboard I've read included at least one warning that running the CPU without a heatsink will cause likely CPU and possible board damage.

In your case, however, I don't think the rest of the board took damage. It appears to use a synchronous VRM for the CPU V_core. So the CPU VRM's controller IC probably prevented anything bad from happening, even if the CPU core shorted out.

With that said, before you try any valuable CPU you care about, perhaps consider getting a cheap AMD Duron or Sempron CPU that the board is compatible with, just in case. Since it appears the board did boot with the recap at some point in time, I think your recap was good - or at least for the caps around the CPU VRM. (On that note, your solder joints don't look that bad, honestly. Most look OK, and a few could just easily be smoothed out by putting a little bit of flux on them and then re-heating them with your soldering iron again.) Of course, if you haven't fully recapped the whole board (that is, most caps 470 uF and up should be change) and there are still bad cap brands left, that could also be the reason why the board is not booting up.

Before anything, though, I suggest you remove the CMOS battery to reset CMOS settings. This should rule out issues stemming from incorrect CMOS settings (like for example, resetting the CPU V_core voltage.)
Then try the board with another CPU (and heatsink) to see what happens.

Also, may we see one or two more pictures of your entire board? (In case we see something that you didn't, possibly. )
Thanks for your lengthy answer! You must be right! It did puzzle me, where did that smoke came from, I looked at the CPU now, it looks like melted sealant or glue(the stuff around the edges of the die) shot from one corner to the left of the board, it probably hit the coil and made it looks like the coils smoked. There was indeed a smell, should be just the CPU then, I can't tell, it must be the first time I burned something outside of a oven...

I indeed only replaced the 4 HM ones, which are the biggest and therefore the easiest to recap I would think, so I guess I was forcing it a bit too. Anyway, I have subzero confidence right now, certainly feel like the more caps I replace, the more chance I will just kill the board, when I get to them all it would be 100% dead

Thanks again and I feel better now as I see some hope. I will get an other set of CPU and board to try it again. I will post more photos of the board tomorrow too.
delide is offline   Reply With Quote
Old 05-22-2020, 05:33 PM   #8
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by ChaosLegionnaire View Post
^this...^ never run the cpu without the heatsink! not even briefly esp. if the cpu and board isnt designed with overheat protection. i know the athlon xp series doesnt have overheat protection to prevent the cpu from being fried.

the proper way to do this is to run it still with the heatsink on the cpu but unplug the cable for the cpu fan. the heatsink should still provide some thermal mass to absorb the heat while the lack of the cpu fan should allow u to touch the heatsink to see how fast it heats up. thats the proper way to do it.

that test xp 1600+ is probably fried. try to get another cheap test cpu to see if it works or test that supposedly fried cpu in another board to see if it works.

u didnt say what psu u are using. are u using a period specific 5v heavy psu? if u are, it means the psu is also quite old and it may be failing. check the psu to make sure it hasnt failed to rule it out as the cause of the non-booting board.

also try to clear the cmos to see if it makes it boot. sometimes either during overclocking or undervolting, if the clocks are too high or the voltage is too low to allow the cpu to work properly, the cmos settings somehow become corrupt and the board cant boot with the bad cmos settings. clearing cmos fixes this problem.

if u have read through the forums, the proper procedure to check if u have soldered the caps on properly would be to try to rock it from side to side perpendicularly across the line of the two legs of the cap. it should resist movement firmly. if it feels flimsy, the solder didnt go all the way into the hole. reflow the joint.

and another joint has too much solder in the bottom2.jpg picture. the left most joint.
Thank you very much for your advices, yes, I really should just touch the heatsink after I removed it! I guess my brain got shorted for a moment before the CPU did!

The PSU is fairly new, it's a bequiet S9 400W, so it should be OK. Yes, the caps feel pretty solid when I tried to move them as you said, but I will try to refill the joint if I got the board to boot up again. The other joint has too much solder indeed, I rarely do soldering at all, it should still be OK? That place actually has a 8 shaped soldering point, so it was easier to work with and the joint will look wider than the others. I used 1mm solder wire instead of buying thinner one. I read that it's a pain to use thick wire, didn't know why but now I kind of get it.

Hopefully I will get the board running again soon, at least my retro PC. Thanks and cheers!
delide is offline   Reply With Quote
Old 05-23-2020, 06:00 AM   #9
dicky96
Badcaps Veteran
 
dicky96's Avatar
 
Join Date: Mar 2017
City & State: Sunny Gran Canaria
My Country: Spain
Line Voltage: 220V 50Hz
I'm a: Professional Tech
Posts: 700
Default Re: Killed my Asus A7V266-E? Need advice

Here are the boardview files for the Asus motherboards that could be very similar to yours

A7A266
A7A266-E
A7V266-M
A7V266-MX

It could be useful to you.
Attached Files
File Type: rar Asus266.rar (1.55 MB, 2 views)
dicky96 is offline   Reply With Quote
Old 05-23-2020, 12:51 PM   #10
RJARRRPCGP
Badcaps Veteran
 
Join Date: Jul 2004
City & State: North Springfield, Vermont
My Country: USA
Line Voltage: 118-127V 59-63.5 Hz-> actualizo: pérdido de voltaje
I'm a: Knowledge Seeker
Posts: 5,124
Default Re: Killed my Asus A7V266-E? Need advice

In that era, "-VM"s have things disabled by Asus. The "-VM"s of that era, are synonymous with crippled!

While the "-E"s OTOH, are synonymous with loaded, especially with the Asus A7N8X-E of 2004. But, even the Asus A7N8X-E, unfortunately, especially in 2004, probably usually have KZGs.

Last edited by RJARRRPCGP; 05-23-2020 at 12:53 PM..
RJARRRPCGP is online now   Reply With Quote
Old 05-23-2020, 02:06 PM   #11
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Thumbs up Re: Killed my Asus A7V266-E? Need advice

I've attched more photos of the entire board, thanks for looking!! Also attached a photo of the CPU, at the top left corner you can see the mark of the "eruption", pointed toward the VRM as well, certainly back up momaka's observations! BTW the die itself still look quite normal too, not that I will put it in again, BUT, just maybe I was not the latest/the last idiot to have fried a XP CPU then guess we will never know

An other CPU is on the way, I will then try to boot it up, this time I will certainly clear the CMOS if it won't boot. If I'm successful, I will then fix the joints, then test again, and then hopeful I will finally be able to put everything back together again! And then doing some more tests!
Attached Images
File Type: jpg top1.JPG (1.38 MB, 22 views)
File Type: jpg top2.JPG (1.17 MB, 16 views)
File Type: jpg upper bottom.JPG (1.36 MB, 13 views)
File Type: jpg lower bottom.JPG (1.57 MB, 14 views)
File Type: jpg cpu.JPG (710.7 KB, 21 views)

Last edited by delide; 05-23-2020 at 02:07 PM..
delide is offline   Reply With Quote
Old 05-23-2020, 02:11 PM   #12
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by dicky96 View Post
Here are the boardview files for the Asus motherboards that could be very similar to yours

A7A266
A7A266-E
A7V266-M
A7V266-MX

It could be useful to you.
Thanks you very much! I have to say I'm far from knowledgeable or smart enough to do anything with it, but I've saved them just in case. And of course someone else may find them very useful!
delide is offline   Reply With Quote
Old 05-23-2020, 02:20 PM   #13
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by RJARRRPCGP View Post
In that era, "-VM"s have things disabled by Asus. The "-VM"s of that era, are synonymous with crippled!

While the "-E"s OTOH, are synonymous with loaded, especially with the Asus A7N8X-E of 2004. But, even the Asus A7N8X-E, unfortunately, especially in 2004, probably usually have KZGs.
Haha, indeed, I used to think it's an economy version of the deluxe version, so I always went for the -E, till I totally get lost in the new generations.

A7N8X-E seems to be very nice indeed, I thought about it as replacement, but then I read that the Nforce chip has trouble with DOS emulation of PCI sound card and gave up the plan.
delide is offline   Reply With Quote
Old 05-27-2020, 07:51 PM   #14
momaka
Badcaps Veteran
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 9,323
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by delide View Post
I indeed only replaced the 4 HM ones, which are the biggest and therefore the easiest to recap I would think, so I guess I was forcing it a bit too. Anyway, I have subzero confidence right now, certainly feel like the more caps I replace, the more chance I will just kill the board, when I get to them all it would be 100% dead
Fear not!
The caps around the CPU are usually the hardest to replace, because they sit on the thickest copper planes. The fact that you got those recapped means you're probably done with the hardest part already. So the other caps should be easier to remove and recap.

If those Nichicon HM caps that remain on the board are still the original ones and they have a date code starting with H01__, H02__, H03__, or H04__, you should replace those as well. Nichicon HM made between 2001-2004 (corresponding to the above date codes) have a nearly 100% failure rate (even when they don't appear bad, sometimes.)

I replace a bunch like that (that appeared OK, but were bad / showing extremely high capacitance due to high internal leakage) in a P4 Gateway PC that was working fine, but always unstable when cold. Replacing all of the Nichicon HM caps on the CPU cured it for good.

Quote:
Originally Posted by delide View Post
BTW the die itself still look quite normal too...
It does... but then you look at the slightly darkened PCB around it, and you know this thing must have overheated.

Quote:
Originally Posted by delide View Post
just maybe I was not the latest/the last idiot to have fried a XP CPU then guess we will never know
Socket 462 CPU have always been tricky and prone to frying. I've seen even experienced PC builders fry them by making a small mistake in the build somewhere. It doesn't take much. Usually, it's the heatsink mounting that can be problematic (for example, the clips on the socket snapping, causing the heatsink to loosen/fall off and fry the CPU instantly.) On that note, I see you have a motherboard/PCB-mounted heatsink retention mechanism on yours. With those, you have to be careful that the CPU heatsink is sitting even on the CPU die and making proper contact at all times. If it isn't, that could also be why you had instability issues.

BTW, it's also really cool that your motherboard has jumpers to set the FSB/voltage/multiplier.

Last edited by momaka; 05-27-2020 at 07:59 PM..
momaka is offline   Reply With Quote
Old 05-27-2020, 10:06 PM   #15
Wester547
-
 
Join Date: Nov 2011
City & State: CA.
My Country: USA.
Line Voltage: 120-125VAC 60Hz.
I'm a: Knowledge Seeker
Posts: 1,244
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by momaka
If those Nichicon HM caps that remain on the board are still the original ones and they have a date code starting with H01__, H02__, H03__, or H04__, you should replace those as well. Nichicon HM made between 2001-2004 (corresponding to the above date codes) have a nearly 100% failure rate (even when they don't appear bad, sometimes.)
They actually appear to be Nichicon HD (1500µF 6.3V, they appear to match the case size in the datasheet as well going by the photos provided). The fifth image in the third post depicts part of the series name (“D” on one sleeve and “(M)” on another). They are rather long in the tooth now, but so are all the lesser quality OST scattered about the board. Hmm... not to be condescending, but haven’t we had this conversation before?

I’d say the 6.3V and 10V Nichicon HM, HN, and HZ made between 2001 and early 2005 (just 2005 for HN and HZ) have a rather high failure rate, but the 16V rated parts seem to hold up better both visually and in terms of their tolerances. Probably due to their thicker oxide film protracting deterioration, and they are usually under less duress unless some MOSFET neighboring them is dissipating too much heat. That said, I would not ever recommend leaving a bad datecode HM, HN, or HZ on any board, no matter the capacitance or voltage rating.
Wester547 is offline   Reply With Quote
Old Yesterday, 08:20 AM   #16
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by momaka View Post
Fear not!
The caps around the CPU are usually the hardest to replace, because they sit on the thickest copper planes. The fact that you got those recapped means you're probably done with the hardest part already. So the other caps should be easier to remove and recap.

If those Nichicon HM caps that remain on the board are still the original ones and they have a date code starting with H01__, H02__, H03__, or H04__, you should replace those as well. Nichicon HM made between 2001-2004 (corresponding to the above date codes) have a nearly 100% failure rate (even when they don't appear bad, sometimes.)

I replace a bunch like that (that appeared OK, but were bad / showing extremely high capacitance due to high internal leakage) in a P4 Gateway PC that was working fine, but always unstable when cold. Replacing all of the Nichicon HM caps on the CPU cured it for good.


It does... but then you look at the slightly darkened PCB around it, and you know this thing must have overheated.


Socket 462 CPU have always been tricky and prone to frying. I've seen even experienced PC builders fry them by making a small mistake in the build somewhere. It doesn't take much. Usually, it's the heatsink mounting that can be problematic (for example, the clips on the socket snapping, causing the heatsink to loosen/fall off and fry the CPU instantly.) On that note, I see you have a motherboard/PCB-mounted heatsink retention mechanism on yours. With those, you have to be careful that the CPU heatsink is sitting even on the CPU die and making proper contact at all times. If it isn't, that could also be why you had instability issues.

BTW, it's also really cool that your motherboard has jumpers to set the FSB/voltage/multiplier.
Thanks for your reply again! I will certainly check those remaining caps! For now I have good news, I got it boot up successfully! It was totally the opposite of straightforward, there were all sorts of strangest behavior, it took me countless boot attempts but then all problems disappeared suddenly! I was so happy to be able to enter the BIOS that I booted a few times more:-) Also ran a few passes of Memtest86 yesterday, so far so good! Thanks for your advices! Will post how it went shortly.

I did reheat and improve the joints while I was waiting for the CPU. I think in the first attempt I made a mistake of not heating up the pins of the caps properly. And one should really practice first if possible. Now I'm just starting to get the hang of it I think. I actually bought the same board together with the other CPU, it hasn't been used since 2008, but it still booted, I will check it later too and probably then recap it completely!

Yes, the clips on the cooler are terrible, very difficult to mount/remove, but easy to damage the board/CPU. I'm using a Zalman cooler now, I still have to be careful indeed, but it's much better than those with clips!

Last edited by delide; Yesterday at 08:21 AM.. Reason: tippos
delide is offline   Reply With Quote
Old Yesterday, 08:32 AM   #17
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by Wester547 View Post
They actually appear to be Nichicon HD (1500µF 6.3V, they appear to match the case size in the datasheet as well going by the photos provided). The fifth image in the third post depicts part of the series name (“D” on one sleeve and “(M)” on another). They are rather long in the tooth now, but so are all the lesser quality OST scattered about the board. Hmm... not to be condescending, but haven’t we had this conversation before?

I’d say the 6.3V and 10V Nichicon HM, HN, and HZ made between 2001 and early 2005 (just 2005 for HN and HZ) have a rather high failure rate, but the 16V rated parts seem to hold up better both visually and in terms of their tolerances. Probably due to their thicker oxide film protracting deterioration, and they are usually under less duress unless some MOSFET neighboring them is dissipating too much heat. That said, I would not ever recommend leaving a bad datecode HM, HN, or HZ on any board, no matter the capacitance or voltage rating.
Wow, I can recognize nothing from my photos! But yes, I read everything you guys said in that thread many times. As I understand it, for his board it's OK to just replace the 4 3300uF caps, somehow I went stupid again, and assumed my board would have the same brand/type of caps as his, so I actually never did check mine and just replaced the same 4 caps, as one of them near the MOSFET just bulged very slightly.

By the way the board is from my build in 2002, I used the PC for about 6 or 7 years, after that I just kept it, even updated it with some more efficient parts, better cooling, new power supply... Otherwise I only turned it on from time to time to see how it's doing(heard that it's better for the hardware/caps). If I remember correctly it has about 15000 hours on it(according to the hard drive), since I won't be using it a lot, I didn't actually plan to recap it, but I will take a closer looker at all caps. Thanks for your advice!
delide is offline   Reply With Quote
Old Yesterday, 09:22 AM   #18
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Here is the story so far, if anyone has boot problems with the board it might be helpful.

With the new CPU installed I checked the wiring, jumpers and everything, even prayed and then turned it on, it was still the same, fan spins, no beep. So I cleared the CMOS and checked everything again, and again still nothing. I repeated the whole process quite a few times, but it's always the same nothing! At that point I believed I was working on a still fresh looking but pretty dead board... I was running out of ideas, but just in case the graphic card went bad, I removed it, powered it on and again still nothing! However as I pressed and hold the power button, I finally heard some beeps before it shut down! So long nothing, I wasn't even sure where that came from! However, as I powered it on again, it went back to nothing!

I can't remember what my plan then was. I did shut down the power supply at some point, the next time I turned the PS on, the board started immediately! And there was the signal for missing or dead graphics card indeed! So the board is not completely dead after all! After a few further attempts, it turns out that the board now only POST when it's onboard power LED is already off and then I turn on the power supply, with the case switch it turns on but never went to POST!

Since I got these stable beeps anyhow, the graphic card was reinstalled. At the next boot via the power supply I got one beep! But still no image, press any key resulted in a beep too. Then at the next boot I got image and made it into the bios, though it showed corrupted text and it has already frozen. Then the next boot the bios displayed message that the last boot failed because of improper CPU frequency settings. I was able to choose a proper clock, save and exit. Then finally, all problems disappeared! POST went normally via case switch and the board doesn't react to the switch of power supply anymore, all problem disappeared, almost can't believe it!

Although the VID is set by the jumpers, I used the jumper free mode for the frequencies, that has caused all these boot issues I think, never knew this can lead to such problems! Even after the reset the jumper free mode didn't work properly it seems, no sure if this has to do with my recap. In hindsight if I just gambled and put the original CPU directly in, it might just worked and save me a lot of nerves! I have now installed the original CPU back, restored some performance settings and done several passes of Memtest86 with no error. So far so good! Soon I will add everything else back and try to boot it into windows and do some more benchmark tests to check the speed and stability! Hopefully other parts of the board work too. Thanks for helps guys!

By the way, a interesting note is that the VID jumpers now work correctly, before the recap I always needed to set the jumper 0.15V lower: for example for 1.2V so that I can get 1.35V in the bios, now the VIP jumpers setting match the bios VID reading. I guess it has to do with the CMOS reset, rather than the recap itself? I've also checked the VRM temperatures while running Memtest86, interestingly the chokes are the hottest by far, at least with the 1700+ at 1.35V. The CPU and MOSFETs don't seem to reach body temp at all, but the coke was clearly higher than 40C/104F, in summer and closed case I expect the difference to be even larger.

Last edited by delide; Yesterday at 09:28 AM..
delide is offline   Reply With Quote
Old Yesterday, 10:02 AM   #19
momaka
Badcaps Veteran
 
momaka's Avatar
 
Join Date: May 2008
City & State: VA (NoVA)
My Country: U.S.A.
Line Voltage: 120 VAC, 60 Hz
I'm a: Hobbyist Tech
Posts: 9,323
Default Re: Killed my Asus A7V266-E? Need advice

Awesome! Great to hear the board is working again.

So it seems the board doesn't seem to read the CPU correctly when set to jumper-free mode and possibly sets the wrong VID/FSB/multiplier, which then causes it not to boot. I wonder if a BIOS fix was ever issued for that, or if this is even a known issue.

I guess it's a good thing that jumpers exist. On that note, this is one of the neat things about socket 462 CPUs - with a bit of work, you can hard-wire any motherboard to whatever VID, FSB, and multiplier you like. Now, as to whether or not the board boots, that's entirely up to what you set it to.

Quote:
Originally Posted by delide View Post
By the way the board is from my build in 2002, I used the PC for about 6 or 7 years, after that I just kept it
Nice!
Come to think of it, that was 18 years ago! Where did time fly?! (I feel old for saying that now.)

My first ever self-built PC was in late 2003, I think (or was it 2004, I can't remember.) Also an AMD/socket 462 with a cheapo Duron 1.4 GHz. But I used it proudly for about 3 years until the PSU became unstable. After this, I've moved onto different PCs, but still kept that PC as well. Many years later, I fully recapped the motherboard, as it was a beautiful red Jetway, but littered with crappy GSC caps. I posted about it here:
https://www.badcaps.net/forum/showth...ghlight=Jetway
Eventually, I also found an Athlon XP 2500+ for cheap, along with fixing up a Radeon 9700 video card. Back in the days, hardware like that seemed like a dream. But many years later, these things cost me barely anything on eBay... though now it seems old "retro" PC stuff (like high-end AGP cards) are climbing back up in value. So maybe a good thing I saved it.
momaka is offline   Reply With Quote
Old Yesterday, 11:53 AM   #20
delide
New Member
 
Join Date: May 2020
City & State: Andernach
My Country: Germany
I'm a: Knowledge Seeker
Posts: 16
Default Re: Killed my Asus A7V266-E? Need advice

Quote:
Originally Posted by momaka View Post
Awesome! Great to hear the board is working again.

So it seems the board doesn't seem to read the CPU correctly when set to jumper-free mode and possibly sets the wrong VID/FSB/multiplier, which then causes it not to boot. I wonder if a BIOS fix was ever issued for that, or if this is even a known issue.
Thanks! But very unfortunately I spoke a bit too early, I tried to power it on moments ago, the nothing came back again! :-( But fortunately, it was straightforward this time, I just reset the CMOS and it boots normally again. So there is probably still something strange going on, if not the CMOS maybe the VRM still have problem?? But once booted up, the Memtest ran without any trouble. Anyway I guess I have to keep the patient on observation for a couple more days, before I can resume and to put everything back.

By the way I just noticed an other problem, due to handling of the board lately, the spacing of the top of 2 replaced caps noticeably changed/reduced compared to my photo from the weekend, I've attached a comparison. The solder joints are all solid, so the difference must be caused by the legs being pulled out from the cap slightly. I do not think I was that heavy handed, probably one of the cap was a bit loose already. Anyway, is that a cause for concern, that the seal maybe compromised, shall I recap those again to be safe?? What do you think?
Attached Images
File Type: png spacing.png (1.14 MB, 10 views)
delide is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



Badcaps.net Technical Forums © 2003 - 2020
Powered by vBulletin ®
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
All times are GMT -6. The time now is 11:02 AM.
Did you find this forum helpful?