can bad caps cause system bus and memory errors?

[ChaosLegionnaire]

i have an asus a8v with the following caps:

4 x 1000uF 16V Panasonic FJ 8mm(Ø) x 20mm(H) @ CPU VRM In Area
6 x 1800uF 6.3V Panasonic FL 8mm(Ø) x 20mm(H) @ CPU VRM Out Area
20 x 820uF 6.3V Panasonic FJ 8mm(Ø) x 12mm(H) @ NB Area and AGP/PCI Slot Areas
9 x 100uF 16V Nichicon VZ(M) A0507 5mm(Ø) x 12mm(H) @ Onboard Sound Chip Area and Fan Headers

i previously recapped only the cpu vrm area of the board which had bad HMs from 2005 and KZGs but i still get freezes and BSODs in windows and kernel panics in linux.

windows throws up a machine check exception 9c error. the BSOD error code appears to be related to system bus or memory errors.

the linux kernel panic message appears to be memory related.

system was tested with memtest and prime95 blend but no error can be detected. however, the readmes for memtest and p95 note that it is possible to pass the test and still possibly have faulty components.

so i have to ask if i could still have bad caps in the NB and RAM areas? could the FJs in those areas be faulty as well and throwing up system bus and memory errors? this board was bought second hand by me. so i dont know the history of the board and how badly it was abused by the previous owner. even good caps can die if abused.

thanks for reading.

[ReeceyBurger123]

Re: can bad caps cause system bus and memory errors?

When capacitors go bad they fail to filter out and smooth any ripple current from the rest of the power supply. This can cause lots of glitches with the epprom and the flash memory. I have seen it on Samsung tvs where bad caps caused the epprom to corrupt due to the memory cycling due to the ripple current. So yes bad caps can cause flash memory to fail due to bad ripple current that the caps used to filter out but no longer do due to be failed.

[stj]

Re: can bad caps cause system bus and memory errors?

near the ram will be a regulator and some caps - look at those.

[ChaosLegionnaire]

Re: can bad caps cause system bus and memory errors?

now another strange phenomenom occured. the system now wont post unless i insert a pci card into one of the pci slots. if the pci card is removed, the system wont post. to me, it looks like the filtering/smoothing caps on the pci card filters out the ripple from the board thus allowing the board to post since it shares the same line. without the pci card, i think the ripple goes out of spec and thus the board wont post.

looks like not just a fault with the NB and RAM caps but with the agp/pci caps as well. well, damn looks like i have a bang up job of replacing all of the 20 filtering caps on the board which is quite a lot. thats gonna take me a few days of work considering my limited free time.

[stj]

Re: can bad caps cause system bus and memory errors?

solder a few caps across the bottom of the board for a test.

[Th3_uN1Qu3]

Re: can bad caps cause system bus and memory errors?

You may also have some BGA issues. The VIA northbridge is large and it sometimes develops bad solder joints due to board warpage.

[ChaosLegionnaire]

Re: can bad caps cause system bus and memory errors?

aaaah~ hope not. cuz i dont think i can reflow the northbridge. its just not an option atm.

now that u mentioned it, i had some problems while gaming whereby the video card would freeze and lockup and the driver managed to recover the display but in vga mode and it said that the video card stopped responding to driver commands and recommended to reboot to restart the graphics card properly. i know its not a problem with the card or psu as it worked in another system flawlessly. might be a bad or unstable HT bus.

really praying and hoping that is caused by bad caps and not the NB. do via boards and chipsets frequently suffer from bga balling problems?

[Th3_uN1Qu3]

Re: can bad caps cause system bus and memory errors?

Your board is old and has a very large BGA on it. Also VIA was always a player in the low-end market, so your board was made to lesser standards than one with Intel chipsets for instance. So it may just be the time to upgrade it, if it keeps giving you trouble.

VIA chipsets did not suffer from BGA soldering issues very often in desktop boards, but they were a disaster for mobile. This is also due to manufacturers' general incompetence, but it's something to keep in mind.

[ChaosLegionnaire]

Re: can bad caps cause system bus and memory errors?

well, i want to keep that board for my personal museum of "some of the fastest working single core athlon 64 gaming rigs".

been doing some poking around and i opened up and found that the psu (fsp amacrox speedpower 500W) i primarily used with that a8v had a teapo primary with capxons and ost rlps on the output side.

surfing around the web, i found a forum where many people from sri lanka who had that psu model complained of the psu failing and taking out their mobo and video card. some also complained that psu produced too much ripple and made their system unstable.

think i will chuck that fsp unit aside and use it strictly for short testing only. im hoping it seems to look more like a bad cap problem than a bga issue.

as an afterword, i have already changed psus before posting this topic but the problems plaguing the board still persist.

[ajwrighter@gmail.com]

Re: can bad caps cause system bus and memory errors?

your problem sounds similar to a problem my bro had on an old rig of his. He had a bad cap at the video card slot and one of his memory sticks were bad too. No matter how many times he did the mem test... if it ever completed (without blue screen) it would always pass. I'd test each stick of ram one by one (preferably in a different system.)