Hardware issue, across the board...

T

TheBlank

Limp Gawd
Joined
Nov 15, 2006
Messages
164
Alright, here's a bit of a mystery for me that I definitely need some input on before I make any kind of decision:

about a year ago, the integrated sound card on my motherboard (specs in signature) died randomly. Took my guitar amp head with it, but somehow left the rest of the board intact. I purchased a SoundBlaster xfi Fatal1ty Champion PCI series card to replace it, and haven't had problems since.

Until...

My DVD-ROM died, refused to play DVDs only CD's. So I replaced it with an LG SATA drive. Simple replacement, swapped the cables, used a molex adapter for the power requirement, and turned on the machine. Only to have the sound card no longer recognized by the system. The lights were on on the board, but no amount of driver installation or rollback would get it to recognize the hardware.

So I pulled the card to reseat it and found green gunk all over the leads. It came off easily with alcohol, but placing it into the same PCI slot did not solve the issue. When I pulled the card again, the same green gunk was all over the leads. I cleaned it and placed it in another open PCI slot. This time the card was recognized, but the front panel (with the all important guitar input) no longer functions. Reseating the cable does nothing.

And now, despite several driver reinstalls (multiple versions) the sound card randomly causes serious system errors that crashes my machine. I'm currently working with creative to see if I can RMA,

However, here's my confusion. How could the simple installation of a DVD player wreak this much havoc? Did the sound card cause that green gunk (which I can only assume was corrosion) or was it my mainboard? I'm wondering if I'm working with a black widow mainboard now, and if I should replace it. First my integrated sound card (and guitar amp), then my DVD player, then my secondary hard drive, then my NEW sound card... methinks something just isn't right here.

I have a dedicated power line for this machine, and I've checked both it and the UPS for voltage issues, and I'm fairly certain I can rule that out as a possible cause.

Also, while running PC wizard 2008, it reads my MHz as 1600 instead of 2667, with VERY rare flashes back up to 2667. I know the new Core 2's throttle, but I can't consistantly get it to stay at 2667, even with heavy use in Photoshop, Illustrator, or any game that I throw at it. Problem with the software, or with the mainboard/processor? Every other program says that it's running at the right clock speed.

Sorry for the book, but I'm going out of my mind here. The prospect of having to shell out for a new mainboard and CPU has me about ready to tear my hair out.

TIA.
 
TheBlank said:
a) about a year ago, the integrated sound card on my motherboard (specs in signature) died
b) I purchased a SoundBlaster xfi Fatal1ty Champion PCI series card to replace it, and haven't had problems since.

c) My DVD-ROM died
d) I replaced it with an LG SATA drive.
e) Now the sound card no longer recognized by the system.
f) I pulled the card to reseat it and found green gunk all over the leads.
g) I cleaned it and placed it in another open PCI slot. This time the card was recognized, but the front panel (with the all important guitar input) no longer functions.

1) How could the simple installation of a DVD player wreak this much havoc?
2) Did the sound card cause that green gunk (which I can only assume was corrosion) or was it my mainboard?
3) I'm wondering if I'm working with a black widow mainboard now, and if I should replace it. First my integrated sound card (and guitar amp), then my DVD player, then my secondary hard drive, then my NEW sound card... methinks something just isn't right here.
4) Also, while running PC wizard 2008, it reads my MHz as 1600 instead of 2667, with VERY rare flashes back up to 2667. ... Problem with the software, or with the mainboard/processor?
Click to expand...

1) You could have bumped something during installation, or something got loose, and is causing problems with a short. Another scenario is your PSU is dying and has been taking your other parts with it.

2) I think its the slot on the mobo.

3) It sounds like your board and/or PSU is dying, lol. However, the question is, if its the mobo, whats killing it? I'm guessing the PSU. I'd replace the PSU and mobo, as well as the soundcard (RMA them all, IMO).

4) This is called EIST... Enhanced Intel Speedstep Technology. It basically lowers the voltage and FSB of your system when such CPU power isn't needed. Once your system needs the CPU power, it fires back up to max speed. There really is no need to change this setting unless you're looking for a stable overclock setting -- after which, its good to re-enable EIST. EIST saves power and does no harm to your CPU... in fact, with less power going into the chip, there's less heat; less heat means a longer CPU life... theoretically.
 
Heh, I didn't even realize he was running WC (I read the sig, but was looking for the PSU, lol)... explains a lot, and I think thats causing his problems... if there is a leak.
 
Thanks for the responses so far.

I really hope it's not that PSU as well, that would just ruin my day just that little bit more, lol.

As for the leak, I don't have any openings in the tubing near the PCI slots, so I'm hesitant to believe that is the case. I'm using Fluid XP+ as well, and with the one leak that I had way early on, it left a white crusty residue everywhere that it dried. I don't see any residue anywhere on the board or in the case, and after a paper towel test all over my WC rig, I've not found anything leaking :-\

And as for the EIST, I knew about that already, I just figured that while it was running higher level processes it wouldn't be kicking in. Even when I'm load testing now it's staying at 1600 mhz. Maybe I need to disable this feature and hope that it doesn't continue. I'd really rather have all of the power I paid for...

Checked and rechecked all of the connections, I don't see anything loose. I reconnected everything just to be sure, and I scanned around looking to see if there were any exposed connectors or metal touching anything. Nothing that I could find right off the bat. Nothing near where I was working on getting the DVD player installed anyway. And nothing suspicious in the area of the PCI slots or the sound card either.

So, the list right now stands at PSU, MB, and possibly CPU? That CPU has been causing me problems since day one. Had to create a whole new WC setup for it because the IHS is not flat, and it doesn't cool as well as I would have hoped. A single rad setup before saw idle temps of 60 and load temps of 100+. Yeah yeah... lap it up, I know. But if it's not working anyway, why go through the trouble? Of course, I guess the speed issue could be completely board related too...

My hair is starting to get a little sore...
 
The green gunk came from somewhere.
As it is exactly where you would expect oxidised copper to be if copper gets damp, that is the primary suspect.
You have had a leak or are having leaks onto live connections with some liquid that can assist in copper corrosion.
 
TheBlank said:
...
And as for the EIST, I knew about that already, I just figured that while it was running higher level processes it wouldn't be kicking in. Even when I'm load testing now it's staying at 1600 mhz. Maybe I need to disable this feature and hope that it doesn't continue. I'd really rather have all of the power I paid for...
...
Click to expand...

How often does that program poll the CPU speed?

If you're stress testing all cores with Prime95, hitting windows pause should bring up the my computer properties, which will show you the current clockspeed.
 
Still not 100% sure on that leak. The angles just don't seem possible, and the fact that I'm using a non corrosive, non conductive fluid suggests otherwise as well. I know that as soon as the fluid enters a non clean environment that potential for free radicals and ions increases and can cause the type of issue that I'm seeing in certain cases. However, wouldn't I be able to see the white residue from a leak near the affected components? Could humidity be causing this by itself? My room gets really hot due to all of the equipment running in it (at least 10 - 15 degrees hotter than the rest of the house).

I rebooted for the umpteenth time and started my load testing again the same way I had done so before. PC Wizard seems to test the clock speed every 3 seconds, and this time when I stressed the system it jumped it to the 2667.5ish that it should be at. I don't know why it didn't do it the last few restarts. I flashed the BIOS as well to see if there was a BIOS corruption or update that might solve the issue, I'm hoping that it holds.

I'd had an issue with an older UPS a few months back, and I'd figured out that that was causing all of the system troubles in my last 2 machines. Bad voltages. The current machine was running on that UPS for a few months before I discovered the problem and corrected it. Could this be an example of residual damage finally beginning to manifest? Possibly systemic damage to the PSU that's taken time to degenerate and fail?

And if that's the case, could I be looking at a possible failure of all of my components due to the same issue, or would it be limited to the PSU, mobo, and any other component that is already experiencing trouble?
 
Take some of that liquid, put it in a cup, put a shiny penny in the cup and leave it and see what happens.
 
The corrosion you found could be from a leak a long time ago, not necessarily recent but its hard to know what damage it could have caused.
It could also be a result of excess long term humidity.
The hotter the PC runs, the less likely that water vapour will be deposited on anything, but the vapour that does come in contact with copper will decay faster.
If it is water vapour, you should find other electronics in the same environment for a long time have suffered in a similar manner.

It sounds like there are too many unknowns to diagnose on paper what is currently wrong though.
I would start by stripping everything down, clean them up as best as possible and dry test (no WC) on an already functional testbed.
 
dandragonrage said:
Take some of that liquid, put it in a cup, put a shiny penny in the cup and leave it and see what happens.
Click to expand...

excellent idea, just started the experiment. I hope it doesn't result in anything, seeing as my water blocks are also copper...
 
Also, it is possible for the additives to evaporate or separate or dry up separately from the water in open air.
 
Oh I know, but that's part of the confusion that I'm having over this possibility because I know that when this stuff evaporates, it leaves a white residue where it dried. I'm seeing none of that anywhere in the system. There is a little on the case near the pump where the first leak happened. I'm guessing I'm going to have to take Nenu's advice and break the whole system down and clean it up and go from there. Maybe then I'll see something that suggests more to me that a leak of my WC setup occurred.

Until then, I'm actually more inclined to believe the humidity theory... even though all of the other electronics in my room, active or otherwise, and near the case, have not suffered similar corrosion.

Really tempted to just scrap the board, PSU, and upgrade the CPU as well in lieu of all the time that I'm going to lose (I use this machine for digital painting for work) breaking it down and trying to find the rotten egg. But then the tinkering side of me is really looking forward to the challenge.

Techno masochism... hah!
 
You must log in or register to reply here.
Back
Top