BSOD STOP 0x7E (0xC5)

TannerS

Distinguished
Sep 23, 2010
28
0
18,530
I have this older computer I been working on, It has been BSOD alot and I can't seem to fix it (I have pic of BSOD if needed)

this is what happens

BSOD with same message each time
Sometimes (rarely) while computer is on
ALMOST always at first start up of each day
Happens before OS loads so NO log in even viewer or memory dump even tho it saying dumping memory
Other then what is listed above, it runs fine after you restart from that BSOD and it does not do it twice in a row during a restart, in other words, turn on BSOD, restart works fine, until next day or few hours

What I have done so far
System recovery - no change
updated drivers by next version in order - no change
Did hard drive diagnostics - fine
Tested all system components from built in software and other software - fine
Memtest+ for 2 days - ram is fine
tried updating drivers again with other drivers not from HP website - no change
Tried running it with only 1 ram stick at a time - both sticks alone, no change
Cleared cmos - no change
Tried sfc /scannow - no luck
chkdsk - no ran yet
We have a wireless wifi usb adapter which has been updated buty not tested without it (the owner needs wifi to do daily activities so taking it out and waiting to see what happens (sometimes bsod come in a few days ) is a problem :p
I cannot remember anything else


So what are the chances the motherboard is dying? or what other test can I do, as someone who wants to be a pc tech any more info on how to diagnostics this pc will be great for developing new skills :)

Well anyways, if anyone can help me, thanks and please in advance and merry Christmas :)
 
Solution
if you see a bugcheck but do not get a log or memory dump:
-check your sata cable connections and port
- enable hotswapping on your sata port if you have that option.

could be the drive delayed too long, and was reset by the controller but hotswapping was not enabled and the drive
could not reconnect. results in a bug check when something critical is needed to be swapped in.
also, results in no logs or memory dump. I debugged a system that would do this and had lots of ram and would run for several hours after it lost connections to the hard drive before it bug checked.

thermal expansion/contraction issue: provide heat to components with a heat gun or hair dryer when the computer is off and cool to isolate the faulty component.

this can be things like legs of a chip that pop their solder off from a motherboard trace, cracks in a circuit trace. often very hard to find without
using a heat gun and freon to isolate with heat/cold and then a optical stereoscope to confirm the crack.
- basically when the system gets cool, the contact can contract away from the copper pad and leave a small gap, the chip still get power and heats up in a few seconds and the problem can not be detected again until the next cold cycle.
I had this happen on a RAM module address line (it caused the physical location of the memory address to change after a few seconds of use, really screwy to isolate)

problems like this are hard to isolate are often are a result of using the new lead free solder that was introduced a few years ago. the solder is more brittle and pops off after a few years of heat/cold cycles. (manufactures have often reduced the warranty periods because of this)
 

TannerS

Distinguished
Sep 23, 2010
28
0
18,530
I heard times heat can cause the solder to not touch the parts but I am lost on how to use the heat gun to find which part is at fault, when I dp it do I do all at once or single parts and what am I looking for? So am I right in thinking it is a mobo issue? Anyone else got any other feed back?
 
normally you would just use a can of Freon to cool components down while they are running to help isolate the problem down to a component. last time I looked at one of these problems, I found that if I let the system cool for 45 mins I could reproduce the crash. took me hours of debugging until I figured out the problem. address line on a memory chip was not in full contact with the pad and the windows kernel was using the memory when it later expanded and reconnected (resulting in physically moving the memory block while it was in use, all the data structures were correct, no corruption but the block was moved into the wrong place in physical memory, when the OS went to get the memory it would go to the wrong place and bugcheck (sometimes)). I think I spent over 30 hours debugging before It occurred to me what the problem was. I even found the memory chip and looked at it under a scope and sure enough the pad and chip leg joint were not in contact when it cooled.

not a fun problem to have.

just to see if it is a thermal problem, heat your system up with a heat gun before you turn it on in the morning, turn it on and see if you still get the problem.

also, in my case I could turn the system off for a hour and the problem would come back on a fresh boot. (system cooled down)
 
if you see a bugcheck but do not get a log or memory dump:
-check your sata cable connections and port
- enable hotswapping on your sata port if you have that option.

could be the drive delayed too long, and was reset by the controller but hotswapping was not enabled and the drive
could not reconnect. results in a bug check when something critical is needed to be swapped in.
also, results in no logs or memory dump. I debugged a system that would do this and had lots of ram and would run for several hours after it lost connections to the hard drive before it bug checked.

 
Solution

TannerS

Distinguished
Sep 23, 2010
28
0
18,530

I think the sata and port are fine but there is no options in bios for hot swapping, very very limited bios, b ut the other suggestions, cool ill try them out. but for now she wants to just buy a new computer if it is a motherboard problem. thanks man