- Mar 11, 2018
- 2
- 0
- 10
Through a series of unfortunate events (don't ask) I bent/dented a small portion of my cpu underside so that two pin contacts are raised slightly away from the normal plane of contact for the motherboard pins. Thankfully my system does still boot and operates mostly normally, but there are serious instability issues that I simply can't tolerate in the long run. Any system crash has a chance of entirely corrupting the current boot drive so that a clean install of windows is required (all attempts at system repair/restore have proven futile on the two occasions this has happened thus far), spec hungry programs such as Adobe AE have a chance of causing the entire system to hang if left open and unattended for long periods of time, and other bugs and errors crop up frequently enough in many other games and applications.
Specific things I have noted that may shed light on the degree of damage: chance to corrupt the boot drive on a system hang or crash rather than rebooting readily, on reinstalling the CPU after the damage occurred the motherboard bios didn't recognise the chip as being the same (''new cpu installed"), the motherboard vga ports (both HDMI and DP) no longer output signal on a boot (I can now only output video from my dedicated GPU) and switching to a motherboard display after boot hard crashes the system. Also confirming my diagnosis that the onboard graphics is unserviceable, since the damage in am unable to install the motherboard VGA drivers, the installer returning the message 'Your device doesn't meet the system requirements'.
There are several questions that I need answered before I can decide on how to proceed, and they are all way beyond my ability to suss out for myself. Any expert advice would be GREATLY appreciated and it may prove an interesting experiment to any of you interested if end up having to go through with my plans for physically fixing the chip (which i will get to shortly).
My first thoughts once I noticed the damage was that the two affected pin contacts would simply be out of reach of their respective pins on the motherboard, but after a bit of research I found that mobo pins are spring loaded -
Does anyone know how far the travel is for these pins on a high end mobo?
In Intel's manufacturing specification for 603 pin sockets they state that the travel for these pins can't exceed 1.52mm, and if mobo manufacturers typically utilise half as much of this allowance I may be out of luck and the pins may well still reach their contacts with the required contact force as I recon the deviation of the bent away contacts from the plane of the rest is between 0.5mm and 1mm, though I can't confirm this till the next opportunity I get to remove and examine the cpu.
If the pins are likely sufficiently in contact still due to the pin spring action then it is likely that the loss of function of these contacts is down to damage within the pcb itself, making the situation entirely irreparable, but I would still really appreciate this next question answered if anyone has any insight at all:
Basically, what sort of stresses could the internal conducting materials leading to the cpu contacts potentially handle. I have zero insight into internal composition of a cpu and I can't find any good sources readily available. I'm guessing it's some sort of trace similar to what one would see on the surface of circuit boards, and if so, could these traces withstand a few tenths of a millimeter of stretching force considering they would be completely enclosed by supporting material?
Furthermore, would those traces, having been stretched a little, compress back to their original while remaining unbroken/sufficiently conductive?
I have devised a means by which I think I can precisely undo the dent so that the two affected contacts are again almost perfectly flush with their fellows. The maximum deviation of this dent is so small to the naked eye that I want to hope it can be remedied - as I say, likely as little as 0.5mm and no more than 1mm - or am I just ****ed?
(I will post diagrams depicting the type of damage now, and actual pictures of the chip at the nearest opportunity)
Worst case scenario is I will have to replace the cpu when I can afford it as I use this pc for work and can't afford crashes during important projects and it will be the backbone of my livelihood for years to come. However, if there is any chance of repairing it I would rather take the risk and save the cached cash for something else down the line. Beyond it being a tragedy to be so clumsy as to damage any cpu, this is an 8700k and it was looking to be the most exceptional overclocker. In my preliminary testing it was running a 5GHz OC fully stable after 24hrs on a P95 torture test using a voltage well below what one would need to draw to hit 5GHz on an average 8700k. It could have been one of those rare 5.3 monsters after a delidd, and I was removing it from the motherboard with that in mind at the time of the 'accident'.
Any insight into solving my conundrum would be greatly appreciated.
Sincerely
he who is suitably mortified
Specific things I have noted that may shed light on the degree of damage: chance to corrupt the boot drive on a system hang or crash rather than rebooting readily, on reinstalling the CPU after the damage occurred the motherboard bios didn't recognise the chip as being the same (''new cpu installed"), the motherboard vga ports (both HDMI and DP) no longer output signal on a boot (I can now only output video from my dedicated GPU) and switching to a motherboard display after boot hard crashes the system. Also confirming my diagnosis that the onboard graphics is unserviceable, since the damage in am unable to install the motherboard VGA drivers, the installer returning the message 'Your device doesn't meet the system requirements'.
There are several questions that I need answered before I can decide on how to proceed, and they are all way beyond my ability to suss out for myself. Any expert advice would be GREATLY appreciated and it may prove an interesting experiment to any of you interested if end up having to go through with my plans for physically fixing the chip (which i will get to shortly).
My first thoughts once I noticed the damage was that the two affected pin contacts would simply be out of reach of their respective pins on the motherboard, but after a bit of research I found that mobo pins are spring loaded -
Does anyone know how far the travel is for these pins on a high end mobo?
In Intel's manufacturing specification for 603 pin sockets they state that the travel for these pins can't exceed 1.52mm, and if mobo manufacturers typically utilise half as much of this allowance I may be out of luck and the pins may well still reach their contacts with the required contact force as I recon the deviation of the bent away contacts from the plane of the rest is between 0.5mm and 1mm, though I can't confirm this till the next opportunity I get to remove and examine the cpu.
If the pins are likely sufficiently in contact still due to the pin spring action then it is likely that the loss of function of these contacts is down to damage within the pcb itself, making the situation entirely irreparable, but I would still really appreciate this next question answered if anyone has any insight at all:
Basically, what sort of stresses could the internal conducting materials leading to the cpu contacts potentially handle. I have zero insight into internal composition of a cpu and I can't find any good sources readily available. I'm guessing it's some sort of trace similar to what one would see on the surface of circuit boards, and if so, could these traces withstand a few tenths of a millimeter of stretching force considering they would be completely enclosed by supporting material?
Furthermore, would those traces, having been stretched a little, compress back to their original while remaining unbroken/sufficiently conductive?
I have devised a means by which I think I can precisely undo the dent so that the two affected contacts are again almost perfectly flush with their fellows. The maximum deviation of this dent is so small to the naked eye that I want to hope it can be remedied - as I say, likely as little as 0.5mm and no more than 1mm - or am I just ****ed?
(I will post diagrams depicting the type of damage now, and actual pictures of the chip at the nearest opportunity)
Worst case scenario is I will have to replace the cpu when I can afford it as I use this pc for work and can't afford crashes during important projects and it will be the backbone of my livelihood for years to come. However, if there is any chance of repairing it I would rather take the risk and save the cached cash for something else down the line. Beyond it being a tragedy to be so clumsy as to damage any cpu, this is an 8700k and it was looking to be the most exceptional overclocker. In my preliminary testing it was running a 5GHz OC fully stable after 24hrs on a P95 torture test using a voltage well below what one would need to draw to hit 5GHz on an average 8700k. It could have been one of those rare 5.3 monsters after a delidd, and I was removing it from the motherboard with that in mind at the time of the 'accident'.
Any insight into solving my conundrum would be greatly appreciated.
Sincerely
he who is suitably mortified
Facebook
Twitter
Instagram