Advice on Skylake/Kaby Lake 1151 socket - some pins look like they've got hot

R

Ron-ski

n00b
Joined
Apr 10, 2021
Messages
9
I've taken a chance on a motherboard that was sold for parts /not working, and the reason given was they didn't have a CPU to test it with, its a Skylake/Kaby Lake 1151 socket.

Looking at the CPU socket there is a group of pins which look like they've got hot, the actual heads of the pins look fine, and none are bent or out of the place, the majority of the discoloured pins are VCCGT which I think are for onboard graphics, and the lower two DDR1 pin's, also some that are just coloured black - I don't know if they are not used or just not labelled. I found the attached pin out in another thread on here, and I've outlined the affected pins although it's very difficult to see, even with 16.5 x magnification.

Before I go putting a CPU in it, could anybody offer any opinions, is this normal, is it an issue, or just case of trying and seeing what happens?


2021-04-10 13.45.25.jpg






2021-04-10 14.01.58.jpg




skylake_kaby_pin_diagram_damaged_pins_outlined.png
 
Thanks for the comments, if you think that one's bad you should see the other one - advertised as a few pins bent, turned out it had been left outside for some time - a suitable refund was obtained.

Anyone have any thoughts on the cause of the over heated pins, or whether it's likely to work - board was most likely in a mining rig, so possibly on board graphics used for display purposes.

I thought I'd be able to pick up a really cheap CPU for testing, but seems those are in demand by miners.
 
So you've already bought all of the parts? No way to know but to try it, worst that can happen is a giant fire and living in a hotel the rest of your life covered in bandages.
 
LOL, that did make me laugh. Yes I have purchased it, I knew it possibly had an issue, so took a chance.

I was kind hoping someone might have seen similar before, and could have offered some insight in to whether it was likely to be an issue or what had caused it etc.

But ultimately the only was is to try it, I doubt (hope) very much it will end in a fireball though ;)
 
Well it looks like it got hot to me and that's probably why it was available, but who knows maybe the CPU fan died and it sat there at 90C for a month before dying?

Fire it up and let us know what happens.
 
it looks like the part of the pins that actually contact the lands on the CPU are still shiny, which is promising. I've seen this scenario a bunch of times and mostly it seems to turn out ok. Just maybe test it the first time with the cheapest CPU you can get ;)
 
Is anyone else guessing this might be someone who tried to push overclocking too far?
 
wandplus said:
Is anyone else guessing this might be someone who tried to push overclocking too far?
Click to expand...
it's possible, but when I've seen this issue posted in the past it was a lot of server/workstation boards (like socket F Opterons and socket 1366 Xeons) that were running under load for long periods of time. If it were damage from an acute overcurrent incident, I'd expect charring & pitting on the tops of the pins where they contact the CPU pads and possibly warping of the pins, but instead I see clean contact points and surface oxidization of the pins which suggests to me a more chronic condition.

I really think it's ok to use (but I'm not buying OP a new CPU if I'm wrong sorry lol)
 
Last edited:
It looks like it was run hot for a long time which could mean some Mobo ICs have suffered electromigration.
This means the chipset could run hot.
Its not necessarily a bad thing if you can keep them cool enough, but certainly something to keep an eye out for.
I would replace the paste under the chipset heatsinks and blow a fan over them. See if they get more than warm.
If they stay cool, its probable there is no issue.

For certain the paste under the heatsinks should be replaced because a hot mobo will dry the paste out.
(assuming it works of course ;))

ps
dont mess with the pins at all, they can bend so easily.
I had a brand new mobo that suffered bent pins after fitting the CPU.
I did nothing wrong yet it happened.
I reckon a faulty socket but you try telling Asus that!
 
Thank you all for your replies, I'm still waiting for my CPU to arrive, its not an expensive CPU, but cost £50 which is about $70 but I was hoping to find a cheaper one, but cheap ones are rather scarce at the moment.

The board is an Asrock H110 Pro BTC+ so unlikely to have been overclocked but highly likely to be running continuously.

I have no intention of trying to clean the pins as the pads look good, if it works, I'll apply new paste under the one and only heat sink, and of course I'll update you all with how I get on.
 
Ron-ski said:
Thank you all for your replies, I'm still waiting for my CPU to arrive, its not an expensive CPU, but cost £50 which is about $70 but I was hoping to find a cheaper one, but cheap ones are rather scarce at the moment.

The board is an Asrock H110 Pro BTC+ so unlikely to have been overclocked but highly likely to be running continuously.

I have no intention of trying to clean the pins as the pads look good, if it works, I'll apply new paste under the one and only heat sink, and of course I'll update you all with how I get on.
Click to expand...
If you want to clean the pins use a small paint brush with some isopropyl alcohol. I don't see any compound in there so I doubt it makes a difference. There are a few very slightly bent pins in there but nothing that would cause issues from what I can tell. Let us know what happens.
 
And the end result is nothing, no explosions, no flames, nothing, just a very faint clicking from the power supply. CPU fan doesn't spin, and no LED's, not even on the RJ45 socket, as they say it's dead Jim. Some you win, some you lose.

Googling would seem to suggest this is a common problem with these boards, but no mention of a fix, someone suggested that a clicking from the PSU indicates a short to earth on the motherboard, but nothing I can see, there's also no sign of failed capacitors, they all visually look fine.

Connecting the components all up to another board, and everything works fine, so at least the CPU is OK.

I'll bung it back on eBay.
 
Ron-ski said:
And the end result is nothing, no explosions, no flames, nothing, just a very faint clicking from the power supply. CPU fan doesn't spin, and no LED's, not even on the RJ45 socket, as they say it's dead Jim. Some you win, some you lose.

Googling would seem to suggest this is a common problem with these boards, but no mention of a fix, someone suggested that a clicking from the PSU indicates a short to earth on the motherboard, but nothing I can see, there's also no sign of failed capacitors, they all visually look fine.

Connecting the components all up to another board, and everything works fine, so at least the CPU is OK.

I'll bung it back on eBay.
Click to expand...
Yeah too bad. Could be a faulty/dead VRM. You could check with a multimeter but why bother. Just make sure to check for bent pins when buying from ebay. That can cause shorts and fireworks, or simply ruin your CPU.
 
If I knew what I was doing it would be worth testing further, I have a DMM, and soldering station (but most components would be far too small for my abilities), but not the best eyesight, and not the steadiest of hands - I'm getting on a bit now.
 
Last edited:
Furious_Styles said:
Yeah too bad. Could be a faulty/dead VRM. You could check with a multimeter but why bother.
Click to expand...
You piqued my interest, I'm going to check some basics this morning - found this excellent set of YouTube video's.



So far I haven't found any schematics for the Asrock H110 Pro BTC+
 
After some investigations today, I don't appear to have any dead shorts, but what I have found with the PSU rocker switch on, pin 9 on the ATX 24 pin connector fluctuates around 1v and every so often goes to roughly 3.3v which seems to coincide with the faint click from the PSU. Pin 9 powers the circuitry that allows the power button side of things to work, and should be 5v, which it is when the ATX connector is unplugged, and the PSU works fine on another motherboard. I also found that the CMOS battery was completely flat, well about 0.7v, so replaced that and unfortunately no difference.

Still haven't managed to find schematics or a board view file for this motherboard :-(
 
Ron-ski said:
After some investigations today, I don't appear to have any dead shorts, but what I have found with the PSU rocker switch on, pin 9 on the ATX 24 pin connector fluctuates around 1v and every so often goes to roughly 3.3v which seems to coincide with the faint click from the PSU. Pin 9 powers the circuitry that allows the power button side of things to work, and should be 5v, which it is when the ATX connector is unplugged, and the PSU works fine on another motherboard. I also found that the CMOS battery was completely flat, well about 0.7v, so replaced that and unfortunately no difference.

Still haven't managed to find schematics or a board view file for this motherboard :-(
Click to expand...
I don't think they're out there. I've looked for other boards and found nothing.
 
Furious_Styles said:
I don't think they're out there. I've looked for other boards and found nothing.
Click to expand...
I've found some board view files for other H110 Asrock boards, just not this one.
Nenu said:
Reset the CMOS now the battery has been changed.
Click to expand...

I tried that quickly earlier, but I pull the battery, disconnect the PSU and put the jumper on and leave it like that until tomorrow night.
 
You must log in or register to reply here.
Back
Top