x170sm-g Grilled pins in a socket on X170SM

[FTW_260]

Hi all.
Yesterday, when starting BF2042, the laptop froze and began to make strange sounds from the speakers.
I turned it off with the button, but it didn't turn back on. (there was no picture on the display)
After several iterations of attempts to diagnose something, and a couple of hours I looked under the CPU and saw the following there:

Some of the pins in the socket are clearly fried, and there are barely noticeable burnout points on the CPU in this place. The photo was taken with phone on a tripod with x7 zoom.

To understand if the MB works, I ordered the cheapest Celeron G5900 and a motherboard based on the H410 chipset for CPU test.
I'm waiting for delivery in a few days, but I have the worst feeling. (
I checked the video card, it works fine in the old P775TM.

[ViktorV]

6 minutes ago, FTW_260 said:

Hi all.
Yesterday, when starting BF2042, the laptop froze and began to make strange sounds from the speakers.
I turned it off with the button, but it didn't turn back on. (there was no picture on the display)
After several iterations of attempts to diagnose something, and a couple of hours I looked under the CPU and saw the following there:

Some of the pins in the socket are clearly fried, and there are barely noticeable burnout points on the CPU in this place. The photo was taken with phone on a tripod with x7 zoom.

To understand if the MB works, I ordered the cheapest Celeron G5900 and a motherboard based on the H410 chipset for CPU test.
I'm waiting for delivery in a few days, but I have the worst feeling. (
I checked the video card, it works fine in the old P775TM.

Post a photo so that you can see the entire socket and the problematic leg.

[FTW_260]

@ViktorV here, now you can see everything:

[ViktorV]

2 hours ago, FTW_260 said:

@ViktorV here, now you can see everything:

Was the laptop connected to an external display? Or have you done anything with the laptop display lately?

[FTW_260]

@ViktorV Unfortunately no.  Normal use on a desk without an external display.  There was also nothing related to the display. I also looked that these pins are related to the display, and I can not find a relationship with anything.(

[ViktorV]

It could be a short circuit in the display cable.

[electrosoft]

19 hours ago, FTW_260 said:

Hi all.
Yesterday, when starting BF2042, the laptop froze and began to make strange sounds from the speakers.
I turned it off with the button, but it didn't turn back on. (there was no picture on the display)
After several iterations of attempts to diagnose something, and a couple of hours I looked under the CPU and saw the following there:

Some of the pins in the socket are clearly fried, and there are barely noticeable burnout points on the CPU in this place. The photo was taken with phone on a tripod with x7 zoom.

To understand if the MB works, I ordered the cheapest Celeron G5900 and a motherboard based on the H410 chipset for CPU test.
I'm waiting for delivery in a few days, but I have the worst feeling. (
I checked the video card, it works fine in the old P775TM.

 

Were you running any type of OC?

Newest BF2042 patch hits the CPU hard with much higher stress on multi-cores (much better performance too it seems).

Hopefully you can figure out if it is potentially the system board or CPU (or both) that got slammed.

 

[FTW_260]

21 hours ago, ViktorV said:

It could be a short circuit in the display cable.

I also thought so, but it seems to me that if this were the case, then the GPU would die.

7 hours ago, electrosoft said:

Were you running any type of OC?

Rather yes. I fixed the multiplier on my 10700 at x46 with TS + 50mV UV. The video card is also slightly tuned to 1990@0.975V. In this mode, the laptop worked fine for two months.

 

Today I took a new celeron.

Before installing it, I tried to pry up the bent legs with a needle and realized that most likely the test with a new CPU would not help.(
The legs have lost their flexibility, and after lifting them with a needle, after closing the CPU in the socket, they again return to the wrong position.
I tried installing a new CPU and doing a test run. The symptoms have not changed, there is no picture.

 

In a couple of days, the MB with H410 chipset should arrive to check the processor, I'll write here according to the results.

In the meantime, I will try to find a trusted service center that will undertake to re-solder the socket on the motherboard.

[ViktorV]

13 minutes ago, FTW_260 said:

I also thought so, but it seems to me that if this were the case, then the GPU would die.

Rather yes. I fixed the multiplier on my 10700 at x46 with TS + 50mV UV. The video card is also slightly tuned to 1990@0.975V. In this mode, the laptop worked fine for two months.

 

Today I took a new celeron.

Before installing it, I tried to pry up the bent legs with a needle and realized that most likely the test with a new CPU would not help.(
The legs have lost their flexibility, and after lifting them with a needle, after closing the CPU in the socket, they again return to the wrong position.
I tried installing a new CPU and doing a test run. The symptoms have not changed, there is no picture.

 

In a couple of days, the MB with H410 chipset should arrive to check the processor, I'll write here according to the results.

In the meantime, I will try to find a trusted service center that will undertake to re-solder the socket on the motherboard.

No need to change the entire socket, it will be enough to solder new legs instead of damaged ones.

[electrosoft]

1 hour ago, FTW_260 said:

I also thought so, but it seems to me that if this were the case, then the GPU would die.

Rather yes. I fixed the multiplier on my 10700 at x46 with TS + 50mV UV. The video card is also slightly tuned to 1990@0.975V. In this mode, the laptop worked fine for two months.

 

Today I took a new celeron.

Before installing it, I tried to pry up the bent legs with a needle and realized that most likely the test with a new CPU would not help.(
The legs have lost their flexibility, and after lifting them with a needle, after closing the CPU in the socket, they again return to the wrong position.
I tried installing a new CPU and doing a test run. The symptoms have not changed, there is no picture.

 

In a couple of days, the MB with H410 chipset should arrive to check the processor, I'll write here according to the results.

In the meantime, I will try to find a trusted service center that will undertake to re-solder the socket on the motherboard.

 

Are Celeron CPUs confirmed working in the X170SM-G? I assume they are but you never know.

Either way, the H410 motherboard will quickly provide clarity.

 

1 hour ago, ViktorV said:

No need to change the entire socket, it will be enough to solder new legs instead of damaged ones.

 

Either this or a small bit of solder on the pads to assure contact for a quick test.

 

Good luck with the H410 tests!

 

 

[FTW_260]

While testing one of @ViktorV theories from PM, I took the laptop off the shelf and tried turning it on again.
And he turned on! with celeron inside) 

There is a picture on display, everything works properly. 
I won’t try on 10700 yet, first I need to deal with the socket.
The motherboard is alive, how glad I am.

48 minutes ago, electrosoft said:

Are Celeron CPUs confirmed working in the X170SM-G? I assume they are but you never know.

To be honest, I never doubted it.
Even on the old P775 there were no problems with starting Pentiums or Celerons.

49 minutes ago, electrosoft said:

Good luck with the H410 tests!

Thank you. Of course, I planned to change the CPU, but not in this way) 

[cylix]

What  was the problem ?  How did you got it working?

[electrosoft]

59 minutes ago, FTW_260 said:

While testing one of @ViktorV theories from PM, I took the laptop off the shelf and tried turning it on again.
And he turned on! with celeron inside) 

There is a picture on display, everything works properly. 
I won’t try on 10700 yet, first I need to deal with the socket.
The motherboard is alive, how glad I am.

To be honest, I never doubted it.
Even on the old P775 there were no problems with starting Pentiums or Celerons.

Thank you. Of course, I planned to change the CPU, but not in this way) 

 

40 minutes ago, cylix said:

What  was the problem ?  How did you got it working?

 

lol, glad to know even bottom barrel 10th gen CPUs will work in there!

 

X170SM/KM are very sensitive to CPU retention pressure and if the retention screws are too tight it may boot but once put under load and heat expansion it can crash and come to a screeching halt. Did it just "boot" or did you reseat/adjust tension?

 

 

[FTW_260]

20 hours ago, cylix said:

What  was the problem ?  How did you got it working?

I just tried to bend the legs to their original position, and started the laptop with a known working CPU. On the third try, but it worked. Thus, the problem is the lack of sufficient clamping of the legs in the socket, which must be solved either by replacing the problem legs, or by completely replacing the socket on the MB.
This is not a complete recovery, I will finally calm down after solving the problem with the MB and testing the good old 10700 fried with legs.

19 hours ago, electrosoft said:

X170SM/KM are very sensitive to CPU retention pressure and if the retention screws are too tight it may boot but once put under load and heat expansion it can crash and come to a screeching halt. Did it just "boot" or did you reseat/adjust tension?

With this I have to agree completely.
Booted only with a light tightening of the screws on the heatsink. While I will not experiment with increasing the pressure, I will do this after repairing the board.

 

[srs2236]

1 hour ago, FTW_260 said:

I just tried to bend the legs to their original position, and started the laptop with a known working CPU. On the third try, but it worked. Thus, the problem is the lack of sufficient clamping of the legs in the socket, which must be solved either by replacing the problem legs, or by completely replacing the socket on the MB.
This is not a complete recovery, I will finally calm down after solving the problem with the MB and testing the good old 10700 fried with legs.

With this I have to agree completely.
Booted only with a light tightening of the screws on the heatsink. While I will not experiment with increasing the pressure, I will do this after repairing the board.

 

Is this something that can happen to desktop motherboards aswell? I wonder what happens when its too tight... Seems odd that it wouldn't work because of that.

[Scruffy]

for fix small amount of bent pins just use small tweezer with microscope and never touch the tip of the pins. just the middle part with a small wood for teeth and small tweezer ( i have fin tweezer in titanium, but on that one u have to apply very small pressure otherwise it will get into an x state and bend pins even more).

have replaced like 50 socket on desktop motherboard.... some just work fine some is totally dead, same techics and so on...

[FTW_260]

1 hour ago, srs2236 said:

Is this something that can happen to desktop motherboards aswell?

I think so, especially with wery large and heavy air-coolers.

 

[FTW_260]

Hi all.
I'm back with good news.
I've picked up H410 MB and searched some memory sticks to test my i7-10700 after my socket problem - and CPU is alive! Booting to BIOS successfully and saw two memory chanels active and get a picture on the display.

I found a man in my city who can only change the burnt legs in the socket separately, and this morning after completely disassembling the laptop (spent an hour on this), I went to him with only one board. (so many screws here )

We spent another hour working on replacing the legs. With the help of the bottom heating and a soldering gun, with great difficulty, the old legs were removed. According to the assurances of the man who did the work, the legs on the server boards are removed with such difficulty.

After that, new donor legs were installed, and adjacent legs were checked for their flexibility. The following picture is the final result after cleaning under a microscope:

Already at home, after the final assembly, I put everything together and was finally able to check everything properly.
Everything works as before, as if nothing had happened. 

[electrosoft]

6 hours ago, FTW_260 said:

Hi all.
I'm back with good news.
I've picked up H410 MB and searched some memory sticks to test my i7-10700 after my socket problem - and CPU is alive! Booting to BIOS successfully and saw two memory chanels active and get a picture on the display.

I found a man in my city who can only change the burnt legs in the socket separately, and this morning after completely disassembling the laptop (spent an hour on this), I went to him with only one board. (so many screws here )

We spent another hour working on replacing the legs. With the help of the bottom heating and a soldering gun, with great difficulty, the old legs were removed. According to the assurances of the man who did the work, the legs on the server boards are removed with such difficulty.

After that, new donor legs were installed, and adjacent legs were checked for their flexibility. The following picture is the final result after cleaning under a microscope:

Already at home, after the final assembly, I put everything together and was finally able to check everything properly.
Everything works as before, as if nothing had happened. 

 

That is great to read! What a fun journey too (well, for us readers).

 

What I gleaned from this is I will be picking up a dirt cheap G5 Celeron to put in mine when I ship it back to have the GPU replaced so nothing happens to my choice SL delidded 10900k in there now. They are only $15 on eBay. 🙂

 

 

[Scruffy]

that's a great news and update 🙂

really happy  that u were able to fix your board and get notebook working again.