Lenovo IdeaPad 3 14ADA05 NVME keeps randomly disconnecting

[pcwhiz24]

This issue has me stumped! I have this Lenovo IdeaPad and the SSD (NVME) keeps on randomly disconnecting. It'll stop being detected in BIOS randomly and when I turn it off and back on sometimes it'll detect again and boot up just fine...only to crash after a few minutes and either boot again or stop being detected.

Things I tried so far:

- Reset the CMOS by removing battery, held power buttons 60 seconds, all that jazz.
- BIOS is fully updated (had to use CHA341 because I bricked it the first time trying to update it)
- Tried multiple working NVME drives and even SATA ones (the ones I used work fine on other laptops/PCs)
- Shoving the NVME drive tight into the slot actually got it to boot for a few minutes but still randomly disconnects

This laptop only has one NVME slot and no other means to boot and I really don't wanna have to spend $100+ replacing a perfectly good motherboard over such a small (yet very important) issue. I spent hours of googling and nobody else seems to have this issue but I did find this:

https://www.youtube.com/watch?v=I4XT9yei3xk

The guy here describes the very same problem I have, with the same model laptop I have (except different color) and looking at the comments it seems reflowing the NVME connector is the fix and as one commenter expressed he didn't show the process so I don't know how to reflow an NVME slot. Do I just hold a heat gun over it for a couple minutes?

[mon2]

Hi. Just posted a possible fix for the poster on YT. Believe your case is different since he is observing that once the SSD is detected, it continues to work till next power cycle. Yours will disconnect while in use so different case. Believe for his case, the suspect is the timing of the SSD reset line = SSD_RST#. This line is driven by a dual input AND gate and one of the input lines to this dual input AND gate is a simple RC circuit where the resistor and capacitor decide the timing to RESET or NOT RESET the SSD upon power up. This concept imho is ghetto. We build hardware for the PC industry had made a similar error on a small batch of boards. The same fault may be here - that is the reset line is kept in reset state too long and the SSD is then NOT enumerated in time upon power up. Sometimes the timing is ok / sometimes not due to the inaccurate RC circuit operation. For his case, suggesting to change the RC capacitor value to a lower value to speed up on the SSD_RST# release time. If timed properly, the SSD_RST# will be a logic '1' and when the BIOS goes to scan the PCIe bus, the SSD will be detected.

For your case, wondering if the issue is with the tension of the contacts / contacts themselves on the PCB. Has the SSD socket seen a lot of action? If not, could be ROHS solder connections giving up contact after xx years in the field. We are servicing a truck ECU with the same fault where the ECU's microcontroller needs to be removed to be reballed and then soldered back again to fix this $$ ECU unit.

Suggestion to the YT OP: Carefully flux and remove the 0201 sized cap @ CC1389 and swap with a lower value like 47pf. This reduced capacitance should release the reset line faster so that the SSD is in a run state sooner and then the mother board bios will be able to enumerate the SSD.

[mcplslg123]

Instead of heatfun,use soldering iron and goos quality flux paste to touch up all the pins od SSd connector.

[tomodachi]

Sad to say but, another Lenovo low temperature soldering (LTS) related issue..
See this article:
https://news.lenovo.com/pressroom/pr...uring-process/

Most probably it caused by cold solder joints at NVME port or loose bga connections under CPU.