Curing my home desktop from locking up in the cold (so far)

May 9, 2023

I've had a long running mystery where my home desktop would lock up if it got cold, where as time developed it seemed that getting too cold was down not much below 68 F (which is hardly cold, at least for Canadians). Back in December of last year I had the idea of 'replacing' the case front panel with just a stand-alone ATX chassis power switch. I hesitated for a bit, but the situation was getting more irritating this past winter and stand-alone ATX chassis power switches are not expensive items. Finally at one point I carried through with this plan (possibly when I had to open up the case after forgetting a critical step in software disk shuffling). Somewhat to my surprise, this relatively simple change seems to have fixed all of the problems.

When I was vaguely planning the change, I had expected to disconnect all of the various cables from the front panel. When it actually came time to do it, I only pulled the primary motherboard front panel connectors on my home desktop's motherboard, which I believe has left USB and audio connected, and I reconnected only the chassis power switch. These days I suspect that this is more than I needed to disconnect and both the power LED and 'HDD' LED are probably safe, but at the time I didn't feel like taking several iterations of testing.

One of the lessons learned for me is that very odd PC symptoms can (apparently) be caused by quite small underlying problems. It seems pretty likely that the underlying cause of my desktop's problems was the power button and possibly the reset button shorting to a 'pressed' state. In retrospect it's easy to see how this could turn my desktop off, but coming back on when things warm up (and presumably the buttons stop being 'pressed') is less obvious. I can imagine a lot of possibilities, including the motherboard having some sort of short detection where if things are held 'pressed' for too long, it concludes something is wrong and refuses to power on until the situation clears.

(I believe the ATX power on behavior is implemented by special motherboard circuitry, instead of being handled in BIOS by the CPU. Possibly all of the real behavior is implemented in hardware, with the BIOS only receiving a 'the user pressed the power briefly' signal when the system is running.)

In the future, I hopefully will remember that other mysterious hardware problems might be dealt with in equally simple ways. Cables can go bad, for example (and I've certainly heard stories of disk problems that were cured with new SATA cables).

Written on 09 May 2023.
« When to use drgn instead of eBPF tools like bpftrace, and vice versa
Mailing lists versus forums, some thoughts »

Page tools: View Source, Add Comment.
Search:
Login: Password:
Atom Syndication: Recent Comments.

Last modified: Tue May 9 22:53:33 2023
This dinky wiki is brought to you by the Insane Hackers Guild, Python sub-branch.