I don't know if this is Ride Runner specific but as that is what I'm running I'll ask anyway.
My setup is a Intel D945GCLF2 system with an M1-ATX automotive PSU. I have configured WinXP to hibernate when it recieves a power button event.
The carpc interfaces with a BMW factory nav setup using a CIS-IBUS box.
Now to the problem. If I hook up the carpc to a monitor, keyboard and lab power supply hibernation works like clockwork - I yank the Acc power and after five seconds hibernation is initiated and the system shuts down in a about 30 seconds.
Connected to the car however, it's a different story. When I turn off the Acc power, the power LED starts blinking just as it should but keeps on until it hits the 1 minute limit where the PSU cuts the power itself. I have no means of seeing if any useful information comes up as I do not have a regular VGA monitor available in the car.
I have verified that permanent 12v and Acc are connected correctly. The only difference between the car setup and lab is that in the car the serial port is also connected (to the iBus inteface).
I tried and reconfigured the PSU to allow for longer timeouts with the only result that it will take longer before power is forcibly cut but hibernate it will not.
Any suggestions? I will try to find a way to connect a VGA monitor in the car but that is a bit difficult for me at the time.
I suppose windows is configures to hibernate on power button. So have you tried bringing the monitor to the car (so that you can see what you see in the lab?)? I did long night hours with a temp 17" in the back seat resolving those problems that only happens after everything is plugged in the car.
the connected hardware may prevent it from hibernating..try with it unplugged (like outside the carpc) to see if it works.
It feels as this is exactly what's happening. The difference in connections between the car or outside the car are:
Originally Posted by guino
* COM-port (on-board, not USB) connected to iBus.
* 1 5V cable drawing power from one of the USB ports.
* VGA connected to the OEM nav instead of a regular monitor.
Furthermore, the lab setup feeds the M1-ATX with 19V whereas the car only has regular 12V.
I will try and do as you suggest and get back with the results.
Hooked up a monitor and found the culprit. It's not hibernation (and definitely not RR) problems, it's PSU problems. Very odd too, but apparently a few have already had similar experiences with the M*-ATX PSU:s:
It turns out that whenever I cut Acc power to the M1-ATX using the ignition key the PC does a hard reset, starts to reboot and finally gets killed by the M1-ATX hard power-off timer. The odd thing is that it only happens if I use the key. If I leave the key in Acc position and instead manually disconnect the Acc wire to the PSU the PC hibernates just fine.
Very irritating... :(