Experimental Physics and
Industrial Control System

[email protected] (Allan Honey) · Mon, 24 Apr 2000 06:07:20 -1000 (HST)

Hi Stephen

My opinion on the fail safe is that it should be done in hardware,
such that the default state of the XVME-240 is the safe state. At
Keck our electronic interlocks were designed that way, albeit we
used the XVME-220. There is no guarantee that an IOC will successfully
reboot and when it doesn't it could require human intervention, which
may not be as prompt as the fail safes require.

At Keck we even have a watch dog which shuts down the telescope
drive. This has occurred numerous times, mainly from unexpected 
resets (power outages, faults in reset ciruits) and occassional 
cpu wedges, some of which were caused by improperly handled interrupts
from XVME-540 and bugs in new software.

regards,
     Allan

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System