Kevan Hashemi
Creator
member is offline
  
Posts: 12
|
 |
AC Power Glitch Problems
« Thread started on: Nov 2nd, 2005, 5:25pm » |
|
LWDAQ devices power up in the sleep state. To increase the radiation tolerance of devices designed for high energy physics experiments, we used a capacitor-resistor network to generate the device's reset signal. If the power supply to such devices fails for a fraction of a second, this network does not work, and the device may wake up and shine a light or a laser. A momentary failure of AC power to a LWDAQ driver can cause such a momentary failure of the device power supplies. Some devices get hot and age quickly if they are left on for days. In particular, enclosed Proximity Masks (A2045) get so hot that some of them fail after twenty-four hours of continuous operation.
http://alignment.hep.brandeis.edu/Electronics/A2045/M2045.html
We find that plugging our LWDAQ driver into an uninterruptable power supply )UPS) stops these reset errors from occuring. Another way of preventing such problems is to have the LWDAQ driver turn off its device power supplies after an AC power supply failure. The driver detects such a failure by means of its own 5-V supply monitor. If the AC power fails, and the 5-V supply drops below 4.75 V, the driver receives a hardware reset. In LWDAQ Driver with VME Interface (A2037A) firmware versions 1 through 9, the device supplies turn on during reset. But firmware versions 10 and up will turn off the device supplies when the driver receives a software, hardware, or power-up reset.
Because the device supplies turn off during reset, we recommend that A2037A users include Diagnostic instrument acquisitions in their Acquisifier scripts, and set the Diagnostic daq_actions to include an "on" command. This way, at the beginning of each Acquisifier script, you are sure to turn on the device supplies. At the end of the same script, you could turn them off again, which would make it even more certain that no power supply glitch could turn on a device.
|
Author
Logged