11-23-2011 02:37 PM
I think by reading the RTC every 2 minutes you are in effect causing more drift since the act of reading could be causing the "random" offset we are seeing. Can you change from reading every 2 minutes to reading every hour and see if you get linear results? If every 2 minutes you get a 2 seconds drift, then every 60 minutes you should get 60 seconds drift.
05-25-2012 02:40 PM - edited 05-25-2012 02:45 PM
I also have a strange problem with the new cRIO-9076 controller. I can normally set date and time (in labview, in MAX or with the "online" method in my web browser). The clock stays and counts normal as long as I don't reset the controller or unplug the power supply. Once it gets back online, the program starts and runs normally, but the date and time are reset (00:00:00 and i think the date starts at 1.1.1907). At first I thought the internal battery had died, but today I tried another cRIO-9076 and got the same result.
I'm using LabView 2011 and NI-RIO 4.1 firmware in my cRIO.
Previously we used a lot of cRIO-9073 and everything worked fine (I used LV2009 back than).
06-06-2012 04:35 AM
I'm afraid that the cRIO-9076 controller does not have any battery backuped RealTime Clock chip. It uses instead the processor to maintain the system clock. This means there is no way to retain the date and time when the power is shut off with that controller. You don't have that problem with the cRIO-9073, which has a battery backuped RTC chip.
What you can do is synchronising your device with SNTP server at startup. Here is how to do that: http://digital.ni.com/public.nsf/allkb/F2B057C72B5
Does that answer to your problem?
06-06-2012 05:10 AM
yes I did that about a week ago using the 3G mobile internet and a nearby NTP server (the default NI NTP didn't work). The cRIO works without a problem since than. Thanks