cRIO scan engine rate

[Kilometers]

Hi Cameron,

If you look in Distributed System Manager at your cRIO on the main page for the device there should be a Scan Engine tab. At the bottom of that tab there should be a fault list. Are you getting any faults populating here? Also what happens if you increase the scan period and network publishing period to something between 25 ms and 800 ms, let's say 200 or 500 ms. Do you see any change in behavior at all? 

[Cameron.Andersen]

Miles,

There is actually an error there.  "-65608 - CompactRIO: The project contains a module in slot 3 that does not match the type of the module in slot 3 of the chassis."

This is strange because I have double checked my project and hardware and confirmed that they agree.  It is a 9208 in the project and in the actual cRIO.  The module also reads values as expected in the DSM and the VI (other than with a slow update rate).  

All of my benchmarking thus far has been in VIs running and indicating on the host PC, so I thought this morning to add some indicators to the RSI control (high speed) loop VI which is native to the cRIO.  I have measurements for the actual loop rate and for the effective update rate of the analog input channel.  With this setup, and with a scan period of 10ms, and a network publishing period of 100ms, the measured loop rate on the RSI control VI is 10ms, and the effective update rate of the analog input channel is a very steady 840ms.  With the scan period at 5ms, the loop rate drops accordingly to 5ms, and the effective update rate drops to a very steady 835ms. With the scan period at 2ms and the network publishing period at 50ms, the loop rate is 2ms and the effective update rate is 834ms.  With the scan period set to 250ms and the network publishing period set to 500ms, the loop rate is 250ms and the update rate is 1000ms (again very steady with no fluctuations).  

There's definitely something about the ~830ms that is affecting the overall update rate.  

Any other ideas?

Thanks!

[Cameron.Andersen]

At last!  A solution!

The problem wasn't the scan engine at all.  The problem was a hardware limitation on the NI 9208 4-20mA input module I'm using for most of my input channels.  In high resolution mode, the update period is 52ms per channel on the 9208, which for 16 channels amounts to 832ms.  This matches well with the measured effective update rate that I've observed.  

I discovered this after "accidentally" observing some much faster updates from certain channels, which I later identified as being from a 0-10V input module.

Thanks for all of the help!

[Mercuri]

Thank you for posting your self-found solution to this. You saved me a lot of trouble as I ran into the exact same issue with the EXACT same module.

Enjoy your kudos.