From 04:00 PM CDT – 08:00 PM CDT (09:00 PM UTC – 01:00 AM UTC) Tuesday, April 16, ni.com will undergo system upgrades that may result in temporary service interruption.

We appreciate your patience as we improve our online experience.

VeriStand

cancel
Showing results for 
Search instead for 
Did you mean: 

Problem multiple models Veristand 2014

Solved!
Go to solution

Hello.

I'm having a strange behaviour with my models running in Veristand. I'm a very experienced user with Labview and Veristand and this is the first time I have seen this problem.

My HW is a pXI8110 with can board configured at 1000Hz.

I have compiled serveral Labview models into Veristand (14 to simulate the ECU plant of a vehicle). There are not so complex. These models read data comming from CAN, execute the code and connect the outputs to other models or CAN. I have also a Custom device with is consuming 50% of the 4 cores.

When I start testing the system I can check that several code into some models is not being executed. I mean, I have 6 can inputs in one model and I can see that all the values concerning 5 inputs are working properly but the code corresponding to the last input is not working properly. I have checked that the mappings are right and I have also forced the last input value without any positive result.

I have also tried to change the models execution order but it still doesn't work. I have increased also the decimation of this model (instead of 10 ms to 100ms) but it doesn't work.

If I try the model in a Veristand PC project it works properly so I'm wondering if it can happens due to a cycle time leak, because the PXI is not able to execute all the model code. 

Do you have any idea about what can I check?

Thanks in advance.

Miguel.

0 Kudos
Message 1 of 4
(3,771 Views)

Hi,

 

If even faulting the channel has no effect you could try a look at http://forums.ni.com/t5/NI-VeriStand/Wrong-channel-value/m-p/2951249/highlight/true#M3814

 

Best regards

 

0 Kudos
Message 2 of 4
(3,756 Views)

Funny this topic is here, I was seeing a somewhat similar issue where there was a growing difference between the number of Tx'd messages and Rx'd for our CAN model. Talking with NI engineers I was advised to implement a custom device:

 

http://zone.ni.com/reference/en-XX/help/372846G-01/TOC46.htm

 

There are differences in how VeriSTAND treats models vs. custom devices. I would look into this page for possible understanding into your issue:

http://zone.ni.com/reference/en-XX/help/372846F-01/veristand/pcl_modes_models/

 

0 Kudos
Message 3 of 4
(3,713 Views)
Solution
Accepted by topic author Allende

Hello.

I have done multiple tests and the model is working again increasing the decimation.

In order to avoid it again I have generated a "code execution time" in each model using counters and I check if this time is greater than the actual execution time parameterized in Veristand. In that case I generate a warning. Then I manually readjust the model execution times and orders. Unexpected results can be obtained if it happens again.

BR,

Miguel.

0 Kudos
Message 4 of 4
(3,706 Views)