0

Has CAN traffic changed from 4.07 to 5.08?

Kevin 1 year ago in IQANdesign updated 1 year ago 2

We have a system with MD3 and XA2, the XA2 expansion CAN is made wireless using Maknetek transmitter and receiver. No issues for around 10 years service. 

Recently upgraded the MD3 to touch screen MD4-5, rest of the system remains identical. We have copied the application file across exactly but are experiencing intermittent drop outs of the wireless bus, resulting in CAN no contact XA2 messages on screen. They are brief and not permanent, only loss for a moment.

The manual for wireless transmitter and receiver states that the bandwidth is limited (understandable) and thus it will miss transmission if too much data is sent. Is there increased CAN traffic on expansion bus when going between 4.07.14.3816 (MD3) and 5.08? The data we are sending and receiving in the software is identical, just wondering if there is more being sent behind the scenes for example module management, diagnostics etc? 

We slowed the polling on expansion bus, and this seemed to help, although because its intermittent we wont know for a week or so. Just seeing if it could be a possibility.

For the title question, Release notes - IQANdesign 5 has this change:

36603Even out CAN bus utilization
JFOUT, GFOUT and APPOUT with send method continuously and a transmit rate > cycle time are automatically distributed in time, reducing bus utilization peaks.

This does not change anything for the XA2. 

The change from MD3 to MD4 could be relevan depending on what buses you use. If you use CAN A or B it should not be a difference, but the MD4 also have the two CAN buses C and D that are handled by a co-processor.

An easy application mistake to make when changing master module could be to forget to set the application cycle time to the value the old application had. 

Thanks for the information. 

The wireless bridge is on CAN A on the MD4.

Are there different components internally (hardware wise), or perhaps the MD3 is less sensitive to CAN disruptions? It appears to be only a momentary loss.

We slowed the polling on the expansion bus to XA2, although it appeared  initially to resolve, it has not fixed our original issue. Although the cycle time was transferred across, we may look at extending this to see if it helps.