huy doan

Delay transfer signal between 2 controller logix

13 posts in this topic

Hello. I have main PLC L73 with redundancy module, and 4 others controller as L71, L72 (S/R A,B, CSU A/B). They connected through control net network as my attached document. I created message to read and write DINT, Real in main PLC L73 to others controller. I can read and write signal but sometime I have delay 20s to receive signal. Has anybody know my issue? How can I fix it? Thanks you

Coal handling configuration.pdf

Share this post


Link to post
Share on other sites

Without doing a detailed study, I cannot say for sure why your data is delayed 20 seconds sometimes, but ControlNet is a scheduled network and your Reads and Writes are utilizing the unscheduled slice unless you did them as Produced/Consumed.  When it comes to data transfer between PLCs I am a believer in READ ONLY NEVER WRITE.  This is because a WRITe a consumes 2X the bandwidth that a Read Does.  You might try putting reads into your L71 and L72 rather than writes from the L73 and see if anything improves.  I suspect it will.

Share this post


Link to post
Share on other sites

Thank you for your reply. However our controller L71,L72, L73 are in different version. When I tried to use producer/ consumer tag, I get an error “ controller not match”. Thats why I created  message to read/ write. I created message write/ read from L73 to L71 and from L72 to L73, I still got same issue, is sometime delay 20-30s to receive signal. I will try to use only read messages in all controllers and see what happen.

Share this post


Link to post
Share on other sites

Here's the why to do read only.  

With a Write from PLC1 to PLC2 we have the following sequence of events:

1. PLC1 to PLC2 - I am going to send you DintArrayX.

2. PLC2 to PLC1 - I am ready to recieve DintArrayX.

3. PLC1 to PLC2 - Here is DintArrayX and it's checksum.

4. PLC2 to PLC1 - DintArtrayX received with the checksum is ok

5a. Inside PLC2 - Checksum is OK write is done.

5b. Inside PLC2 - the checksum doesn't match what I calculated - Repeat Steps 2-4.

With a read from PLC2 to PLC1 you have the following sequence of events:

1. PLC2 to PLC1 - Please send me DintArrayX

2. PLC2 to PLC1 - Here is DintArrayX with checksum

3a. Inside PLC2 - CHecksum is 0k Read is done

3b. InsidePLC2 - Checksum is NG repeat Steps 1 & 2

So 1/2 the traffic for the same amount of data transferred when using reads in all PLCs/PACs.

Share this post


Link to post
Share on other sites

Thanks BobLfoot. I tried to use read messages in all controller, but it is same issue. It look like communication drop each 20-25s and reconnect. I don’t have experience with control net. So I don’t know, maybe something wrong in this network configuration. 

Share this post


Link to post
Share on other sites

I'm no controlnet expert - {I just co-led a project install a Logix V15 Cotnrolnet Redundant system for baggae handling at a major US Airport and achieved 30ms scan time max}, but in your shoes I'd start with the following.  

1. List all PLC CPU Revisions 

2. Check the Firmware in each Controlnet Card for revision level and cpmpatability to the CPU their associated with.  {Dirty little secret, some controlnet firmware versions, don't like certain PLC firmware versions, this is why RA maintains a compatability matrix}  I recommend using it.

3. Re-check my Controlnet Hardware , Hardware issues {loose connections, poor power supplies} can plague a system.

Share this post


Link to post
Share on other sites

Thanks Bob. Yes, I found others controller and local controlnet module from other users identity mismatch. Before I used producer/ consumer tag to transfer, But controller mismatch. Thats why I use unschedule message. So I hope this fault not affect to this issue

I am worried to reschedule network now. 

Share this post


Link to post
Share on other sites

if you have identity mismatch errors by your plc of the cnet card this could definitely cause the reset every 30 seconds or so.  Is it just one plc or all that are showing identity mismatch?

Share this post


Link to post
Share on other sites

Hi Bob. It show all others CPU, and cnet identity mismatch in RSnetworx but Module indicate OK . Because other PLC and cnet use different version.

PLC1, Cnet1-  Cnet2, PLC2. I guess problem is PLC1 and PLC2 Both PLC configure Cnet2 and PLC in different version.

So I think Cnet2 should be not scheduled both side, then can send message to transfer signal. 

Thank you for your reply

Share this post


Link to post
Share on other sites

@huy doan PLC1 and CNET1 should be checked and proved compatible.   

Then PLC2 and CNET2 should be checked and proved compatible.

After that scheduling everything from RSNetworx connected to PLC1 should be all you need to do and things should work.

Share this post


Link to post
Share on other sites

Thanks Bob. Because our system have builded from others suppliers and they use different software version. As you talk we need upgrade Studio, RSNetwors for all to same version? It really inconvenient.

I don’t understand what they mean: unschedule message and producer/ consumer schedule.

Share this post


Link to post
Share on other sites

OK - from the simplest overview I can give.  ControlNet is deterministic.  Assume you have 3 ControlNet nodes a Scanner and two slaves.  The Scanner Controller divides each second into four 250 ms windows.  Window1 is for the Scanner to send its scheduled data to slaves 1 & 2.  WIndow2 is for slave1 to send its scheduled data to Master and Slave2.  Window3 is for slave2 to send it's scheduled data to Master and Slave1.  Window4 is for anyone with an unscheduled message to claim and send their message.  Oldest message waiting to go gets Window 4.

Now I've used a 1 second window and 250ms slots for ease of understanding.  What actually happens during network scheduling is each window is sized so that all scheduled data is transmitted once from each user. and a remainder window assigned to unscheduled.  Now if the unscheduled message is large it may take several windows to complete. 

Add to this whst you describe in your case the PLC2 and CNET2 are incompatable versions or not proerly configured.  and the problem gets worse.

Share this post


Link to post
Share on other sites

Thanks BobLfoot,

We have 2 controlnets. 

The first net is PLC1-Cnet(Node3)  -  Cnet(Node 50)-PLC2: This communication worked properly even PLC1 and PLC2 incompatable version and do not need to schedule PLC2 in PLC1 network.

The second net is PLC1-Cnet(Node4)-Cnet (Node14)-Cnet(Node16) -  Cnet(Node45)-PLC3: This communication have problem as I mention above. Actually when I remove Cnet(Node45) to disconnect with PLC3, I found the controlnet led A,B sometime blinking (It mean communication drop this time). When I remove Cnet(Node16,NOde45),  PLC1-Cnet(node4)-Cnet(node14) led A,B not blinking. So I guess we have problem from Node14 to Node 16. 

From Node 14 to Node 16 is 2km and We use repeater 1786 RPFM and fiber optic is Multimode. Normally, multimode fiber just design for short distance. 

Have you think multimode fiber is problem?

Thank you very much

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now