Communication Errors on PanelMates

Randall

Member
Join Date
Apr 2004
Posts
4
We are experiencing some communication errors with a couple of our PanelMates in our plant and can't seem to find a way to clear them out. Some background on our network.. We have 7 Modicon compact 984 units with 6 Cutler Hammer PanelMates all connected through Modbus Plus. The PanelMates are 1700 color flat screens. The communication problem is occuring only at one section where we have two PanelMates side by side in a cabinet. Because of operator requests these PanelMates were stationed side by side and their controls are identical. The only thing different are the node addresses. You can clear the errors by taking one of the two PanelMates offline. We have ran new conduit, new cable; In fact, we now have four cables instead of the original two with a 15' jumper between them. We also have a repeater installed before the run to the PanelMates.
Even though these are flat screens, could it be possible that having the PanelMates so close together, that this could be causing the problems?

Thanks!
Randall
 
Yes, the network has a minimum of 10' recommended on cable which we had before using a 15' jumper between the two PanelMates. We now have a considerably longer run between the two. Still this has not helped, this is why I wonder about a minimum physical distance between nodes.

One more thing I forgot to mention in my first post, we have changed PanelMates too. We have changed PanelMates and the Modbus Plus interface cards and still the problem remains.
 
Have you done a net scan of the Modbus Plus network? Does it show all of the connected devices as individual addresses?

What type of Modbus Plus connectors are you using and what brand of cable? The ASMBKT style connectors require a minimum 10' cable length of between connections, while the 990NAD style can use the drop cable length to meet the minimum. Also, many installers do not properly trim the conductor ends when using the ASMBKTs, the excess cable length can act as transmitters/antennas.

Have you swapped one of the problematic Panelmates with a known good one?
 
Thanks Jim,
Yes we have swapped PanelMates and interface cards. I believe we are using the ASMBKT connectors, these are the ones with the vampire teeth. The cables ends are trimmed to the connector and the drain wire is dead center. We only have about an inch and a half of exposed cable to the connectors so I don't think we should be getting too much of an antenna effect here. I can't give you the number off the top of my head but, we are using the recommended Belden Modbus cable, it has Modicon stamped all down the line. I will try to get you that number asap.
We did scan and everybody shows up. We have even changed addresses with out any change. One funny thing that I have noticed is that the errors always seem to be worse on this one particular PanelMate (left side). This one also has the higher address. We have swapped this unit out numerous times with the same results. We have swapped the cables with the same results. Always on the unit on the left side.
Currently, for about two hours now, I am showing no errors but it will come back. It's only doing this of course because I have been watching it!
 
Since your communication problem is in the same cabinet, I would focus on something in that cabinet,

1. Make sure everthing is properly grounded, even installing a ground rod next to the cabinet.

2. If there are any relays or contactors (even General Purpose Relays) make sure they have surge suppressors on them.

3. Check for any relays chattering on/off.

4. Check for any loose wires in terminal strips.

It may take some detective work, but there probably is something in that cabinet causing the problem not related to the Panelmates.
 
Grounding

When you check your grounds, make sure each device has it's own dedicated wire back to a common ground point in the cabinet.

A dedicated and isolated ground rod should never be installed, per the NEC all ground rods must be bonded together. One primary purpose of grounding is to bond everything together to prevent ground loops and voltage differences, not to connect things to earth.
 

Similar Topics

Hi! I'm newer to PLC stuff as it seems to come up at the worst possible times at work. We have an old SLC5-04 connected to a Panelview 900. Over...
Replies
8
Views
784
Hey guys, I'm messing around in my rslinx classic program trying to get familiar with the Porgram, under the Driver Diagnostic General tab I...
Replies
0
Views
1,176
Hi, I am dealing with a problem with a vendor that has gone unresolved for a long time. We have a slc 5/05 processor communicating over ethernet...
Replies
9
Views
2,254
As part of an upgrade I am working with 7 different SLC programs. I had the first program in a test controller in my office running and...
Replies
6
Views
2,523
Our company developed a driver (around 2002) to read/write bits and analog data around 2002 for the Series 90-30 364/374 CPUs over SRTP. We have...
Replies
0
Views
1,279
Back
Top Bottom