Any ControlNet gurus here?

Does anyone know if it is possible to use the routing tables in a Control Logix gateway to route a ControlNet message? What we're trying to do is route messages anywhere on a complex network made up of ethernet, DF-1, and ControlNet links by just using the link IDs from the gateway. So far the only way we've figured out how to do it is if we supply the routing information ourself in the ControlNet packet, which kinda defeats what we are trying to do.

I'm thinking that it is possible since we can go into RSLogix and configure a multihop message using just the link IDs by clicking on the RS Linx check box. Unfortunately the PLC we have is too early a rev to use this kind of message so we can't actually test it and see what kind of message it produces.

Any info would be greatly appreciated.

The reason why you have not received any answers as of yet is simply because we don't understand the question.
For example:
Ethernet is just a layer 1 and 2 of the 7 layer stack.
DF1 is an AB communication driver used on channel 0, RS-232.
ControlNet is an AB high speed deterministic network.
How can you tie all these components together?
You can't.
I think that what you are after is routing of DF1 message packets encapsulated over TCP/IP by some sort of tunneling device such as Lantronix or Moxa and also you need to route CIP protocol over
TCP/IP ???
In case of DF1, the tunnelling devices will only allow communications over Ethernet TCP/IP from point A to point B. Now that is a shortcoming of the simple tunnelling device. You could use the
source and destination addresses within the DF1 and hold the IP addreses in the memory and then cross reference. The problem is that
you will not be able to route to SLC5/05 or to ENI module since they use the complex EtherNet/IP header.
You would have to use your routing device on all DF1 devices.
In terms of the CIP routing. The EtherNet/IP protocol is well documented. I presume that you would use the PCCC encapsulation
described in volume 2 of the spec. To compose the EtherNet/IP header is no small task.
More detailed explanation from you with strict adherence to proper networking terminology is needed.


terminology is necessary

JT said:

The reason why you have not received any answers as of yet is simply because we don't understand the question.

Click to expand...

Didnt understand the question!!! I didnt understand the answer either

Sorry, meant DH+ where I typed DF-1.

Here's the basic situation. Consider a plant that has a lot of different types of PLCs (SLC 5/04's and different models of PLC 5's mostly). If you start connecting all of these together with DH+ links you end up saturating the links since they are only 57K and the inter-PLC communication gets very much bogged down. Bumping up the communication to 230K helps a little, but not much, and a lot of equipment doesn't run very smoothly at 230K so that's not really a solution. One thing you can do is buy a bunch of Control Logix chassis and populate them with DH+ cards and ethernet cards. This allows you to segment up your DH+ networks and bridge them together with ethernet links. This works fairly well, once you go through the royal hassle of figuring out how to set up the routing tables in the Control Logix gateway and figure out how to do the message routing on the PLCs.

However, you still have a problem, in that each local DH+ link is fairly low bandwidth. One solution to this is to start using ControlNet links in place of the DH+ links. Since ControlNet is much more configurable, you can bump up the data rate considerably and get rid of some of the bottlenecks in the system.

You can also start using ethernet capable PLCs like the PLC 5/80E.

So now to the root of the problem (my problem). We make a device which sits on the DH+ network and exhanges data with the different types of PLCs. We can use the message routing capabilities of the Control Logix gateway to access pretty much any PLC in the entire system, simply by specifying the link ID of whatever link it is on and the node number of the PLC. This works all fine and dandy, except that it sits on the DH+ network which is a bottleneck. So, we're trying to do the same thing over ControlNet.

We can access local PLCs with no problem. However, when we try to go through the gateway from ControlNet to another network the only way we've figured out how to do it is that if we supply all of the routing information. We've done this and successfully routed to a PLC on a DH+ network from the ControlNet network. It seems kind of silly to have to configure all of the routing information in the Control Logix gateway, then configure all of the routing information again on our system, so what we're trying to do is take advantage of the routing tables in the Control Logix gateway. The ControlNet spec isn't much help. When I go into RS Logix, I can configure a message block that uses the link IDs, which looks like exactly what I want to do, except do it from our system instead of the PLC 5. If I could just figure out what a PLC 5 sends out when you configure it this way I'd be all set. Unfortunately, our PLC 5 is of too low of a revision and won't accept the message block.

So, what I'm looking for is info on how to send routed ControlNet packets using the link IDs set up in the Control Logix gateway's routing tables.

I hope this makes sense, and sorry if my terminology is a bit off.