ControlLogix gateway drops DH+ nodes

[Amish 0]

Hello everyone, I have a ControlLogix gateway that has been periodically loosing communicaionts with the DH+ nodes since we put in a new conveyor drive with a ControlLogix processor. If I kill power on the conveyor drive, the gateway stops dropping the nodes and everything is fine. I got on the web interface for this gateway, as well as two others, and the only difference I can see between the three is that with this one, the ENBT card in the gateway is initiating CIP messages with the ENBT card in the conveyor drive. On all the other gateways, the only CIP messages are generated from our SCADA servers to the gateways. One gateway ever shares a switch with a controllogix conveyor drive, and is not trying to message it. I don't know why there would be a difference. I couldn't find a way to disable this from the web interface or Logix 5000, but maybe I'm just missing it? Or, is there something I need to change in the switch? Thanks for your help.

[Amish 0]

Could it have something to do with the default gateway address? It is the only gateway with one defined, but it is just the switch address, it doesn't go to another network or anything.

[newpageboba 7]

The number one recommendation for all DH+ issues is to very that the installation is done correctly. 90% of the time it is an installation issue. Follow Allen Bradely's specifications.

[Amish 0]

Ok, that may be true. But could an issue on the DH+ side of the gateway really cause the Ethernet side to establish CIP message connections with another 1756-ENBT? The fact that I could stop these connections by killing power on the controller with the other ENBT, and my DH+ nodes would stop dropping out made me think that the DH+ had nothing to do with the issue. However, the other day, we killed power on one of the DH+ nodes (there are 3 nodes on the network, this is the closest to the gateway). When this occured, the CIP messages stopped and communication was constant to the remaining two DH+ nodes. That has me even more confused. It is almost like there is an error in DH+ communications associated with this problem node, and as a result of this, the gateway is establishing CIP message connections with the ControlLogix conveyor drive. Is this a known function of these gateways? I'll keep digging. Thanks.

[Amish 0]

I fixed the issue today, but a question remains. I found out I can use Linx to configure the gateway. I had just been using the BOOTP utility for the ENBT card, and the selector switches for the DHRIO card. So I get on with Linx and go to the DHRIO card, and go to the routing table. Sure enough, there was a link established to the ENBT at the questioned conveyor drive. I realized this was a connection for network sequence between the Logix belt drive on Ethernet and the other on DH+. Since this was no longer configured after replacing the 5/05 with the Logix controller, and we have an inductive pickup for backup sequence, I got rid of this link, and the DH+ nodes stopped dropping out. So, why does the setup work with a SLC 5/05, but not a ControlLogix with a 1756 ENBT card? CIP connections and multicast messages? Do I need to make sure I have IGMP snooping enabled? Thanks.

[dmargineau 6]

The usage of managed ENet switches with the IGMP snooping enabled is highly recommended for mixed protocols networking or considerable implicit communications traffic. Another solution would be updating the CLX firmware to Rev.19.11 and the 1756-ENBT bridges to Rev.5.1; this scenario enables the user to Unicast the "single pipe" connections without the need of using of the IGMP Snooping Enabled managed switches as the network "traffic cop".