DeviceNet Lost Communication


Thread Starter

Chris Pawlik

I have a few DeviceNet networks. On the two of them I'm losing communication to the same node. Number of nodes is 12, distance about 200', 24 VDC power supply seems to be fine. I'm losing that node for short period of time and then is online again.

Master scanner talks to SLC 5/05. Slaves are Allen-Bradley: E3Plus, 1336 VFDs, BUL 825 Smart Motor Manager, PowerMonitor 3000 and I/O unit DSA4/2. Does anybody experience similar problem?

I'm using RSNetworx for DeviceNet. Is there any other software I can use for diagnostic of the DeviceNet? What about DeviceNet meter?

Thank you.
What kind of comm, is it ethernet, serial?? Make sure your program settings are correct at slave device, baud rate, half, full duplex, address. Make sure everything is properly grounded, and comm cable is shielded properly on 1 end. Does the slave device have a signal ground, which is different from earth ground, and is it connected? I have run into these problems before on 1 device going off/online in a daisychain serial comm.

Trevor Ousey \(list\)

Generally most problems I have come across are cable or power supply related. I assume you have the baud rate set to 125k, check the terminating resistors are correct (2 x 121 ohms). You can test this with a multi-meter, you should read around 60 ohms between can-h (white) and can-l (blue).

Which cable are you using? One of the flat trunkline cables we have gives us trouble with a Panelview, my preference is to use the round shielded cables as it seems more reliable. Check that there is a ground the shield at one connectors. Tighten spare connector terminals on the 'double' connector style plugs.

I have also found that you do need a good clean power supply, I have changed several power supplies to good quality units and higher current rating. And it is not a good practice to use the same power supply for general 24Vdc stuff.

Several of our DNet nodes on our PED (stand alone) welders had this same intermittent problem. These welders all use Dense Packs set at node address 16. They would run fine for a while and then lose communication between the Dense Pack and the robot controller. After a short period of time they would come back on line. One of the factory reps came in and suggested we run a seperate ground from the metal frame of the junction box that houses the Dense Pack to the shield (center pin) at the DNet connection. They said that this was their standard fix. Sure enough it worked - no more problems. Hope this helps.