DeviceNet Com Module possible issue

Hello All,


I wonder if some of you can give me some insight in this problem that I have. Here are the symptoms


1. 1756-DNB/A DeviceNet Comm Module is in the same chassis as the ControlLogix PLC


2. RsNetWorx on first scan detects only nodes 0, 1,3,4, and 5. Node 2 indicates an identity mismatch, help file suggest to delete the node from the network graph and re-browse. Same issue


3. I power down the PLC chassis and removed the terminal block from the DeviceNet comm module, then reinsert. Just plain ol unplug and plug.


4. Power up the chassis, fire up RSNetWorx and went online, now the network sees node 0, 1,2 and 3 only


5. Repeat step 3 and 4


6. Now network only sees node 1 and 3


On the LED scroll on the DeviceNet Comm Module I see code 78, 5, and 4. Before the power cycle, I only see code 77, means mismatch data type according to the help manual. 77 went away after the power cycle.





This led me to believe that maybe the DeviceNet Comm Module is going bad. As Node 0 is the PLC that is on the same chassis as the comm module, if RSnetWorx sees the PLC intermittently then maybe the module is going bad?


Anyone encounter a similar issue with DeviceNet Comm Module? Any steps you have taken to further troubleshoot this issue?

It is possible that the 1756-DNB is failing slowly.

It is far more common for there to be loose or corroded conductors, or missing termination resistors, or induced electrical noise.

Start by verifying that there are just two terminating resistors: power down the network and measure resistance between the white and blue wires. If it's anything but about 60 ohms, there's a problem with termination.

Can you describe the devices, especially Nodes 4 and 5 ?

Error 77 is very uncommon in a system that has been commissioned and is running and has not been modified, since it's a data size mismatch and has to be the result of mis-configuration.

Error 77 being replaced by Error 78 essentially means "the device was reporting a data size mismatch before, but now it's not communicating at all".

Ken Roach said:

It is possible that the 1756-DNB is failing slowly.

It is far more common for there to be loose or corroded conductors, or missing termination resistors, or induced electrical noise.

Start by verifying that there are just two terminating resistors: power down the network and measure resistance between the white and blue wires. If it's anything but about 60 ohms, there's a problem with termination.

Can you describe the devices, especially Nodes 4 and 5 ?

Error 77 is very uncommon in a system that has been commissioned and is running and has not been modified, since it's a data size mismatch and has to be the result of mis-configuration.

Error 77 being replaced by Error 78 essentially means "the device was reporting a data size mismatch before, but now it's not communicating at all".

Click to expand...

Hello, I have checked and Do see that there are two terminating resistor between the blue and white wire at two nodes, and measured the resistance between the blue and the white wire, with the power off, (Can + and Can -).


The resistance came out to be 120 ohms. I did this between the green and white wire at every node.


So this does indicate a termination issue?

40mpg said:

Hello, I have checked and Do see that there are two terminating resistor between the blue and white wire at two nodes, and measured the resistance between the blue and the white wire, with the power off, (Can + and Can -).


The resistance came out to be 120 ohms. I did this between the green and white wire at every node.


So this does indicate a termination issue?

Click to expand...

Or a cabling issue.

Each of the resistors should be 121 ohms (call it 120 for convenience). With two of them in parallel, albeit at opposite ends, the overall resistance should read half that, or ~60 ohms.

A measurement of 120 ohms normally suggests one of the the resistors is not properly connected


Since you have confirmed the presence of both, if we assume they are properly connected the reading suggests there is a break in either the white or blue wire at some point in the circuit -- you're reading across one resistor or the other depending on which side of the break you're on, but never both.


First verify the resistors are good individually -- remove each one and check its resistance when disconnected from the circuit, it should read ~120 ohms. Assuming both are correct resistance, then re-install one of them and move down the line checking resistance at each node to find where you are losing it.


Note that the resistors should be installed at the *end* nodes of the trunk line.

Somehow DeviceNet termination discussions often end up being absurdist plays about the word "two".

There should be two 120 ohm resistors, total. One should be at each end of the network trunkline. Because they are in parallel, you should measure about 60 ohms.

Maybe one of yours is not properly installed.

I recommend removing them both, measuring them while they are uninstalled to verify that they are really 120 ohm resistors, and re-installing them.

Ken Roach said:

Somehow DeviceNet termination discussions often end up being absurdist plays about the word "two".

There should be two 120 ohm resistors, total. One should be at each end of the network trunkline. Because they are in parallel, you should measure about 60 ohms.

Maybe one of yours is not properly installed.

I recommend removing them both, measuring them while they are uninstalled to verify that they are really 120 ohm resistors, and re-installing them.

Click to expand...

Hello, Thank you so much for the troubleshooting pointer.


The issue was a fried resistor at the end node. Visibly burnt, replaced it and the system worked like a charm.


This exercise helped me understand DeviceNet connections in the field.


Thank you everyone again.

Ken Roach said:

Somehow DeviceNet termination discussions often end up being absurdist plays about the word "two".

Click to expand...

That goes in the "memorable quotes" Hall of Fame .

I wrote down a complete devicenet troubleshooting guide long time ago. The only problem is that is written in Spanish. If anyone need it just send me an email requesting it.

Is it:

When you see a devicenet network, leave.

Engineerj22 said:

Is it:

When you see a devicenet network, leave.

Click to expand...

@Engineerj22 and @andrewwei :LOL.
I was a fan (still) of devicenet from small networks up to three o more networks with more than 50 devices on each branch and It always worked and still working.
Now I am retired now but I visited a plant this week on monday and one o those networks is still working as a charm, I just installed three new AC drives from yaskawa and It was like 1,2,3.
Do not worry you are not the first nor the last who do not like devicenet.

I joke, there's a steel mill that has a couple MCC cabinets full of E3 plus devicenets. It's not terrible to work on. I even figured out you can switch from E3 to E300s in E3 emulation mode (both devicenet) almost painlessly.

@Engineerj22. :
Yep, I did it the other day.