Powerflex 40 Comms Loss

bjkallmeyer

Member
Join Date
Jul 2018
Location
Erlanger
Posts
5
Hello. I have a wierd one, or at least I think a wierd one. It all started about 3 weeks ago.....

I have a powerflex 40 with a 22-comm-d module. It is on a devicenet (obviously) network of 20 drives. It is node 16, near the middle of the network. It all the sudden started dropping out, which caused a running feedback fault on the PLC. What's strange is sometimes it will show a fault 81, sometimes it will not. When it doesn't actually trip the drive out, operator can ack the fault and then keep on running. When the drive actually trips on fault 81, then maint has to be called to reset the fault. The way our plc is set up, is pretty much as soon as we don't see a running feedback from the drive, then it triggers a plc/HMI fault for the fan it was controlling. There is a comms loss timeout setup in the drive as default for 5 secs. So my thinking was we were missing that running feedback bit for less than 5 secs. I've added a fault delay to the program, to allow them to run if the running feedback doesn't stay on for more than 5 secs.

So, the event happens completely at random. It can happen 2 hours between, and up to a week between faults. So, I started out by replacing the 22-comm-d, no go. I then disconnected the HIM module thats mounted on the front of the bucket. No go. Went ahead and replaced the HIM module as well as the base and cable. Still no go. I then replaced the drive. Nope. I replaced the dnet cables going to/from the next node. Upgraded firmware on DNET card as well as PLC firmware (needed it anyway). And....nope. Replaced the terminating resistors on the dnet network. Added the recommended grounding to the I/O common (as mentioned in rockwell technote) And still no better. This is the only drive on a network of 20 that is experiencing any issues.

Now I've added some historian points to try and catch this gremlin, but even that's a little funky. I can see the fault status on the trend, but I have a running feedback tag that never goes low when the fault happens. Thinking maybe a stale tag when this experiences issues. I've just recently added some of the status bits from the 1756-dnb (DeviceFailureRegister, DeviceIdleRegister, ActiveNodeRegister, ScannerStatus, ScrollingDeviceAddress, ScrollingDevicesStatus, DeviceStatus[16]. The last fault we experienced, where it actually tripped the drive on fault 81, my new historian data didn't show this node as experiencing any issues.

So that's where I'm at. All these possible fixes came from myself (drive, HIM, cables, resistors and firmware) and Rockwell tech support. I am now at a loss. It appears there is some type of interference that is still present. Next step from Rockwell is to add a motor choke on the output of the drive, which is currently on order.

Sorry for the long story, but I feel too much is better than too little. Let me know if anyone can be of some assistance
 
Have repairs been performed on any other drives on the network prior to this fault showing? Perhaps the problem isn't wit this drive, but another drive on the network is causing issues with this drive.
 
there was one drive (node 20) that was experiencing a fault 81 in this same time frame. That was found to be that the remote him module connection was flaky, and the HIM module would not light up. I replaced the HIM module and base on that drive but resulted in same thing, so believe the DSI port on that one went bad. I unplugged the him cable going to this drive, until it can be replaced. Since unplugging that one, I hadn't had an issue with it. That's the only thing I'm aware of
 
there was one drive (node 20) that was experiencing a fault 81 in this same time frame. That was found to be that the remote him module connection was flaky, and the HIM module would not light up. I replaced the HIM module and base on that drive but resulted in same thing, so believe the DSI port on that one went bad. I unplugged the him cable going to this drive, until it can be replaced. Since unplugging that one, I hadn't had an issue with it. That's the only thing I'm aware of
 
sparkie, do you think it could somehow be related? It does seem odd that they were both experiencing issues in the same time frame. The drives are physically located several bucket spaces away from eachother. It shouldn't make sense, but I've seen some strange stuff over the years. At any rate, the one drive with flaky HIM connection does need replaced regardless. Just trying to find an explanation as pretty much everything surrounding the "original post drive" has been replaced. I know I can alter the "A105" setting to pretty much make it go away, but I would rather try and find the root cause.
 
Is the devicenet cable possibly picking up noise? When I have seen this with bucket mounted drives on D-net, it usually affected multiple drives.

I have also seen problem crop up when there is a motor with insulation breaking down and the increased noise going to ground affects the drive to which the motor is grounded first.

At one time about 10 years ago we experienced an issue with a batch of bad ribbon cables between the d-net module and VFDs, but I think those were PowerFlex 70s and I think you have replaced all the similar parts in your case already.

Since you have exhausted many other causes, I think a megger test on the motor could at least rule out a motor insulation problem and if you have access to the tools and skills, that is what I might try next.
 
I'm waiting on an o-scope from another plant to maybe see if we do have some interference on the dnet side. But like you said, usually if you have noise on the dnet cables, then it effects multiple drives. Not just single one out. I have megged the motor as well, from the bucket to the motor and locally at the peckerhead and both passed with flying colors. Maintenance is scheduled to replace this motor as well, just to eliminate of one more piece.
 

Similar Topics

Hey so I have an issue that is driving me insane. I have 6 AB powerflex 527 drives configured in a DLR network over CIP Safety. They go back to a...
Replies
1
Views
1,706
I am trying to control a Rockwell Powerflex 525 via Modbus RTU (RS485) using an ABB AC800M DCS controller, and am having some issues getting...
Replies
2
Views
496
Hello, I just replaced a blown PowerFlex 755 drive. I set the IP address from the HIM to the correct network address. The drives has no faults...
Replies
7
Views
3,416
I'd like to be able to detect if the Ethernet Comms falls over between my PLC Ethernet card and any of the 5 PF525 inverters I have connected to...
Replies
7
Views
4,741
Setup: - Schneider Electric M340 PLC w/ NOE 100 Module installed - Allen Bradley PowerFlex 753 VFD w/ 20-COMM-M Modbus Adapter installed -...
Replies
2
Views
4,525
Back
Top Bottom