PLC5 to PF700 w/20-comm-e Problem

[Mark Snodgrass]

We have a Power-flex 700 drive with a 20-comm-e communications adapter card being commanded from a PLC5/40E. The communications work fine during normal operations, the problem occurs when the ethernet cable is unplugged, power is cycled on the drive, or a power outage (anyone have a 100MVA UPS? ).

The messaging errors out for up to 15 min before it starts talking normally. Thankfully this is not a critical motor, but we want to be able to convert our other drives to ethernet communications and are hesitant to do so till this problem is resolved.

I have tried talking to AB tech help and they did not have an answer.

I have looked on the adapters webpage and I can see it is receiving and sending. I can also connect with Drive Executive and see the drive parameters, so I know the ethernet connection is good.

Has anyone else had a similar problem, and were you able to find an answer?

Thanks in advance
Mark

 

[Mark Snodgrass]

No one else had this problem before?

 

[The Plc Kid]

One issue that I see is that PF 700 drives were never meant to be controlled per se from a plc 5 or slc series ethernet setup.

Ethernet on plc 5 and slc is primarily for comms. Like connecting rs logix or maybe bringing in data from a hard wired drive.

Most drive control from plc 5 or slc is done with RIO,Devicenet or Controlnet. Not ethernet.

Make sure you do not control any critical or dangerous drives using plc 5 or slc ethernet comms.

 

[Ken Roach]

You don't see this problem much because RA very strongly recommends against this method of drive control.

Just because it's technically possible to control a PowerFlex and 20-COMM-E using a messaging controller (PLC-5E, SLC-5/05, MicroLogix 1100) doesn't make it a good idea.

The PLC Kid is right on; you absolutely should not implement more drives on Ethernet with a PLC-5 sending messages to them. RIO, DeviceNet, and ControlNet are the right ways to connect drives to PLC-5 controllers.

I'd have to see the way you configured your MSG instructions to determine if there's something about them that is filling up a buffer, but the 15 minute time sounds like a TCP Timeout. You would really have to capture the Ethernet traffic (Wireshark is the standard tool) to find out.

 

[Mark Snodgrass]

Yeah, I kind of figured that it may just be due to the PLC5 being such an older platform that ethernet messaging is not really reliable on it. All of our other drives are controlled via RIO and other than the random DPI port fault, we have no problems with them. We were just trying to see if we could upgrade all of our comms to the ethernet (hell, we still have some comms via DH+).

We chose this drive to experiment with as it is not process critical and we set it up to stop with any loss of comms. Luckily, we don't break connection often and the messaging runs flawlessly once it establishes itself. We just cant use it on other drives due to downtime issues it would cause if the connection is broken for some reason.

We are upgrading most or all of our mill to contrologix over the next few years and I know they work well over the ethernet. Our newest boiler has contrologix and it works well, except for a few environmental issues which we had to correct (condensate in conduit directly over the processor).

Thanx for your responses
Mark