Troubles!!

Good Morning all

I'll start at the beginning.

1756-L62 , talking to my plant network through a 1756-enbt/a FW. 4.003.

A week ago the machine was started up after a week long shut down, No work was done to this machine during this shut down, only basic cleaning.

Machine starts up and runs for a few hours, now it has a feedback error on axis 12 of a 14 axis kinetix 6000 system. Through troubleshooting with Rockwell tech we determine that the motor and cables should be replaced as the pulses were being miscounted between the motor and PLC.
The motor, MPL-B430P-MK22AA is now older and I had to replace it with
MPL-B430P-MK72AA. This was the number given to me by tech support. Along with this new motor, new cables also had to be installed to conform with the new connector style.

Installing the new motor, no problem, installing the new cables went well except for the fact that the wiring was different on the old cable compared to the new one. I went according to the wiring guidelines in the manual and have verified it several times since. Rockwell assured me that swapping out this motor would be painless and I would not need to configure anything. I called also the OEM for the machine and they said the same thing.

So after installing the new motor and cables, I power the machine up and realize that now not only do I still have the feedback error on the drive , I also have a Sercos card in the rack that is flashing all of its LEDs, and an ENBT/A that will not talk to my plant network.

The ENBT/A is invisible to my Linx and not able to respond to ping upon power up over my regular wireless connection. I can patch in to the switch in the cabinet and ping it as well as see it on linx, I can do everything I would normally do with it when I'm connected over the network but only when I'm patched in. The other thing is that there are no LED flashing green or yellow on the ENBT Ethernet connector. all the other devices on that switch are flashing lights but not the enbt.

Fast forward to this morning, nothing on the machine has been touched since I left it last. Now, over my plant network I can ping it, and linx can see the enbt but will not look past it. I can look at the enbt's configuration, status but it will not connect to the PLC. If I try to go online I get the error message "Failed to go online with the controller. Unknown Application Framework subsystem error."
On linx, all the items in the rack are red X except the enbt, Also, now the LED are flashing on the Ethernet port for the enbt connection to the switch. As well, when Im online with the processor, I can see the other PLCs that I have set up as producer/consumers are intermittently displaying yellow triangles and saying connection attempts failed.

Normally I would just say there is a comms issue and keep on that track but Im getting side tracked and perhaps mislead by the fact that one of my sercos cards is flashing all of its led green and when patched in to the PLC and online I see that in my motion groups folder, in the project manager window of
RSL5000 version 19.11, all the servo groups are showing yellow triangle and showing a axis status as sercos ring faults.

As well, when in linx and looking at the sercos card status, it shows "idle" while all the other cards in the rack show "active".


I am trying to get a Rockwell tech here but the regional guy is away and I'm looking at paying some serious donairo to bring a guy in from another region from Rockwell.

Worst of all this, I am seriously lacking any servo or motion skills, no experience with it at all but I do have a course starting This Monday on Motion and servo with Kinetix. Go figure.




So any first thoughts? Am I simply looking at two issues here, one with an enbt and one with a sercos problem, or could there be something common causing both these issues.

Thanks for any input.

SOO Odd.. I just connected wirelessly after about the 25th try through linx.

No error this time.

Now, If I could only solve this Sercos problem!

Thanks

And it kicked me off again. @#%&#~!!!

The error it gives is "can't communicate with linx."

So,,, I have the Sercos issue fixed. I had a miswire in the drive end of the feedback cable. I did wire it correctly but I wired it according to the WRONG wiring diagram.
I would have never guessed that a miswire on one feedback cable would cause 14 drives to be faulted.
That's ok now.

However, still I am plagued by this intermittent communication situation.
If I brute force it and repeatedly click go online, it will eventually let me on but only for a few minutes.

This machine has been here for 2 years with no communication issues yet.

Thanks

PLC Pie Guy said:

And it kicked me off again. @#%&#~!!!

The error it gives is "can't communicate with linx."

Click to expand...

If you have something misbehaving on the network you can get poor or intermittent connections. Try to see if you can connect with only your laptop and the ENBT connected to the switch. If that works, then start adding the other devices to the switch. This won't fix your problem, but may help rule in or rule out what is causing the problem.

This advice is not based on any facts, more on a feeling .. and likely reflects a bunch of biases that I have ...

This sounds like a flaky switch to me. TCP services connecting, running for a bit, timing out at some layer of the 7 layer stack, and not having the error passed back through the layers in any meaningful way. If you have spare switches, I'd replace one piece at a time - standard troubleshooting stuff. If your switches use the wall-wart style power supplies - replace them as well as the switch. Those power supplies have tripped me up many times.

If you are using switches fed by 24VDC, change out the power supply, check for stable 24V out of the power supplies, a sag on the input of the power supplies ... network switches act *WEIRD* when they don't get good power.

Good luck and keep us posted!

I would try increasing the "System Overhead Time-Slice" setting in the controller properties, "advanced" tab... it defaults to 20% reserved for system operation and comms, but you may need more to allow programming software access.

You can go to town with this, setting it very high, but make sure that "During unused...." the option is checked for "Run Continuous Task" (default).

SO the machine is running well and has through the night. The last time I was online with it yesterday from my desk over LAN, I left it online after being kicked off many times during the morning. It was still connected this morning for like 18 hours. Just for fun I went offline, and now it doesn't want to connect anymore over the very same LAN connection it just held for 18 hours.
Its still talking to its produced/consumed peers over the network but just only wants me to connect through the local Ethernet switch inside the cabinet.

The machine is running so I can't do much now, but when it sees a down day I intend to look at firmware revision of the enbt. Currently at 4.003 but it can go to version 6.
Also, I am replacing the old el cheapo Ethernet switch for a new one.

Daba, I am also going to look into the "System Overhead Time-Slice" setting.

Somebody did mention checking the DC supply o the switch, it is a
solid 24.4VDC. Just for the record.

thanks for all the replies

Try running a continuous PING test at the location you were last online...

Ping 111.222.333.444 -t

...see if you get consistent times, or if there are frequent long response times.

Consistent is OK, variable not so good...

Just a thought - although no changes to this machine , were any other changes made during shutdown that might affect the network - ie new machine, duplicate IP address on another PC or HMI added , someone with a laptop working on another piece of equipment and logged off for the 18 hr run ?