Powerflex 753 with Devicenet lost connection after power loss

Hi all
I am working on a retrofit project.

There is one backup pump with Powerflex 753 drive controlled by 2 different PLCs(one 1756-L63 ,one 1756-L71 , have 1756-DNB on both sides)

We are using 20-COMM-D adapter on Powerflex 753 and work well via Devicenet.
However, there is one requirement from the customer to simulate in the case of power loss.

When the drive was on standby and cut power, there is no issue, and could be back online after the power was back.

But when we cut the power during the drive is controlled by PLC and recover the power back, the drive will disconnect and can't back online automatically.
I also use RSNetWorx to check the DeviceNet network, and the drive can't be scanned at the time.

The 1756-DNB and the drive also didn't indicate any error.



Any suggestions would be greatly appreciated

It sounds like you have a DeviceNet-based "hot standby" system, where one controller runs the PowerFlex over the network and the other inhibits its I/O connection until some kind of switchover occurs.

Unfortunately there was never a well-supported method for doing this with DeviceNet. The protocol never had a redundancy object or well-defined redundancy mechanism.

Some users created their own. Some relied on a limited Rockwell example, which I think was only ever well-tested with certain CompactLogix.

I can't tell if your system is purely software-based, or if it includes some kind of physical disconnect to remove the "standby" controller from the network.

What is the scope of this "retrofit"; are you replacing an older A-B drive with this PowerFlex 753 ? I presume you didn't design the original "standby" system.

I am surprised that you can't browse the drive when it's not making a logical connection to the controller; A-B drives have always had a very robust unconnected message manager that allowed them to talk to other devices even while under I/O control of a PLC.

That makes me think maybe there's a physical disconnect that drops the DC power or the CAN signals to one of the 1756-DNBs as a "switchover" mechanism.

Ken Roach said:

What is the scope of this "retrofit"; are you replacing an older A-B drive with this PowerFlex 753 ? I presume you didn't design the original "standby" system.

I am surprised that you can't browse the drive when it's not making a logical connection to the controller; A-B drives have always had a very robust unconnected message manager that allowed them to talk to other devices even while under I/O control of a PLC.

That makes me think maybe there's a physical disconnect that drops the DC power or the CAN signals to one of the 1756-DNBs as a "switchover" mechanism.

Click to expand...

Hi , thanks for reply.
The system was running with 1336 series and we had this application in different production line, we also did some upgrade on other line.
You are right, it is kind hot standby concept, and it was design to share this drive to different PLC. But due to the limitation of production , we will only stop one PLC at a time and keep same structure. Therefore it still running on Devicenet.



I found this situation only happened when the control power of the drive are totally off ,and power back at very close time.(in this drive, it around 10second ,I can see the HIM screen is off) When this happened, the drive is totally disappeared from the DeviceNet. I also check the drive status from HIM, there is no error occurred.


If the power source back later , for example 15 second , it still working well and could be found on Devicenet.

I assumed it is a hardware bug. In our test procedure , it required to cut the power off for 10 second then turn the power on.
For other line, the drive usually can keep standing around 20 second after power off. But I don't know the reason this drive consumed the power more.

Anyway,I am wondering is that possible to find some way to "wake it up"?

Seems I can't do anything from PLC because the drive is totally offline on the Devicenet.

I have an idea to do some logic in drive's Devicelogix to monitor two PLC connection and to restart drive by itself. not sure it will work or not.

Thank you very much for those details !

I think you are testing the the hypothesis that the PowerFlex 20-COMM-D module does not fully reboot or goes into an no-communication mode when power is only removed from the main control board for about 10 seconds.

I think that you need to examine it very carefully during this condition.

Look at the LED indicators (PORT, MOD, and NET A) and record their color and flashing condition.

Use a HIM module to see if you can connect to the 20-COMM-D over the onboard DPI network, even when it does not appear to be "visible" over DeviceNet.

If the problem has to do with the 20-COMM-D never fully powering down and "remembering" its I/O connection to a different 1756-DNB scanner, I would expect it to still have an I/O timeout and cause a fault on the drive, and to be browsable over the network because it still has a good Unconnected Message Manager feature.

I look forward to reading your description of how the 20-COMM-D diagnostic LEDs indicate its status after the 10-second power loss.