BRX fixed on wrong CPU type

[Kristjan]

We have a BX-DM1E-18ER3-D set up in the dashboard but when connecting, it claims to be a BX-DM1E-M-D even though we go online with the disk project. We are unable to fix this by changing CPU type and writing to PLC.

NetEdit shows correct type. Restoring factory settings through NetEdit doesn't fix the issue. Clearing memory doesn't fix the issue. This CPU is still certain that it is a BX-DM1E-M-D...

Is this a known issue? Any solution other than physically changing the PLC?

[BobO]

We have a BX-DM1E-18ER3-D set up in the dashboard but when connecting, it claims to be a BX-DM1E-M-D even though we go online with the disk project. We are unable to fix this by changing CPU type and writing to PLC.

NetEdit shows correct type. Restoring factory settings through NetEdit doesn't fix the issue. Clearing memory doesn't fix the issue. This CPU is still certain that it is a BX-DM1E-M-D...

Is this a known issue? Any solution other than physically changing the PLC?

There was a known issue with the board to board interconnect on -10 CPUs that could cause misidentification, but I know of no issue with -18s. It sounds like a failure of some kind, but not one I'm familiar with. We're happy to repair and return ASAP...would love to do so actually, so we can understand the failure. If you want to do that, PM me contact info and I'll get that moving.

[Controls Guy]

A current batch of 10ED13s no issues at all so far!   

[BobO]

A current batch of 10ED13s no issues at all so far!   

Good to hear. This whole thing has been frustrating and embarrassing. You ship 1000s of these with essentially the same connector, no issues. Then, boom. Welcome to the roaring 20s.

[Controls Guy]

Connector mfgr probably lost all their people during Covid, then restaffed by cleaning out the local McDonalds.   

[BobO]

Connector mfgr probably lost all their people during Covid, then restaffed by cleaning out the local McDonalds.   

That happens more often than some would like to admit, although with what they are paying at McD's these days, that might be harder.

[Kristjan]

I initially posted this issue on a BRX -18. Since then, I received it from my customer and it is correctly identified on my workbench. So we cross our fingers and hope it has matured beyond the type conflict 

Now I have the same issue with a BX-DM1E-10ER3-D, the same type that has had some known faults (see post from BobO above). Latest firmware and memory cleared and DoMore Designer still telling me it is BX-DM1E-M-D. Even resetting to factory defaults doesn't help. NetEdit, on the other hand, is identifying the correct type 

This specific PLC was OK just a couple of weeks ago...

Are there any known batches with errors? The PLC I have has serial number 02222A4B306273. Could that be from an affected batch?

[BobO]

Yes, -10 CPUs had an issue with the board to board interconnect. Send it back to us and we'll fix it.

During COVID, we were unable to get the original connector, and sourced the connector from a different vendor. While the specs were identical, the new connector didn't stay properly mated. At first we though it was an assembly problem, but eventually figured out that the new connectors simply didn't work like the original. We re-speced connector and the problem was resolved.

[Kristjan]

Thanks Bob. It's a bit of a hassle to send a single PLC so we'll just keep it for now and use it for some test and data processing purposes.

[BobO]

Thanks Bob. It's a bit of a hassle to send a single PLC so we'll just keep it for now and use it for some test and data processing purposes.

If the boards are undamaged, you can open up the case, squeeze the board to board connector back together, and then either hot glue or tape the connectors to keep them from separating. In some cases, the uneven application of power due to the connector issue has damaged the CPU card. The boards are usually undamaged though.

[Controls Guy]

With one troublesome unit we planed about 0.015-0.020 off the face of the female connector before hot gluing because it felt as if the pins were barely engaging and some angular lead-in on the mating contacts was trying to separate the connector.   No issues so far.

[BobO]

With one troublesome unit we planed about 0.015-0.020 off the face of the female connector before hot gluing because it felt as if the pins were barely engaging and some angular lead-in on the mating contacts was trying to separate the connector.   No issues so far.

Yeah...we're pretty sure the wipe design was wrong, and that was the primary difference between the two manufacturers. Lots of magic formulae got lost during the pandemic.

[Kristjan]

Hot glue did the trick. The B2B interconnector easily falls out.
I already fixed a couple of -10s in my lab, both of them with HW rev 2A 4B / 0222.
Is it only this HW revision that is affected?

This seems to both affect the CPU identification in Do-More designer and the I/O behavior. Sometimes the connector has sufficient contact to identify the I/O board and in other cases the CPU doesn't see the I/O board at all.

[BobO]

Hot glue did the trick. The B2B interconnector easily falls out.
I already fixed a couple of -10s in my lab, both of them with HW rev 2A 4B / 0222.
Is it only this HW revision that is affected?

This seems to both affect the CPU identification in Do-More designer and the I/O behavior. Sometimes the connector has sufficient contact to identify the I/O board and in other cases the CPU doesn't see the I/O board at all.

Anything in that same housing had the potential problem, so -10, -M, EBC, DMIO, etc. The I/O is also on the other board, so yes, both discovery and I/O behavior could be a bit dodgy. As long as the connector stays mated, there should be no issue.

This whole problem has been frustrating and embarrassing. We had shipped 1000s of those units previously with no problems, and then all of the sudden it was a problem. As I mentioned, we thought it was an assembly issue at first, but eventually we figured out the alternative source of the connector didn't work like the original. We used the hot glue approach for production and repair until we used up the supply, then we moved to connector with longer pin engagement.

Between C19, inflation, chip shortage, etc, this has been a humdinger of a decade so far. Starting to better understand why my grandparents (who were kids during the Great Depression) were the way they were.

[Controls Guy]

Starting to better understand why my grandparents (who were kids during the Great Depression) were the way they were.

No doubt!