SIMATIC MMC - Read Error or defective

Hi all.

I have a problem with Simatic Micro Memory Card used in a Simatic CPU 315-2 PN/DP.
Recently I get quite a lot of feedbacks from our sites that the PLC is in STOP mode, and will not return to RUN mode(we have more or less the same program running on different plants). Our service personell tries to connect to the CPU/PLC both with ethernet, MPI and Profibus, without any luck. Only solution so far has been to replace the MMC and download the program again.
However, when the memory cards have been returned to me, it is possible for me to delete the MMC card via an USB prommer. That is, when I try to open the MMC in the USB prommer, I get an error message:

"EPROM: An error occured reading the MMC - 33:8578 ...a Read Error has occured... try to erase the MMC... if problem persists, it's likely that the MMC is defective"
Now, if I delete the MMC via the USB Prommer, I can copy the program to the MMC again, insert it in the PLC and switch the PLC in RUN mode. Then everything is OK.

Does anyone know what can cause a problem like this with the MMC's?
I don't think it is due to writing to the memory, althoug I do use SFC83/84 in some plants, but then the function is only called when asked for by operator.

The problem described above often happens after a power failure on the plant, but not always.

Best regards
eibe

eibe said:

Our service personell tries to connect to the CPU/PLC both with ethernet, MPI and Profibus, without any luck.

Click to expand...

Do they scan for reachable devices via "edit ethernet node" for example ?
Can the find the CPU by its IP address ?
Can the find the CPU by its MAC address ?
Even if the MMC is defective, the CPU should remember its IP address.
If the CPU isnt reachable at all, it appears to me that the CPU itself has been zapped, not just the MMC. I am thinking noise or lightning strike.

eibe said:

I don't think it is due to writing to the memory, althoug I do use SFC83/84 in some plants, but then the function is only called when asked for by operator.

Click to expand...

You must also block the program or user from changing the blocks that are being backed up with SFC84.
I only allow the use of SFC84 when the machine is in stop, and then I bypass the rest of the program while SFC84 is running.

Hi
Thank you for your answer.

Do they scan for reachable devices via "edit ethernet node" for example ? YES
Can the find the CPU by its IP address ? NO - which I find strange
Can the find the CPU by its MAC address ? NO


You must also block the program or user from changing the blocks that are being backed up with SFC84.
I only allow the use of SFC84 when the machine is in stop, and then I bypass the rest of the program while SFC84 is running.
Thanks, I will look into this to see if it can be a problem. However, it's not in all programs where I have these problems...

Forgot something:
Can the find the CPU by its IP address ? NO - which I find strange
Can the find the CPU by its MAC address ? NO

But it works fine when they change the MMC without changing the PLC...

Can they find the CPU by its MAC address after doing a memory reset by the switch ?

Can they find the CPU by its MAC address after removing the MMC and cycling power ? (even without an MMC, the CPU should be reachable by the IP or MAC address).

Anyway, that is not so important. If it happens, they have to insert an MMC card that isnt defective anyway.

There were some issues with some CPU firmware having some serious bug, so that could cause the program to be corrupted. But I cannot imagine it can actually affect the MMC card.
So I still think noise or lightning strike as the real cause of the problem.

Hi Jesper.
thank you once again for your feedback.
Siemens has adviced that I update the PLC firmware to the latest, which is v3.2.10.
One of the CPU (plant) which have had this problem has v3.2.7... I'm currently trying to find out what firmware the other CPU's have.

However, Siemens cannot say whether this (v3.2.10) will solve the problem.

The failure often occurs after a power failure on the plant - but again, not after every power failure, and not always after a power failure. We do have 24VDC input/output converter in our cabinets to isolate and secure 24VDC supply... I suspected that the problem was due to a brown out, but the personnel on site assures me that there has been no alarms on the 24VDC power supply system.