H2-CTRIO's losing configutation

[RobG]

I have 3 H2-CTRIO's in a 9-slot base with a D2-250-1 CPU. The counters occupy slots 5,6 and 7. Whenever power is interrupted and the PLC is powered back up at least one or more of the counters loses it's configuration. I must then write the configutaion file back to counter to restore it. Total power budget for the base is within the specs. Firmware is up to date. I've considered installing a UPS to keep the PLC powered during outages. Any ideas?

[BobO]

Really?!? Weird.

The configurations are stored in both RAM and ROM, but the ROM is only read and validated at startup. You said the firmware is up to date, did you get it that way, or have you updated them? The firmware and config are stored in the same part.

We have never heard of this that I am aware of. Since you have more than one failing module, that tends to point to something else...like base power supply. However...the reset chip on the module should kill them if the power drops to about 4.5v, which is inside the spec for the flash parts. I have no clue.

I'd like to swap modules with you and see if we can duplicate the problem here.

[Greg]

RobG, I've actually seen this once or twice over the years. In all cases I know of this was caused by really dirty power (i.e. power spikes or excessive rippling). So a UPS isn't a bad idea, but power filtering might work too.

[BobO]

Hmmm...ok...so this is what happens when you give the boss a login and turn him loose on the forum. He lies.

[RobG]

Strange but true. I updated the firmware after this first occurred. It didn't make any difference. The power supply is close to the max at 2490 mA. At the moment only 2 of 3 counters are being used. The third one was designed in for part of the equipment that is not currently used. I could remove the unused counter but that wouldn't tell me if it's the smaller load on the power supply or a different problem. I could also try installing the counters in a base with no other modules but the CPU and if that works I guess that would point to the power supply. If that seems to be reasonable approach I will try that and if that fails then I guess swapping modules with you would be the next step.

(One thing I failed to mention in the initial post - there is also a remote base.)

[Greg]

Yes, the closer you are to the power supply limit, the more susceptible it is to rippling and spikes. I recommend you just filter the power, or reduce power requirements (if you can)... or... both.