Remote I/O problems

[neildobson 1]

I have a strange problem where the remote I/O is freezing (the inputs change state but the image table is not updated in CPU). My application is quite complex. I have four main racks populated with a variety of digital and analog I/O. This is expanded by using the D4-450CPU as remote I/O master (the 25-pin port) which connects to an additional two remote slave racks also containing various digital I/O modules. Finally for good measure the main rack contains an H4-ECOM ethernet module which connects to a PC which does logging/data capture functions. This application has been running quite successfully for the past 7 years. However from day one we have had some mysterious I/O problems where certain ladder rungs do not operate correctly. I have spent some time on this recently and the problem is the remote I/O. One of these racks has its inputs and outputs configured as X1000 and Y1000 respectively. So the following example should work fine: ----[X1000]----------------(Y1000) You would naturally expect when bit 0 of remote inputs turns on (X1000), the first remote output bit 0 (Y1000) would also turn on. This does work. However after a while, the I/O is no longer updated and if Y1000 is already ON, then if we lose X1000, Y1000 still stays ON. Conversely, if Y1000 is OFF (because X1000 is off), then no matter if X1000 switches ON, Y1000 stays OFF. The problem appears to be that the CPU forgets to refresh the I/O image table from the remote I/O racks and the whole lot is just frozen. Even more strange is that to fix the problem we fire up DirectSoft and as soon as the link appears (green icon) in the launch window, everything fires up normally and I/O refreshes fine for a while. I guess there is no easy answer for this, but I have a number of things to look at: 1) The remote I/O cable. This appears to be wired correctly and is only 5metres end-end, fitted with terminating resistors. I will check screening, resistor values etc. 2) Remote I/O configuration. This has been checked over and over and as far as I can tell this is correct. The I/O is configured as per the example in the D4 user manual. The unused slots are zeroed and C740 is set. Everything is set-up using first-scan bit SP0. Is the remote I/O information in the D4 manual correct? I see it differs slightly from the remote I/O manual. 3) Remote I/O status. I have seen no errors on any of the slave racks. Perhaps the status bits should be logged? 4) D4-450 as remote I/O master. How reliable is this. Any problems reported? Would it be better to use a D4-RM instead? 5) H4-ECOM. There is some heavy duty logging going on. Could this cause issues in the CPU? If anyone has had any remote I/O problems, particularly if using the D4-450 as remote master, or had any other problems when using the H4-ECOM I would be interested to hear from them. Regards,