S7 - Strange goings on in S7GRAPH

I've solved my problem, so this is not really critical any more, but I'd sure as heck like to know what's going on here, so that I don't get bitten some time in the future.

I've got a very simple S7GRAPH program that I use to set all my motor-driven switches back to a safe state when I do a RESET. In the RESET program a bit is set to call the S7GRAPH program ("M_DO_LTS_RESET"). After the sitches have been set-up the last two instructions clear the "M_DO_LTS_RESET" bit and set the "M_LTS_RESET_OK" bit.

If I do both of these together as in this screen dump, it works.



However, initially I only cleared the "DO" bit in this step and set the OK bit in the last step. I tried it both with a transition after 100ms and also with the transition controlled by the "OK" bit being set, as shown here. In both cases, the program would appear to run correctly through and shut down, but the "OK" bit wasn't set.



Anybody any ideas what's going on here?

Where is "OK" reset ? Are you sure "OK" is not just coming on for one scan ?

It's reset in the RESET program, before the "DO" bit is set to cause the LTS_RESET FB to run, called from OB1.

I've checked and double checked in X-Ref, including MB/MW & MD all around M101.3. When the fault was present, I could set the "OK" bit in a VAT and it would stay set. As I said, the only difference is that I added the setting of the "OK" bit to the second-last, Step in the program (without removing the "SET" command in the last Step).