Thanks very much for posting that RSS program.
I am a notorious curmudgeon on the topic of PCCC messaging to control AC drives, so please don't take any criticism personally. I'm friendly and helpful; just ask my dog.
One issue is that all of the MSG instructions were programmed in the classic "fanning the trigger" mode where they run independently and as fast as possible, re-triggering as soon as each one finishes by un-latching the .EN bit whenever the .DN or .ER appears.
This works for simple testing, and can be necessary when you're really pounding on MSG instructions to get data between controllers fast.
But I recommend strongly against it for this sort of control. Recycling timers and more careful MSG sequencing is better for the PLC's comms bandwidth.
Another issue that might be responsible for the 5-10 second time between commands appearing to take effect is that I believe that the PowerFlex might not accept 1-word writes to N41:0,1,2. It might be that it was accepting it once every time the PCCC command function timed out. Because your logic was re-triggering the MSG instruction as fast as possible, you never got the opportunity to visually see the .ER bit or the .ERR error code.
The proper way to do it is to write to N41:0, Length 3. That unused N41:1 word doesn't hurt anything, and the Command bit word and the Reference word arrive in one packet.
In addition, I noticed that you used Copy Word to move the Feedback value to a FLOAT type data table. That's going to give you a weird result because CPW copies the raw bit pattern, rather than automatically converting from Integer to Floating Point.
The PowerFlex 525 uses Reference and Feedback values in units of 0.1 Hz, so the value "123" represents 12.3 Hz. You can handle them in those units, or just divide by 10.0 and put the result in a FLOAT, or do any scaling to RPM, inches/second, or furlongs per fortnight.
I'll post the amended RSS file in a few minutes. There's plenty of room for me to have made a mistake because I don't have a MicroLogix 1400 or PowerFlex 525 to test on, so I'd like your feedback about whether the modified logic works.