Dead time in aspen dynamics

[leon_aurel]

I did a control simulation in aspen dynamic. I added some PIDincr block, connected it with PV and OP, initialized it , set direct/reverse action. Then i ran it, and the simulation ran ok. Then i tried to tune it using close loop ATV. after tuning, i found that close loop ATV tuning gave ultimate gain and ultimate period. I found that it is weird since i didn't assign any dead time block.
Based on theory, i will get ultimate gain and period only if my system exhibit more than second order or first and second order with dead time. This happened every time i did tuning, whether it's level controller, pressure or temperature controller.

Does aspen dynamics automatically add dead time everytime we do close loop ATV tuning??

[nabucodonosan]

To obtain good results that can be transferred to a real control system, it is important that you accurately model the loop dynamics.
It is particularly important to model lags and dead times in the control loop.
For any sensor, the response of measurement is not instantaneous. It is more realistic to include measurement lags.
Temperature of Liquid 0.5min (First order lag)
Temperature of Gas 1min (First order lag)
Composition 3~10min (Dead time)
Yuo must insert a deadtime element on the floesheet upstream the temperature control (or composition control). Remember: is important installing the controller initially without the deadtime element to avoid initialization problems that sometimes crop up if you attempt to install the deadtime and the controller all in one shot.
When in your flowsheet is present "Deadtime" + "PIDIncr" everything is ready for the releay-feedback test with "Cloosed loop ATV" as the "Test method"