Location of Thermocouple in Distillation Towers 1

[lgmarti]

What is the best location for a tray thermocouple? I have a thermocouple located at the outlet of the tray downcommer. The temperature fluctuates by 10-15 degrees and is hard to control. Should the thermocouple be on the vapor or the liquid phase? In the downcommer where I can guarantee there is always a liquid level?

 

[CARF]

Dear lgmarti,

Your problem looks like an oscillating control system. With proper tuning of the PID controller parameters you me get rid of the 10 - 15 degrees fluctuations.

A nice site is:

http://www.controlstation.com

But of course with any control system it starts with a good measurement. Personally I would never use thermocouples but Pt100 elements, much more reliable in my opinion. What you could do as well is measure pressure. Since you have saturated vapour (liquid / vapour system) pressure and temperature are 1 to 1 related, just as in saturated steam.

Please give some more detail about the control strategy,
CARF

 

[lgmarti]

Thanks for your reply. I do not know much about controls so please excuse my ignorance. The only thing I know is that the thermocouple is located in tray #2 from the bottoms (TC701); and that TC-701 controls the steam flow controller (FC405). The flow controller flows the setpoint very well but the TC setpoint is almost never met. A friend of mine said that I should adjust the tunning in the TC point by making T1 bigger. He then pointed out that that points are REGHG in the highway and that I can't change the equation (whatever that means!). I do not know anymore. I have heard of pressure compensating the tower's temperature readings. However, the pressure in this tower is very steady. We only have one pressure transmitter in the overhead. Any help is appreciated. This is my e-mail
martymartinez126@hotmail.com I will check out your link. I need a crash course on controls!

 

[aspearin1]

I've had the fortune of studying under Dr. Cooper, the inventor of Control Station. I can vouch for the Demo and the tutorial guide. The utilization of both can "simplify" (conceptually) complex control issues that are more commonly described with lengthy differential equations and Laplace transforms. If you don't know control, it's a good start. But I wouldn't call it a "crash course" either.

ChemE, M.E. EIT
"The only constant in life is change." -Bruce Lee

 

[CARF]

Hi lg,

The TC701 (Temprerature Controller) is generating setpoints for the FC405 (Flow Controller). This is called cascade or master-slave control. The fast flow control loop is controlling the fast fluctuations in steam flow, whereas the slow temperature control loop is telling the FC405 what steam flow-setpoint is required to reach the desired temperature (= temperature setpoint). Very simple.

Now please don't listen to that friend anymore; the two most likely reasons you do not reach the required temperature setpoint are:

1) Too small steam control valve (check if control valve is 100% open all the time!).
2) Bad TC701 controller tuning (PID-parameters).

If the temperature fluctuates in a sine like wave it is called an oscillating controller and you need to adapt
the PID-parameters in the Controller or Process Control Computer (PLC).

I strongly suggest you connect a datalogger on this system and have a good look on the historical data!

Hope this helps,
CARF

 

[Yordif]

I can't speak to control loops but as far as acurate temperature measurement there is a company called Gayesco in Pasadena, Tx. They do multipoint, flexible, thermocouple temperature measurement for a multitude of hydroprocessing systems as well as distilation and contactor towers. They can be found at Gayesco.com