Control system of chiller with propane as cooling medium

[Dendrobium1975]

Dear members,

We operate a Propane Refrigerant Unit to provide chilling requirement for some chillers with various purposes. In those chillers (kettle type), shell side is continuously filled with propane in a certain level while the process fluid flows within the tube.
There are two different control systems applied to those chillers.

1.) For a chiller which is used to cool an MDEA-water mix from 43 deg C to 32 deg C, the control system allows the process fluid outlet temperature to be controlled by adjusting the backpressure of propane (by throttling a control valve in the outlet of the shell side/propane side) while the shell side level is controlled by adjusting the propane supply (by throttling a control valve in the inlet of the shell side/propane side).

2.) For a chiller which is used to cool light hydrocarbon (C1-C6) from 11 deg C to -23/-31 deg C, the only control system applied is the shell side/propane side level control by adjusting the propane supply (by throttling a control valve in the inlet of the shell side/propane side). For this chiller, the process side outlet temperature is not controlled but just monitored by a temperature indicator.

I’m still wondering why different control should be applied for those two chillers. Any helpful answer will be very appreciated. Thanks in advance.

 

[CJKruger]

Dendrobium1975, I think the easiest method to control the hot fluid outlet temperature is with a bypass around the chiller.

Cilliers

http://www.korf.co.uk

 

[Domz]

Control schematic is decided based on process requirements. In your case there are two control parameters one is level and the other is temperature. Since maintaining the level in the exchanger is important for this procees both the exchanger has the level control.

1) If we need to maintain the temperature of a process fluid at a particular set-point or to have different set-points based on plant load then we need to have a controller. hence you have temperature control in your first chiller.

2) If there is lot of room for temperature to fluctuate and if it doesn't cause any upset to the plant then temperature monitoring is sufficient. By selecting a temperature indicator we reduce one potential mistake that can be caused. For e.g all the controllers need a set-point. There is a possibility of operator entering a wrong set value. Moreover controllers needs proper tuning(PID). If it is not set properly there is a posibility of valve to open/close which inturn can upset the process as well as cause the valve to fail.

 

[Dendrobium1975]

CJKruger and Domz,

Big thanks for your helpful response.

Cheers