4 replies to this topic

[sgkim]

Hello..

You don't need to make streams for reflux, condenser, drum, reboiler, or pumps etc. To run the simulation with them all is too tedious to get converged.

Just specify the condenser of the column as "partial condenser" and a side stream from 2nd stage as overhead liquid product if required. And note that the reflux ratio is based on the ratio of liquid flow to vapor product from the partial condenser.

Then run the column. The duty of thermosiphon reboiler and vapor/liquid flow rates, can be extracted from the "Tray Properties" and "Tray Compositions" on "Results" Menu after the simulation.

If you want thermal rating further, you run "Sizing>Shell & Tube>Thermosiphon Reboiler".

Stefano/050901

[sgkim]

Thx for answer,

I managed to run the simulation the way u proposed,it has converged..but I want to know how to do it as well if I separate condenser and reboiler,because I have another dist column that has top stream enter reactor and series of exchangers before condensator-I would like to simulate that as well,it has to be separated then,I can't include reactor which is used to eliminate C2H2 with H2 to get more C2H4

(1) If your top stream vapor travels only for heat-exchange, you run the column simuation without actual network, get the condenser duty and temperatures. After you get the column converged, add such heat-exchangers, convey the column data (reflux ratio, temperature, etc) to the new configuration, then run the new flow scheme. But you have to consider that the condensate will drop every time the vapor passes thru the heatexchanger. What's your idea on this 'condensate' ? You probably consider a "common condensate collector?

(2) On the other hand, if the top stream run thru the reactor for <reaction> you have to specify the column as <no condenser>, <no reboiler>, add separate condenser or reboiler, separator(flash drum, etc) and divider then connect by arrows. The reflux rate, however, shall be fixed by your own <ratio> set by the divider "manually". You have to try a lot to run, check, and get the "reasonable" value from the simulation data of the column and reactor converged together. Verification of the "resonable" data of column and reactor should be made from case studies.

sgkim/051001