Common mistakes in process design/simulation

[mohamedellejmi]

Guys,

through my carreer, I had to mentor some new process engineers and I noticed that most of them make almost the same mistakes... Here are some:

1- When designing scrubbers or separators, people take the Standard Flow for Actual flow and they calculate huge volumes
rule of thumb : Actual volume =~ Standard volume / pressure (barg)

2- When dealing with gas compressors (especially reciprocating), people have the tendence to believe that compressor discharge pressure is set by the compressor it self. Which is totally wrong!
Discharge pressure is the backpressure exerced by downstream equipment.
(In case the downstream equipment is a control or a globe valve, the discharge pressure changes with flowrate, which is controllable on the compressor it self)

3- When using HYSYS for designing distillation columns, people have the tendency to use short-cut column and to inject the resulting tray number, reflux ratio, feed position, reboiler temperature,... into the column model and run it...!!!
They spend a week trying to converge the column... Waste of time!!!
I recommend you to take only the number of trays out of the short-cut method + 2 to 5 trays and take the feed position.
Then, you should specify the light key content in the bottom product and the heavy key content in the overhead product, and if necessary, specify one of the product's flowrate based on the inlet composition.
This will converge immediately... then it's up to you to OPTIMIZE!

4- Some people rely on HYSYS to determine the H2S content of a stream... WRONG! HYSYS doesn't deal correctly with H2S solubility in water and in HC. Be Careful!


Feed us with your experience guys... Tell us about the common mistakes you know!

[mkhattaby]

Thanks alot , I personalyy experienced similar mistakes with my direct reporters .

[subramanian.R]

Its really a valuable information . Pls everyone share ur experience . This error almost all process will make ,its for sure.....

[mohamedellejmi]

Another mistake I came across yesterday...

When designing a process where gas Chillers/Coolers are used, and when calculating the energy consumption, people have the tendency to use the Chiller duty (say 1MW) as an energy consumption.

However:

- if the chiller is a propane refrigeration unit, the duty to be considered is the energy absorbed by the Propane compressor's driver + lube oil pumps (if electrical) + condenser fans.

- if the chiller is just an Air Cooler, the duty to be considered is only the energy absorbed by fans.

- if the chiller is just a sea-water/gas heat exhanger, the duty to be considered is only the energy absorbed by the sea-water pump.


Guys, we need your experience... come on!

[uetian98]

EOS Selection Criteria from myside.
Hope this helps.

[talha_sangi]

from your question i understand that u are talking about simple coolers/heatexchangers, in which we only enter process condition of one side.
if yes,
then the calculated duty given by heaters/cooler is only for process fluid whose condition is given in the heater by you. it doesnot tell you that how much flow requirement and how do you add or remove that heat from process fluid side.
this is only susggestion.

[Ahmed El-said]

In simulation
A mix of therrmodynamics (shifts) that isn't compatible.
Can give occurence to sudden temperature jumps which looks strange
Thermodynamics selection. E.g. when H2S is involved one needs SOUR (ProII) or similar to get closer to reality. Seen one used a straight SRK in which there was 10 wt% H2S in the feed. Thermodunamic methods are correlations (only) and are never perfect, but are the best methods available.
Water solubility in HC fractions is another area of errors. These are nearly always misleading. Solubility curves are always advisable to use (if they exist).
Running distillation columns below weeping points. The column will converge in the calculation but in real life, "it swings like...."
Not accounting for fouling properly and not retrofit area excess into fouling.
Not doing an ASB (AS Built) simulation to retrofit actual equipment to behaviour. Should be standard procedure but is seldom done.

In Design
Not doing a proper job on startup/Shutdown cases which can turn out to be governing cases for parts of the unit.
Not retrofit the hydralic pressure drop profile causing control valves to operate at low openings.

A few of the things that comes to mind

[aria]

Dear all,
This is a wonderful topic, it would be great if we could create a wiki like page to create a document for this topic i.e "Common mistake in process design/simulation". I am sure it will be one of those page with several thousands hits and will be helpful to students and practicing engineers.

[aria]

Dear all,
This is a wonderful topic, it would be great if we could create a wiki like page to create a document for this topic i.e "Common mistake in process design/simulation". I am sure it will be one of those page with several thousands hits and will be helpful to students and practicing engineers.

[omli]

people must make the difference between design and optimization, at the begining we try to design the unit or a part of process system, own calculations and effort are needed, it is different from optimizing an existing case, here we must enter all equipements and installations properties and run in dynamic mode when modifying conditions , but when designing we start in steady state after objectif is fixed , we optimize, we must put in our mind that hysys and any other simulator is only a tool which can be used correctly only when good practice of engineering fundaentals and states of arts
sorry my english is bad hope mssage will be received

[omli]

for exemple if i need to make a simulation in which there is a column, i must have a minimum informations about trays calculations, relation between column hight and the number of trays, tray spacing, fouling, relation between reflux and head product purty, heat exchangers.....

when we talk about co2 removal we must know that it is recomended at hight pressure and low temperature ((because it an exothermal reaction) and the regeneratio must be at high temp (reversibility of the rection) and low pressure ( henry law) without forgeting corrosion effect and......

conclusion : the simulation come latter after a good database is made

[alexcv]

Thanks