8 replies to this topic

[Clayton]

Ok, so I'm trying to add a pump around to an existing crude column simulation that I have. I'm using HYSYS. I wanted to find the duty for the PA so I used Packie's method by finding the del_T_gap and the del_T_50% and I used the material balances from HYSYS to find my internal reflux flows. I use the flow from the tray as the Ln and the Naphtha product flow as the Dn to calculate my internal reflux as I have seen done in - Petroleum Refinery Distillation - RN Watkins

The PA is being returned between the Naphtha(overhead distillate) and the Kerosene(highest side stream) and taken from the bottom stage of the kerosene stripper.

The problem I am having is that when I calculate the del_T_Gap between the Naphtha and the Kerosene from the assay I get 57oF but when I use the reflux and the del_T_50% of 158oF(from assay) I get a del_T_gap of 45oF. The number of trays between the return tray of the Kerosene and the Naphtha is 9.

Does this lower temperature gap mean that proper fractionation is not possible between the Kerosene and Naphtha streams or am I doing something wrong???? I've done this in HYSYS and it works out pretty well using random duties but i can't explain it.

[Zauberberg]

Clayton, you should walk outside in the plant, and measure the temperatures on both shellside and tubeside fluids, flowrates you can obtain from DCS (pumparound flow is always measured and controlled). Laboratory will do the density measurement for you, and later on you should apply the density correction at temperature levels you have observed in the plant. It's that easy. If you get calculated heat exchange on both sides of pumparound exchangers with the difference not greater than 10%, this is your pumparound duty.

If you are simulating imaginary tower, refer to "Petroleum refinery distillation" by Watkins, and "Elements of petroleum processing" by Jones. There you will find a lot of information how pumparound duty is distributed across the tower - depending on the number of pumparound circuits, their location in the tower, and available side-cut products.

Good luck,

[Clayton]

The existing tower does not have a PA between the naphtha and the kerosene side stream. I am required to justify putting a PA between them. I have been using those two books as you have recommended them before. The Watkins was a bit more helpful than the Jones so I am using the method for a type A tower in the Watkins (a tower with a reflux).

The problem I am having is that when I find my degree of separation using Packies method and compare it to the degree of separation from the assays of the kerosene and the naphtha the DOS using packies is smaller.

eg
DOS using Packies method = 42 oF

DOS using the assay data = 57 oF

does this tell me that a proper separation is not possible even though it has been accomplished in HYSYS

Thankyou for your help

[Zauberberg]

Assuming properly characterized crude feed, HYSYS will always give you more reliable results - in my opinion.
You said you have been requested to justify the installation of additional pumparound between the naphtha (overhead or side cut?) and kerosene product. What is the reason behind additional pumparound installation? Do you want to utilize additional heat available in the CDU tower at the expense of reduced separation efficiency, or there is a requirement due to tower heat balance control problems? Let's make these things clear from the beginning.

By installing a circulating reflux between two adjacent products, there is always a loss of separation efficiency between these two products - as compared to the base case when there is no pumparound. Upload your model, or at least provide a skecth of CDU tower, and then we can focus on the subject.

[Clayton]

The naphtha is overhead referred to as BS or base stock in the model. The engineers at the plant told me to simulate the existing column and then simulate a PA between the kerosene(top side draw) and the BS(overhead). They mentioned reducing the heat load on the condenser but weren't very clear as to the reason.

I wanted to justify the adding of the PA by the method outlined on page 72-73 of "Petroleum Refinery Distillation-R.N.Watkins" by showing that the degree of separation attained using Packies method and the mass balances provided by HYSYS to the existing column would be more than the degree of separation between the 95% of the (Naphtha or Bases stock)Overhead and the 5% of the Kerosene.

The reason I would like to justify it on paper and not just use HYSYS and say well hey it works is that this is a project I am working on and I don't think that saying it works on HYSYS is enough work for me to merit a good grade.

I will call the engineers tomorrow and verify the reason they want to add the PA.

The model provided is of the existing column as best as i can model it with out the PA.

Can you provide a site where I can upload the model as you did before. It seems that I am not allowed to upload this type of file and I do not know if i can use the link you provided me in an earlier conversation.

[Zauberberg]

Adding a pumparound will definitely reduce the separation efficiency, i.e. the gap between 5% Kerosene and 95% Naphtha distillation points.

You can upload the file (zip it first) on: http://www.engineeri...32/Default.aspx
If it doesn't work, send it to: zauberberg@verat.net

[Clayton]

Sorry i took so long to reply
was preparing for exams

(to Zauberberg)
So i just wanted to know how you justified that flow rate and Dt

[Zauberberg]

I've just used those data which appeared to be close to reality, based on my experience and other column parameters you have provided. Of course, this may be different from actual design - depending on available heat sink(s), tower internals capacity, and heat removal requirements.

[Clayton]

i think i have developed an acceptable model
seems so hard to just give up the project and move on to other things now

but thanks for the help it was greatly appreciated feel free to ask me about any outcomes of this project if just for the sake of curiosity