Chemical Engineering Calculations for Pressure Vessels 4

[5997]

Hi everybody,
Can anybody tell me how to perform chemical engineering related calculations for pressure vessels, both vertical and horizontal.
Thanks in advance.

 

[TD2K]

That covers a lot of ground, how about some details what you are looking for?

 

[5997]

ok, to start with,
a. how do I calculate the residence time
b. What is/what is the criteria to select NLL or normal liquid level for vertical vessels
c. how to assess CD, the drag coefficient
Thanks for the quick reply

 

[TD2K]

1. In its most basic form, residence time is the volume of the vessel divided by the flow rate (vol divided by vol/time gives you time). For multiphase vessels, it's usually the volume occupied by each phase divided by the flow rate of that phase. Depending on the specific system, there can be other definitions.

2. Setting NLL is a whole discussion in itself. Low, normal and high levels can be set based on residence/storage requirements or control system requirements, it really depends on the system you are talking about. For say a reboiler steam condensate pot, you might size it for only 2 to 4 minutes between low and high level because you only need enough time for the control system to respond to changes and for level, that is relatively quick. If you have a vessel on the other hand that is feeding a distillation column, you are going to want to have more surge capacity because if there is a change in the process flow into that vessel, you don't want to quickly have to change flow rate to the column so the feed vessel is designed with the volume to catch that process changes and then slowly adjust the feed to the column.

3. This sounds like a question for a flare KO drum. API 520 (or it's API 521) has a chart for CD, it's an iterative calculation but I don't remember the exact details. A flare drum will likely have a NLL close to or equal the low liquid level because you want to have as much room to catch a slug of liquid as possible but that's a specific case.

 

[SNORGY]

Might be some benefit to reviewing GPSA particularly Volume 1 Chapter 7.

 

[5997]

Thanks TD2K for the response.
About #3, I was confused how you are gonna find which equation applies on it, whether stokes law, intermediate or Newton's equation. That of course depends on the fluid Reynold#.

Vessel sizing is a bit intricate and its really hard to find the stuff on google for example.
Thanks guys for your input. What is GPSA snorgy.

 

[TD2K]

GPSA - Gas Processors Suppliers Association. They have a handbook that a great reference, highly recommended.

http://gpsa.gpaglobal.org/databook/

My copy is likely about 20 years old and it's one of my most frequent references. I have also added stuff to various sections over the years so it doesn't bear a lot of resemblance to the original document (it's now in 5 binders)

 

[5997]

thank you everyone again

 

[LeoChem]

There is a table you can use to facilitate your calculations

 

[5997]

Oh thank you.
While this thread still seems hot, I have an other question which I'm sure you would know.If I want to increase flow in the existing pump can I do that by increasing the discharge line dia?

 

[Montemayor]

5997:

Regarding your pump query: Are you a chemical engineering student or a non-degreed "engineer"?
The answer to this very basic question is NO!

 

[5997]

v=Q/A, so increasing dia of the line for a pump will actually decrease the flow rate.

 

[TD2K]

Umm, no. While your formula is correct, it's an incorrect conclusion.

A larger line might allow you to get more flow from a pump but it depends on the system.

Take a simple system where a centrifugal pump is delivering water through a long pipeline to an open area. As you increase flow through the pipe, you need more pressure at the inlet to the pipe for that flow. That pressure is supplied by the pump. The pump also produces less pressure as the flow increases. The flow rate through the system is where the pump produces the pressure required to flow the water through the pipe. If you search this site, you'll find discussions on pumps curves and system curves which further discuss this.

Take however a system where you have a control valve maintaining the flow through your pipe. If you had a larger pipe, the pressure drop through the pipe would decrease but since the control valve is keeping a constant flow, the control valve has a greater pressure drop across it. Bottom line, no flow increase.

Take a system where the pump is a positive displacement type of pump which puts out a fixed flow rate. A larger pipe will not increase the flow from the pump.

So, if the pump is a centrifugal pump and the system is maxed out, a larger line may increase your flow IF the piping is taking a significant portion of the total dynamic pressure drop.

 

[5997]

Oh thank you TD2K.