5 replies to this topic

[chemsep05]

I know that sometimes in practice the diameter of the column is smaller at the top of the column, is there some kind of way that you know when is a good time to change the diameter of the column at the top to a smaller diameter? Is there some type of rule of thumb, calculation or other way to decide when you can do this?
Thanks for the help

[Art Montemayor]

Chemsep:

If you are a Chem Eng Student with Unit Operations/distillation already under your belt, I am very surprised at your question.

Don't take my comment as a put down or a reprimand. My point is that your distillation prof or instructor should have taken you through the methodology of determining the diameter of a distillation column. This is done by using the Mott Souders - George Granger Brown Equation (or one of its variations):

Max. Allowable superficial velocity = K [(dL - dg)/dg]^0.5

where,

K = an empirical constant;
dL = Liquid density;
dg = gas (vapor) density.

Knowing the superficial velocity in that section of the tower, you can calculate the allowable diameter since you also know the gas or vapor rate going up the tower.

This equation is one of the really basic and great ones, developed over 75 years ago by two of the really great chemical engineers that ever used a slide rule. Souders went on to lead engineering in Shell Oil; Brown was (and has been) known as an outstanding academic in chemical engineering at Michigan.

Note that all you need to know is the vapor mass rate up the column and the liquid & vapor densities. The value of K is published in literature and references, depending on the application.

Neat.

[chemsep05]

We've done several projects on sizing columns, I was just wondering about the columns I see in practice that have a smaller diameter partway up the column, In the program Chemsep, you can specify the two diameters the same or the top to be smaller, My professor briefly went over that and didn't explain when you would do that, We're designing a chemical plant right now with 3 towers and I'm wondering if this could benifit my design

you can see in this picture what I'm talking about, http://www.gsnmagazi...Plant_Night.jpg

I wondering about the use of these columns
Thanks Again

[Art Montemayor]

I guess I failed in putting across my main point.

You are essentially stating that you leave the sizing of column diameters up to your computer program. What I've tried to teach you is that whatever else your professor is telling you, if you want to be a practical, thinking, and effective engineer instead of a computer clerk or technician, then you should revert back to the basis of how and why the diameters are determined the way they are. If you do not require the same diameter in your Rectification section (the Upper section, I believe), but rather a smaller diameter AND the economics and the structural requirements both show improvements in applying a smaller diameter, then by all means investigate that avenue. That is why I always make sure young engineers know the basis of the design and how/why the equations were derived and what they represent. When you know and understand the basics you actually dominate the application - regardless of whether you have a computer or not.

Regardless of what the stupid computer (here, I am not being insulting; the machine is indeed a stupid instrument. It only does what it is told) tells you, YOU are the engineer. Sometimes, the answer is very simple and practical: if you reduce the diameter, you may not be able to install the required tray manways, tower ladders, and platforms. If that is the case, then it is far more practical and utilitarian to employ the same diameter - regardless of what the computer program says.

My main point is this: chemical engineering students seem to be thinking that engineering design is all predicated and based on what a computer program (especially an impressive one, like HySys or Aspen) calculates. This is simply NOT the case. I predict that if you rely solely on computer printouts as an engineer, you will be a total failure in your chosen profession. You are supposed to know how (and when) to use your INGENUITY - that's the generic word for engineering and what is expected of us.

Whether a reduction (or expansion) in your tower diameters is better for your project depends on many factors. Some of them are:

1. Does it produce the specified product quantity and quality?
2. Does it do so in an economical and safe manner?
3. Does it lend itself to easy, safe operation and maintenace?
4. Does it meet the parameters of the project's Scope of Work?

And forget whether it looks good or is aesthetically attractive or fashionably "chic". Engineers are not expected or paid to be interior or exterior decorators. We are paid to make the thing work, work safely, and produce the maximum return on the investment made.

[siretb]

Art has explained it quite right. The governing parameter for the column diameter is the capacity-ralated factor What Art describes is equivalent. Let C=U * sqrt(rhog/(rhol-rhog)) U velocity, rhol and rhog densities; F=U sqrt(rhog) is also used very often, and is also capacity related.
The reason why many columns will have smaller upper sections than lower sections is usually because the vapor rate G (kilomoles/hour) is lower in the upper section than in the lower section. This is often because the feed, located in the middle is somewhat subcooled, so just obove the feed the real vapor rate G drops. Also, the liquid rate increases below the feed. so, in a sense, the overall traffic is heavier in the lower section than in the uppere section. This is not always the case.
In your case, because you want to do a design, you know the G (vapor rate) and L (liquid rate) at different locations. Calculate the capacity factor C , and see whether it is much different below and above feed(s). The rest is common sense.

[chemsep05]

I agree fully, computers are dumb, in my earlier courses we did all of the calculations by hand but, now we've been using hysys and it really rots your brain, I get the idea now why the top of the column would be a smaller diameter, Thanks for the help