Shell and Tube Heat Exchanger Design 1

[Shockz]

Hello all, I have been recently tasked to design & manufacture and water to water shell and tube heat exchanger. I was wondering if anyone had any working knowledge of these devices and would explain were I must start. I am aware you need the air coefficient and you base most of your calculations around that value? Also does anyone know any supplier of heat exchanger parts as I am having little to no luck finding anything. Thank you.

 

[marty007]

The design of shell and tube heat exchangers can be a complicated process. It isn't too hard to find correlations for the tubeside heat transfer rates and pressure drop since it is simple flow in a pipe (with entrance/exit pressure drops). The shellside is a much more complicated issue however, and is dependent on many many factors (tube pitch, baffle spacing, baffle cut, ...).

In my experience, shell and tube heat exchangers are designed using commercial software. The two major packages out there I am aware of are ASPEN Exchanger Design and Rating, and HTRI Xchanger Suite. Both (if used properly), can take into account all of the issues I mentioned above.

For off-the-shelf exchangers, there are a few companies out there (Basco exchangers come to mind).

Again, in my experience (oil/gas, chemical), shell and tube exchangers are typically custom designed using the packages mentioned above.

 

[crshears]

Hmmm...water to water heat exchanger...air coefficient...what am I missing?

Is this for a one-off specific project or on behalf of a heat exchanger manufacturer who will mass-produce them? It makes a difference...

Heat exchanger parts? Not likely to find many, other than for the tubes, maybe; these would be available off the shelf, for the most part, depending on the required properties i.e. open or closed system [applies to both sides], pH, freshwater or salt, etc. In my experience most HXR manufacturers either have their own shops or outsource for the tubesheets, shells, end castings, and so on, so if this is for a one-off, I wish you good fortune finding parts.

 

[abeltio]

Speaking from experience, manufacturing shell and tube heat exchangers is the "science of drilling holes" in the tubesheet and baffles.

I would use Process Heat Transfer, Donald Q. Kern for a first approach at sizing it and have an idea of how big the HX is.

Then, give it to a professional. I've spent three years learning the details of designing and building S&THX and there are MANY design considerations that are not obvious to the untrained mind, to name a few:
01. which TEMA type is the most suitable?
02. materials?
03. corrosion protection?
04. hydraulic test flanges and covers?
05. supports? (vertical or horizontal installation) - consider space to pull tube bundle?
06. which fluid inside the tubes?
07. required location of nozzles?
08. how many tube passes? (affected by 07)
09. what is the allowable pressure drop?
10. paint? does it need to be zinc plated?
11. how much does the tube bundle weight? are skates to pull it needed?
12. risk of contamination? are double tube (bayonnet tubes) needed?
13. why am i doing all this instead of calling an expert?





saludos.
a.

 

[Shockz]

Thank you all for your input you've given me a lot to think about so far. And to marty007 I'm not familiar with those software packages and are they easy to use and how much?

And this is just a side project to get a better understanding of the S&T HX and it will be my first design project where I will be designing everything from scratch.

And my main concern would be where to start with regards to the mathematics? For instance, I have quite a few heat transfer equations and machine design equations and just curious where to begin really. The likelihood will be that I will be using water at a room temperature and having water leaving the reservoir and entering a pump which will leading it into the cylinder.

In the cylinder I hope to have water being heated by a heat mat, the water passing in the pipe will hopefully receive heat energy from the water in the cylinder and go back to the initial reservoir and heat that up and gradually over a period of time the temperature of the water will increase.

I know this would seem trivial to the likes of ye experienced individuals but this is my first design project and would like to gain valuable knowledge and experience by attempting this the right way from the start.
Thanks

 

[davefitz]

Schockz,
I think you might gain a lot of insight from the monte python skit re: how to play the trumpet like an expert: you just blow into one end and move your fingers up and down.

In the case of a heat exchanger design, you just go to engineering school , graduate after taking courses in heat transfer , fluid mechanics, and math, then work for a few years for a HX manufacturer.

For a water to water HX with no phase change, a shell and tube HX is not as competitive as a plate and frame HX.

"Whom the gods would destroy, they first make mad "

 

[marty007]

Shockz - Those software packages aren't cheap (>$5000 per year).

The basic theory is relatively simple, it's heat transfer across a pipe. It's the development of the convection coefficients and pressure drop on either side of that pipe that are non-trivial. Various research bodies have spent a lot of money and time over the years developing correlations (HTRI for example).

With regards to the mathematics, the commercial software packages take an approach similar to finite element procedures. They break the heat exchanger being analysed into small interdependent pieces that are then solved iteratively.

davefitz is right on with the plate and frame HX for a water-water applications, the design of these are a whole other can of worms. There are far fewer companies that fabricate plate and frame exchangers, and from what I understand, they keep their design procedures to themselves.