There are no rules for this - only scope definition, need, engineering criteria, and horse sense. For example, if you have a high pressure fluid transfering with a low pressure fluid, it makes engineering sense to put the HP fluid in the tubes - since you know that the hoop stress in the tube is more resistant than that of the shell because of the diameter difference.
When you have a condenser application, you certainly want to know if you're going to be draining - as in a total reflux condenser. If so, then the shell side is the obvious answer for the vapor when you have multiple tube passes.
If you have a corrosive fluid, it normally goes in the tube side because it comes in contact with less exotic metal than if if it were in the shell side. This is common sense and makes for a more economical solution.
There are also design questions on some of the more exotic TEMA configurations - for example, the shell side is the answer for putting the process fluid if you need an extremely low condensing pressure drop - as in vacuum operation. The pressure drop requiement is also a factor in a natural circulting thermosyphon reboiler. A one-pass tube side is the choice for the process fluid rather than the shell side.
I hope these examples explain to you how these decisions are made without fast and hard rules. I like it better this way because it makes you use your ingenuity - which is what engineering is all about.
If you have access to a copy of the TEMA standards, they give some guidelines for what fluid should go on the tube shell versus the shell side. If you go through it, each recommendation seems reasonable on its own (eg. higher pressure fluid should go on the tube side)
But as Art alludes to, you'll find in real life one fluid is recommended for one side by a guideline and the next guideline can say it should go on the opposite side (drat those fluids). There's no cookbook I've seen that takes all of the experience factor out of the work.
At that point, you (and others) will need to weigh the pros and cons based on experience and plant practices (and success/failure in that plant ;-)) to decide which side(s) you want to assign to which fluid(s).
There is an article in the March 1998 issue of Chemical Engineering Progress that discusses the allocation of sides. It was written by Rajiv Mukherjee and is entitled "Broaden your Heat Exchanger Skills". He mentions the consideration of: viscosity,corrosiveness,fouling,pressure,flow and temperature range.
I have one additional comment if you are designing for a fixed tube sheet exchanger. If there is a low allowable pressure drop problem,use the tube side and use one tube pass. Then you are not fighting turning losses that use up pressure drop.
To Srfish Gulley, allow me to clarify that a one tube pass unit is not necessarily a fixed-tube type, it can also have a floating head. I think it is types XXS or XXT.
I would only like to add that the fouling and cleanliness of the fluid should be given strong consideration. This fluid should always be placed on the tubeside. The heat exchanger should also probably be of single pass design, removable heads and with tubes 0.75" or larger....
Also, as a general rule, whenever there is a phase change of the liquid (vapor to liquid or liquid to vapor) place that stream on the shell side. (The exception to this rule is tank/reactor mounted condensers)
Common sense (horsesense ?) is the best guide....
Please download the helpful guide offered on the YUBA Heat Transfer website......
To MJC, I've seen lots of horizontal S&T vaporizers with the vaporizing stream in the tube side. For example, furfural recovery mixed streams in lube units, in which the solvent tends to polymerize and coke up; the shell side with hot oil or with steam.
