Thermal Shock in Reboilers?

[Basil]

On one of our units, we currently have a reboiler which uses tower bottoms (T=700 F tube side) as the heat source for the reboiled process (shell side).

The issue we are having is leaks on the exchanger body flanges during start-up. The root cause (I believe) is that no reboiled process exists in the shell until adequate temperature is in the column to get the process to draw-off. So you have 700 F material on the tube side, and nothing on the shell side until it all of a sudden starts coming in. I believe the thermal shock of receiving the "cold" reboiled process into the hot process creates the leaking we are seeing.

I have not seen this type of thing happen with the steam reboilers we have elsewhere, probably because the temperature gradient isn't as great.

Have other people seen this problem during this time frame of start-up? Does anyone have any other theories? Any solutions? Maybe going to B16 bolts or another gasket type?

 

[25362]

Large delta T's aren't desirable for a variety of reasons:

- mechanical thermal effects that may distort the body of the tubular or its internals. If an expansion bellows is provided on the shell, it may adopt the shape of a banana when some tubes are heated more than others;
- may promote stratification of warmer and colder fluid fractions.
- may reach critical (film boiling) boiling conditions.
- may induce fouling.

This is very general, to be able to analyse your situation a better description of the type of unit you are referring to, the number of tube passes, the shell side arrangement, the tube pitch, etc., etc., as well as some information on the fluids concerned should be given. Then, probably one of the experts reading the details may give you good advice.

 

[JOM]

Basil,

There was case of a reboiler at an Australian natural gas plant (1998) which was leaking due to abnormal temperature differences.

It was not during startup, but during a major process upset that introduced very cold temperatures into one side of the exchanger. The temp differential caused warping of the end flanges, leading to severe leaking, giving maintenance a very difficult problem.

I think "thermal shock" really refers to a sudden increase in temp differential leading to increased and unresolved stresses in the steel. Cheers,
John.

 

[davefitz]

Sound like yu have guessed the right cause.

As a solution, the unit would need to be repiped with added flexibility added. One expensive method is to provide flexibility by "twisting " the tube array ( as was done at the Clinch River heat exchangers ), if opposed tubesheets are required. But the conventional solution might be to use a U-tube designed exchanger with a single tube sheet.