F-shell is not a good choice from a heat transfer performance point of view - unacceptable leakage across the baffle in actual plant operations. Usually the baffle is damaged after maintenance which involves pulling out the tube bundle for cleaning. Choose more shells hooked up in series on the shell side to get the desired temp approach. In each shell, temp approach should be such that Ft, the LMTD correction factor, is >0.8

There are no limitations that I'm aware of for design pressures when using a TEMA F shell with a longitudinal baffle. The only limitations I've heard of are

1) F shells should not be used when the shell-side pressure drop exceeds around 7 psi due to the potential for shell-side bypass.
2) F shells should not be specified for tube bundles that are extracted frequently for cleaning due to the potential for damaging the sealing devices during bundle extraction/installation.
3) F shells should not be specified for applications with a large shell-side temperature range due to the potential for thermal leakage through the directly baffle (this one is probably debatable).


-Christine

Thought about this a bit more...

4) F shells should not be used for exchangers that require very small surface areas (e.g. < 200 sq. ft.), as hairpin exchangers are preferred here.
5) F shells shouldn't be used for heat exchangers that have only 1 tube-side pass as this negates the 2-for-1 advantage of using F shells.


F shells also work best in exchangers where the shell-side heat transfer coefficient is controlling (they can markedly improve the shell-side heat transfer coefficient) and when there are only 2 tube-side passes (pure countercurrent).


-Christine.

Quote (Christine74)

4) F shells should not be used for exchangers that require very small surface areas (e.g. < 200 sq. ft.), as hairpin exchangers are preferred here.

Weirdly enough, I recently sized a heat exchanger for a cryogenic application with a BFU designation. Total SA < 200 sq ft but this was chosen because;

V. low water temperature (dT of 4 degC)
Desired Ho >> Hi to control tube skin temperatures.

I very briefly looked at a hairpin unit but struggled to raise skin temperatures above -120 degC at the cold end and the resultant steady-state ice thickness may have resulted in the flow path being fully blocked on the shellside.