The numerical examples presented in the preceding sections are for the case where the cell effectiveness is constant. This implies that the number of transfer units of a cell NTU_c is constant. To accommodate the shell nozzles, the designer is obliged in many cases to use a spacing between the baffles and the tube sheets which is different from the spacing between the neighboring baffles. In such cases, and in many other cases, the overall heat transfer coefficient and the cell heat transfer area are different in different cells. This leads to different numbers of transfer units NTU_c and different cell effectiveness values E_c. However, the calculation procedure does not change.

The cell method was applied in Section 110 for one of the most common shell configurations (TEMA: E shell) and extended in Section 116 for the case where the shell is provided with a longitudinal baffle (TEMA: F shell). The cell method can be applied equally well for other shell configurations with segmental baffles e.g., shells with divided flow (TEMA: J shell). The heat exchanger may be divided into a number of cells; the inlet and outlet temperatures in the different cells are then related to one another, as in the cases discussed, by using the information obtained from the flow configuration in the layout of the heat exchanger considered.

The cell method can also be applied to predict the thermal performance of a shell-and-tube heat exchanger with segmental baffles, when leakage and bypass streams are present. This requires however a knowledge of the leakage and bypass streams from hydrodynamic considerations as well as a knowledge of their influence on the shell-side heat transfer coefficient. An evaluation of the influence of leakage streams on the thermal performance of shell-and-tube heat exchangers for different ratios of the leakage stream to the main shell-side stream has been carried out by Gaddis and Schlünder (1976).

NOMENCLATURE FOR SECTION 1.6.12