Page 434 - Design and Operation of Heat Exchangers and their Networks
P. 434

Experimental methods for thermal performance of heat exchangers  417


                 If the axial dispersion coefficient approaches infinitely large values, that
              is, Pe¼0, the fluid temperature approaches a uniform distribution. This is
              the case of stirrers, for example. It is easy to obtain the real-time fluid tem-
              perature response:
                                         ð 1+ Ba 1 Þe a 2 τ    1+ Ba 2 Þe a 1 τ
                                                       ð
                          θ Pe¼0 x, τð  Þ ¼ 1                           (8.122)
                                                  ð
                                                 Ba 1  a 2 Þ
              in which a 1 and a 2 are the eigenvalues of the governing equations for Pe¼0
                              1
                                      ð
                      a 1,2 ¼   ½ NTU 1 + BÞ +1Š
                              2B
                               1  q ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2
                                        2
                                                          ð
                                   NTU 1+2Bð   Þ + 2NTU + 1 NTUBÞ (8.123)
                              2B
                 For B¼0, Eq. (8.122) reduces to
                                                NTU       NTU  τ

                            θ Pe¼0,B¼0 x, τð  Þ ¼ 1    e NTU + 1        (8.124)
                                               NTU+1
                 The effect of the axial fluid dispersion on the outlet fluid temperature
              dynamics is shown in Fig. 8.6. It illustrates that the initial temperature rise
              at the outlet of the heat exchanger is more sensitive to the axial dispersion.
              For the case that the value of NTU is about 2, the enlargement of the initial
              temperature rise due to the axial dispersion is the most significant, as is
              shown in Fig. 8.7. Only for small values of NTU, the axial dispersion effect
              can be neglected. If the working fluid is a liquid, the effect of the axial dis-
              persion becomes more significant. The outlet fluid temperature responses of
              the fluid with B>0 are given in Fig. 8.8.
                 The effect of axial heat conduction in the wall on the outlet fluid tem-
              perature dynamics is similar to that of the axial fluid dispersion. The differ-
              ence is that it does not change the initial temperature rise if the thermal
              capacity of the fluid is negligible, as is shown in Fig. 8.9.
                 The earlier solutions are the temperature responses to a unit step change
              in inlet fluid temperature. The same method can be used to obtain the
              dynamic responses to other inlet temperature variations such as the expo-
                              0
                                                              0
              nential variation θ ¼ 1 e  τ=τ ∗  and cosine variation θ ¼ 1  cos 2πτ=pÞ
                                                                       ð
              (Luo, 1998). The temperature response to an arbitrary inlet fluid tempera-
              ture variation can be calculated by means of the Duhamel theorem (Grigull
              and Sandner, 1990), which yields
                                   τ
                                   ð
                                                              0
                          θ x, τð  Þ ¼ θ U x, τ  τÞf τðÞdτ + θ U x, τÞθ 0ðÞ  (8.125)
                                      ð
                                                        ð
                                   0
   429   430   431   432   433   434   435   436   437   438   439