Page 282 - Fluid-Structure Interactions Slender Structure and Axial Flow (Volume 1)
P. 282

PIPES CONVEYING FLUID: LINEAR DYNAMICS I1              263


               (2) w3







               21
                          u = 2.0


                         -8
                  20
                                19



















             Figure 4.36  (a) Variation  of  the  direct  receptance  all for  a  theoretical  harmonically  forced
              system  with  u = 2  and  varying  values  of  w  (nozzle  area parameter  a, = A/Aj = 3, j3 = 0.203,
              y  = 5, zero  dissipation, and  N  = 4). (b) Same  as  in  (a),  but  u = 2.2, 2.4, 2.6,  2.8  and  3.0.
              (c) Inverse  direct  receptance  at  6 = 0.15,  close  to  the  instability  boundary  of  the  experimental
             system  (a, = 1.27, ,8  = 0.387, y = 250) for  a  number  of  values  of  the  flow  parameter,  R, and
                     varying  frequency parameter, f, defined in the text (Bishop & Fawzy  1976).


              oscillate at its critical frequency, wcf  = 15. When w < w,f, energy flows from the pipe to
              the driving mechanism, and the displacement leads the excitation - which is not possible
              for passive systems. The phase lead continues until w = w,f, when no energy flows to or
             from the driving mechanism, since all the energy required to achieve an infinite amplitude
              is supplied solely by the fluid. For w  > mcf, the pipe is forced to oscillate more rapidly,
              and the displacement lags behind the force; hence the receptance curve is now below the
             real axis.
                A  number  of  other,  special  and  interesting  features  of  these  receptance  curves  are
             discussed  by  Bishop  & Fawzy,  among  them:  (i) the  vanishing  of  the  receptance  at  a
             finite  w; (ii) the  migration  of  the  starting point  of  the  receptance  curve  along  the  real
              axis towards  the origin.  The first point  suggests that  the system may have  some purely
              imaginary  antiresonance  eigenvalues  - which  means  that  these  are  the  resonances  of
              a  system  with  the  excitation  point  (at  the  downstream  end  of  the  pipe)  constrained  to
              zero,  i.e.  those  of  a clamped-pinned  pipe.  Indeed,  it  is  known  that  the  eigenvalues  of
              the  clamped-pinned  system  are  purely  imaginary  up  to  a  critical  value  (in  this  case
   277   278   279   280   281   282   283   284   285   286   287