Page 64 - Instrumentation Reference Book 3E
P. 64

Shop-floor viscometers 49
            Newtonian fluids is very similar to (2.1 1) and is in   so that the viscosity 7 is determined from
            fact given by
                                                         7 = Mgth/xA                   (2.16)
                    2
                c = 3 7ra3-Y71(”i)            (2.12)   Clearly, this technique is only applicable to “stiff”
                                                      systems, i.e., liquids of very high viscosity.
             where the shear rate y is given by
                Y = RoiOo                     (2.13)   2.4  Shop-floor viscometers
             which is (approximately) constant throughout the
            test fluid, provided  00 is small (<4”, say). This is   A number of ad hoc industrial viscometers are very
             an important factor in explaining the popularity   popular at the shop-floor level of industrial prac-
             of cone-and-plate flow in non-Newtonian  visco-   tice and these usually  provide a very simple and
            metry.  Indeed,  it  is  apparent  from  (2.12)  and   convenient method of determining the viscosity of
             (2.13) that  measurements  of  the  torque  C as  a   Newtonian  liquids.  The  emphasis on  the  “New-
            function of rotational speed 00 immediately yield   tonian”  is  important  since  their  application  to
            apparent viscositylshear-rate data.       non-Newtonian  systems  is  far  less  straightfor-
              Sources  OF  error  in  the  cone-and-plate  vis-   ward  (see,  for  example,  Walters  and  Barnes
            cometer have been discussed in detail by Walters   (1980)).  Three  broad  types  of  industrial  visc-
            (1975) and Whorlow (1980). Measurements on all   ometer can be identified (see Figure 2.7). The first
            fluids are limited to modest shear rates (< 100 s-l)
            and this  upper  bound  is significantly lower  for
            some fluids like polymer melts.             TYPE       1   EXAMPL€
              Time-dependent effects such as thixotropy  are
            notoriously difficult to study in a systematic way.   ROTATIONAL
                                                          DEVICES
            and the constant shear rate in the gap of a cone-                U
            and-plate viscometer at least removes one of  the
            complicating  factors.  The  cone-and-plate  geom-
            etry  is  therefore  recommended  for  the  study  of   FLOW  THROUGH
            time-dependent effects.                     RESTRICTIONS
              For the rotational viscometer designs discussed
            thus fa, the shear rate is fixed and the correspond-     ROLLING   FALLING   RISING
            ing stress is measured. For plastic materials with a   FLOW  AROUND   BALL   BALL   BUBBLE
            yield stress this may not be the most  convenient   OBSTRUCTIONS
            procedure, and the last decade has seen the emer-
            gence of constant-stress devices, in which the shear
            stress is ccintrolled and the resulting motion (Le.,   Figure 2.7  Classes of industrial viscometers
            shear rate:i  recorded.  The  Deer  rheometer  is the
            best  known  of  the constant-stress devices and at   type comprises simple rotational devices such as
            least three versions of such an instrument are now   the Brookfield viscometer, which can be adapted
            commercially available. The cone-and-plate geom-   in favorable circumstances to provide the appar-
            etry is basic in current instruments.     ent viscosity of non-Newtonian  systems (see, for
                                                      example,  Williams  (1979)).  The  instrument  is
            2.3.4  Parallel-plate viscometer          shown  schematically in  Figure 2.8. The  pointer
                                                      and  the  dial  rotate  together.  When  the  disc  is
            In the  parallel-plate  rheometer,  the  test  fluid  is   immersed,  the  test  fluid  exerts  a  torque  on the
            contained  between  two  parallel  plates  mounted   disc.  This  twists  the  spring  and  the  pointer  is
            vertically;  one plate  is  free to  move  in  the ver-   displaced  relative  to  the  dial.  For  Newtonian
            tical  direction,  SO  that the  flow is  of  the plane-   liquids (and for non-Newtonian liquids in favor-
            Couette  type  and  approximates  that  associated   able circumstances) the pointer displacement caE
            with  Newton’s postulate.                 be  directly  related  to  the  viscosity  of  the  test
              A mass ilid is attached  to the moving plate  (of   sample.
            area A), and this produces a displacement x of the   The  second  type  of  industrial  viscometer
            plate in a time t. If the plates are separated by a   involves what we might loosely call “flow through
            distance h, the relevant shear stress T is given by   constrictions”  and  is  typified  by  the  Ford-cup
                T  = Mg/A                     (2.14)   arrangement. The idea of measuring the viscosity
                                                      of a liquid by timing its efflux through a hole at
            and the shear rate 7 by                   the bottom of a cup is very attractive. It is simple
                                                      to operate, inexpensive, and the apparatus can be
                y = x/th                      (2.15)   made  very  robust.  Historically.  the  cup  device
   59   60   61   62   63   64   65   66   67   68   69