Siren Design and Evaluation in Mud Flow Loops and Wind Tunnels - Measurement While Drilling Signal Analysis, Optimization and Design - page 253

Next consider the single transducer shown at the bottom of the tubing,
again, for the top drawing of Figure 9.9b. This transducer will measure the
initial signal plus all reflections and reverberations. It will be extremely difficult
to understand the meaning of this measurement. However, since the single
transducer location “x,” say x = - L/2 as suggested earlier, is known, along with
values for Z, L, c and the measured pressure p 1 (- L/2,t), it is a straightforward to
calculate the source strength p s from Equation 3-A-14. This p s should equal the
value obtained from the above differential pressure measurements.
The middle drawing in Figure 9.9b also shows a speaker test, now attached
to a squirrel cage blower which is not blowing. Again, we have an open-open
system. Note that the blower basically changes the length of the wind tunnel.
Thus, if a single transducer test is performed when a blower is attached, the
proper physical source location of “x = 0” must be determined before Equation
3-A-14 can be properly used. One would need to adjust the location of the
speaker (by trial and error) until the p s value determined using Equation 3-A-14
agrees with that obtained from differential pressure measurement.
In the bottom drawing in Figure 9.9b, a siren replaces the speakers used
earlier. The stator-rotor gap location should be identified as the “x = 0”
determined in the above paragraph. Unlike simple speaker tests, a number of
fluid-dynamical complications arise. First, a swirling vortex flow will be found
downstream of the rotating rotor. Thus, it will be necessary to place flow
straighteners as shown, in order to remove the strong pressure oscillations
associated with this flow - only then will differential pressure measurements be
accurate. Single pressure transducer measurements, for this very same reason,
should be obtained upstream of the siren stator in order to minimize vortex
noise. Flow straighteners may also be required at the blower outlet to reduce
wind noise, as in Figure 9.3. In both cases, because wind is flowing, background
turbulent noise will be found. Equation 3-A-14 is used in the same manner as
before. Again, redundant methods for p s signal strength determination provide
error checks, and neither method alone is likely to provide absolute accuracy.
We now assume that the engineer has developed a suitable test matrix of
parameters. For example, this might include two-lobe, three-lobe and four-lobe
sirens operated at different flow speeds and frequencies. In addition, each siren
prototype may be evaluated with different taper angles, rotor-stator gaps,
different convergence and divergence angles in the central hub, different rotor
circumferential clearances with the collar housing, and so on. Each of these
configurations should also be evaluated in the short wind tunnel for torque and
erosion tendencies. Wind tunnel data should be extrapolated to field conditions
for muds flowing under actual pump speeds, in order to determine if electric or
hydraulic motors can be found for the design. A good acoustic test procedure is
suggested below which addresses a number of key engineering questions which
arise and summarizes the steps required to fully characterize the siren.