Siren Design and Evaluation in Mud Flow Loops and Wind Tunnels - Measurement While Drilling Signal Analysis, Optimization and Design - page 240

are the ones transmitted to the surface, the hydraulic component must be
eliminated to ensure accuracy in characterizing the acoustically significant 'p.
We had previously suggested the use of the gray colored “return section”
shown in Figure 9.3 to provide a closed-loop system useful in lost circulation
jamming evaluation. There is actually no reason why a second siren (with its
own flow straightener) driven by a second motor cannot be installed in this
section. If differential pressure transducers are used to measure 'p in each case,
the measurements at each siren are not affected by the presence of the other
siren. This test mode provides for some time and labor efficiencies since the
volume flow rate Q is measured only once.
9.2.1 Siren torque testing in short wind tunnel.
Two important properties are important for high-data-rate mud pulse
telemetry. First, because the siren must stop, start, increase or decrease rotation
speed very often, the stopping and starting torques associated with turning
should be low - or at least low enough that the driving electric or hydraulic
motor does not find it problematic. Be aware that torque increases quadratically
with flow rate. Mud pumps used in the field do not pump at constant speed.
Thus, unplanned flow rate increases downhole may prevent a marginal siren
from working at all - in fact, a “stable closed” siren will block the oncoming
mud flow, prevent turbine operation, and build up dangerous pressures in the
drillstring. Thus, torque measurement, and siren design for low torque, are two
important objectives for short wind tunnel use. The second siren property
important to high data rate is strong 'p signal strength. Because carrier
frequencies need to be high, attenuation will be an important design
consideration - strong 'p signals are required so that pressure waves can be
detected faraway from the source with minimal error. We will discuss testing
methods and telemetry design approaches in the context of “intermediate” and
“long” wind tunnel testing later. As we will discuss in Chapter 10, nature may
limit the maximum 'p that can be created for a given siren pulser. Fortunately,
this can be increased by clever use of telemetry schemes using constructive
interference, by using sirens-in-series, and so on.
In Chapter 7, we discussed the “stable open” or “stable closed” nature of
many siren designs. Sirens with rotors upstream tend to be stable closed and
possess restoring torques that are strong when an attempt is made to hold the
rotor open - simply try holding open such a rotor in the wind tunnel !
Downstream rotor configurations may be stable-open, stable-closed, and often,
partially-open. The restoring torques which tend to keep closed rotors in the
closed position tend to be much smaller than those for upstream rotor designs.
All of the comments in this paragraph are based on actual empirical observation.
We will focus on downstream siren rotor designs, that is, the design developed
by this author in the early 1980s currently used in commercial MWD tools.
Figure 9.4a provides a schematic of such a siren where, for clarity, the outer