Kaman KD-5100 - Manual (Page 37)

37
Application Variables and Caveats
There are some application variables that will also affect performance. The effects listed are not
considered in the results of this tech note. These effects must be considered separately.
Sensor Loading: Sensor loading by conductive materials that are incidentally in “view” of the
sensors can affect the results significantly and must be considered on a case-by-case basis. The
discussions in this tech note assume that incidental materials do not load the sensor.
Cosine Error: This is error that occurs when the target movement is from tilting as in a fast
steering mirror assembly and can become significant at larger measuring ranges.
Cross Axis Sensitivity Errors: This error occurs in 2 axis measurements of tip and tilt (common
in fast steering mirrors) and can become significant at larger measuring ranges.
Setup Error: The system can become very sensitive to the null gap position when setup for very
small ranges. Changes in the null gap will affect both the sensitivity and temperature coefficient.
Usually only significant problem on ranges of <
± 10mils (< ± 0.25mm).
Target Effects: Eddy current sensors are significantly affected by the target material resistivity
and permeability. Aluminum targets are considered in this tech note. In general the KD5100 is
best used with non-magnetic (relative permeability --
µ
r
=1), low resistivity targets. The effect
this has on the sensor is dependent on the operating frequency and coil diameter.
Cable Length: The sensors used by Kaman are passive. This means that the cable is an integral
part of the sensor and affects the measurement performance. The data presented assumes a 2
meter cable length. Actual results may vary with different cable lengths. Long cables are
especially bad for performance because they degrade the effective Q of the sensor and increase
the inherent temperature coefficient of the sensor coil. Long cables will also cause problems
with thermal drift and variations in the output caused by cable movement.
Optimization: In the data presented the circuit was optimized for each range. This means that
the component values in the circuit may be different for each specific measuring range. The
system will not get the performance shown simply by changing the range and recalibrating – it
would require factory optimization. Also you could optimize on a specific parameter (say
temperature) and achieve better performance in that category – allowing the other performance
parameters to be worse. The data shown for a specific range is a compromise of all the
performance parameters. At each measurement range listed the linearity, temperature
coefficient, and sensitivity listed are achievable simultaneously.