FGM PDR

1
Automated Absolute Measurement based on Rotation
of a Proton Vector Magnetometer
Volker Auster, Ronald Kroth, Olaf Hillenmaier,
Markus Wiedemann Magson GmbH, Berlin, Germany

Different Approach compared to Auto-DIF and
GAUSS
2
Fundamentals

• Requirement on ABSOLUTE measurement
• Field vector has to be embedded in geographical
  reference system
• Elimination of systematic errors
•  
• Fundamentals of the proposed methods
• A PVM measurement is an absolute measurement of
  the field component directed to the magnetic axis
  of the coil.
• Two components and the total force F are
  theoretically sufficient for a unique
  determination of the field vector.
•  
• The problem
• How do we know whereto the magnetic coil axis is
  oriented?

3
Our Approach to rotate a PVM

PVM equipped with telescope, CCD and level tube
will be rotated by motor and worm gear about
vertical axis. Measurements are done each 30
measured precisely by angle decoders. Close to
the nominal azimuth mark direction pictures are
taken.
4
Orientation vs. vertical direction

• Inclination of rotation plane versus horizontal
  plane is determined by tilt sensor measurements
  (resolution 0.5arcsec, offset of tilt meter can
  be neglected due to rotation)
• Inclination ? of coil axis versus rotation plane
  can be calculated from constant term A0 of sine
  fit

5
Orientation vs. azimuth mark

• Telescope, digital camera and image processing
  shall be used to determine the orientation with
  respect to the azimuth mark
• Advantage of this method compared to laser beam
  options are the flexibility regarding targets and
  distances
• Misalignment of optical axis versus mechanical
  and magnetic coil axes have to be considered.

• Both alignment errors can be determined by
  rotation of the bias coil
• Optical axis error will be adjusted (only
  horizontal one is relevant)
• Misalignment of magnetic axis will be rotated
  into vertical direction, which can be eliminated
  by the rotation of the table
• Due to the long cylinder geometry of bias coil
  and telescope, the axes differences are constant
  (proven over more than one year operation)

6
The rotation table
High mechanical precision required

• Worm Gear
• globoid structure to keep the contact as large as
  possible
• Slide bearing
• Both parts are made from AlZnMdCu1.5
• Both parts manufactured on the same 4 axes CNC
  machine
• Surface of bearing anodized with Teflon
  contribution
• Bearing clearance 0.01mm!

Excentricity error in arcsec
7
Drive unit and angle measurement

• Rotation angle is measured by two decoders
• (1) on the worm axis
• Res 1/1120000 (1.16 arcsec)
• (2) directly at the table
• Res 1/2400000 (0.54 arcsec)
• Error sources
• Worm gear
• Excentricity
• Decoder accuracy
• Error determination
• PVM replaced by standard Theodolite (magnetic)

8
The optical system

9
The control unit / software

10
Results of the measurement

Amplitude 0.16nT Phase
378,90 Mean Value 48957,17nT
11
Results of the measurement

Amplitude 18468,92nT Phase
-1,98 Mean Value 1083,23nT
12
Results of the measurement

Mean Value 0.00nT St. Deviation 1,02nT
13
Results of the measurement

14
Summary

• Basis for the presented automated system are
• Well proved PVM design (homogenous coil for bias
  fields),
• Experiences with manually performed absolute
  measurements in various configurations (e.g. on
  top of a Zeiss Theodolite) since 1980.
• High precision mechanics (Steingross
  Feinmechanik, GER)
•  
• High reliability and comparable low costs can be
  achieved by usage of well tested commercial
  parts Industrial PC, Industrial Camera, Tilt
  sensor (Fredericks, USA), DC Motor (Faulhaber,
  GER), Incremental decoder (Kuebler, Numerikjena,
  GER)
•  
• Comparison with Auto DIF and GAUSS
• The PVM measurement has the advantage that
  components are measured directly and the total
  force F is already included.
• The optical system is more flexible compared to
  laser approaches.
•  
• Acknowledgment Our Mechanical Engineers