In 2000 the ENEA undulator U50 was moved from Frascati to Dresden, but up to now we had no the possibility to control the magnetic structure of this equipment for damages originating by shocks during the transport. Corresponding measurements are normally made point by point using a Hall probe. Thereby great care must be taken for a precise mechanics and for the minimization of nonlinearities and calibration errors of the probe. Such measuring technique is not available at our institute. To find new irregularities of the magnetic field B_y(z) arising from the transport we used the pulsed wire method [1, 2], which seems to be a much simpler technique than the Hall probe is. Parts of the equipment for this method and the software for data acquisition were delivered by ACCEL¹.

The pulsed wire method enables an in situ measurement of the first and second field integrals I₁(z) and I₂(z) of the magnetic structure depending on the current pulse form [1,2]. To measure these functions we used a CuBe wire with a diameter of d = 50 mm and length of L = 7500 mm. The tension was produced by a mass of m = 85 g, resulting in a tensile force of P = 0.83 N and a tension of T = 4.24·10⁴ N/cm². In a first attempt an additional calibration magnet outside the undulator was not applied. The table 1 contains the electric parameters of the measurements.

The measured integrals I₁(z) and I₂(z) in arbitrary units for a gap width of g = 25 mm are shown in fig. 1. The field component B_y(z) is the derivative of I₁(z) and was calculated by the data acquisition program.

The fig. 2 allows a comparison with the results obtained at ENEA with the Hall probe technique before the transport to the FZR was carried out. In general one can conclude that no additional field errors arised.

Fig. 1 First field integral I₁(z) (part b) and second field integral I₂(z) (part c) measured with the pulsed wire method. The derivative B_y(z) was not measured.

Fig. 2 The field component B_y(z) in the middle plane (a) measured with the Hall probe technique. The functions I₁(z) and I₂(z) are obtained by integration of B_y(z).

Finally, fig. 3 shows the second field integrals without and with an application of a homogeneous external field of B = 3.3 G to correct the behaviour of this function. A bending effect can clearly be seen, which will help to guide the electron trajectories through the waist within the optical resonator. More quantitative investigations will be done in near future.

Fig. 3 The integral I₂(z) obtained by the pulsed wire method at a gap of g = 30 mm without (a) and with an additional homogeneous field (b) of B = 3.3 G in the exit part of the undulator.

¹⁾ACCEL Instruments GmbH, Bergisch-Gladbach

References

[1]	R.W. Warren, Nucl. Instr. Meth. A 272 (1988) 257
[2]	A. Geisler, M. Ridder and Th. Schmidt, Delta-Rep. 94-8,
	University of Dortmund, Dept. of Accelerator Physics, unpublished

 IKH 05/22/01 © P. Gippner