Development and operating experience of a 1.1-m-long superconducting undulator at the Advanced Photon Source

Development of superconducting undulators continues at the Advanced Photon Source (APS). Two years after successful installation and commissioning of the first relatively short superconducting undulator “SCU0” in Sector 6 of the APS storage ring, the second 1.1-m-long superconducting undulator “SCU1” was installed in Sector 1 of the APS. The device has been in user operation since its commissioning in May 2015. This paper describes the magnetic and cryogenic design of the SCU1 together with the results of stand-alone cold tests. The SCU1’s magnetic and cryogenic performance as well as its operating experience in the APS storage ring are also presented.

Figure 1

Superconducting undulator SCU1 in the APS storage ring.

Figure 2

SCU0 undulator magnet core.

Figure 3

SCU1 undulator upper magnet core. Core components: core body (1); magnetic poles (2); magnetic poles with the end tabs (3); G10 side spacers (4); spring pins (5); conductor turn around pins (6). The flanged pipe is for liquid helium cooling.

Figure 4

Winding scheme of the SCU1 magnet. There are 53 conductors in a full main coil pack (blue), 38 main coil conductors in the second to last grooves, and 15 main coil conductors in the end grooves. There are 15 conductors (red) wound over the top of the main coil to form the corrector coil in the last two grooves at either end.

Figure 5

Helmholtz-style coil to correct the dipole vertical field along the length of the device.

Figure 6

Dipole coils installed upstream and downstream of the undulator magnet inside the cryostat that can be used to compensate the second field integral.

Figure 7

Temperatures of the SCU1 components during the cooldown process: the 1st stage of 4-K cryocooler (blue); beam chamber (red); LHe tank (yellow), and the magnet top core (green).

Figure 8

Quench currents during training of cores.

Figure 9

Magnet excitation curve.

Figure 10

Measured undulator field profile at magnet current of 450 A.

Figure 11

Measured phase errors as a function of magnet current.

Figure 12

Measured first and second field integrals during quench test.

Figure 13

First week of SCU1 operation. The ring was in top-up mode at 100 mA.

Figure 14

Comparison of SCU recovery after a beam-dump-induced quench, as measured by the LHe tank pressure. Prior to the quench, the average magnet temperature was 4.23 K for SCU1 and 4.34 K for SCU0, and the main coil current was 450 A for SCU1 and 668 A for SCU0.

Figure 15

Measured vertical position of the vacuum chamber inside the cryostat over a one-year period using the beam-based method.

Figure 16

Magnet core temperatures before and after a quench, measured by six (SCU1, left) and four (SCU0, right) temperature sensors on two magnet cores, shown in different colors. Prior to the quench, the main coil currents were 426 A for SCU1 and 666 A for SCU0.

Figure 17

Vertical-field integrals of initial correction as measured using the stored beam, before final beam-based adjustments. The first integral is on the left and the second integral is on the right.

Figure 18

Effect of induced quench on the beam orbits at all the insertion devices (IDs). Left panel: Slow orbit correction only. Right panel: Fast orbit feedback (FB) turned on.

Figure 19

Measured odd-harmonic SCU1 tuning curves (450 A maximum) of monochromatic flux through a $0.5\times 0.5{\mathrm{mm}}^2$ aperture at 27.5 m compared with those of 11-mm minimum-gap permanent magnet devices U33 (continuous tuning curve envelope) and U23 (even and odd harmonics).

Figure 20

Comparison of measured and expected SCU1 flux.