|
|
|
A second RF cryomodule in the SOLEIL storage ring!
|
|
|
This second cryomodule, containing two superconducting accelerating cavities, was delivered to SOLEIL on May 20th, 2008. With both cryomodules in operation, it will be possible to increase the intensity of the electron beam in the storage ring from 300 mA up to 500 mA.
|
 |
 What is the SOLEIL cryomodule and what is it for?
|
|
The SOLEIL cryomodule is a cryostat containing a pair of superconducting accelerating cavities operating at a frequency of 352 MHz. The cavities are made of copper with a niobium coating on their inner surface. They are kept in a bath of liquid helium at 4.2 K, at which temperature niobium is a superconductor (resulting in negligible dissipation of RF power in the walls).
In the SOLEIL storage ring, the two cryomodules will supply a 4.4 MV accelerating voltage, and 575 kW of RF power (at 352 MHz) at the maximum current of 500 mA, so as to procure a long lifetime to the electron beam and to compensate for its energy losses in form of synchrotron radiation.
Each of the four accelerating cavities is powered by a 180 kW solid state amplifier at 352 MHz, consisting of 727 elementary modules of 315 W (link to RF amplifiers). The RF power is transferred through waveguides and fed into the cavities by coaxial antenna-type power couplers (Fig.1). 
Figure 1 : Section view of the SOLEIL cryomodule containing a pair of superconducting cavities One of the challenges for a high-current storage ring such as SOLEIL (500 mA) is to maintain a very intense electron beam with extremely small dimensions (longitudinally and transversely) and in very stable conditions both in position and energy. These qualities of the stored electron beam are essential to guarantee the brightness of the radiated photon beam.
At high current, the excitation of higher order modes (HOM) of the accelerating cavities can generate instabilities of the electron beam. In order to eliminate such a problem, the impedances of these parasitic modes are strongly attenuated by four HOM couplers which consist in superconducting loops coupled to coaxial lines terminated by 50-ohm loads(Fig. 1).
|
|
'In house' design of the first cryomodule
|
|
In 1996, during the study phase of SOLEIL, the decision was made to develop a new concept of superconducting cavities with strong attenuation of the HOMs. A prototype cryomodule containing two superconducting cavities was built in the context of a joint CEA-CNRS-CERN project. Initial tests of the prototype with RF power but without beam loading were carried out at CERN in 1999. It was then tested on the ESRF storage ring at Grenoble, where it contributed to the storage of a 170 mA electron beam at 6 GeV, providing an accelerating voltage of 3 MV and a power of 190 kW per coupler. Late in 2002, based on these results, this prototype was chosen to be the cryomodule 1 (CM1) for SOLEIL, after bringing in a few improvements. It was sent back to CERN for modification, where it was completely disassembled, and then reassembled and tested (2003 - 2005). Finally, after testing at performance levels far above the specifications (2.5 MV per cavity and 200 kW in total reflection on each coupler), it was taken to SOLEIL and installed in the storage ring in November 2005 (Figures 2 and 3).
The set-up process went smoothly, and the current could be increased quite quickly during the commissioning of the ring. | 
Figure 2 : Using a craving crane to move the first cryomodule into position in the storage ring at SOLEIL (November 2005)
| 
Figure 3 : The first cryomodule installed in the storage ring
(January 2006). |
|
|
And here is the second cryomodule…
|
| In parallel with the setting up of the first cryomodule, a contract was signed in 2005 with the German company ACCEL for the manufacturing of a second cryomodule: the twin brother of the first one.
The CM2 was built and assembled in a clean room at ACCEL. Low-power tests of the cavities, in a vertical cryostat, were carried out on the test bench at CERN. CM2 was delivered to SOLEIL on May 20th 2008 and immediately installed in the storage ring (Fig. 4). After installation inside the tunnel, several actions are required before the electron beam can again circulate through the ring: connecting the waveguides to the power couplers, connection, pumping and baking out of the transition vacuum chambers, pumping of the superconducting cavities, cleaning and conditioning of the cryogenic lines which feed in liquid Helium and liquid Nitrogen and recover the cold gases, and then cooling down, using the cryogenic plant which supplies both cryomodules. Once cooled down to 4.2°K, the CM2 will be 'de-tuned' in order to allow the electron beam to pass through without significant perturbing interactions.
The next step is RF conditioning, which consists in transferring some RF power through the couplers of both cavities. This must be done very gradually, because the intense electrical fields can produce electron emissions from the various surfaces, mainly at the power couplers, bringing the cavity vacuum pressure up. This RF processing can require up to several weeks for each of the 2 couplers. The last step is the increase of the power transmitted through the couplers in the presence of the beam.
It is only once the second cryomodule will be able to store up to 300 mA on its own (without CM1) that the electron beam intensity will be increased from 300 up to 500 mA in 'multi-bunch' operation, using the two powered cryomodules.
This will be done very gradually, in stages, because we must on the one hand check that all the equipment of the ring can tolerate this power increase (in particular the absorbers such as crotches that intercept the radiated power at the dipole outputs, or the front ends shutters), and check that the stability of the beam is preserved, and on the other hand the radiation controls around the storage ring tunnel must be done again at 500 mA, as well as the ones around the shielded hutches of the beamlines.
Finally, with both cryomodules in operation, the 500 mA stored current will allow to reach the expected high brightness of the order of 1020photons/s/mm2/mrad2/0,1% Δλ/λ (for an energy around few keV).
Next news in 2009!
| 
Figure 4 : CM2 being lowered into the storage ring (Wednesday May 21st 2008). |
|
|
Augmentation du courant vers 500 mA
|
|
C’est dans ces conditions qu’a été réalisée la montée en courant au delà de 300 mA. Le dimanche 19 octobre, a eu lieu une première tentative qui a permis d’atteindre 400 mA pendant un peu plus d’une heure. Lors de l’augmentation du courant à 450 mA, une brutale remontée de pression au niveau d’une bride de la chambre à vide, dans une section droite dédiée à des onduleurs, a empêché d’aller plus loin.
Lors de la seconde tentative, le dimanche 16 novembre, un courant de 455 mA a pu être stocké pendant une heure, mais la défaillance du second système d’accord en fréquence du CM2, n’a pas permis d’aller au-delà.
A 455 mA, le système RF a délivré sans difficulté une puissance de 450 kW au faisceau et tous les paramètres de l’accélérateur (pression, température, stabilité de position et d’énergie, …) ont montré des enregistrements tout à fait normaux.
Une intervention sur le CM2 est prévue pendant l’arrêt de fin d’année, pour réparer les systèmes d’accord en fréquence ; en parallèle, la validation sur banc de tests d’une version améliorée est prévue pour début 2009. Finalement, avec les deux cryomodules complètement opérationnels, il devrait être possible, dès le 1er trimestre 2009, de stocker jusqu’à 500 mA en mode utilisateur et ainsi atteindre, une brillance de l’ordre de 1020 photons/s/mm2/mrad2/0,1% Δλ/λ (pour une énergie autour de 1 keV).
|
|
|
|
|
