International collaboration has always played a major role in particle physics experiments. Could this also be applied to the accelerators themselves? Here, Albrecht Wagner,chairman of the directorate of the DESY Laboratory, Hamburg, looks at an innovative scheme for accelerator task sharing.
During the past 50 years, high-energy accelerators have not only become major research tools for nuclear and particle physics, but also influenced many other fields of science and industry by providing a powerful source of synchrotron radiation and other beams. New accelerator concepts have been the key to both an increased understanding of nature via fundamental research and the growing application of accelerators and accelerator techniques in other fields. It is therefore important to continue to develop new accelerators and to maintain accelerator expertise worldwide.
However, the size and cost of future large accelerators will most likely outstrip the resources of a single region, and building them will require a new approach. One way is via the framework of an international collaboration. A collaboration for a major accelerator facility must meet the following challenges:
* maintain and nurture the scientific culture of the participating laboratories;
* maintain the visibility and vitality of each partner.
Furthermore, all participating countries must be willing to invest and to commit themselves through long-term agreements. The proposed solution is a Global Accelerator Network (GAN).
Gaining through GAN
Scientists and engineers from laboratories and research centres around the world could form a network to integrate their scientific and technical knowledge, ideas and resources, and focus them on a common project – a merger of worldwide competence.
The GAN would allow participating institutes to continue important activities at home while being actively engaged in a common project elsewhere. All of the participants could demonstrate a visible level of activity, thus maintaining a vital community of scientists and engineers, and attracting students to the field of accelerator research and development. Last but not least, the network approach could facilitate the thorny problem of site selection for new large accelerator facilities.
The approach is based on the substantial experience gained in the construction and operation of large particle physics experiments at the LHC and LEP (CERN), HERA (DESY) and Fermilab’s Tevatron. In these projects, multinational teams, motivated and united by a common research goal, share the responsibilities of a large experiment. In this way, many groups, mostly from universities, become technically and financially responsible for the design, construction, operation and understanding of parts of the detector, which may be small but are nevertheless vital to the success of the experiment. Much of the work would be done at the home institutes.
These experiments have continually grown, with those for CERN’s LHC collider being comparable in manpower terms with major laboratories, and in complexity with large accelerators.
On the other hand, most accelerators so far have been built and are operated by only one laboratory. An important exception is the HERA electron-proton collider at DESY, where major accelerator components were designed and built in laboratories in other countries. However, once installed, responsibility for their operation and maintenance was handed over to the host laboratory, DESY. The LHC at CERN has evolved along similar lines.
In the GAN framework, new accelerator facilities, as well as experiments and beamlines for synchrotron radiation, would be designed, built and operated by an international collaboration of “partner” laboratories and institutes.
The machine would be built at an existing laboratory – the “host” – to capitalize on available experience, manpower and infrastructure. The host state would have to underwrite a major part of the finance and to make a clear commitment to support the project throughout its duration. (In the case of CERN, the host states are not the principal sponsors of international facilities built on their territory – the organization as a whole is responsible.)
Each partner would take responsibility for certain components of the project, designed, built and tested at home before being delivered to the host site. This responsibility would be maintained even after delivery. Component maintenance, operation and development would be carried out as much as possible from the home institutes, using modern communications technology. For this the partners would need to maintain duplicates of accelerator components for testing, checking and development. In some institutions, “copies” of the accelerator control room could even provide for highly efficient round-the-clock operation. At the host site, a core team, under guidance from all partners, would provide the necessary on-site technical support.
Sharing the cost
With GAN, major capital investment and operation funding would be taken up inside the partner states. Operational costs (mainly electricity), excluding manpower, would be shared by all partners according to a predefined arrangement. Most manpower would remain in the partner institutions, except during periods of installation and overhaul, and during collaboration meetings.
Details of the collaboration and management structures, together with the exact sharing of responsibilities between partners and the host, have yet to be worked out, but examples can surely be found within existing arrangements.
Remote control and diagnostics, allowing off-site partners to participate on site actively, are the key GAN features. While this would be an innovation for accelerators, there already exists substantial experience worldwide in the remote operation of large technical installations.
In major particle physics experiments, sub-detectors are frequently monitored and run remotely. A synchrotron radiation facility in Hiroshima, Japan, is operated under remote control from Tokyo. Large telescopes for astronomy are operated remotely – experiments on satellites and on distant planets are routinely operated from control centres on Earth. In industry, remote diagnostics and operation have become standard, even in nuclear power plants.
Many technical issues, including hardware- and software-related items, such as multiple control rooms, modular components and spare parts, standardization of systems and software, common data bases, common documentation, optimal communication and adequate protection against unauthorized access, are examined in an initial proposal (Willeke et al.1999).
The financial implications of a GAN now need to be appraised, especially to understand the additional costs resulting from remote operation. Several human aspects are also involved. How can the desired “corporate identity” be attained? How much manpower is needed at the host site and at the partner institutes? What scientific sociology will emerge? Many of these issues resemble those that have already been encountered in large experiments, which will serve as useful role models.
Whatever the challenges, a GAN could provide the framework for the construction and operation of future large accelerators, which would otherwise be impossible to realize. As a first step, the International Committee for Future Accelerators (ICFA) has set up a task force to study the model and its implications.
F Willeke et al.November 1999. Proposal for an Inter-Laboratory Task Force on Remote Operation.