• D.R. Myers
• 29^th April, 1998

LHC Experiments Joint Controls Project
Report on the Initial Phase

1. During the first phase of the project it will be necessary to set up the Project Team, establish an initial set of work items with their definitions, and refine the operation of the project in collaboration with its members. The first iteration of this work should be completed as soon as possible, but not later than the first quarter of 1998.

2. The experiments to come forward with a motivated set of milestones for the control of each sub-detector. These should include the dates by which decisions must be made on the choice of low-level hardware, buses, etc., plus the requirements for the availability of test software. An overall date must be fixed by which a definitive solution for controls has to be available. The lists of sub-detector milestones should be established by the end of the first quarter of 1998, and should be updated at regular intervals.

3. Based on the work done in establishing the milestones for sub-detector control, the experiments should also produce a list of short term needs. It is clear that some sub-detectors are more advanced than others and may need interim solutions. The aim is to ensure that any ad hoc solutions adopted can be adapted to the solution finally chosen. In addition, collaboration with sub-detectors and test beams in some cases may allow the testing of possible hardware and software solutions. The short-term needs should be established by the end of the first quarter on 1998.

4. There is at present under way a Technology Survey and this should report its findings at the latest by the end of the second quarter of 1998. Intermediate reports will be made.

Item 1 will be covered in the next section and items 2 and 3 in the section which follows. Item 4, on the Technology Survey, is covered under the section on Sub-Projects.

The Project Team (M1)

A Project Team has indeed been set up and holds regular plenary meetings which are attended by up to twenty participants. Of these, apart from IT/CO group members, ATLAS regularly has two or sometimes four people present, LHCb has three, and ALICE and CMS have one or sometimes two participants. Members of IT/CE and other CERN groups also attend when they can bring expertise in particular areas. So far there are few participants from sub-detector groups.

Over the first three-month period a total of nine plenary meetings of the Project Team have been held. These have been used to discuss sub-projects/work items, as well as a report on the TIS-4000 evaluation, a technical introduction to PLCs[2], two interim reports on the Technology Survey and an overview of the controls issues for a particular LHC sub-detector (the ATLAS Tile Calorimeter). Future presentations are foreseen on the outcome of the Technology Survey and on particular industrial SCADA[3] products.

Planning Information (M2/3)

As the Joint Controls Project has to provide suitable and timely solutions not only for all the sub-detectors in a single experiment, but actually those in all four experiments, knowledge of the milestones and constraints of the sub-detectors as laid out in point 2 and 3 above is essential in order to formulate a strategy. The situation regarding these items was reviewed at the Project Team meeting of April 8^th. LHCb reported that they would put no constraints on the work of the project over the next two years. No details were available from ALICE as the Controls Coordinator reported that his information on the milestones of the sub-detectors was as yet incomplete. He would not object to choosing a SCADA system now, but believes that a decision could be put off for two years assuming that a satisfactory intermediate solution can be provided. Neither was information available for CMS or ATLAS, but the ATLAS Controls Coordinator would like to use input from the June workshop and the results of the Technology Survey to enable a plan to be agreed by the autumn assuming the project is to continue. At a subsequent meeting, a representative of the ATLAS Tile Calorimeter was invited to present the situation for this sub-detector, which would appear to be on a critical path and in fact may already have got to a point where incorporation into an integrated ATLAS DCS could prove difficult.

Sub-Projects and Work Items

The names in brackets are those of the sub-project leaders.

1. Infrastructure (D.R. Myers)
In order to make available all reports, minutes and other documentation a Web site[4] has been set up on a server managed by IT/CO group. Documents may be submitted in ASCII, MS-Word, FrameMaker and PDF formats. Ideas for automating the installation of information have been proposed by Pere Mato and discussed with IT/IPT, but these can only be followed up if additional resources are found.

2. User Requirements (A. Daneels)
Writing down the controls requirements for physics experiments seven years before they run is not an easy matter. Currently, there are three User Requirements Documents (URDs) dealing with supervisory-level functionality of the control systems, including a draft for ATLAS which dates from November 1995, a draft for CMS from mid-1996 and a recent common URD for ALICE and LHCb. Writing these documents has helped everyone to think about the problem. However, the documents' form, and part of their contents, differ, and they have not all been brought up-to-date. CMS and ATLAS have informed the Project Team that there seem to be no major incompatibilities with the contents of the ALICE-LHCb document so, although this is currently pending, there is hope that it should eventually be possible to achieve a unique URD for controls.

In addition to the supervisory-level URDs, more details must be collected on the types and quantities of equipment to be controlled by each sub-system as no quantitative information at this level is currently available to the project. A questionnaire was produced several years ago which could be used for this purpose.

3. Technology Survey[5] (A. Daneels)
The survey was under way when the Joint Controls Project was started and is mentioned specifically in the list of milestones. It is proceeding very satisfactorily and is on-time. A date has already been proposed for a full report to be made to the Project Team, and it is also foreseen to present a shortened version of this at the Controls Workshop. Two people from IT/CO are involved in the survey.

4. Prototype ATLAS TRT Gas System[6] (R. Barillere)
As for the Survey, this work item was already under discussion and later subsumed by the Joint Controls Project. It is being carried out in collaboration with members of the Gas Working Group (GWG) and involves one staff member and one cooperant from IT/CO. The ultimate aim of the GWG is to design a standard gas system for use by as many sub-detectors as possible and these will, of course, need control systems. Although the TRT gas system is somewhat specialized (it recuperates the Xenon), it was still thought to serve as a good way to continue a collaboration with the GWG in order to ensure that ultimately they have appropriate standardized elements for gas control. The prototype system is scheduled to be ready by June, and so far good progress is being made.

5. CAN Field Bus[7] (H. Burckhart)
The working group on Field Buses at CERN has recommended use of either CANbus, Profibus or WorldFIP. ATLAS in particular, as well as ALICE and CMS, have expressed an interest in CAN. ATLAS is developing a CAN node, the Local Monitoring Board (LMB), as a low-cost solution for acquiring data from sensors which is radiation tolerant and can operate in a magnetic field. One of the sub-project goals is to use the development of this box as a vehicle for investigating the CAN Open high-level communications protocol. Commercial equivalents to the LMB are also being looked at, including aspects of functionality, price and radiation resistance. A third work package under this heading involves the connection of equipment via CAN to a stand-alone BridgeView SCADA system. The project currently has participants from ATLAS, IT/CO group and IT/CE group.

6. Architecture[8] (P. Mato)
Despite the fact that we do not have a single URD for all four experiments, it was felt that sufficient was understood about requirements in order that work could start on architectural design, and in fact this is probably a key issue for further progress. So far four meetings have been held and have attracted a dozen or more participants. Proposals have been made for architecture on the levels of hardware, task structure and the data model. In addition, a list of issues have been identified which need further discussion and a presentation has been made on the current state of the architecture for the ATLAS DAQ. The goal, if possible to be reached by the summer, is to write a short document on a suitable architectural design with recommendations for the resolution of the issues raised. Where issues cannot be resolved, they must be noted and the arguments presented for each possible course of action.

7. Liaison with Level 3 Safety Group (W. Tejessy)
A working group under the auspices of ST/MC group has been started to look into aspects of Level Three Alarms. This involves danger to personnel, where CERN has legal obligations. As there exist obvious connections between high-level alarms and low-level alarms signalled by the control systems, a liaison has been established between the Joint Controls Project and the Safety Group.

8. EPICS[9] and TIS-4000[10]
Although work on the evaluation of the EPICS and TIS-4000 SCADA systems was essentially completed before the Joint Controls Project was started, all the relevant documents and reports have been published within the same Web structure.

9. Workshop (D.R. Myers)
The idea to hold a workshop on the topic of controls was floated already last autumn. After long discussions to agree on a suitable date, the 3-5 June was eventually chosen. Despite the very short time scale, it was felt to be essential to hold the workshop before the summer in order to help get members of the sub-detectors involved with the Joint Controls Project. Arrangements are now well advanced: an agenda has been fixed, a Bulletin and Guidelines for Speakers written and circulated, a Web site set up, speakers invited, and several rapporteurs selected.

Future Sub-projects

Whilst the topic of future planning has been deliberately omitted from this document, it should be mentioned that a presentation was made on an "ALICE Test and Evaluation Station", with the suggestion that this might be of general interest. The item will require further consultation. There is also a long list of items which should ultimately be considered, including liaison with the LHC machine and Magnet groups; selection of standard hardware components; and standard controls for high and low voltage power supplies, crate fan trays as well as rack power and cooling. However, of more immediate importance, are the detailed evaluation of systems selected by the Technology Survey and, perhaps, investigation of industrial developments like OPC[11]. These issues will be discussed in a separate document dealing with the next phase of the project.

Staffing

The staffing situation is delicate, and the two posts promised to IT/CO group have been reduced to one. Efforts are being made on an ad hoc basis to get help from other CERN groups where there is an overlap of interest, such as for PLCs, Field Buses and so forth. So far, despite the reasonable attendance at Project Team meetings, there are as yet few commitments from the members of the experiments actually to work on common sub-projects. One should also mention that although the project goals include the recommendation of hardware components, IT/CO group has very little expertise in this area.

Conclusions

In the three months which have past since the inaugural meeting of the Steering Group a substantial amount of progress has been made. A number of practical sub-projects are operational, and the sub-project on Architecture as well as the Technology Survey are of great importance for achieving the long-term goals. In addition, a workshop has been organized on a very tight time scale which should help to involve physicists from the sub-detectors and address a number of fundamental design issues.

On the subject of requirements the situation is most uncomfortable, but understandable. The primary objective of the project is to "define, select and/or implement as appropriate the architecture, framework and components required to build the control system" for four experiments, two of which are exceptionally large. It is unlikely that such an ambitious goal can be achieved with the level of information currently available about the needs of the sub-detectors. A very active rôle must be played in understanding what the sub-detectors in each experiment are doing and planning with them a strategy to make available components as they need them, but without compromising the ultimate maintainability of the control systems. Simply selecting a product is insufficient.

In summary, no technical reasons have surfaced which could compromise the goals of the project, but an increased commitment from all those involved will be required to ensure its success.

Acknowledgements

Although many people have contributed to the progress achieved so far, the support provided by Mme. C. Nebout with the Web and the Workshop must be emphasized. Advice on setting up the Web site was obtained from IT/IPT group.