Minutes of the 5^th Timing Working Group Meeting (TimWG) Held on Wednesday the 11^th June 1999

  Present:

Excused:

          Adriaan RIJLLART,

Agenda

• Approval of previous minutes
• Beam Instrumentation: slow and fast  requirements. Jean-Jacques SAVIOZ
• PS requirements for LHC. Julian LEWIS
• Next speakers
• Next meeting

Approval of previous minutes
 
Michel JONKER commented that if it would be possible to drive the RF by functions as the power converter than there would be no need to link timing event to fixed values of the momentum. The minutes of the 4th meeting have been updated with this comment.

Regarding the possibility to freeze the ramp at an unpredicted moment, Ruediger SCHIMDT commented that, in his opinion, the freezing of the ramp is not one of the essential requirements but it could be useful to have such a facility. He explained that in the case that a ramp would be frozen, he would not expect to continue ramping with beam. Therefore he suggested the following procedure:

a) the ramp is frozen (as an example, due to some beam observations  that the operator would like to look at the orbit),
b) if the beam (or a part of the beam) survives the freeze, some observation would be done,
c) when this is finished, the power converter would be ramped to the end of the ramp in order to have the same magnetic history. The beam might be dumped before the restart of the ramp,
d) the RF is free to do what it is most convenient for it.

Beam Instrumentation: slow and fast  requirements
 
Jean-Jacques SAVIOZ presented the actual status of the beam instrumentation timing requirements (see slide attached). He classified the different timing needed into five categories and summarized the different timing necessary for each type of instrument supported by the BI group:

• Millisecond based machine timing,
• Beam synchronous distributed command,
• Revolution frequency,
• Bunch frequency,
• RF frequency.

He explained that the BPM (Beam Position Monitor) and BLM (Beam Loss Monitor) electronics will be located in the tunnel under the magnets but that the final position (mid of half cell or beginning of half cell) is not yet determined and depends mainly on the radiation level, cabling cost and timing distribution constraints (propagation delay to be compensated). He explained also that the SL/BI group is actually evaluating the possibility to use the same kind of approach as SL/PO group for the control of their different equipment (WorlFIP fieldbus + VME gateway). He expressed that the local timing distribution is a SL/BI activity and is not in the mandate of the TimWG working group. Finally, he evaluated the possibility to reuse the actual BST system installed in the LEP for the LHC and presented the necessary modifications to be done in order to cover the SL/BI requirements for LHC.

During the following discussion , Philippe BAUDRENGHIEN raised the question of how the first turn measurement of each batch could be implemented and how the synchronization with the RF system could be done. It appeared rapidly that different solutions are possible:

• A measurement prepulse can be send directly by RF synchronously with the batch on which the measurement will be done,
• A measurement prepulse can be distributed via the BST,
• Bucket number and batch number on which measurement will be done can be distributed via the control system and SL/BI retrieves the correct measurement trigger prepulse from those information.

In general, the flexibility of the beam measurement possibilities during MDs has been raised by the working group. This topics will be put on the agenda of a further meeting.

PS requirements for LHC
 
Julian LEWIS explained that there are few timing requirements from the PS (see slides attached). He explained the C-Train timing generation in the PS and the timing event generation and distribution based on a central MTG for the PS complex and distributed TG8 modules around the different machines.  He summarized the functionality of the TG8 card used in the PS and the advantage of the double twisted pair communication which permits to reduce the jitters on the event to +/- 2ns. Gary BEETHAM confirmed that this functionality is actually not used in the SPS. He explained also the telegram mechanism used in the PS which permits, in addition to the coarse timing,  to generate locally timing events.

He presented a Timing Surveillance Module (TSM), developed at TRIUMF, which is able able to monitor the distribution of up to 32 timing events distribution and perform post mortem analysis with a precision of 50ns. He suggested that this kind of module could be used for dump request monitoring.

He made also a presentation on the PS cycle management system. He explained that the PS complex is continuously playing scenario composed of sequence of beam processes and  that the PS cycle management system is able to reply to a request, to take the decision to play a new scenario (to modify its current scenario) and to commit its decision. The estimated delay between a request and the beam availability is typically 30s. He presented also the actual status of the SPS2001 project timing aspect and the possible extension of the selected solution for the LHC central timing.

AOB

Gary BEETHAM showed of first summary (see slide attached) of the slow timing requirements. He asked the members of the working group to control and complete if necessary, the document.

Next meeting

Next Meeting will be held on Friday 25 June 1999 at 10:30 AM in 864 1C01.

Agenda

1. Approval of previous minute
2. Timing Policy for Electricity Control. Arthur Swift
3. Timing Requirements for the Experimental Areas. Guy Baribaud
4. Next speakers
5. Next meeting
6. AOB