Meeting of Sept. 26, 1996

Agenda:

• more on CDF beam stability [Henryk]

• event generation needs [Boaz]

• news about trigger simulation tools [Phil]

• constraints for signal routing: what are consequences/requirements on signal arrangement of FTC and SVX signals if we want SVX preprocessor? [Fred, Marvin]

• about old simulations [Ron]

• resource needs for event generation (project disk space, computing power, farm-time,...)

People present at meeting:
John Ellison, Phillip Gutierrez, Boaz Klima, Ron Lipton, Meenakshi Narain, Doug Norman, Henryk Piekarz, Horst Wahl

MINUTES OF MEETING:

1. Event generation needs [Boaz Klima]
   Here is the document presented by Boaz:
   Monte Carlo Samples for Impact Parameter Trigger
   • 10,000 events each sample??
   • b Physics
     • 
     • 
     • Background for previous two processes - CDF data?? (understanding fakes)
     • , where and ( 's from B's)
     • , where and ( 's from B's) **
     • , where (prompt 's)
     • , where
   • Top Physics
     • 
     • or **
     New Physics
     • , where 5 GeV(?)
     • , where is the Stop Quark (or the Top Squark...)
   ** - to be generated first??

   Discussion:
   In the discussion, it was pointed out that there exists no single Monte Carlo data sample with the full DØdetector after the upgrade. Therefore it was decided to go ahead and generate 10000 events each of the two physics channels which are considered of highest priority (marked by ``**'' in Boaz' table above), even though the latest version of the SVX detector is not yet implemented in GEANT. Boaz will babysit this generation. Meena will look after the GEANT simulation.

2. Computer resources needed
   
   • disk space:
     One event after full simulation needs about 0.5 Mbyte, i.e. for 10k events 5Gbytes are necessary. For the two event classes that we want to generate with highest priority, this means 10 Gbyte. This means that we need 2 large (9Gbyte ) disks in the near future. A request to that effect will be submitted.
   • computing:
     The farm at UTA can generate about 5k events full GEANT per week. Other possible resources will be investigated.
3. More on beam position monitoring by CDF [Henryk Piekarz]
   

   Here is the document presented by Henryk:
   Monitoring and correcting beam position in the interaction region:
   At the interaction point (IP) the colliding beam position and size are determined by operation of the low beta quads.

   The DFG's (Dipole Field Generators) can be set to control the beam position within a range of up to +/- 400 microns.

   The DFG dynamic range is +/- 10.3 V with an output step size of 5 mV thus giving about +/- 2000 steps. Each step corresponds to a theta value of about .

   Given the low beta region around D0 and CDF (beta = 1) the smallest change in the IP position is of the order:

   

   which indicates that the IP position of the colliding beam can be set with a very high precision to any value within the range of .
   So, as the beam can be moved with such a high precision the questions are:

   1. With what precision can the IP position be measured?
   2. Is it feasible to correct the DFG's so the IP position can be controlled within range smaller than impact parameter resolution?
   3. Is it necessary to make DFG's corrections in real time, or is it sufficient to do them every 5 or 10 min?
   4. Is it feasible to make the initial Tracker positioning to better than 400 in order to place it within the dynamic range of the DFG's correction?

   AD (a)

   IP position measurement:
   

   A.

   Beam Position Monitors
   The BPM's (from AD) provide information on the beam position with a precision of +/- 150 . This is insufficient for the operation of the SVT Trigger with a projected impact parameter resolution of some 30 .

   B.

   Silicon Vertex Detector
   The Silicon Vertex Detector as part of the Tracker should be able to provide the IP positioning information considerably better than the expected impact parameter resolution (30 ).

   AD (b)

   IP position control with DFG's:
   

   Last Spring an application program was developed (Lyndon Carmichael and Paul Derwent, AD) that utilized the CDF IP position measurements at the interaction region to control the operation of DFG's. This was done using the following 4 ACNET devices:
   

   

   Results from February 1996 test showed that the IP could be moved in both XZ and YZ planes within range higher than 100 and with a tolerance of about 5 . The slope of the beam was controlled to better than 50 micro-radians.

   AD (c)

   frequency of DFG corrections:
   

   The data from CDF arrived every 5 min. while the DFG controllers were updated every 10 min.. It turned out that the 10 min. updates provided acceptable stability of the beam position. Nevertheless the plan is to move the DFG's feedback control system from "application" to a "central service process" with operations in real time and the DFG's corrected only if the observed beam offsets were higher than e.g. 20 .

   AD (d)

   tracker positioning:
   

   To facilitate the initial placement of the CDF central detector and the initial alignment of the SVXII detector to the beam axis, a set of measurements in the B0 Collision hall during the commissioning period at start of Run II was proposed.

   During the commissioning period the CDF central detector will not be in the collision hall. This enables installation of additional instrumentation to measure precisely the IP position.

   The goal is to align the SVX II detector to beam axis to better than 100 micro-radians and the detector center to the beam center better than 500 microns.

   The instrumentation will consist of:

   • 1 Horizontal and 1 Vertical Flying Wire Assembly, installed +/- 1m from the nominal IP position,

   • 2 Horizontal and 2 Vertical BPM's, installed on both sides of the drift region 1 m from the low beta piece (it means BPM's are 10 m apart).

   The proton beam has Gaussian width of:

   

   so at +/- 1m the beam has sigma = 123 microns.

   With BPM's 10 m apart and the flying wire assemblies using 30 carbon fiber wires the absolute position of each monitor needs to be known only to better than 1 mm in order to achieve goal of 100 micro-radians alignment.

   As most instrumentation is available it is expected that CDF will have to contribute of the order of $10-15k only to perform above measurements.

   SUGGESTIONS FOR DØ:

   1. Make sure that the plan for assembly and alignment of the Silicon Tracker takes into consideration requirements of the SVT Trigger,
   2. Make plan of alignment of the Tracker with respect to the beam that will satisfy needs of the SVT trigger. Should we adopt the CDF plan?

   3. Investigate the DØ connection to the accelerator DFG's feedback program of low beta quads controls. The following ACNET devices were designated for DØ:
      

      

4. Trigger simulation tools
   Phil has been in touch with Cecilia Gerber to learn about trigger simulation tools. He will generate a short recipe-like description about how to run the relevant software, and will also provide a compendium of where to find documentation.
5. Some open questions
   • information about ``old'' simulation (i.e. of DØnote 2984)
     (how was it done, what software was used, where is the code, where are the events)?
   • track finding algorithm to be used in preprocessor:
     Phil will dig out some work that he has done for SDC, to see how much of it is applicable to our task. He will also get in touch with Marvin to get a better understanding of his ideas.