temporary file to be inserted correctly

Energy loss of electrons on passing the VLE setup

The energy loss dE of an electron per dipole magnet is given by dE[keV] = 26.5 E[GeV]³ B[T] s/(2 pi r). Given a Bs=3.8 Tesla*m and the curvature radius of r=16.7 m with a path length s=2 m within the magnet, the energy loss of an electron with 10 GeV passing a dipole is about 1 MeV (0.01%) and therefore negligible.

Pion/electron separation

Cherenkov

A Cherenkov counter is installed in either the straight (10m) or the VLE (2m) section of the H8 beam line. The question is, how efficient/pure is the separation of the particles. An electron can only be faked by a background signal in the photomultiplier, which again is excluded, if no trigger signal at the same time is occurring. This leads to the assumption, that an electron can not be faked. Otherwise, if the detector efficiency is not 100 %, the loss of a photon signal caused by an electron can lead to a fake pion, indicated by the trigger signal. What ever the efficiency of a single Cherenkov counter is, the use of several can enhance the overall performance. We state, that we are only interested in the confirmation of being a pion or an electron, the absolute composition of the beam is of no interest.
The detector efficiency is calculated by taking into account the number of produced photons (determined by the gas, pressure, length of the counter), the transmission and collector factor of the radiator and light guide, as well as the conversion efficiency of the Cherenkov counter.

TOF

Does make TOF with the two triggers located at Spec1 and Spec4 make sense?

The distance between Spec1 and Spec2 is about 6.5 m, which is traversed in about 2 10^-8s. In order to separate the particles with a precision of 10^-4, we need trigger counters with a resolution in the order 10^-12. The trigger counters used have a resolution of 10^-9s. TOF separation is not a possibility.

A clean electron beam

Inserting a absorber plate before the Cherenkov could clean the beam to the needed level. Even a second one after SPec4 could addtionally improve the situation. Do not place an absorber in between Spec1 and Spec4, in order to minimize the error on the momentum analysis.

Beam Instrumentation

JS: We agreed on connecting Spec1, SPec3 and Spec4. Space is kept for SPec2 if needed in the 2004 run. All planes will be connected. Preferred is the connection of same planes to a single TDC. If connected to two different TDC, it is possible, that the stack numbers get shifted and the correlation of coordinates for a single particle are not assigned correctly.

Spectrometer Resolution

The figure shows a schematic drawing of the spectrometer in the VLE setup of H8. Also indicated is Spec1, where this one is not installed presently. So far it seems sufficient to have only Spec1, Spec3 and Spec4. According to d1, the distances to the other Spec's the offset is defined similar. Using all four Spectrometers the best resolution can be obtained by using the error on the angle DQ=Ö2 Dd (1/L1+1/L4). The error on the momentum measurement is given by

Dp/p = 1/Q DQ ≈ 0.46 %

, with a resolution Dd=300 microns of the wire chambers and the distances L1=3.325 m and L4=1.0 m, tilted by Q=120 mrad.

As only Spec1, Spec3, and Spec4 are installed, the resolution is worse, but still acceptable. Assuming, that the deflection is point like and has to be located in a plane at center of the dipole, one can calculate this point by using Spec3 and Spec4. The momentum is determined using Spec1 as well. The resolution is given by

Dp/p=2/Q Dd (1/L1 (1+L3/L4)+1/L4) ≈ 0.85 %

, where L1=4.235 differing to the previous example as no Spec2 is present.

H8beamelements.xls (indicators for trigger coincidences still missing)

Old pages

http://sl.web.cern.ch/SL/eagroup/beams.html#h8

http://ab-div-atb-ea.web.cern.ch/ab-div-atb-ea/Beams/H8/h8_gen.htm