Jim, Thank you for looking this over. Some of the information you request I can give right away. Hrachya, you can add to any of this if you wish. Richard J. Jim Stewart wrote: > I would add an introduction to say what the purpose of the pair spectrometer is. I understand it is to cross calibrate the tagger and to monitor the linear polarization of the photon beam. You should then be able to define the basic requirements of the spectrometer in terms of these goals. Yes, we need this. This has to come first. > If you want to cover the whole range of the tagger then you > need to detect e+/e- pairs from photons with energies from 25% to 90% of 12GeV (11.8 to 3 GeV). Do you mean 25% to 95% (3 - 11.4 GeV)? > If we go with a two magnet design for the tagger then the momentum region where the particles pass between the two tagger dipole magnets is especially important. Some plots of the cross section and rates then make sense. I can supply these. > The Hall-B Pair spectrometer only detected particles very near the horizontal plane. Do you know why this is? The angular distribution of high-energy bremsstrahlung and pair-production is naturally collimated on the order of 100 microradians or so. The natural scale for the transverse momentum of these pairs is the mass of the electron. > I think you need to address what usable range of particle > momentum you want to measure and then with what resolution. This together with a rate and acceptance estimate defines the pair spectrometer. With the above introduction you can make an assumption on an acceptance and then estimate what thickness the radiator needs to have to make the measurements explained in the introduction in a reasonable time. I think it will also then be clear why you need to be able to vary the radiator thickness. > I think this is a reasonable approach, and it sounds like we have all of the essential information we need right now to just sit down and write these paragraphs. > Now one can reasonably define the magnet and detector requirements. To me the two are tightly coupled. I can live with a shorter magnet if I have more precise track information. With a 2m long 1.3 Tesla magnet with poles 0.5 m wide, you get your required position resolution using 1 to 2 cm wide scintillators. Am I wrong but does this imply with a 1m long magnet the poles only need to be 250mm wide but the detectors need to be between 5mm and 10mm (detector plane in same position)? This still sounds reasonable. > The displacement is quadratic in B*L, right? So halving the L for the same B means all of the transverse dimensions scale down by a factor 4. > Just out of curiosity if you were to use the microscope for your detector what type of magnet would you need? The microscope has 1mm square fibers. I only bring this up because I like to reuse existing designs if possible. I don't know how many channels you would need in this case. > Actually, the microscope fibers are 2mm square. It is an interesting suggestion: what could be saved by reducing the size of the PS magnet and using a fiber-based high-resolution detector like the microscope? > I don't know how JLAB works but at some time someone may ask us to look at the cost of running a big magnet compared to investing in a factor of two more readout channels. Elke knows all about costs and can tell me if this is nonsense. > Arguments about running vs construction costs make sense, but from what I have seen, overall construction budget envelopes tend to constrain everything. > In your design you have two layers of detectors. Why? > Overlapped designs optimize resolution for fixed number of readout channels. In our case, we need to optimize for high rate capability. The BH cross section is immense so I don't think that backgrounds will be an issue, but it should be simulated. > Did you think about background? Do you need to collimate the scattered particles? I think background studies would be very interesting. The Hall-B pair spectrometer used a 40um double sided silicon sensor. Do you know why they needed such a high precision? > The silicon tracker was used for polarimetry. Franz is the one responsible for that device, and can fill you in more on that. The BH process has an azimuthal asymmetry in a linear-polarized beam, which can be exploited to measure the polarization. We have pushed to have beam polarimetry included in the beamline package, but did not manage to persuade the collaboration to do it.
S/MIME Cryptographic Signature