Minutes of 11th FFIR meeting on 6/7/2000

The meeting was held in a room of 3-425 at KEK, 10:00-11:20, 6/7/2000. We discussed on (1) iron structure of 3Tesla solenoid, (2) magnetic field calculation of conventional QC1/QC2 for the dump line, (3) dump line update an others.

(1) iron structure of 3Tesla solenoid: (transparencies, 13 pages of pdf, 2.28MB) or tiff, 2.46MB Yamaoka presented an optimization of iron thickness of the structure for 3Tesla solenoid. The previous design which was based on the published (Green book's) one of 2 Tesla solenoid has overall dimensions of 13.5m (Z) x 16.8m (R). The major design principle was to have uniform magnetic fields in a volume of CDC, i.e. R<1.55m and |z|<1.55m. The uniformity was 1.1%.

In this report, he compared 4 designs (A,B,C and D). A and B are slightly improved ones by reducing the barrel iron thickness, so the overall sizes are 13.5m(Z) x 15.8m(R). However, the uniformity could be improved to 0.86% by adjusting coil-current densities. Difference between A and B is due to different ways of connection between endcap and barrel iron slabs. B has much more smoother field lines than A since all the iron slabs are connected without being intercepted by air-gaps due to muon chambers. It is indeed nice idea as clearly seen in figures. The another 0 is that B has much smaller leakage fields, i.e. 1/100 of A at beam line.

Therefore, next designs of C and D are based on configuration of B. Since leakage field of about 100 gauss can be acceptable for vacuum ion pumps at the beam line without any magnetic shield, he reduced the iron thickness up to the leakage field of 100G (at z=10m from IP) while the uniformity was kept to 0.93%. So, he got a design C. The overall size becomes to 13m(Z) x 14m(R). Keeping the uniformity (1.2%) and leakage field, inner radius of the endcaps can be increased from Rin=0.47m to 0.75m in order to make a space for support tube. So, C has Rin=0.47m and C-2 has Rin=0.75m .

Forgetting the leakage field and further reducing the iron thickness, he got a design D which still has the uniformity of 1% while the D-2 (Rin=0.75m) has that of 1.3%. The overall sizes are 12.3m(Z) x 13.6m(R). The leakage field goes up to 500G at Z=10m.

There were discussions about how much the leakage field can be allowed. Most sensitive device may be electromagnetic valves, especially in their control-circuits, for superconducting QCs, which can only tolerate up to 50G without magnetic shield. With proper magnetic shield, they are expected to be OK up to ~500G. Finally, our present conclusion is the design of C-2 whose dimensions ( 13m(Z) x 14m(R) and Rin=0.75m ) will be passed to JIM simulation. The Green-book's ones was 15.8m(Z) x 14m(R), Rin=0.5m.

(2)magnetic field calculation of conventional QC1/QC2 for the dump line : (transparencies, 25 pages of pdf(Japanese),484kB) or ps.gz(Japanese), 371kB Miyamoto finally completed magnetic field calculations of QC1 and QC2 by ANSYS. At the same time, he determined shapes of the poles as well as coils. There has been no actual design for QC2, that is, the same dimension as QC1 has been assumed for QC2 in JIM. Therefore, previous QC2-design has a few contradictions such as overlapping the compensation coil etc. . For QC1, one can see thinner pole tips than the previous ones, while QC2 has smaller outer radius. Now, both QC1 and QC2 have enough spaces for synchrotron radiations. To get a high field gradient of 225T/m (QC1), Fe-Co alloy was employed instead of pure or Belle-iron with smaller aperture for QC1 because of the high permeability. Non-linearity of field gradient was calculated to be less than 0.3x10^{-4} within |x|<1mm where horizontal beam size is about 0.2mm. So, there must be no problem for the non-linearity.

According to his updates, new geometries of QC1 and QC2 were installed in JIM. You can see them by postscript or pdf files, where, the outmost circle corresponds to inner radius of the compensation coil and 4 figures of QC1 and QC2 at both ends with 8mm horizontal crossing.

(3) Dump line update: (transparencies, 5 pages of pdf,57kB) or tiff, 117kB Kubo updated the dump line design with taking account of QC1&QC2 magnetic fields along extracted beam line. First, the fields, which is still old-version of Miyamoto's calculation, were decomposed into multipoles up 07 16-order for SAD-calculation. The extracted beam is further bent horizontally from 8 to 10 mrad at exit of QC2. He added two horizontal bends for dispersion free in horizontal direction (eta_x=0) at the second focal point, while the beam is vertically bent for measurement of the energy. The updated one can also have the same energy resolution as the previous design. While he tried to estimate energy losses, he failed to transport off-momentum particles by using SAD. So, there is probably some problem concerning particle-tracking at off-centered quadrupole magnets in SAD calculations. He is debugging this problem.

(4) Others

The next meeting will be on 21 June, 2000 13:30 - 15:30 at 3 gokan, 425.