To: Distribution 2 Jan 96 From: Martin Nordby Subject: Minutes of Near IR Engineering Meeting of 22 Dec 95 Forward End Small-Radius Region A 3-D layout of the Forward End (Downbeam HEB) small-radius region has been initiated to block out space for the structural support elements for the Support Tube, and for routing of the SVT, ATC, and Forward End D.C. cables and utilities. The SVT ribbon cable coming out of the S.T. will connect to thicker gauge ribbon cable to minimize signal losses. The target length for the thinner cable is 3 meters, with a total cable run from Transition Card to rack of less than 15 m. The thicker cable is about twice the cross-section as the thin ribbon. Alex Grillo will look into the connector size, and space needed to connect/disconnect. One design for the support of the Support Tube now being investigated seriously is a "Cantilevered Raft" design, with earthquake brackets holding the front of the Raft to the detector door. This requires cuts in the Q2 magnetic shielding plug for support members. Since this is a "permanent" support design, not requiring removal for F-Calorimeter removal, it must fit inside the bore of the F-Cal. Thus, it must fit tightly around any Q2 magnet design. Dave Humphries is now starting to look at the mechanical design of a Q2 permanent magnet, and should have a good idea of the needed envelope in a few weeks. More room is needed on the out-board end of Q2 for the tune/skew quad rings' rotation mechanism, so the raft will be flared to leave more room for this. Q2 Septum Chamber Lou Bertollini reported on preliminary thermal analysis of a re-designed Septum Chamber to fit around the Q2 permanent magnet design. The chamber is 5 mm thick OFE copper, reduced to 4 mm on either side. Cooling water circuits are joined (E-B welded?) to the side being struck by synchrotron radiation, as near to the incident stripe as possible. Initial values for the water flow was: 10 ft/sec, 25 degC inlet, with natural convection to 21.1 degC ambient air as the boundary conditions. Max power density due to SR is 453 W/cm^2 (from a 2 amp LEB). This produced a maximum temperature in Q2 of 33 degC and an azimuthal temperature distribution of 3 degC. These values are marginal, at best (see below), but were for an un-optimized chamber design. Lou Bertollini and Dave Humphries will look at ways to reduce the temperature effects, including: --Include the Q2 outer aluminum collar in the analysis, and look at changing the inner collar to aluminum, from stainless steel. This should greatly reduce the azimuthal temperature distribution. --Make chamber more elliptical rather than octagonal. This better matches the beam stay-clear, and maximizes room outside the chamber for cooling water or a possible thermal shield. --Check on B.S.C. values, to ensure that the chamber cross-section includes BSC's for solenoid off and on configurations --Check that the vertical position of the Q2 magnet is correct. Lou and Dave will iterate on this design in January. Q2 Magnetic Shielding Dave Humphries investigated the effects of adding a 1 mm thick iron cylinder around the Q2 P.M. to shield the HEB from the external fields from the Halbach quad. For a single 5 cm long slice, the peak fringe fields occur where the iron shield stops. However, for longer slices, there is an increase in the positive fringe field in the center of the slice, and a higher negative field at the ends. Extrapolating to the length of Q2, the integrated fields at the HEB centerline are: Un-shielded: -50 Gauss Short model: -120 Gauss Long model: +10 Gauss + 4 Gauss (Vanadium Permador shield) Extending the magnetic shield beyond the ends of the magnet should clearly help reduce the stray fields, but may affect the harmonics of the internal field. Also, these results are qualitatively different for the two lengths. Dave will continue to investigate this with intermediate-length models, and look at the strength and harmonics of the internal quad and dipole fields with the magnetic shield. Another alternative is to put the magnetic shield around the HEB vacuum chamber. However, this introduces a magnetic asymmetry in the region around the P.M. magnet, and may produce bad harmonics inside Q2. Dave will also look at alternatives for this. Dave also reported on the effects on Q2 of increasing the BSC by 1.6 mm, as required by the Q2 design presented at the Mini-Review last week. This increase in the inner radius of the vacuum chamber and, hence, the inner radius of the P.M. material, requires a lengthening of the dipole blocks by 2-3 cm, and the quad blocks by 5-6 cm (for a 10.6 T/m quad gradient). Some of this can be gained back by increasing the outer radius of the blocks, and by trimming down the inner support collar thickness. Permanent Magnet Temperature Effects Hobey DeStaebler and Mike Sullivan reported on the effects on performance of temperature variations in the three permanent magnets (B1, Q1, and Q2). For Sm2Co17 P.M. material, the remanent field strength varies -3.0 E-4/degC. Assuming a 3 degC fluctuation in the overall temperature in the magnet (0.1% change in magnet strength), effects on the magnets are: B1: parasitic crossing and Q2 beam separation decreased by 0.1%. This is not a problem. Q1: Focal point at I.P. moves by 1.8 mm. This is small compared to 17 mm bunch sigma in Z. Also, the LEB tune is changed by 0.01, which is 1/3 of the beam-beam tune shift. This is too big of a shift, and warrants a temperature stability of more like 1 degC total. Cornell's problems with the P.M. quads centered on this change in tune shift. The thermal time constant of the magnets was sufficiently long enough to make identifying the problem extremely difficult. Mike Sullivan looked at introducing a (10 degC)sin^2(theta) azimuthal variation in block temperatures to Q1 and Q2 magnet slices. For both magnets, the dipole and sextupole harmonics shifted by 6. E-4 at 60 mm radius. None of the other harmonics changed significantly. To reduce these harmonics back to the 1 E-4 level, the azimuthal temperature variation should be held to less than 2 degC, with 1 degC being the target. These minutes, and agenda for future meetings, are available on the Web at: http://www.slac.stanford.edu/accel/pepii/near-ir/home.html