To: Distribution 13 Oct 95 From: Martin Nordby Subject: Minutes of Near IR Engineering Meeting of 13 Oct 95 Alignment Requirements A general discussion of alignment requirements was kicked off by Stan Ecklund presenting results of dynamic aperture studies. LER Q1 and Q2 Alignment: X, Y: +/-200 microns if both quads move together +/-100 microns if motion is in opposite directions Z: +/-1 mm HER Q1, Q4, and Q5 Alignment X, Y: +/-300 microns if all three magnets move together +/-50 microns if Q5, Q4, Q1, Q4, Q5 move: (+,-,-,+,+,-) Z: +/-1 mm Since Q1 is not strongly focussing the HEB, its misalignment may not significantly effect the HER dynamic aperture. This will be investigated in future simulations. Closed-orbit calculations by Stan show that the orbit distortions are qualitatively the same for movement of the first three quads of the LER. Thus, it may be hard to know which is the source of an orbit problem. However, the study clearly showed that the closed orbit is significantly affected by changes in the position of these quads. Vibration requirements for Q1 are equally tight: X: 6 microns (1-10 Hz) Y: <1 micron (1-10 Hz) These are numbers produced by John Seeman and presented at a PEP-II tolerance meeting last month, and re-iterated by Stan at this meeting. SVT Alignment For the SVT, Roy Kerth presented the following numbers: X, Y accuracy of gimbal rings with respect to Q1: 100 microns X, Y measurement precision: 100 microns Stability: <50 microns motion relative to Drift Chamber (T = 6 hours) The relative alignment will be monitored by a system developed by Pat Burchat. Between B1 and the SVT, this system will be used primarily as a "collision monitoring system." However, between the outside of the Support Tube and the ends of the Drift Chamber, the system will monitor motions to determine if, and when, new calibration constants are needed for track reconstruction. One concern centered on the ability to actually position the SVT gimbal ring to 100 microns. Because it is supported off of a cantilevered B1 magnet, the inaccuracy of the adjustment system of B1 will be multiplied by the moment arm of the cantilever. Martin Nordby will look into modifying the current adjustment concept to improve the alignment accuracy. Q1 Mini-Review The end of the meeting was used as a mini-review of the mechanical design of the Q1 and B1 magnet support system. Q1 Magnet Collars The Q1 magnet is built in 5 cm thick "pineapple slices," surrounded by an aluminum collar. The entire slice is potted in epoxy to hold the P.M. blocks and prevent them from moving under magnetic loads from neighboring blocks and slices. The slices are joined together by dowel pins to ensure good alignment, then the entire package of 24 slices is held together with 8 high-strength stainless steel rods. Since the slices repel each other with 31 psi pressure, the rods are tensioned to prevent any separation of slices under the 3300 lb repulsion force. Also, when the carbon Support Tube is slid over the Q1 assembly, it must be cantilevered off the back of the Q1 magnet. This produces a large bending moment in the magnet, and a 76,000 psi tensile stress in the rods. Q1 Magnet Supports The Q1 magnet is supported off of the Support Tube at four points: at 3:00 and 9:00 o'clock positions at the in-board end of Q1, and again 13" in from the outboard end. This minimizes the total effect of the sag of the quad on the beam. However, the sag is only 2 microns, so this is not a problem. The supports are mounted to Q1 after the Support Tube is put in place. It is pinned and bolted to the Q1 collars through holes in the Tube, then held and positioned by adjusting screws. The support system has been sized for the design load of 2300 lb per support (seismic load), and fits within the outer diameter of the Support Tube, while not obscuring much space for cables inside the Tube. B1 Magnet Support The B1 magnet is cantilevered off the front of Q1. This support must hold B1 during alignment, and provide the 100 um placement accuracy needed for the SVT gimbal ring shoulder. This is done using three screw-adjustable pads to position the magnet off the front face-ring of Q1. Opposing hold-down bolts and belleville spring washers hold the load as the pads are being adjusted, so positioning is not lost during adjustment. The support ring for B1 is stiff enough to have large access holes cut in it for hand-access to the BPM feedthroughs and pump cables. It is also out of the way of the SVT transition cards which are near the back end of B1. Martin Nordby will look into modifying this design to ensure that the adjustment screws have a fine enough adjustment to position the SVT gimbal shoulder. Also, this entire system must be passed by the SLAC alignment group for feedback.