To: Distribution 18 Nov 96
From: Martin Nordby
Subject: Minutes of the IR Engineering and Physics Meeting of 8 Nov 96
Hard-Copy Distribution:
Bob Bell | 41 | Nadine Kurita | 18 |
Gordon Bowden | 26 | Jim Krebs | 41 |
Pat Burchat | 95 | Harvey Lynch | 41 |
David Coward | 95 | Tom Mattison | 17 |
Scott Debarger | 17 | James Osborn | LBL B71J |
Hobey DeStaebler | 17 | Andy Ringwall | 17 |
Jonathan Dorfan | 17 | John Seeman | 17 |
Stan Ecklund | 17 | Mike Sullivan | 17 |
Alex Grillo | 95 | Uli Wienands | 17 |
John Hodgson | 12 | Mike Zisman | LBL B71J |
Hank Hsieh | LBL B71J | ||
David Humphries | LBL 46-161 | ||
Roy Kerth | LBL 50-340 | ||
David Kirkby | 95 |
Electronic Distribution:
Curt Belser | Kay Fox | Jeff Richman | Joe Stieber |
Lou Bertolini | Fred Goozen | Natalie Roe | Jack Tanabe |
Catherine Carr | J. Langton | Ross Schlueter | Rick Wilkins |
Al Constable | Georges London | Knut Skarpaas VIII | Fran Younger |
David Coupal | Joseph Rasonn | Ben Smith |
Q1 Prototype Results
Mike Sullivan reported on results of the Q1 Prototype assembly and magnetic measurements. During assembly, one quad P.M. block failed on a bond line. It was not clear why this happened, since it could have been due to shear loads applied by the fixturing, or by bending stresses from neighboring blocks. These bending forces moved the blocks around more than expected, so they needed to shim the blocks to hold them in the correct azimuthal position.
After initial assembly, the magnet was measured using the new vertical swept-coil system developed by Zach Wolf and Dave Jensen. Initial harmonics were in the 10^-3 range (at the coil radius of 6.35 cm), which agreed with Mike's predictions, based on MBUILD models of the magnet which used expected block position errors.
To improve on harmonics, the quad blocks were moved radially, based on an algorithm developed by Mike. He picked the four or five blocks requiring the most motion, moved them, then re-checked harmonics. The maximum needed motion was +0.03 inches to -0.04 inches, which was right at the max range of motion available (also predicted by MBUILD).
After 40 iterations, the final magnet harmonics were all <10^-4 at the coil radius, for harmonics n=3-18. At these harmonics, the needed block motion for correction is <0.001 inch, and the repeatability of coil measurement is starting to dominate the results. This demonstrated the need for precision motion at the 1 mil level, and for a low-noise coil, which does not produce its own harmonics due to mis-alignment, or fabrication errors. Andy is working on modifying the fixturing, and Zach plans to re-check his coil to ensure accuracy.
To check the mechanical stability
of the magnet, the assembly was rolled around the building, blocks
were pushed around, and the set-up was prep-ed for epoxying. This
appeared to move blocks at the 0.5-mil level, and affect harmonics
at the 10^-5 level, which was in the noise of the measurement.
The assembly is being potted now, and a final check will be made
after the quad arms are removed.
B1 Magnet Design Update
Andy Ringwall reported on the design status of the B1 Magnet. Since the prototype dipole blocks went together fairly smoothly, the B1 magnet slices have been modified to assemble like the Q1 dipoles. This has the P.M. blocks glue directly to the faceted inside of the support ring.
The first 3 slices of B1 are clam-shelled to allow them to be assembled after the Vertex Vacuum Chamber is welded in place. The split plane is vertical, which means the half-slices repel each other, but it was not clear if this is the preferred orientation (Andy will look into this).
The first 8 slices (including the clam-shells) are bolted together using #8 or #10 socket-head cap-screws. These have their heads buried in the larger of the tapered slices, and threads tapped into the collar of the smaller slice. The magnet is assembled from small to big slice, then slices 9-13 are joined together using threaded rods (similar to what will be used for Q1).
The collar for slice #1 (smallest slice) includes a shoulder on which the SVT gimbal ring mounts. The slice also has tapped holes for set screws which position the B1 Chamber front end. Since the gimbal ring hides the screws, Chamber alignment will be done first, then the gimbal ring is mounted.
Between B1 and Q1 space is very congested. Below is a full list of services in this area:
B1 Chamber cooling--2 inlets (3/8" tube) and 1 outlet (5/8" tube) come of the chamber, and out through the B1 Magnet support Bell.
Ion Pump HV cable--this uses a custom connector, to save space, but must be protected, and may need RF shielding to avoid bunch-gap kHz-level pick-up.
BPM semi-rigid coax cable--4 cables. Also need access room to disconnect from BPM. This is rad-hard, low-loss BPM cable.
SVT cabling--power and signal lines are all routed on twisted-pair ribbon cable. This come off 51 transition cards, mounted uniformly around the B1 Magnet, then are folded and routed azimuthally to their correct position to be routed out along the Q1 Magnet.
SVT water manifolds--at 12 and 6 o'clock the SVT water lines come together in a manifold, then are routed out along Q1.
SVT nitrogen manifold--this distributes dry N2 in small tubes into the SVT.
Tungsten shielding--still not known how much is needed, but Andy is trying to provide for 360° coverage of 2 cm thick tungsten. This will cover all but 4-5 cm of z-space between magnets, with no notches in the horizontal plane.
B1 Chamber supports--need X, Y, Z, and Roll adjustment and support for the B1/Q1 Chamber assembly. Knut Skarpaas and Andy are working on this design now.
B1 Bell Housing support--this supports
the magnet by cantilevering it off the front of Q1, and also provides
adjustment for alignment in Z, Roll, Pitch, and Yaw. This housing
is slotted to allow assembly of the B1 Chamber, and to let cooling
tubes and cables come through from the chamber.
The B1 Chamber services will come
out through the Bell housing radially, and extend through the
SVT ribbon cables into the clear radial space outside the cables.
They will come out at their correct azimuthal position, then will
bend and run down the side of the Q1 magnet.
B1 Chamber Status
Knut Skarpaas reported on progress on the B1 Chamber design. The Q1 Chamber and B1 Pump Can have been modified to move the final Pump Can weld farther from the Q1 Magnet. This allows it to be welded in-situ, and with the Bell Housing in place, making assembly possible (although the weld is difficult).
Tom Mattison felt that the ion pump HV cable should be coax to minimize noise from beam pick-up. Knut will look into this. Also, the stainless steel parts of the chamber (mostly near the pump can), may need to be plated. On the inside, plating would minimize power deposited in the pipe from H.O.M.'s. On the outside, plating would reduce transmission of kHz-level noise to the SVT and Drift Chamber. Tom will look into what is needed here, given the limited space.
Given the tight space between B1
Magnet and Chamber, and between the proposed background detectors
and the Vertex Chamber, Tom will look into modifying the detector
geometries to increase clearances. This is needed, since there
will not be room for the final position sensor array, and accurate
location and gap measurements will not be available during initial
assembly.
These minutes, and agenda for future meetings, are available on the Web at:
http://www.slac.stanford.edu/accel/pepii/nearir/home.html