To: Distribution 16 Sep 96
From: Martin Nordby
Subject: Minutes of the IR Engineering and Physics Meeting of 13 Sep 96
Hard-Copy Distribution:
| Bob Bell | 41 | David Kirkby | 95 |
| Lou Bertolini | LLNL L-287 | Jim Krebs | 41 |
| Gordon Bowden | 26 | Harvey Lynch | 41 |
| Pat Burchat | 95 | Tom Mattison | 17 |
| David Coward | 95 | James Osborn | LBL B71J |
| Scott Debarger | 17 | Andy Ringwall | 17 |
| Hobey DeStaebler | 17 | John Seeman | 17 |
| Jonathan Dorfan | 17 | Knut Skarpaas VIII | 18 |
| 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 |
Electronic Distribution:
| Catherine Carr | Nadine Kurita | Natalie Roe | Rick Wilkins |
| David Coupal | Georges London | Ross Schlueter | Fran Younger |
| Fred Goozen | Joseph Rasonn | Joe Stieber | |
| Rick Iverson | Jeff Richman | Jack Tanabe |
Vertex Vacuum Chamber
Knut Skarpaas reported on the design of the Vertex Vacuum Chamber. The 0.059 inch annular water gap is difficult to analyze for heat transfer and flow, because it falls in the gap between standard models for enclosed flow in circular ducts, and flow between infinite flat plates. The latter predicts that flow is in the transition region between laminar and turbulent flow. Knut is putting together a flow prototype to model this. He will look at pressure drop across the Chamber and manifolds, flow uniformity and eddying, and heat transfer.
Since the goal is to use a sub-atmospheric water system, Knut is trying to reduce pressure drop wherever possible. He has increased tube diameter for inlet and outlet tubes to 3/8 inch. There is room for these inside the Support Tube, and the flow prototype will show whether this is sufficiently large.
The prototype will also test the impact of the epoxy paint on heat transfer, since the outer tube is only joined to the inner ribs by this paint.
Knut and Fred Goozen have modified the DMZ stay-clear region between
the SVT and Vertex Vacuum Chamber on the back end to make room
for the water manifold. At closest approach, the two devices are
separated by 0.081 inches. The manifold is right at the DMZ line,
while the SVT inner RF shield is 0.041 inches away. If the B1
Magnets pitch or yaw, this gap between devices closes. For 4.5°
pitch/yaw, the gap closes by 0.04 inches. Since this can only
happen during installation or an earthquake, it was agreed that
the DMZ could be used for this clearance.
Near IR Cooling Plans
Martin Nordby reported on plans for cooling the Near IR components in the Support Tube. There will be two independent chilled water systems. One is dedicated to the Vertex Vacuum Chamber, while the second supplies chilled LCW to all other components. To minimize temperature changes with time and with azimuth, most of the parallel cooling circuits are split into parallel left/right circuits, and are independently controllable, using feedback from thermocouples on the devices.
There will be five cooling circuits on each side of the I.P.
| Circuit | Independent Parallel Circuits | |
| B1 Circuit | --B1 Chamber S.R. cooling
--B1 Chamber Temp Stabilizing | |
| --B1 Magnet Temp Stabilizing | ||
| Q1 Inner Cooling | --Q1 Chamber Cooling | |
| --Q1 Inner Thermal Shield (Left/Right) | ||
| Q1 Outer Thermal Shield | --Q1 Outer Thermal Shield (Left/Right) | |
| Q1 Quad Trim Coils | --Quad Trims | |
| Q1 Bellows | --Bellows Flange Cooling | |
| Total Per Side: |
These cooling circuits will be supplied by a dedicated chiller, located (most likely) in Building 625. This will supply ~18°C LCW to the Piers under the Q2/4/5 Raft. On each Pier there will be a distribution box, where the parallel circuits are split off, with metering valves and in-line re-heaters for each parallel circuit.
A question arose regarding the heat capacity of the various devices,
and the expected response time of the cooling circuit. If the
heat capacity of the device and water in the cooling system produce
a long enough time constant, it may be hard to control temperature
of the device. However, by using in-line heaters on the IR hall
floor, the water volume has been minimized. Heat capacity of all
devices except the magnets should be low enough that the time
constant is short (10-15 minutes, max). For B1 and Q1 magnets,
more work is needed to understand the heat transfer dynamics.
SVT Cooling Plans
Roy Kerth described cooling plans for the SVT. The heat-producing electronic chips mount on aluminum nitride cards, which are clamped to buttons protruding from the cooling rings on the SVT cones. FEA anlaysis and tests predict the temperature drop along the alum nitride card to be 20 °C, with half of this coming from the clamped connection at the buttons. To produce an average temperature of 20 °C, the water will be supplied at 10 °C and the chips will run at 30 °C. Roy expected that the temperature of the silicon will be set by the ambient temperature inside the Support Tube, which is established by the air purge.
Total power from the system is 250 Watts, and total water flow is 2.4 Liters/minute, through four circuits. There are four inlets and outlets, running at 12 and 6 o'clock out of the Support Tube. The outlets are ganged together just outside the S.T., and all lines drop down to floor level at the Pier. This is where the chiller is located. A separate loop runs back up to beamline level, where a partial vacuum is pulled on the system. This maintains the entire system below atmospheric pressure. Pressure drop across the entire system is 30 inches of water, so the system will be run at a gage pressure of -50 inches.
Solenoid valves are installed on each of the four parallel circuits
to allow remote shut-off, if chip power supplies must be turned
off. It is not clear if this is needed.
Q2 Magnet Update
James Osborn reported on the status of the Q2 Magnet design. His current design is a laminated magnet with 0.06 inch laminations and 2 inch thick endplates. All four corners are chamfered to reduce the diameter of the circumscribing circle. James is planning to laser-cut the laminations to reduce cost. Tolerances on the the pole tips and coil pockets are 0.001". Uli Wienands relayed his bad experience with laser cutting lam's for past magnets he has built. The cutting process could not maintain the tight tolerances needed. James will investigate this with manufacturers.
The nearly-complete modification to the lattice and beam stay-clear locations thorugh the IR have increased the beam separation through Q2. James has used this extra room in the septum for a 3 mm thick iron shield around the HEB, and a 1-2 mm air gap. Designs for the main and trim windings are unchanged.
The current core length is 24 inches. This produces a magnetic length which is 4 cm shorter than the Design Review length, which makes room for a Harmonic Correction Ring and SK1 between Q2 and Q4. The coils are sized to produce the correct gradient for E = 3.1 GeV +/- 10%, so the current density is comparable to the Design Review values.
Scott Debarger exresssed concern about the stiffness of the laminated
magnet, and the tight space inside the Q2/4/5 Raft for any rigid
support system. James is planning to make the magnet stiff enough
to be supported solely by a kinematic support, with no additional
straigtening support needed. He is including beefy straps on all
sides of the magnet to ensure the magnet is straight and stiff.
These minutes, and agenda for future meetings, are available on the Web at:
http://www.slac.stanford.edu/accel/pepii/near-ir/home.html