8 August 1995
To: Distribution:
From: Stan Ecklund
Subject: "IR +- 2m Engineering"  Meeting Notes 4 August 1995

Tom showed BaBar solenoid calculations of the radial field 
component at the non-DIRC Q1 location.  The current design with 
225 mm radius hole in the end plug gives a 0.4 Tesla radial field 
at r=150 mm.  Modifying the hole with a 300 or 400 mm chamfer can 
reduce this a factor of 2 or 3 but also makes the field in the 
detector non-uniform.

Stan reported on data from CLEO provided by Dave Rice on their 
experience with radial fields.  Their SmCo5 quadrupoles see 0.2 
Tesla radial fields from the solenoid.  The quadrupole self fields 
are about 1 Tesla for a total of 1.2 Tesla.  The recently 
remeasured and adjusted their quadrupole to minimize harmonics.  
Most of the magnet typically had 2-3 gauss errors out of 6000 
gauss.  The region of the magnet in the radial field had errors of 
4 to 5 gauss, about twice as large.  Block measurements by Steve 
Herb showed demagnetization at 1.2 Tesla reverse field.

Our present design needs to survive larger reverse fields; about 1 
Tesla from quad self field, 0.2 Tesla from superimposed dipole, 
plus the radial solenoid component of order 0.4 Tesla.  We need 
more information on the properties of Sm2Co17 to resolve this 
problem.  Tom noted that the offset quad (not hybrid) design has 
about the same magnitude of a demagnetization problem because it 
extends to larger radii where the radial fields are larger.

Harvey asked if the solenoid end cap can be moved outboard enough 
for the radial field to miss Q1.  It would have to move 2x 0.37 m.  
This would complicate the solenoid coil design or field uniformity 
and Q2 shielding at the DIRC end.

Martin discussed the need to increase the strength of the carbon 
fiber portion of the support tube.  The CDR thickness (0.5% 
radiation length) calculates to give a 5 mm deflection under 1 g 
gravity with latest numbers for Young's modules and a quasi 
isotropic skin.  The baseline has to move to at least 0.62% r.l. 
and possibly to 0.9% r.l. from buckling considerations.  The 
direction of the fibers can be tuned for strength and less 
deflection at the cost of a lower buckling threshold.

The stainless steel portion of the support tube can be reduced 
from 3/4 inch to 1/2 inch thickness, giving more radial space for 
cables.

Martin showed a new crossection through a hybrid Q1.  The water 
cooled dipole trims have been replaced with smaller "proximity" 
cooled solid conductor.  This change and the thinner stainless 
leaves 15.3 mm outside of Q1 for SVT cables and clearances.  No 
tungsten shielding is present.  We need to decide if this works; 
i.e. enough clearance to assemble and align, no shielding for 
backgrounds, workable Q1 containment design, cable space, e.t.c.  
The time scale is end of August for a decision on the support tube 
diameter.

Roy reported they are close to a final SVT layout, having six 
silicon sizes.  On the backward end he needs more room; to be 
worked out offline.  The forward clear acceptance is 350 mradians.

Roy and Alex are looking at making up custom cable harnesses from 
individual twisted pairs.

Action Items:
1. Hybrid Q1: [Tom, Mike, Stan, Max, Artem.]
     Demagnetization properties of REC (Sm2Co17) materials.
     Required strength of Q1 dipole trim.
2. Cables and utility routing. [Fred]
3. HOM RF calculations with new mask geometry.  [Eddie Lin]
4. Check that a rotated BPM misses synchrotron rad fan. [Mike]
5. Shielding calculations [D. Coupal]
6. Required thickness of Q1 containment. [Martin]
7. Required thickness of Support Tube [Martin, J. Hodgson]
    calculate support tube deflection;
    earth quake deflection.
8 Beam generated noise to the drift chamber. [D. Coward]

Next Meeting: 13:00 Friday  11 August 1995 in the Yellow room with 
TV to LBL

attachments: transparencies