To: Distribution 13 Sep 96

From: Martin Nordby

Subject: Minutes of the IR Engineering and Physics Meeting of 6 Sep 96

Hard-Copy Distribution:

Bob Bell41David Kirkby 95
Lou BertoliniLLNL L-287 Jim Krebs41
Gordon Bowden26Harvey Lynch 41
Pat Burchat95Tom Mattison 17
David Coward95James Osborn LBL B71J
Scott Debarger17Andy Ringwall 17
Hobey DeStaebler17John Seeman 17
Jonathan Dorfan17Knut Skarpaas VIII 18
Stan Ecklund17Mike Sullivan 17
Alex Grillo95 Uli Wienands17
John Hodgson12 Mike ZismanLBL B71J
Hank HsiehLBL B71J
David HumphriesLBL 46-161 Orrin FacklerLLNL L-291
Roy KerthLBL 50-340

Electronic Distribution:

Catherine CarrNadine Kurita Natalie RoeRick Wilkins
David CoupalGeorges London Ross SchlueterFran Younger
Fred GoozenJoseph Rasonn Joe Stieber
Rick IversonJeff Richman Jack Tanabe

Fields in Q2

John Seeman reported on further refinement of comparisons between the Q2 / solenoid testing he had done, and the analysis done by Orrin Fackler. For these tests, the quoted solenoid field was measured one inch off the front of the core, near the center of the bore. This corresponds to the 2.4 m location in Orrin's calc's. At a radius of 4 cm, the calculations predict Bz = 250 G, and Bmod = 260 G. This difference has not been accounted-for, and may be offset by other discrepancies, so should not be counted on. However, future field quotes should be for Bz, at this one inch location.

John also reported that the octupole coils in Q2 produce a 16-pole field which has the same polarity as that induced by the solenoid. There was some hope that the polarities would be cancelling, but, since they add, the limit of the octupole field is 375 G, to maintain the 16-pole field below the spec for Q2.

Comments on Q2 3-D Analysis

Stan Ecklund reported on attempts to correlate Fran Younger's 3-D analysis of Q2 with the tests done by John Seeman, et al. The 3-D analysis showed that the radial field entering the pole tips from the solenoid field is sinusoidal with the poles. Assuming that all Bz becomes Br and enters the pole tips:

int(Br(z)*dz) = Bz*r0 / 2 (for r0 = 4.78 cm)

This yields 186 G-cm for r0 = 4.78 cm, and 129 G-cm for r0 = 4.23 cm

The latter reference radius was used by Fran in his 3-D calculations. He predicted 117 G-cm integrated octupole This agreement suggests that 90% of the Bz entering the bore produces integrated octupole.

The simple sinusoidal model also agress to 30% with the tests done on the HER quad.

BaBar Magnetic Analysis

Orrin Fackler reported on ongoing work to optimize/understand the Q2 shielding plug finger design. To satisfy curiosity, he ran a model of the forward plug with solid fingers, using KEK Venus steel. Bmod = 6000 G in Q2, compared with 187 G for the same configuration with fingers.

He then thickened the fingers at their root, but this shunted more flux from the first to the second finger, and from the second to the third, which pushed more out into Q2. Bmod = 211 G in Q2 for this configuration.

To bound the problem, Orrin ran a case with mu = 20 kOe to eliminate any saturation effects. Here, Bmod = 32 G in Q2. This is lower than an earlier simulation where he reduced the solenoid field to 1.7 kG, ran the calc's, then multiplied results by 10X. This showed that, even at 10% of field, parts of the fingers were still being saturated.

Another set of runs investigated the sensitivity of field in Q2 to the shape of the top end of the B-H curve. This part of the curve has been extrapolated, since reliable test data was not available. Following is a list of stray fields for various curve extrapolations:

Curve Extrapolated
Field in Q2
Pure Fe (measured)
168 G
KEK(lower curve, extrapolated)
187 G
KHI extrapolated data
202 G

This shows that sensitivity to poor definition of the B-H curve above 100 Oe is only at the 10% level.

Next, Orrin modified the plug geometry to fit the Q2/4/5 Raft design in its present incarnation. Specifically, he opened up the flare of the third finger to miss the flare of the Raft, and straightened the second and third finger to make the parts more build-able. Neither had any effect on stray field in Q2 (at the 5 G level).

Finally, Orrin looked at the effect of reducing the main solenoid field on the stray field. For the two types of steel listed below, stray fields changed significantly with main field magnitude:

Stray Field in Q2:

Solenoid Field
KHI Steel
KEK Venus Steel
17 kG
320 G
187 G
15 kG
115 G
114 G
13 kG
67 G
62 G

This suggests that reducing the solenoid field can never help more than to bring Bmod down to 60 G. Orrin feels that a bucking coil near Q2 should bring the field down to <100 G. This is based on his past experience with adding bucking coils to this region.

Next Steps:

--Look at a water-cooled copper bucking coil, running at 10 A/mm^2. Plan to support the coil off the Raft, and minimize its cross-section to reduce the quantity of steel which is removed (Orrin Fackler).

--Look into using annealled steel, either KHI or KEK Venus steel.

--Map Bz and Br over the entire length of Q2, to better understand the analysis results, and correlate with tests.

These minutes, and agenda for future meetings, are available on the Web at: