To: Distribution 28 Oct 96
From: Martin Nordby
Subject: Minutes of the IR Engineering and Physics Meeting of 25 Oct 96
|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||Eddie Lin||17|
|Roy Kerth||LBL 50-340||Lew Keller||41|
|Curt Belser||Rick Iverson||Jeff Richman||Jack Tanabe|
|Catherine Carr||Nadine Kurita||Natalie Roe||Rick Wilkins|
|David Coupal||Georges London||Ross Schlueter||Fran Younger|
|Fred Goozen||Joseph Rasonn||Joe Stieber|
Near IR H.O.M. Analysis
Eddie Lin reported on work in analyzing the RF behavior of the Near IR vacuum system, between Q2 Chambers. Given a Q(external) = 1600, and with every 32nd bucket filled, power deposited by a bunch is completely dissipated before the next bunch arrives. Total power dissipated for this scenario (assuming all buckets are filled, and I = 3 Amps) is 9.2 watts in the beryllium vacuum chamber.
The Q(external) is fairly high due to the long taper of the beampipe. Specifically, the back (in-board) face of the B1 Chamber mask closest to the beryllium chamber is fairly gradual. Steepening this would open up the structure, minimizing the trapping effect. More power may be deposited in the region during nominal conditions, but the area would be less susceptible to trapping resonant modes.
The main resonant frequencies of the region are all TE11 modes. The TE116 resonance shows that an additional 80 watts could be deposited in the chamber. Other TE11 modes show little decay, due to the high Q, especially for the lower frequency modes. If a beam harmonic falls on one of these modes, the power dissipated into the beryllium chamber could exceed 5000 watts.
There are four options for dealing with this possibility. First: hope. Since all of the resonances are only ~2 Mhz wide, the probability that the power of a beam harmonic falls directly on one is fairly small (1% or less). The general consensus was that luck is not reliable enough for our needs.
A corollary of this is to still do nothing, but, after luck fails us, rely on the ensuing heating of the chamber to de-tune the cavity. However, to shift the cavity frequency by 2 Mhz, the volume must change by 5 X 10^-4. This corresponds to a 0.1 mm change of radius, which requires a very large change in temperature, which cannot be tolerated.
Option two is to change the Q(external) by sharpening the in-board face of the mask slope. This seems like the most viable option. Mike Sullivan and Knut Skarpaas will on this.
Option three is to add loss-y material to damp the resonance. This does not change the resonance, it just reduces the peak magnitude by removing energy at a discrete location. However, space and cooling are difficult to come by, so this option would be very difficult to implement. Reserve this as a last resort.
Finally, the fourth option is to change the excitation of the
H.O.M. by changing the geometry of the entire region. Given the
stringent requirements of S.R. masking, rapidly changing BSC's
for two beams, and tightly-fit permanent magnets on the outsides
of the chambers, this is not an option.
Q2 Shielding Plug and Raft Design
Scott Debarger reported on progress on the Raft design. The PEP-II envelope has been made completely left/right symmetric, which has affected (read: "increased") some of the stay-clear radii at the 2-3 mm level. The conical transition in front of Q2 has been changed to move the "small-finger" bucking coil to the outside of the support (but still inside the stay-clear).
In this region, the cross-section of the free space around the septum chamber is 95 cm^2 each above and below the chamber. This appears adequate to allow all machine and SVT water lines to be routed out along beamline. Also, all machine magnet and I&C cables would come out through this region. It is not clear of there is room for Drift Chamber cooling water. All water and services would drop down through a hole in the top of the top plate of the transition cone, and run under the Q2 magnet. The knee in the transition cone has been located to ensure that there is space for all these services in this region.
Lew reported that Orrin is looking at thickening up the flange on the out-board face of the Q2 Shielding Plug. This flange started at 10 cm, and now wants to grow to 20 cm, with an additional 10 cm of Z-space used for the "Q4 bucking coil." Martin and Scott were concerned that this would encroach on the in-board support for the Q2/4/5 Raft, which had been located as close to the back face of the door as possible (leaving 8.4 inches of clearance for earthquake motion of BaBar). This would also eliminate the possibility of using a column for this support, since the notch in the skid-plate of the door is no longer in the correct position.
Lew and Orrin will try to trim this flange thickness down, as
they re-optimize the finger shape. Orrin has been focusing on
modifying the finger geometry to account for some loss in steel
when room was taken for the "small-finger solenoid"
and the ensuing change in the Raft. Also, he is looking at using
standard KHI steel for the first two fingers. Since they are heavily
saturated, their magnetic behavior should not be significantly
degraded by using a lower quality steel.
SK1 and Q2 Harmonic Corrector Ring
David Humphries has been updating the geometries for both of these magnets using the latest BSC values from Mike Sullivan. One consequence of the increased beam separation in this region is that the is more room for permanent magnet material for SK1. However, the needed integrated gradient for this magnet has increased 10% due to a change in the solenoid compensation scheme. The resulting magnet is 10.5 cm long, with 8.05 cm of active P.M. material, which produces a gradient of 124.52 kG/m.
To further increase septum space for P.M. material, the standard octagonal "quad chamber" shaped beampipe through the Q4 LEB chamber is extended through SK1 and the Q2 HCR. Inside the Q2 magnet, this transitions to a more elliptical shape.
Around both SK1 and the Q2 HCR, there is a magnetic shield which
protects the HEB from stray fields. David expects this shield
to fit over the flange on the LEB beampipe, and be segmented in
Z to increase its reluctance and decrease stray solenoid field
shunted from Q2 to Q4.
The Q2 Harmonic Correction Ring is designed to be a 24-element ring of 1.3 cm diameter P.M. rods, on a 5.95 cm radius circle. David expects a conservative length for this is 6 cm. However, this does not fit in the space between Q2 and SK1. There are a few options to work around this. First, he could go to a 16-element corrector. This would have less capacity to correct for higher-order harmonics, which are arguably tougher to correct on the magnet steel itself. Also, feed-down harmonics for the 24-element HCR were already at maximum allowable limits, so a 16-element HCR may have problems with feed-down.
Another solution is to stay with a 24-element HCR, but increase the circle radius, making room for larger diameter elements. This, too, would decrease capacity for correcting higher harmonics, but should have less of a problem with feed-down harmonics.
Finally, David and James Osborn have talked about the possibility of shortening Q2 again. It has already been whittled down quite a bit, so there is not much extra margin to shorten it. The big issue is current density, since none of the steel is near saturation. James will run through this option with David and Fran Younger.
Next Step: David and James will look at the options above. Specifically, James will continue to refine his 3-D and 2-D analyses of the magnet to better understand both the expected allowed and un-allowed harmonics. If higher-order harmonics dominate, due to tolerances of coil placement, for example, then shortening Q2 and lengthening the HCR may be the best option. However, if lower-order harmonics are the bigger problem, a larger diameter HCR may adequately handle these, in conjunction with chamfers on the endplates.
The field analysis of the HCR has been done with reference radii
of 4.5 cm and 5.0 cm. However, since the physical aperture of
the quad chamber is 4.5 cm, this is the appropriate value to use.
At this radius, all feed-down harmonics are less than 2 X 10^-4
of the Q2 quad field, and the corrector capacity per unit length
is very good, with 2 X 10^-3 level harmonics corrected for up
to n = 14 (per 1 cm of corrector length).
These minutes, and agenda for future meetings, are available on the Web at: