To: Distribution 12 Feb 96 From: Martin Nordby Subject: Minutes of Near IR Engineering Meeting of 9 Feb 96 Q2 Exterior Fields Tom Mattison reported on his continued investigation into the Q2 stray fields. Last week he showed how the magnitude of the exterior field goes as 1/(radius)^(n-1), and the 3-D end stray fields extend on the order of one radius beyond the end of the magnet. When the field is integrated along a line parallel to the magnet axis, the total equals the expected 2-D field. For Q2, though, this line of integration (along the HEB axis, for instance), is not parallel to the Q2 magnetic axis. Tom looked into the effect of this and found that, for small radii, where the field magnitude is high, small changes in angle of the line of integration do not produce a large effect. At larger radii, and large (100 mradian) slope angles, the difference increases to 25% of the nominal field magnitude. This is still a small effect, given how quickly the field is changing with distance from Q2, and is due to the nearly complete cancellation of errors from one end of the magnet to the other. Essentially, this effect does not look large enough to worry about. Tom also investigated the effects of a non-unity permeability on the stray fields. At the extreme, if a small iron cylinder with infinite-permeability were inserted into the quad bore, it would distort the quad field, and reflect it back beyond the outer radius of the magnet. The field would scale as 1/r^3, and vary as cos(3*theta). For finite-permeability material, the effect would not be as extreme,.but would still be present. This is essentially what the quad is: for a given cylinderical slice of the material, all material radially inward of that slice produce this effect on the field generated by the slice. This prediction is borne out by results of analyses done by both David Humphries, and Stan Ecklund. Mapping the field around a quarter of the magnet, just outside the outer radius, the field changes periodically, with two frequencies: it oscillates from block to block,. but also varies as cos(3*theta). Stan's analysis clearly shows this effect for permeability greater than 1, with no effect at all when it equals 1. As a figure of merit, Tom estimated that the field magnitude varies as (mu-1)/(mu+1). Q2 Magnet Design David Humphries reported on progress on the Q2 design. Last week he reported on a problem with the inner radius of the magnet encroaching on the BSC for the LEB. To rectify the problem, David increased the inner radius of the material, and changed the shape of the cone to better match the BSC. To account for the material lost, he refined the mechanical design of the harmonic corrector ring and the rotating quad rings to find space that was not being used efficiently. The new design (V. 5) has an 8 mm longer dipole section, and a 27.5 mm longer quad section, but still stays within its Z-envelope of 2770 mm to 3600 mm. Clearly, the design is getting tighter, with less available space. David is starting to look at NdFeB as an alternate material for the magnets. The higher remanent field would produce the same strength magnet with 5-10% less material. However, the temperature coefficient of the remanent field is 3 times that of SmCo, and it is less radiation hard. A study unearthed by Hobey DeStaebler, shows that NdFeB losses magnetization when subjected to high doses of neutrons. However, the critical energy of the neutrons was different, and the test did not include other radiation sources Q2 would see. Hobey suggested running a Monte Carlo simulation (radiation physics group?) to understand the type and magnitude of radiation, then possibly running a test, using the same protocol at the Naval Postgraduate School, where the original tests were performed. David has also started investigating the stray fields in the HEB, using an iron shield around the HEB beampipe. Initial values show that the shield does, indeed help, but the field "spills over" the end of the shield. Update of Q2 P.M. "To-Do List" Q2 Chamber -Layout new chamber cross-section through Q2 by 26 Jan 96 (Bob Holmes) Status: Cross-section layed out per Q2 V4 design -Run thermal analysis of cross-section (Bertollini/Nordby) Status: Analysis pending at LLNL. ETA is 16 Feb 96 -Work up new 3-D model of entire Q2 chamber by 2 Feb 96 (Bob Holmes) Status: First-cut complete. Working on details of Q4 magnet interface. Bertolini, Humphries, Swan, and Nordby to discuss details, week of 12 Feb Q2 Magnet -What is max allowable sextupole of stray Q2 field in HEB (Weinands, Seeman) Status: P.M. Q2 harmonics with Lattice Group for simulation. ETA week of 12 Feb 96 -For Q2 quad magnet, first intrinsic harmonic of magnet is either n = 18 or 26 depending on segmentation (16 or 24-blocks, respectively). What is maximum allowable value for these harmonics. This is driving the choice for segmentation (Ecklund, Seeman) Status: P.M. Q2 harmonics with Lattice Group for simulation. ETA week of 12 Feb 96 -What is maximum tune range needed for Q2. Version 4.0 design allows for +10% tune. Is this needed, or is +5% sufficient. What is needed for possible future crossing angle collision at higher energies (Seeman) Status: Design for +/- 5% tune only. This means that current V5 Q2 design can be weakened by 5%, and spare volume used for Q2 Chamber and increasing margin. -Should nominal condition for Q2 (and Q1, for that matter) be the solenoid-off configuration (currently being used), or solenoid-on (perhaps a more logical choice, since this is supposed to be the standard operating configuration). Changing this will buy 1-1.5% strength back (Seeman) Status: For Q2, this does not help at more than the 1% level, per Seeman, so maintaining solenoid-off configuration seems reasonable. For Q1, this may help more, so solenoid-on configuration should be new baseline. -Investigate the effects of adding a 1 mm shield around the HEB beampipe, with and without a shield around Q2. Look at how to balance asymmetry of magnetic circuit outside Q2 (if this is needed), and effects on harmonics of Q2 main field (Humphries) Status: Just started looking at this. More results by 16 Feb 96. -Look at using a doughnut on the ends of Q2 to contain stray fields. Most of the stray field is due to end-effects, so a disk may help minimize this (Humphries) Status: This is, indeed, helpful. End disk replaces the overhanging cylindrical shield to reduce stray end fields. Humphries will incorporate this into the P.M. Q2 design. -Look at integrated harmonics of stray fields at HEB, for the 3-D analyses that have been done (Humphries) Status: David is working on this. Possible results by 16 Feb 96. Since field is dropping off quickly through the HEB, and since the two beamlines are not parallel, the meaning and definition of harmonic values needs better understanding -Measure stray fields in CESR, SSRL, SSC P.M. quad slices (Ecklund, Wolf) Status: Zach Wolf should measure this next week. Stan to follow this up for results by 16 Feb 96. General -Is a mirror plate needed for Q4. The Q4 steel begins at 372.5 cm, but the mirror plate begins at 366.2 cm (6.3 cm closer to Q2). If this could be removed, it would free room for a flange pair for the Q2 chamber (Johanna Swan). Status: A conventional, thick mirror plate is NOT needed, per Jack Tanabe and Johanna Swan. But, plan on a thin (0.25" thick), remove-able (if necessary), mirror plate with an outer diameter equal to the coil diameter. This should help minimize the stray end fields of Q4 in the LEB beampipe. Also, the Q4 coil will be shortened by ~1" to increase z-space for Q2 Chamber flanges. -What are the harmonics of the solenoid field near Q2, where the field bends out radially (Seeman, Ecklund, Sullivan) Status: Stan Ecklund has solenoid field values from Orrin Fackler, and Mike Sullivan will work on this when he returns from China. -How does the Q2 HEB shielding affect the solenoid field. Is more field channeled out of the shielding plug and into the vanadium permendur HEB shield. If so, this could saturate the shield and channel more field into Q4 (Mattison, Fackler) Status: David Humphries and Tom Mattison to develop some shielding options and work with Orrin Fackler on 2-D models. Hopefully, this will be parasitic use of Orrin's time, around the IFR steel modelling work. -How could Q2 be better shielded from the solenoid field, especially in the Cantilevered Raft design, where there are two slots cut in the Q2 shielding plug (Mattison, Fackler, Nordby) Status: [From discussions after the meeting] Martin Nordby and Scott Debarger to investigate alternate cantilever designs to minimize 3-D solenoid field leakage These minutes, and agenda for future meetings, are available on the Web at: http://www.slac.stanford.edu/accel/pepii/near-ir/home.html