E-rhic-ir-l mailing list

["Parker, Brett" <parker AT bnl.gov>]

Greetings,

 

In anticipation of our next IR Workgroup meeting on April 5, 2019, I have uploaded a PowerPoint file with information relevant to our forming a baseline official lattice that will be the focus for our common study. Note this file is not a full final presentation but rather is intended to give everyone a heads up on some of the more critical magnet parameters.

 

The file can be found on the SharePoint site eRHIC IR documents area under the 2019.04.05 folder at URL: link to IR_meeting_5apr19_official_lattice_heads_up.pptx. Note I was careful to include the required monospaced fonts in this file so all text should display properly in the SharePoint web viewer interface and for anyone who downloads this file.

 

Some comments:

 

Slide 1: The beta* is as shown but the actual Twiss functions are tentative until a full match is done to give the required optics functions at the Crab Cavities (located beyond this region after the B2APF and B2BPF dipoles and any require matching quads). The intent here is to follow the region shown with at a minimum a defocusing quadrupole and a focusing quadrupole to transform the beta_x up to the target 1600 m level. It is unfortunately obvious that without a more complicated transformer (which will take up a lot of space in order to bring the Crab Cavity back to a multiple of 90 degrees phase advance with respect to the IP) it will be impossible to get the slope of the dispersion to be zero at the Crab Cavities. Note this non-zero slope was the case for the previous preCDR match and with simple optics we can only look to minimize the magnitude of the slope.

 

 

Slide 2: The basic IR geometry matches the Mad/Bmad convention that all quadrupoles are centered on a natural reference orbit as Guillaume's tentative geometric solution already assumes. However there is a perturbation to the 275 GeV orbit due to the fact that Q2PF is reverse rotated by 4 mrad and offset by 5 mm. The effect of this rotation/offset is corrected with a non-zero value for the C1XPF corrector dipole field before Q2PF and adjustment of the field of the B1PF main dipole after the quad. This -0.96 mm bump restores the orbit to the unperturbed reference geometry at the downstream end of B1PF up to an almost negligible change in path length due to this bump.

 

Note the B1PF location shown, much the same as Bob's last iteration, is shifted further from the IP than Guillaume has previously assumed. So Guillaume will have to do a readjustment of the ring matching in any scenario. This will be addressed in greater detail later.

 

For the 100 GeV and 41 GeV optics cases the non-scaling of the B0 spectrometer causes an additional orbit deviation which are both brought back to the design orbit by a combination of C1XPF and B1PF as shown. Here the specific example of the 41 GeV floor geometry is shown. Because the orbit is always brought back to the reference value in B1PF and does not impact its aperture specification.

 

Slide 3: Here we see the magnet parameter table for all three energies, 275, 100 and 41 GeV with the required clear beam pipe half-aperture (radius) is show along with the final field strength or gradient and the effective "pole tip field" according to Bob's convention. Note is this table we distinguish between the scaled field strength, which can be used to calculate the nominal Mad/Bmad reference trajectory and the correction to this field (her denoted as field strength error given to Bmad) to make the previously discussed 3-bumps that bring us back to the reference orbit at the downstream end of B1PF in spite of the Q2PF offset and B0 spectrometer non-scaling. Of course for the magnet design we need to take into account the total field that these magnets are expected to deliver at each energy.

 

Slide 4: Here we show a closeup of just the forward proton magnet apertures overlain with various trajectories generated at 275 GeV. A careful eye should note the reverse rotation of Q2PF. This 4 mrad rotation pus 5 mm X-offset adds about 11 mm of additional separation between the electron and hadron beam pipe apertures at the front end of Q2PF while keeping the downstream end of Q2PF centered on the +/- 4.75 mrad (roughly 1.3 GeV transvers momentum)  outer acceptance limits at 275 GeV. This tilt also helps to properly accommodate the excursions for the circulating beam at 100 and 41 GeV. The 10 sigma proton beam profile is quite a bit smaller than the 1.3 GeV acceptance band and is not shown.

 

In optimizing these magnet apertures and rotations/offsets I decided not to rotate/offset Q1PF since a reverse rotation does not do much good for the acceptance at the upstream end but it certainly can change the acceptance at the downstream end of Q1PF (subtle detail is that Q1PF is defocusing while Q2PF is focusing).

 

Slide 5: Since there can in principle be concern that Q1PF "fits" alongside the electron beam pipe, we show here a proof of principle, fully 3D, design for a NbTi actively shielded, Direct Wind magnet that with reasonable assumptions of the beam pipe and outer cold mass structure. I say in principle because this design goes beyond the field strengths that have been used to date for Direct Wind magnets; however, this is a topic for a new LDRD proposal to investigate just how far we can take Direct Wind magnets performance wise. It is expected that Direct Wind is the most cost effective approach for making this magnet, but as a fall back position we intend to investigate making the inner coil structure using the conventional collared coil Rutherford technique (shield coil would still be Direct Wind as in the FOA Fast Track Actively Shielded Prototype R&D).

 

Frankly with my existing software setup I have only had time to crank out a Direct Wind design; the Rutherford coil alternative will probably only happen after the April cost review.

 

Summary and open issues:

 

This lattice and magnet design avoids having to use Nb3Sn conductor while providing space for the second electron focusing quadrupole (not yet shown) along side the C1XPF corrector at a location compatible with keeping within the present synrad cone acceptance the drives the rear side magnet designs (i.e. keep present rear side designs). The Q2PF and B1PF have very similar positions and apertures to those Bob has previously shown and therefore have the same challenges/drawbacks. Still this solution does support the modifications requested by Holger to support his April cost review presentation. That is he can now state that for the magnets he feels are questionable in the presently frozen design to be presented in April, we have identified mitigation measures we can pursue that are backed up by a proton forward lattice that meets know physics acceptance requirements. That is if Holger thinks it makes sense to break up B1PF into two magnets to meet these/Bob's new lattice requirements, he is fee to do so.

 

However there is an important open issue, the 275 GeV B1PF field shown in the previous table was deliberately adjusted to give roughly the same ring geometry match as Guillaume's present solution after taking into account the difference in the new bend center location for B1PF in the present/Bob's lattice (up to very minor changes in path length due to shifting B1PF). This B1PF field does not however guarantee that we have sufficient separation between the forward charged particles measured at the Roman Pots and the nominal horizontal outer edge of the ZDC neutron detector. That is why I stress that the present lattice is a step towards fixing upon a new official reference design where some further study is needed as soon as possible and this is why I am circulating these slides before the next meeting as a heads up for where we next need to go.

 

If, as I strongly suspect, we need to increase the B1PF bend angle and/or move B2APF/B2BPF further down stream, this would force Guillaume to make a more aggressive change to his present ring geometry matching solution. This could in turn impact his choice of hadron beamline offset and inclination at the IP which in turn impacts Steve's electron beam line and would need to be checked by Doug to see how we fit into the tunnel. It will be interesting to see Doug's presentation which should hopefully show Guillaume's present layout to see just how much "wiggle room" we might have if the B1PF bend angle has to change significantly. Of course any B1PF change impacts Holger's/Brett's B1PF magnet design solution.

 

But to summarize, the above outlines the draft version of the official revised working lattice which we will discuss at the April 5^th IR meeting.

 

Sincerely yours, Brett Parker

 

 

 

-----------------------------------------------------------------

.signature_basic

                    Brett Parker, SMD/AM

                     Building 902A

                     Brookhaven National Laboratory

                     Upton, NY 11973, USA

    BNL Office Tel: (631) 344 3231

    BNL Office Fax: (631) 344 2190

USA T-Mobile Cell: (631) 873 8589

   Skype Username: ILCBNL

   (Caller ID is linked to my T-Mobile cell phone number)

         eMail: mailto:parker AT bnl.gov

 Home Page URL: http://www.bnl.gov/magnets/Staff/Parker/

-----------------------------------------------------------------