[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

*Subject*: [bds 89] Re: emittance growth in upstream chicane optics from schuler@xxxxxxxxxxxx*From*: 田内 利明 <toshiaki.tauchi@xxxxxx>*Date*: Tue, 05 Apr 2005 00:36:10 +0900

Date: Mon, 4 Apr 2005 14:47:39 +0200 (MEST) From: Peter Schuler <schuler@xxxxxxxxxxxx> To: "Walker, Nicholas John" <nicholas.walker@xxxxxxx> Cc: "'Peter Schuler'" <schuler@xxxxxxxxxxxx>, "Woods, Michael B." <mwoods@xxxxxxxxxxxxxxxxx>, "Meyners, Norbert" <norbert.meyners@xxxxxxx>, "Moffeit, Kenneth C." <moffeit@xxxxxxxxxxxxxxxxx>, bds@xxxxxxxxxxxxxx Subject: RE: emittance growth in upstream chicane optics In-Reply-To: <5AF7BC8A6CDCD511880500508BB9E7E6056796CE@xxxxxxxxxxxxxx> Message-ID: <Pine.LNX.4.58.0504041209200.29806@xxxxxxxxxxxxxxxx> References: <5AF7BC8A6CDCD511880500508BB9E7E6056796CE@xxxxxxxxxxxxxx> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT X-AV-Checked: ClamAV using ClamSMTP Hi Nick, indeed we have to be weary of relying on scaling laws, even if we start with the correct scaling expression. The reason is that there are significant differences between the 3-magnet energy chicane and the 4-magnet polarimeter chicane: (a) different chicane geometry: in your 3-magnet energy chicane, the central magnet bend (which you call theta) is twice as large as the bend angle theta/2 in either of the two outer magnets. Now you can imagine splitting the central magnet into two identical halves, so that you get effectively a 4-magnet chicane, where each of the four magnets now generates a bend angle of theta/2. The situation is then still different from the genuine polarimeter chicane, which has four magnets of identical length, while the pseudo 4-magnet energy chicane has two half-length magnets in the middle which operate at twice the field. So each of your pseudo half-magnets will generate twice as much synchrotron radiation as either one of the outer ancillary magnets. Therefore I conclude that our 4-magnet chicane generates only 2/3 as much synchrotron radiation as your 3-magnet chicane, for the same energy and deflection angle theta/2 in the ancillary magnets. (b) there is another difference between the energy and the polarimeter chicane, which affects the scaling of these devices in a major way: while the field setting of the energy chicane is (presumably) scaled with beam energy, so as to maintain a constant deflection, we plan to operate the polarimeter chicane with a fixed field setting at all beam energies, as the Compton spectrum will then always be mapped into the same detector region. Consequently the deflection angle will decrease with increasing energy for our chicane, and the scaling of synchrotron radiation and emittance growth will be quite different. I had anchored the two together for the same theta/2 and beam energy, by applying the 2/3 factor. All we then need is a valid scaling law, that accounts correctly for the operational differences. So I will ponder your corrected version and see what difference it makes. But I certainly agree with your warning that we should be cautious and establish the actual effects through tracking in a realistic lattice. I hope that our SLAC colleagues will be able to do that soon. Cheers, Peter On Mon, 4 Apr 2005, Walker, Nicholas John wrote: > Dear Peter et al, > > Since my name keeps appearing in these discussions, I thought I'd better go > back and check the scaling law I original jotted down now five years ago. > > Peter - the scaling law on my original note/memo is rubbish. Not sure where > I got that from. > > After some thought, I believe the correct scaling for the emittance growth > is more like E^6 theta^5 L_c / l_b^2, where E is the beam energy, theta the > bend angle(s), L_c the length of the chicane and l_b the magnet length. This > scaling law is correct for the three bend chicane geometry in the attached > note (my original note updated - yet again!). > > One should note that the local matched optics also plays a significant role. > I have added an additional section to my note explaining and showing this > effect. > > So the moral of this story is: beware of scaling laws and always check using > tracking in a realistic lattice. (Which I believe the SLAC people are > doing.) > > Regards, > Nick. > > ------------------------------------------------------------------------ > Dr Nicholas Walker DESY ILC Project > phone: +49-40-8998-4570 Deutsches Elektronen Synchrotron > mobile: +49-160-9892-6113 Notkestr. 85 > fax: +49-40-8994-4570 22607 Hamburg > e-mail: nicholas.walker@xxxxxxx Germany > > ------------------------------------------------------------------------ > > >-----Original Message----- > >From: Peter Schuler [mailto:schuler@xxxxxxxxxxxx] > >Sent: Friday, April 01, 2005 5:06 PM > >To: Moffeit, Kenneth C. > >Cc: Peter Schuler; Woods, Michael B.; Walker, Nicholas John; > >Norbert.Meyners@xxxxxxx > >Subject: emittance growth in upstream chicane optics > > > > > > > >Dear Ken, > > > >hmm, only 5 mm chicane dispersion (at 250 GeV), instead of 20 mm. > > > >That makes indeed a lot of difference, to the detector, vacuum > >chambers, etc. I had assumed 20 detector channels with a hor. > >aperture of 10 mm ea., covering a range of x = 20 -> 220 mm > >from the beam axis. > > > >Scaling simply all x-dimensions by a factor 1/4 would give us > >very narrow detector channels of only 2.5 mm nominal, which is > >not comfortable. Maybe 5 mm would be possible. Also, I don't > >want to get closer to the beam than x_min = 20 mm. So that > >would decrease the maximum covered energy. > > > >You indicate considerations about emittance growth as the > >origin of this much more docile chicane design. This surprises > >me, since we have actually investigated this point, based on > >earlier work by Nick Walker on the TESLA energy spectrometer. > >If you look at page 18 of the talk Norbert presented, we > >estimate an acceptabe emittance growth of 2.5%, but only at > >E_cm = 1 TeV, and totally negligible at lower energies. So, > >unless our estimate is bogus, it does not seem to be justified > >to decrease the dispersion of the upstream chicane so > >drastically. My gut's feeling is that we could absorb a factor > >of 0.5. But 0.25 would really hurt. > > > >Even worse, and I would go as far as saying "unacceptable", > >would be only 1.7 meters between the inner dipoles. We assumed > >a center to center distance of 10 meters, which gives us > >nominally 8 meters of space for a typical magnet length of 2 > >meters. The reason for this space requirement is simple > >geometry for the laser beam insertion/exit. We went through > >this exercise already with our Tesla design, and the arguments > >have not changed and are still valid. You want to accomodate a > >crossing angle of 10 mrad, and you must keep the optics away > >from any synchrotron radiation. These conditions are met with > >our design, which accomodates a vertical beam crossing, but > >retains a minimal magnet gap height of only 20 mm for all > >dipoles. With a space of only 1.7 m, you would either have to > >increase the height of the gap by some unacceptable factor, or > >you would have to resort to horizontal beam crossing, which > >will jeopardize the optics and is therefore also not acceptable. > > > >I have not had time to look at all the details of the current > >beam line elements, but I will take a closer look. Obviously, > >there are already enough issues of contention. Clearly the > >actual amount of emittance growth in the chicane should be > >checked with high priority. > > > >Cheers, > >Peter > > > > > >********************************************************************* > >* * > >* K. Peter Sch¸ler Tel.: ++49-40-8998-2015 * > >* DESY-HERMES Fax ++49-40-8998-4034 * > >* Notke Strasse 85 * > >* D-22603 Hamburg e-mail Peter.Schuler@xxxxxxx * > >* Germany * > >* * > >********************************************************************* > > > >On Thu, 31 Mar 2005, Moffeit, Kenneth C. wrote: > > > >> Dear Peter, Mike and Norbert, > >> > >> Peter, please forward this email on to Norbert Meyners. > >> > >> I noticed in Norbert talk that he had assumed the dispersion at the > >> upstream Compton Chicane IP was the same as for the extraction line > >> dispersion (2 cm). Actually it is only 5mm. It was shown > >correctly in > >> my drawing of 24 Nov 04. This changes quite a bit the > >geometry for the > >> detector with the backscattered e- offset from the beam line only > >> ~4.5cm instead of the 17.8cm for the extraction line polarimeter. > >> Assuming a beam pipe radius of ~1cm we only have 3.5 cm for the > >> cerenkov cells. Can the upstream detector be made to work in the > >> smaller horizontal dimension? The reason for the low > >dispersion number > >> is to avoid emittance growth in the final focus. > >> > >> Note, there is only 1.7 meters between BPC2 and BPC3 at the Compton > >> IP. Is that adequate for the laser light to collide? We can > >bring the > >> laser light in/out on the outside of magnets BPC3 and BPC2. Norbert, > >> can you redo the geometry drawings of your talk with the 5mm > >> dispersion? > >> > >> I updated my 24 Nov 04 drawing to refect the upstream beam optics > >> values from Mark Woodley. The new version of the drawing with the > >> correct S location and nomenclature is attached. Below I > >give the web > >> address for the Woodley optics deck, and have extracted the section > >> around the compton chicane for display below. Note the > >chicane starts > >> with bend labelled BPC1A. The 3rd column gives the s location in > >> meters at the end of magnet and the number 0.121359E-03 is the bend > >> angle the magnet. Note, he breaks each magnet, eg BPC1 into 2 parts > >> BPC1A and BPC1B for optics reasons. > >> > >> Regards, > >> Ken > >> > >> > >> > >> >From Mark Woodley, > >> > >> > >http://www.slac.stanford.edu/~mdw/ILC/Beam_Delivery/20050316/ebds1.opt > >> ics > >> > >> $ %16s %16s %e %e > > %e %e %e %e > > %e %e %16s > >> @ XIY %e 0.00000 > >> @ XIX %e 0.00000 > >> @ QY %e 8.84391 > >> @ QX %e 8.06409 > >> @ CIRCUM %le 2550.52946695 > >> @ DELTA %e 0.00000 > >> @ TYPE %08s "OPTICS" > >> @ COMMENT %44s "e- Beam Delivery System 1 [20 mr] (ILC2005)" > >> @ ORIGIN %20s "MAD 8.51/15 Windows" > >> @ DATE %08s "16/03/05" > >> @ TIME %08s "13.03.11" > >> > >> > >> * NAME KEYWORD S L > > K0L K1L K2L K3L > > K4L TILT TYPE > >> > >> "MMOVER" "MARKER" 708.695 > >0.00000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "QPFF3" "QUADRUPOLE" 709.695 > >1.00000 0.00000 -0.309720E-01 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPFFB" "DRIFT" 710.195 > >0.500000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "POLCHIC" "LINE" 710.195 > >0.00000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "PCBPM1" "MONITOR" 710.195 > >0.00000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPCBPM" "DRIFT" 711.595 > >1.40000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC1A" "SBEND" 713.095 > >1.50000 0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC1B" "SBEND" 714.595 > >1.50000 0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPCBBO" "DRIFT" 730.695 > >16.1000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC2A" "SBEND" 733.695 > >3.00000 -0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC2B" "SBEND" 736.695 > >3.00000 -0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPCBBI" "DRIFT" 737.395 > >0.700000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "PCBPM2" "MONITOR" 737.395 > >0.00000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "POL" "MARKER" 737.395 > >0.00000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPCBBI" "DRIFT" 738.095 > >0.700000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC3A" "SBEND" 741.095 > >3.00000 -0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC3B" "SBEND" 744.095 > >3.00000 -0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPCBBO" "DRIFT" 760.195 > >16.1000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC4A" "SBEND" 761.695 > >1.50000 0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPC4B" "SBEND" 763.195 > >1.50000 0.121359E-03 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPCBPM" "DRIFT" 764.595 > >1.40000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "PCBPM3" "MONITOR" 764.595 > >0.00000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "POLCHIC" "LINE" 764.595 > >0.00000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "DPFFB" "DRIFT" 765.095 > >0.500000 0.00000 0.00000 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "QPFF4" "QUADRUPOLE" 766.095 > >1.00000 0.00000 -0.426096E-01 0.00000 > >0.00000 0.00000 0.00000 "~" > >> "BPMQ079" "MONITOR" 766.095 > >0.00000 0.00000 0.00000 0.00000 > >> > > > > ********************************************************************* * * * K. Peter Sch¸ler Tel.: ++49-40-8998-2015 * * DESY-HERMES Fax ++49-40-8998-4034 * * Notke Strasse 85 * * D-22603 Hamburg e-mail Peter.Schuler@xxxxxxx * * Germany * * * * address at SLAC from 30. Aug. - 31. Oct. 2004: * * * * K. Peter Schuler Tel.: (650) 926-4678 * * SLAC-MS66 * * bldg. 212 - room 3 e-mail: schuler@xxxxxxxxxxxxxxxxx * * 2575 Sand Hill Road * * Menlo Park, CA 94 025, USA * * * ********************************************************************* ------ End of Forwarded Message

- Prev by Date:
**[bds 88] Emittance Growth in Polarimeter Chicane vs dispersion** - Next by Date:
**[bds 90] diagnostics section options for ATF2, etc.** - Previous by thread:
**[bds 88] Emittance Growth in Polarimeter Chicane vs dispersion** - Next by thread:
**[bds 90] diagnostics section options for ATF2, etc.** - Index(es):