SiD Machine Detector Interface (MDI) Tom Markiewicz/SLAC ALCW 2015, KEK, Japan 23 April 2015

SiD Machine Detector Interface (MDI)

Tom Markiewicz/SLAC

ALCW 2015, KEK, Japan

23 April 2015

MDI Activities (I)

• Backgrounds and Detector Design– Primary particles produced in the beam-beam interaction

• Disrupted e- or e+ beam (off energy, off angle), beamstrahlung photons, e+e- pairs, hadrons

– Muons from beam halo scraping in the collimation system– Synchrotron radiation (SR) photons, especially from the final doublet– Secondary particles in detector after primary particles strike something

• Beam gas interactions• e+e- pairs striking BeamCal cause backgrounds in VXD, Tracker and Calorimeters• Neutrons streaming back from beam dumps

• Accelerator Design– Work closely with the BDS “Beam Delivery System” group

• Collimation system, Final Focus, Extraction Line

– Work with the LEP “Luminosity, Energy & Polarization” group– Work with those selecting IP spot parameters

• Detector Design– Beampipe and BeamCal in particular– Detector resident masks, apertures

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLAC 2 of 40

MDI Activities (II)

• Engineering– Work with the magnet designers: QD0 and QF1 especially

• May be possible to shorten QD0/QF1 &/or change crossing angle

– Work with mechanical engineers: • Location, support & alignment of everything within r<30cm out to z=9.5m• Push-pull mechanical and cryogenic systems

– Work with civil engineers: underground IR Hall, above ground Assembly Hall

• Multidisciplinary:– Radiation protection: self-shielding, “PACMAN” design, etc.

• Shielding design and radiation calculations

– Vibration measurement & control, especially of QD0– Alignment, especially of QD0– Metrology

• Frequency Scanning Interferometer to re-establish alignment after a push-pull

– Cryo: • guaranteeing system, lines, interconnects consistent with detector design and push-pull operation

– Vacuum for z<200m– Feedback to stabilize/maximize Luminosity: “FONT” BPM and Kicker– Beam commissioning with or without detectors


MDI People

• The Future– Everything that has been done to produce the LOI/DBD will need to be re-done in the

context of the Japanese site specific design• Many results used for the design are >7 years old• Details have changed yet not been incorporated into a self-consistent collection of results• Design changes: new L* desired by AD&I team• “Value Engineering” should be applied as ILC project comes on the mass shell

– A new team must learn how to use existing tools or develop better tools– Many projects have not begun or have been abandoned for lack of available

personnel

• Past contributors (partial list):– Takashi Maruyama & Lew Keller (backgrounds)

– Mike Woods, Mike Hildreth(NDU), Eric Torrence (UO), Ken Moffeit (LEP Issues)

– Mario Santana (FLUKA radiations calculations)

– Wes Craddock (Solenoid & Yoke design; Cryo)

– Marty Breidenbach, Phil Burrows(OU), Marco Oriunno, TWM (all)

– Brett Parker, Andy Marone & Bill Sporre (BNL QD0)

– Bill Cooper(FNAL), Mike Sullivan, Sasha Novokhatski (Beam Pipe, Vacuum, HOM Heating)


-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0 1 2 3 4 5 6 7 8 9 10

SiD 4.1/9.5m FD with FB Kicker Behind QD0(Door Closed)

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLAC

HCAL Door Yoke PACMAN

QD0 CryostatQF1

Cryostat

QD0 L*=4.1m

QF1 L*=9.5m

QD0 Service Pipe

FB Kicker

BPMs

BeamCal

PolyCarbonate

LumiCal

W MaskBeampipe

ECAL

Movers

Beampipe Spider

Support

Bellows & Flange

117cm156cm

Valves/Pumps

5 of 40

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0 1 2 3 4 5 6 7 8 9 10

SiD 4.1/9.5m FD with FB Kicker Behind QD0(Door Closed)


HCAL Door Yoke

QD0 CryostatQF1

Cryostat

QD0 L*=4.1m

QF1 L*=9.5m

QD0 Service Pipe

FB Kicker

BPMs

BeamCal

PolyCarbonate

LumiCal

W MaskBeampipe

ECAL

Movers

Beampipe Spider

Support

Bellows & Flange

117cm156cm

Valves/Pumps

2.8m

6 of 40

-8

-6

-4

-2

0

2

4

6

8

0 1 2 3 4 5 6 7 8 9 10

SiD Schematic(Door Open 2.8m)


HCAL

Door Yoke

QD0 L*=4.1m QF1

L*=9.5m

QD0 Service Cryostat

FB Kicker

PACMAN

Barrel Yoke

Solenoid Cryostat

HCAL

Cryo Chiller

7 of 40

“Recent(?) Past” MDI Projects

2011-12 for DBD:– Add real vacuum components to beamline to

make sure it is buildable– QD0 Support Tube and Magnet Mover System– Frequency Scanning Interferometer System– Vacuum and High Order Mode Calculations

Older:– Self Shielding Calculations (2010?)– PACMAN Engineering (2008?)– Beam-Beam Backgrounds, DID/Anti-DID,

Apertures (2006 and for “SB2009”)


Hot Topic Again

8 of 40

Granada LCWS’11 Engineering ModelGo from PPT Engineering to COTS parts


G. Anzalone/SLAC

Transitions Critical

9 of 40

More Aggressive Bellows Design


Assume functionality of double multi-convolution

bellows provided by bellows outboard of the

VXD support

Need to ensure adequate clearance between

beampipe and all FCAL components

Suggested that this bellows “comes for free” if

back end of cone appropriately engineered

Handshake with QD0/FCAL Support Scheme Essential

Rejected Conservative Design2-multi & 1-single convolution

bellows to allow FCAL position & VXD angle adjust

22% Evts in LumiCal lost

10 of 40

Vacuum: Still Assumes QD0 Cryo-Pump Only,But…..


(“Virtual”?) Single Convolution Bellow in front of BeamCAL

Recent suggestion to eliminate individual beampipes within BeamCal and the

Low-Z absorber for better conductance

Previous VersionImplication to reduced BeamCal acceptance

not yet considered

11 of 40

Space Behind BeamCal Increased from 100mm→277mm to accommodate BPM


W-Mask to protect QD0 from Pairs

Common or Individual bellows

4-electrode 100mm stripline

1um resolution BPM(2nd BPM Now Requested)

Increased separation to accommodate

bellows, BPM and flanges

12 of 40

Transverse Space Behind BeamCal Tightly Constrained


Electro-mechanical Design of Fast Feedback BPM & Kicker

by S. Smith/SLAC 2006

13 of 40

Evolution of QD0-QF1 region(L* = 3.5m Design; needs updating)


•Valve/Pump Out/RGA assemblies near QF1 end•QD0 Service Line to 2K chiller extended maximally to rear•Support tube behind QD0 extends to allow 2.8m door opening transitioning to a half-cylinder for access

Door plates Door Ring5683mm

QD0 Service Line

Wedge Alignment System in Pockets in Door Ring

Back end of QD06933mm

Back end of Support Tube

Valve Assembly550mm FB Kicker8090mm from IP

FB BPMG. Anzalone/SLAC

Front QD03283mm

Front QD09236mm

14 of 40

Plan & Elevation Views of Disconnect & Pumpout ValvesRequired for Rapid Push-Pull


1-1/2” Gate Valves

Formed Bellows & Disconnect

Flanges

Angle Valve for Roughout Pump

G. Anzalone/SLAC

Ion Pumps

All Vacuum Parts from MDC Catalog

15 of 40

Integration of the QD0 cryoline

Hilman Rollers

QD0 service cryostat

He2 cryoline

Ancillary platform

T. Markiewicz/SLAC2015.04.23 SiD MDI ALCW2015 16 of 40

M. Oriunno/SLAC

BNL Design of QD0 Support and Alignment System

• ANSYS analysis of QD0 outer can in either original 0.25” or suggested 1” thickness shows unacceptable deformation of cold mass support structure

• BNL designed an external support tube and a compact mover system


Bill Sporre/BNL

Bolt Hole Rotation Pt.

17 of 40

Support Tube Analysis Weight Distribution When LumiCal & BeamCal Supported on Extension of QD0 Cryostat

Support Tube 980 kg

QD0908 kg

M. Oriunno/SLAC2015.04.23 SiD MDI ALCW2015

Needs Updating after L* Change to 4.1m

Displacements (mm)VM Stresses (Kgf/mm2)

Door Closed: 100um deflectionAt LumiCal

Supports

M. Oriunno/SLAC



Displacements (mm)VM Stresses (Kgf/mm2)

2.8 mDoor Open 2.8m: 6mm deflectionAt LumiCal

VXD Tracker Supports

M. Oriunno/SLAC



QD0 Wedge Design Concept

Total Pad Travel as is = .475in

Height of pad and distance of displacement will be changed pending analysis on sagging of beam line.

Conceptual design only at this point

Motor & Drive Rod system not yet done

Limit SwitchesPotentiometer

Bill Sporre/BNL

2015.04.23 SiD MDI ALCW2015

Pads located in cutouts in iron ring

to which door plates are welded

216<r<620mm

QD0 Wedge Design Concept

Bill Sporre/BNL

2015.04.23 SiD MDI ALCW2015

Drive Shaft & motors also need work

Frequency Scanning Interferometry

• Pioneered by Armin Reichold et al at Oxford for ATLAS– Oxford has commercialized a system bought by LCLS to

monitor adjustable gap undulator: many heads

• In SiD context, R&D program by Keith Riles et al at U. Michigan 2003-2007 (for SiD tracker) & 2011-2012 (for QD0)

– Established an 8-channel optical table in Michigan lab

– Verified 1-D, 2-D displacement measurements with sub-micron precision – Available equipment limits 3-D displacement test to few-micron precision, but should be fine– Developing corner cubes and customized beam launchers – Currently compact and light enough for MDI application, but not for SiD tracker


FSI Grid for Monitoring QD0


Keith Riles, Hai-Jun Yang/U. Michigan

QD0 QF1

24 of 40

IP Vacuum

• VACCALC: “ A Method for Calculating Pressure Profiles in Vacuum Pipes”, SLAC-PEP-II-APNOTE-6-94– The outgassing rate is taken to be 0.1 nTorr•l/s/cm2.


If cryo-pump only 136 nTorr• If add 10 l/s pump 69 nTorr• If add 20 l/s pump 46 nTorr

M. Sullivan/SLAC

25 of 40

Beam-Beampipe Interactionat the IP and in QD0


Sasha Novokhatski/SLAC

Wakefields and Bunch field as beam passes BEAMCAL and LUMICAL

Beam-induced wakefields result in power loss due to trapped higher order modes and resistive wall heatingThese have been calculated by A. Novokhatski for the current IR geometry using MAFIA and NOVO codes:See: http://ilcagenda.linearcollider.org/materialDisplay.py?contribId=2&materialId=slides&confId=5596

Effects are very dependent on exact geometries, materials, shielding schemes and contact resistance

26 of 40

http://ilcagenda.linearcollider.org/materialDisplay.py?contribId=2&materialId=slides&confId=5596

Summary of HOM Heating at IP

• Average power of the wake fields excited in IR is around 30 W for nominal parameters (6 kW pulsed)– 90% from modes excited in pipe (geometry (R/Q) & frequency

dependent)

– 10% from resistive wall heating

• In the QD0 region there is an additional ~4W from resistive losses in the pipes and wakefields, excited by pipe diameter changes, due to the shielded bellows– Flange edge size, contact resistance, coatings important

• Heating from BPMs and kickers must be added


Full 3D LOI Beam Pipe(no bellows, flanges, or pump ports included)

Sasha Novokhatski/SLAC 27 of 40

Self-Shielding and PACMAN

• PACMAN shields IR Area between the detector and the beam tunnel• Current Concept has a SiD-mounted piece and a ILC/SiD common

piece to accommodate different detector sizes and QD0 support schemes

• Figures of merit are “Total Dose” and “Dose Rate”– Relevant Japanese authorities need to be educated about the nature

of ILC pulsed beams

– Dose Rate limits (which seem artificially low given the short time scale of an “accident” (1ms)”) need to be re-evaluated

• Handling of PACMAN sub-pieces will likely determine underground crane requirements

• Design of UG IR Hall, tunnel lip that holds QF1, interface with platform motion system, QD0 support and push-pull all argue for active work in this area



Electrical motor, low friction hinges

Current SiD “PacMan” Rotating Hinged Design

M. Oriunno, SLAC 29 of 40

T. Markiewicz/SLAC

Geometry: pacman and SiDB-B’

SiD elevation at beam plane Small PACMAN

SiD elevation at beam plane Large PACMAN

SiD elevation at beam plane Large PACMAN

50 cm steel +

170 cm concrete

120 cm steel +

250 cm concrete

M. Santana, SLAC, LCWS 20102015.04.23 SiD MDI ALCW2015 30/40

MSL - SLAC 31

20 R.L. Cu target in IP-9 m. Small pacman.

9 MW9 MW

BEAM PLANE PENETRATION PLANE

9 MW9 MW

µSv/event- 1000 µSv/event- 18000 mSv/h

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLACM. Santana, SLAC - 2010 LCWS Beijing 31/40

MSL - SLAC 32

20 R.L. Cu target in IP @ 9 m. Large pacman.

9 MW9 MW

BEAM PLANE PENETRATION PLANE

9 MW9 MW

µSv/event- 10 µSv/event- 180 mSv/h

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLACM. Santana, SLAC - 2010 LCWS Beijing 32/40

Provisional conclusions

• Small pacman and pacman-cavern interface are sufficient in terms of dose per event.

– Beam aborted after 1 Train• However, the dose rates for the small pacman are very high:

– Proven mechanisms should be installed to:• avoid these accidents to occur• shut off beam after 1 train (200 mS)

– Possible Debates• The large pacman complies with all criteria.• The penetrations in the pacman don’t require local shielding.• The shielding of the detector may be insufficient to comply

with dose rate limit. Exclusion area?• More studies ongoing (mis-steering…)

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLACM. Santana, SLAC 33/40

Backgrounds & Forward Calorimetry

• All studies done by T. Maruyama in GEANT3 (pairs, SR) and FLUKA (Neutrons from Dump, Dose Rates)

• Last looked at for “SB 2009” IP Parameters and presented early 2011

• Have not kept up with changes to nominal beam parameters, beam energy variations, or changes to FCAL

• Questions of DID/anti-DID implementation, “bent solenoid”, new L*, beampipe variants, etc. should be investigated


SiD Forward RegionLumiCal20 layers of 2.5 mm W +10 layers of 5.0 mm W

BeamCal50 layers of 2.5 mm W

3cm-thick Tungsten Mask13cm-thick BoratedPoly

Centered on the outgoing beam line

EC

AL

Beampipe+/- 94 mrad (detector)+101 mrad, -87mrad (ext. line)

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLACT. Maruyama SLAC 35/40

Pair edge and Beam pipe design

Pt=0.032 Pz0.43

Pt

(GeV

/c)

Pz (GeV/c)

• Pairs develop a sharp edge and the beam pipe must be placed outside the edge.

• Find an analytical function of the edge in Pt vs. Pz space.• Taking into account the crossing angle and solenoid field,

draw helices in R vs. Z space. SB2009 500 GeV TF


SiD beam pipe and pairs edgeMessage: Aggressive Beampipe

500 GeV RDR Nominal 500 GeV RDR Low P


Detector-Integrated-Dipole

• Total pair energy into BeamCal is half.– 10 TeV/BX vs. 20 TeV/BX with NO-DID.

• Pair distribution is symmetric and more confined.• No. photons backscattering to the tracker is half.

We want Anti-DID. DID

SiD Solenoid


VXD hit density / train

0.1

1

10

100

Hits

/mm

2/tr

ain

54321

VXD Layer

1000 GeV TF 1000 GeV NoTF 500 GeV RDR 500 GeV TF 500 GeV NoTF 350 GeV TF 350 GeV NoTF 250 GeV TF 250 GeV NoTF

100 hits/mm2

6 hits/mm2

• Detector tolerance

• Use generic 1% pixel occupancy

• Dependent on sensor technology and readout sensitive window.

– Standard CCD 20m x 20m

• 2500 pixels/mm2

• 6 hits/mm2/sw (assuming 1 hit 4 pixels)

– Fine pixel CCD 5m x 5m

• 40000 pixels/mm2

• 100 hits/mm2/sw (assuming 1 hit 4 pixels)

T. Maruyama SLAC

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLAC 39/40

Summary

Every area mentioned would benefit from increased collaboration and fresh ideas

Prototypes and experimental verification of design will eventually be reqired.

2015.04.23 SiD MDI ALCW2015 T. Markiewicz/SLAC 40/40

Documents

SiD Machine Detector Interface (MDI) Tom Markiewicz/SLAC ALCW 2015, KEK, Japan 23 April 2015