1. CERN LCD MAGNET MEETING SiD SOLENOID UPDATE
Wes Craddock
SLAC
May 18, 2010
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

2. SiD MAGNET GENERAL STATUS and UPDATE SINCE THE CLIC OCTOBER 2009 WORKSHOP
Baseline Design is Based on CMS Design, Construction Techniques and Conductor
Iron End Cap HCAL Field Calculations
Fringe Field Studies
Brett Parker (BNL) has made Opera 3D calculations incorporating the DID
ANSYS 3D modeling work with the DID coil has started.
Still pursuing advanced aluminum stabilizer development.
Preliminary Detailed Engineering Design with Overall Construction Drawing Due at the End of 2010 (SLAC Objective).
Reduction in available manpower at SLAC. Wes Craddock assigned 1/3 time to CDMS. John Weisend moves to Michigan State University.
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

3. SiD SOLENOID MAJOR DESIGN TASKS (Same as October 2009 CLIC Workshop)
3D MAGNET FIELD CALCULATIONS FOR SOLENOID & DID COMBINED FIELDS AND FORCES
2D MAGNETIC FIELD CALCULATIONS FOR STRAY FIELDS
DESIGN OF THE DID COIL AND INTEGRATION WITH THE SOLENOID
ASSEMBLY AND INSTALLATION PROCEDURES
OVERALL TOLERANCES AND FINAL SIZE SPECIFICATIONS
STRUCTURAL ANALYSIS OF COIL PACKAGE, SUPPORTS AND VACUUM SHELL
ENGINEERING DRAWINGS
CRYOGENIC INTEGRATION OF THE SiD SOLENOID WITH QD0
CONDUCTOR R&D
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

4. SiD MAGNETIC FIELD ANALYSIS STATUS
IRON END CAP HCAL STUDY --- DONE
FRINGE FIELD REDUCTION --- Struggling for better results
3D ANALYSIS WITH DID COILS --- 3D ANSYS starting.
Brett Parker has Opera 3D model
POWER SUPPLY Modified CMS design
DUMP BREAKERS
DUMP RESISTOR
May 18, CERN LCD Magnet Meeting
SiD Solenoid
Wes Craddock / SLAC

5. SiD Iron HCAL ENDCAP STUDY (Presented LCWS Beijing, March 2010)
ANSYS was used to see if it was worthwhile to use iron in the EndCap HCAL.
POSSIBLE ADVANTAGES:
Improved field uniformity
Reduced number of solenoid amp-turns or greater superconductor stability.
Slight reduction in material cost
DISADVANTAGES:
Magnetic forces on the HCAL with increased construction and engineering costs
Substantially greater difficulty in magnetic field mapping
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

6. SiD ANSYS 2D FEM MODEL (showing Iron EndCap HCAL)
Air extends
much further
IRON
AIR
COIL AL
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

7. SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

8. Fe
SiD Solenoid
Wes Craddock /SLAC
March 29, LCWS 2010

9. SiD Solenoid
Wes Craddock / SLAC
May 18, CERN LCD Magnet Meeting

10. SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

11. SiD Iron HCAL ENDCAP CONCLUSIONS
SUMMARY OF RESULTS:
Magnetic forces are a large but manageable 250 T (towards the Door) to -400 T (into the solenoid) during solenoid ramping to full field.
An iron HCAL EndCap reduces the operating current from A to A ( only a 4% reduction).
Field uniformity is improved.
DISADVANTAGES:
Magnetic forces on the HCAL with increased construction and engineering costs.
Substantially greater difficulty in magnetic field mapping.
CONCLUSIONS:
It is doubtful whether the improved field uniformity is enough to offset, the very substantial increased difficulty in field mapping and to a lesser extent the forces.
However, this option exists if it is considered to be absolutely essential.
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

12. SiD FRINGE FIELD REDUCTION
Trying for 100 G at 1 m (LOI)
More typical values are 300 to 500 G at 1 m above the Door
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

13. SiD FRINGE FIELD REDUCTION
A few of the many iron profiles tried for fringe field reduction
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

14. SiD FRINGE FIELD SUMMARY
Have not found a really good design yet. Still have many more options to try. The ANSYS model is segmented in many different areas so this is easy and fast to solve (35,000 elements.
If lower fields are really needed everywhere, adding enough iron (cost) will always work.
Still fairly optimistic that a good design is still possible.
Good News: Most iron configurations have fringe fields above the door/barrel with minimum values at the mid-plane where most all the electronics would be or could be located.
There is little difference in fringe fields between Iron End Cap HCAL or “Air” End Cap HCAL geometries.
SiD Solenoid
Wes Craddock
May 18, CERN LCD Magnet Meeting

15. 3D ANALYSIS FOR DID COILS
Simplified
DID Model
An ANSYS 3D model that includes the DID coils has begun.
This model will be used to compare the OPERA 3D that Brett Parker (BNL) has created and solved.
This ANSYS model will permit direct / easy coupling of DID forces into structural analysis. It can eventually be used for transient analysis and coupling of the solenoid to the DID
The ANSYS model uses the very new and improved SOLID 236/237 edge-flux formulation elements.
Race track coils with rounded ends have been created.
SiD Solenoid
Wes Craddock
May 18, CERN LCD Magnet Meeting

16. SiD SOLENOID POWER CIRCUIT
May 18, CERN LCD Magnet Meeting
SiD Solenoid
Wes Craddock / SLAC

17. SiD SOLENOID POWER CIRCUIT
SiD Power Circuit Based on the CMS Design
Differences Between SiD and CMS
Power supply operates in only 1 quadrant, positive voltage & positive current.
The CMS supply is two quadrant, positive and negative voltage with pos. current.
This means CMS uses more complex thyristors but can voltage control ramp down.
SiD uses simpler and more reliable free wheeling diodes.
SiD uses a water cooled resistor.
SiD has no changeable buswork for current reversal.
Fermilab is looking at the grounding /ground monitoring scheme.
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

18. SiD CONDUCTOR OPTIONS (Other than baseline CMS conductor)
Dilute high purity Al alloy or high purity Al matrix with superconducting cable and two high strength wire ropes
Dilute high purity Al alloy or high purity Al matrix with
superconducting
Rutherford cable
CMS conductor must carry 130 kN = 94 MPa (avg) of hoop load
With the assumption that all the CMS hoop load is carried by its structural aluminum with a safety factor of 2.4 on 4 K yield, the average required 4 K aluminum alloy yield stress for SiD is 225 MPa unless option 2 is used.
Note: High purity aluminum has a 4 K yield of 10 MPa = 1500 psi
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

19. HIGH PURITY/IMPROVED STRENGTH ALUMINUM STABILIZER OPTIONS
Dilute Aluminum alloys using Nickel, Cerium, Yttrium and Scandium
Perhaps two dozen other Al binary systems have been studied but virtually none have both 4 K strength and electrical conductivity measured.
Cold working is essential to increasing strength but is also detrimental to electrical conductivity.
Other materials such as the industry standard TiB2 grain refiner.
Multiwalled carbon nanotubes.
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

20. ALUMINUM / CARBON NANOTUBES
Aluminum reinforced with carbon nanotubes is the most intriguing of the nanocomposites. It comes in single walled and multiwalled. Multiwalled is the cheapest(~ $500 to $5000/kg) and is the only one considered here.
At the present time there are only about 20 published papers discussing aluminum carbon nanotube composites.
Carbon nanotubes are incredibly strong and stiff.
Much debate and theoretical work is being done on carbon nanotube electrical properties. Some reports show them to be ballistic conductors. Some reports show that they can have a critical current carrying capacity of 109 A/cm2.
Electrical resistivity of Al carbon nanotube composite
99.5 % Al + 10 wt % nanotube
30 nm X 200 nm to micrometers long
1% and 4% nanotube graphs are very similar
From C.L. Xu et. al. Carbon 37 (1999)
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

21. ALUMINUM / CNT MANUFACTURING
Although CNTs may hold the greatest promise for high purity aluminum reinforcement, they are the most difficult material to produce at least on an industrial scale.
To date all Al-CNT composites have been produced by powder metallurgy. This is mostly likely very difficult on an industrial scale.
Molten aluminum does not wet CNTs
CNTs come tangled
CNTs stick together by Van der Waals forces and agglomerate at grain boundaries.
Try the liquid metal route using one or more of the following techniques:
Plate the CNTs; electroless Ni or Cu; vapor deposit Ti, Nb, NbTi, Ce? Y?.
Disperse the CNTs with electromagnetic stirring and/or ultrasonic cavitation.
Rheocasting (semi-solid casting)
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

22. CNT R&D
A SLAC Lab Directed R&D (LDRD) proposal was submitted again this year. The $125 K proposal was reduced in scope from the previous year to study only carbon nanotubes.
Wes Craddock will visit the master alloy company KB alloys prior to the June SiD meeting at Argonne. Will discuss liquid processing and other topics such as the Conform/Conklad extrusion process.
Argonne is also interested in helping with this work (perhaps CNT plating).
LLNL has acquired a couple samples of 99.5% pure Al reinforced with carbon nanotubes from Yonsei University in Korea. They intend to test them.
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

23. AL-Sc Alloys and Conklad Extrusion
Scandium is unique among all possible elements that could be used to create dilute high purity aluminum alloys.
Simply stated Sc is a powerful grain refiner, provides the greatest increment of strength per atom, and is the most powerful antirecrystallization element that can be added to aluminum. Al-0.26wt% can raise the recrystallization temperature of aluminum to 600 C. Unfortunately, Sc costs $5500/kg (purchased as 2% master alloy). For SiD this means $1.14 M additional cost. However, lesser amounts would probably be needed since extrusion temperature is only 430 C and rare earth metals can partially substitute for Sc.
This material might be best coextruded with the Rutherford cable by the Conklad method (as in ATLAS).
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting

24. CONCLUSIONS (Same as presented at the October 09 CLIC workshop)
There are no show stoppers to the SiD solenoid.
CMS and ATLAS design philosophy and engineering experience are the basis of the SiD solenoid design.
Much work remains to produce a first pass set of engineering drawings with not much manpower.
Advanced conductor development is an exciting area of many avenues of approach, potentially benefiting all large detector magnets as well as other areas such as high field MRI magnets.
SLAC would enjoy a collaboration with CERN, INFN and other institutions.
In the USA, LLNL, ANL, Univ. of Texas A&M, and small companies such as Milward and KB alloys (master alloy producers) have expressed substantial interest in collaboration.
SiD Solenoid
Wes Craddock /SLAC
May 18, CERN LCD Magnet Meeting