Of MICE Spectrometer Solenoid 2 M Tartaglia 18 January 2013 The Spectrometer Solenoid M1 M2 E1 C E2 40 cm diameter warm bore 26m long cryostat SS2 under test SS1 under repair ID: 779735
Download The PPT/PDF document "A Summary of the Fermilab Magnetic Measu..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
A Summary of theFermilab Magnetic MeasurementsOf MICE Spectrometer Solenoid 2
M. Tartaglia
18 January 2013
Slide2The Spectrometer Solenoid
M1 M2 E1 C E2
40 cm diameter warm bore
2.6m long cryostat
SS2 under test; SS1 under repair
All coils wound on one machined Aluminum (6061-T6)
Mandrel; Outer Aluminum Bands
Slide3Solenoid Power, Control & Monitoring
Testing and Measurements used 3 power supplies:
Configured to power M1-only, M2-only, E1+C+E2 in series
Slide4The Measurement System
Hall Probe: Senis 10T probe (S/N5406)
Field Readout:Keithley 2700 Mux/DMM
Student developed Labview systemProbe positioning SystemProbe mount with close pair of bearings shaft sideLong G10 shaft (sections) with continuous scaleScribe line to keep probe angular orientationStainless Steel guide tube (2 joined sections)
Tested in TC1206 at 4 T to ensure non magnetic jointGuide Tube positioning plates at each endTube positioning slots Machined to 0.1mm precision
G10 capture rings each end fix the Z position reference
Slide5The Measurement System
Slide6The Measurement System
3D Hall probe calibration checked against NMR probe up to 4 T at Fermilab MTF in Tevatron dipole TC1206; stainless steel guide tube and seam also verified as non-magnetic .
Slide7The Measurement System
Coordinate Systems
Probe Readout:
+Z is from M1 towards E2+Y is up
+X is to the left as viewed from M1 looking toward E2Probe Position: +Z is from M1 towards E2; measured wrt M1 end of the guide tube (fixed by capture rings wrt end plates)“X100” is really X= -100 mm in probe frame“Y100” is really Y= -100 mm in probe frame
Field: Bz points from E2 to M1 (along –Z); Opera matches Data
+Bx
+By
+Bz
Slide8The Measurement System
Magnet Current
Readout
Visual front panel display on each PSInternal shunt recorded by LINUX monitoring sys.Digital values agreed with nominal settings (<1%)Measurement Procedure (2 people)Start a new file for each Z-scanEncode coils, current, date into file nameMove Guide Tube to desired location, adjust ref. angle
Manually position probe in Z (moving in or out)1 or 2 cm steps, except when centering bearings near endAdjust probe angle to align scribe markManually enter current, Z position to Labview GUITrigger reading of Hall probe voltages (avg
~3/min)Automatically recording 10 measurements for each Z step
Slide9The Measurement History
6/11/2012
0 A noise, offset levels
Checkout 10 A M1+M2, X0Y06/120 A noise, offset checkM2 50A scans, X0Y0; X100Y0; X0Y100ECE 50A scans, X0Y0; X100Y06/13Shaft jammed in tube; removed and repaired (Z shifted!)
M1 50A scan, X0Y0ALL 185 A started (3 points), then Quench6/14ALL 150A scans, X0Y0; X100Y0; X0Y100
Slide10The Magnetic Model
Opera2D is adequate (assuming cyl. symmetry)
Independently modeled in Opera3D by Marc B.
As-Built Geometry and Winding Data UsedMICE note 207, table 5 for SS2 (SS1 is VERY close)Dimensions known at room temperature to 50μmThermal Contraction Coefficient ? I used 3*10-3This gives 7.5 mm shrinkage over the magnet length!
4*10-3 is probably a better number for Aluminum…
Generate {Bz, Br} vs Z to match each data setFor Specified coils at 50A or 150A, at R=0 and R=100mmFine 1mm spacing in Z, to make best match to data
Slide11The Magnetic Model
Slide12The Data Analysis
Use MATLAB to calculate dBz, dBr (data-model)
For each data set, determine Z
offset to give best agreement to match Bz (by eye; to ~± 2mm)For each Zdata, the program:calculates averages, errors of the 10 Bx, By, Bz data measurements
calculates BxCor= Bx + α ·Bz, where α=-.016 (later -.018)Due to constant X-Z tilt of the probe
(this is ~ 1 degree)Loops through array of model points, finds index of (Zmodel-Zoffset) that is closest to Zdatathen calculates dB=Bdata-B
model for that ZdataWrites a new excel spreadsheet of B’s and dB, vs ZRun program separately to generate dBz and dBr
Plot data, model, differences in Excel
Slide13Summary of Results
Zoffsets, from best match of Bz
Date
Coils
CurrentPosition
Zoffset, cm6/11M1M210X0 Y0
(-40.0)6/12M2
50
X0 Y0
-40.0
6/12
M2
50
X100 Y0
-40.2
6/12
M2
50
X0 Y100
-39.8
6/12
ECE
50
X0 Y0
-39.8
6/12
ECE
50
X100 Y0
-39.8
6/13
M1
50
X0 Y0
-41.4
6/13
ALL
185
X0 Y0
QUENCH!
6/14
ALL
150
X0 Y0
-41.66/14ALL150X100 Y0-41.66/14ALL150X0 Y100-41.6
Shaft stuck
Slide14Summary of Results
Bz vs Z
M1M2
on axis
Slide15Summary of Results
Bz vs Z
M1 on axis
Slide16Summary of Results
Bz vs Z
M2 on axis
Slide17Summary of Results
Bz vs Z
M2 off axis
Slide18Summary of Results
Bz vs Z
M2 off axis
Slide19Summary of Results
Bz vs Z
ECE on axis
Why the slope?
Slide20Summary of Results
Bz vs Z
ECE off axis
Why the slope?
Slide21Summary of Results
Bz vs Z
ALL on axis
100 G shift in measured Bz
1A change is
2.25T/150A
=150 G
Slide22Summary of Results
Bz vs Z
ALL off axis
Why the slope?
Slide23Summary of Results
Bz vs Z
ALL off axis
Why the slope?
Slide24Summary of Results
Bz vs Z: Maximum dBz(data-model)/Bz
Date
Coils
Current
PositionM1M2ECE6/11
M1M210X0 Y0
~1%
~7 %
6/12
M2
50
X0 Y0
2.5%
6/12
M2
50
X100 Y0
2.5%
6/12
M2
50
X0 Y100
3.0%
6/12
ECE
50
X0 Y0
0.5%
6/12
ECE
50
X100 Y0
0.5%
6/13
M1
50
X0 Y0
0.5%
6/14
ALL
150
X0 Y0
0.5%
1.0%
3.0%6/14ALL150
X100 Y0
0.5%
1.0%
3.0%
6/14
ALL
150
X0 Y100
0.5%
1.0%
3.0%
Small discrepancies might be improved by a) Z scale (use CTE=4 10-3), b) slight radial offsets (next section)
Slide25Summary of Results
Bz vs Z: Sensitivity to radial position?
For small R, not very:
look at M2-only, at peak Bz: at R=0.1m for dR=10mm, dBz/Bz~5/375=1.3%(ECE: even less)
Slide26Summary of Results
Br vs Z
: Sensitivity to radial
positionM1 example @ 50A
Br vs Z at R=0.1m
Br vs R at peak (Z=-0.3m)
Br ~linear with R;
dBr/dR= 9.1 G/mm
@R=0.1m
Slide27Summary of Results
Br vs Z: Bx and By offsets
Bx(0A) ~ +4 to +5 G By(0A) ~ 0to -5 G
Slide28Summary of Results
Br vs Z:
M1 on axis - easily
affected by probe tilts
Consistent with radial offsets ~ 1mm from magnetic axis
Slide29Summary of Results
Br vs Z:
M2 on axis – peak dB/dR ~ 5 G/mm @ R=0:
dX ~ +2 mm, dY ~ -2.5 mm
Slide30Summary of Results
Br vs Z: Data vs Model
M2 off axis – Bx at X=-100mm (compare to –Br(0.1m))
Slide31Summary of Results
Br vs Z: Data - Model
M2 off axis – peak dBr/dR = 6.7 G/mm @ R=0.1m
dX~0 (<1mm); By affected by probe rotations!
Slide32Summary of Results
Br vs Z: Data vs Model
M2 off axis – By at Y=-100mm (compare to –Br(0.1m))
Slide33Summary of Results
Br vs Z: Data - Model
M2 off axis –
either dY~ -4mm (or some coil ellipticity?) Bx affected by probe rotations
Slide34Summary of Results
Br vs Z
:
Data vs ModelECE on axis – peak dB/dR ~ 9.4 G/mm @ R=0: dX ~ +3 mm, d
Y ~ -4 mm
Why the slope?
Slide35Summary of Results
Br vs Z: Data vs Model
ECE off axis – Bx at X=-100mm (compare to –Br(0.1m))
Slide36Summary of Results
Br vs Z
:
Data - ModelECE off axis – dX ~ +5 mm, d
Y (probe rotation)
Slide37Summary of Results
Br vs Z: Data vs Model
ALL on axis – dBr/dR=
18.5, 29.3 G/mm at R=0 @ E1,E2
dX~ +1.3,
+2.1 dY~ -7.4, -7.3 mm
Why the By slope?
Shift in Bx due to
2
mrad change in probe tilt
Slide38Summary of Results
Br vs Z: Data vs Model
ALL off axis – Bx at X=-100mm (compare to –Br(0.1m))
Slide39Summary of Results
Br vs Z: Data - Model
ALL off axis –
dBmodel /1mm = {25.7, -11.0, 7.5, -4.5, 17.5, -29.0}T/mm
dX ~ {4.7, 1.6, 14.1, 4.2, 7.3, 5.5 } mm (30 G corr. for Bx shift = 1 mrad)
(probe rotated
in Y
)
Slide40Summary of Results
Br vs Z: Data vs Model
ALL off axis – By at Y=-100mm (compare to –Br(0.1m))
Slide41Summary of Results
Br vs Z: Data - Model
ALL off axis –
dBmodel /1mm = {25.7, -11.0, 7.5, -4.5, 17.5, -29.0}T/mm
dY ~ {-5.8, ҉, -16.7,
-14.7, -15.4, -14.1 } mm
(probe rotated
in
X)
Slide42Summary of Results
Br vs Z:
d
X, dY (in mm)
Date
CoilsCurrentPosition
M1M2E1
E2
6/11
M1M2
10
X0 Y0
6/12
M2
50
X0 Y0
+2.0
-2.5
6/12
M2
50
X100 Y0
<1
6/12
M2
50
X0 Y100
-4.4
6/12
ECE
50
X0 Y0
+3.6
-4.3
+3.6
-6.7
6/12
ECE
50
X100 Y0
+4.8
+5.0
6/13
M1
50
X0 Y0
~1
~1
6/14
ALL
150
X0 Y0
+1.3
-7.4
+2.1
-7.3
6/14
ALL
150
X100 Y0
+4.2
+7.3
+5.5
6/14
ALL
150
X0 Y100
-14.7
-15.4
-14.1
Not completely sure what conclusions we can draw from these numbers
Slide43Summary of Results
Br vs Z:
There appears to be a small slope dX/dZ and dY/dZ ~1-2 mrad wrt meas. axis
Did cold mass move after the quench (or at 150A due to greater ext. forces)?
(or, what else might be going on?)
8 mm shift !
On axis/
Off axis
difference
Slide44Lessons Learned
Improvements To Make
Stiffer
beam or tube; Key/slot to fix probe angleDigital encoder for Z-position (e.g., “string pot”)Bearings along the length, better shaft centeringBetter yet, Motorize and automate the scansPiezo motor - Has to operate in high field.Short probe holder between bearings, drawn by motorAdditional Data to Take
Each coil powered separately, on & off axisWider range of currents, better current monitorsCapture current from the shunt directly, with probe V’s