2D ANSYS analysis of the QXF structure Mariusz Juchno QXF internal meeting 19 February, 2013
2D ANSYS analysis of the QXF structure
Mariusz Juchno
QXF internal meeting19 February, 2013
Keypole material and size
• Keypole slot added
19/02/2013Mariusz Juchno 2
L
W/2 D
• I0 = -19670 A• Keypole
– D = 4.64 mm (from HQ drawing)– L = 10 mm– W = 6, 10,15 mm (8 mm hole needed)
• Shellth = 27 mm• Interf = 650 um• Pbladder = 41.7 MPa (for interf + ~100 um)• Special cases to illustrate possible adjustment
– (*) Interf = 600 um -> Pbladder = 38.9 MPa– (**) Interf = 625 um -> Pbladder = 40.4 MPa
Parameters (155 T/m, 90% of Iss)
19/02/2013Mariusz Juchno 3
Keypole study summary
19/02/2013Mariusz Juchno 4
Mat W Coil σeq max
warmCoil σeq max cold
Iron σeq max warm
Iron σI max cold
Coil pcont
Al 6 mm 101 164 208 233 -4.5, -12.4
10 mm 102 164 199 234 -4.3, -12.6
15 mm 104 165 190 234 -4.0, -12.8
G10 10 mm 106 167 195 233 -8.1, -14.7
15 mm 109 168 190 232 -9.3, -15.7
SS 10 mm 100 162 202 234 -2.4, -11.1
15 mm 100 162 190 235 -1.1, -10.7
Ti 10 mm 101 161 201 235 -1.2, -9.6
15 mm 102 160 190 235 0.7, -8.8
G10 * 10 mm 99 164 185 223 -3.6, -11.8
G10 ** 15 mm 106 167 183 228 -7.0, -14.1
• Keypole width variation – effect only visible in case of iron σeq max at warm
• Thermal contraction and elastic modulus plays important role for coil contact pressure (preload)
• Best candidates:– G10
• Simplifies insulation scheme• Bigger thermal contraction -> intercepts less force• Might allow to reduce preload• Risk of loosing contact with collars
– Titanium• Can be integrated in the pole• Smaller thermal contraction -> intercepts more force• Requires more preload
19/02/2013Mariusz Juchno 5
Keypole study summary
Sensitivity study
Mariusz Juchno
QXF internal meeting19 February, 2013
Interference
19/02/2013Mariusz Juchno 7
Interference
19/02/2013Mariusz Juchno 8
-50 0 50180
190
200
210
220
230
240
250
interference [m]
[
MP
a]
Maximum stress in the iron
eqv
warm - bladder
eqv
warm - preload
I cold - max gradient
-50 0 50-16
-14
-12
-10
-8
-6
-4
-2
0
interference [m]
p cont
[M
Pa]
Pole contact pressure
layer 1 - mid node
layer 2 - mid node
-50 0 5038
39
40
41
42
43
44
45
interference [m]
p [M
Pa]
Bladder pressure
bladder pressure
-50 0 5090
100
110
120
130
140
150
160
170
180
interference [m]
[
MP
a]
Maximum stress in the coil
warm - bladder
cold - cooldown
cold - max gradient
• Pbladder adjusted to always have +100um more than interference
19/02/2013Mariusz Juchno 9
Shell thickness
Shell thickness
19/02/2013Mariusz Juchno 10
-2 -1 0 1 2180
190
200
210
220
230
240
250
shellth [mm]
[
MP
a]
Maximum stress in the iron
eqv
warm - bladder
eqv
warm - preload
I cold - max gradient
-2 -1 0 1 2-16
-14
-12
-10
-8
-6
-4
-2
0
2
shellth [mm]
p cont
[M
Pa]
Pole contact pressure
layer 1 - mid node
layer 2 - mid node
-2 -1 0 1 239
40
41
42
43
44
45
shellth [mm]
p [M
Pa]
Bladder pressure
bladder pressure
-2 -1 0 1 290
100
110
120
130
140
150
160
170
180
shellth [mm]
[
MP
a]
Maximum stress in the coil
warm - bladder
cold - cooldown
cold - max gradient
• Iron σI stress at cold slightly more sensitive than in interference case
• OD is fixed• Pbladder adjusted to have
fixed interference
19/02/2013Mariusz Juchno 11
Pad thickness
Pad thickness
19/02/2013Mariusz Juchno 12
-10 -5 0 5 10180
200
220
240
260
280
300
320
padth [mm]
[
MP
a]
Maximum stress in the iron
eqv
warm - bladder
eqv
warm - preload
I cold - max gradient
-10 -5 0 5 10-15
-10
-5
0
5
padth [mm]
p cont
[M
Pa]
Pole contact pressure
layer 1 - mid node
layer 2 - mid node
Yoke(busbar slot)
Pad(corner)
• Gain (when decreasing padth):– Reduction of iron σI stress at cold– Improve the contact (L1 more than L2)– Should improve keypole contact
• Loss (when decreasing padth):– Increase of stress in the coil– Increase of stress in the iron at warm
(pad cornet – might not be important)
-10 -5 0 5 1080
100
120
140
160
180
200
padth [mm]
[
MP
a]
Maximum stress in the coil
warm - bladder
cold - cooldown
cold - max gradient
19/02/2013Mariusz Juchno 13
Vertical key position
Vertical key position
19/02/2013Mariusz Juchno 14
-20 -15 -10 -5 0-14
-12
-10
-8
-6
-4
-2
0
2
keyy [mm]
p cont
[M
Pa]
Pole contact pressure
layer 1 - mid node
layer 2 - mid node
• Gain (when shifting the key down):– Reduction of coil stress at cold– Reduction of the pole contact offset between
layers– More space for the bladder
• Loss (when shifting the key down):– Increase of σI stress in the iron at
cold (mostly busbar slot)– Chance for losing keypole contact
-20 -15 -10 -5 0160
180
200
220
240
260
280
300
keyy [mm]
[
MP
a]
Maximum stress in the iron
eqv
warm - bladder
eqv
warm - preload
I cold - max gradient
-20 -15 -10 -5 0100
120
140
160
180
200
220
keyy [mm]
[
MP
a]
Maximum stress in the coil
warm - bladder
cold - cooldown
cold - max gradient
19/02/2013Mariusz Juchno 15
Key length
Key length
19/02/2013Mariusz Juchno 16
0 5 10 15 20-13
-12
-11
-10
-9
-8
-7
-6
-5
-4
keyh [mm]
p cont
[M
Pa]
Pole contact pressure
layer 1 - mid node
layer 2 - mid node
• Gain (when lowering the bottom face):– Reduction of coil stress at cold– Reduction of the pole contact offset between
layers– Smaller chance for losing keypole contact
• Loss (when lowering the bottom face):– Increase of σI stress in the iron at
cold (mostly busbar slot)
0 5 10 15 20140
160
180
200
220
240
260
keyh [mm]
[
MP
a]
Maximum stress in the iron
eqv
warm - bladder
eqv
warm - preload
I cold - max gradient
0 5 10 15 20100
110
120
130
140
150
160
170
180
190
keyh [mm]
[
MP
a]
Maximum stress in the coil
warm - bladder
cold - cooldown
cold - max gradient
19/02/2013Mariusz Juchno 17
Pad Engagement (?)
19/02/2013Mariusz Juchno 18
Pad Engagement (?)
-10 -5 0 5 10100
110
120
130
140
150
160
170
180
eng [mm]
[
MP
a]
Maximum stress in the coil
warm - bladder
cold - cooldown
cold - max gradient
-10 -5 0 5 10180
200
220
240
260
280
300
320
eng [mm]
[
MP
a]
Maximum stress in the iron
eqv
warm - bladder
eqv
warm - preload
I cold - max gradient
-10 -5 0 5 10-14
-12
-10
-8
-6
-4
-2
eng [mm]
p cont
[M
Pa]
Pole contact pressure
layer 1 - mid node
layer 2 - mid node
• Gain (decreasing eng):– Control over the pole contact
pressure (not significant)– Space for axial rods
• Loss (decreasing eng):– Plasticization of the pad
corner (artifact?)
19/02/2013Mariusz Juchno 19
Busbar slot angle
19/02/2013Mariusz Juchno 20
Busbar slot angle
-6 -4 -2 0 2100
110
120
130
140
150
160
170
180
busdeg [deg]
[
MP
a]
Maximum stress in the coil
warm - bladder
cold - cooldown
cold - max gradient
-6 -4 -2 0 2180
200
220
240
260
280
300
busdeg [deg]
[
MP
a]
Maximum stress in the iron
eqv
warm - bladder
eqv
warm - preload
I cold - max gradient
-6 -4 -2 0 2-13
-12
-11
-10
-9
-8
-7
-6
-5
-4
busdeg [deg]
p cont
[M
Pa]
Pole contact pressure
layer 1 - mid node
layer 2 - mid node
• Gain (when increasing):– Small reduction of σI
stress in the iron at cold
Design CriteriaWork in progress
11/15/20122nd Joint HiLumi LHC - LARP Annual
Meeting - H. Felice 21
• Pole-coil contact at 155 T/m (90% of Is),
pcont ≥ 2 MPa in midpoint
• Max bladder pressure < 50 MPa (better 40 MPa?)
• Bladder should open the interf=interfnom + 100μm
• σeq coil max ≤ 150-200 MPa at 4.3K and 155 T/m
≤ 100 MPa at 293K
• All components σ ≤ Rp 0.2
• For iron at 4.3K (brittle) σI ≤ ~200 MPa
Material Rp 0.2 [MPa]
293 K 4.3 K
Al 7075 480 690
SS 316 LN 350 1050
NITRONIC 40 353 1240
MAGNETIL 180 723
Ti 6Al 4V 827 1654
Parameters tuning
19/02/2013Mariusz Juchno 22
Parameter Reference Modified
Material Al Ti G10 G10
Kpw 15 mm 15 mm 15 mm 12 mm
Interf 650 um +0 +0 +0
Shellth 27 mm -2 -1 -2
Padth 42 mm -4 -2 -2
Keyy 27 mm -2 -1 -1
Keyh 12.7 mm +4 +2 +2
Eng 33 mm -6 -5 0
Busdeg 30o +2 +2 +2
Structure state (90%(1), and 80%(2) Iss)
19/02/2013Mariusz Juchno 23
Reference ModifiedAl Ti G10 G10
15 mm 15 mm 15 mm 12 mm
Coil σeqv (b) 104 107 110 105
σeqv (k) 71 75 77 75
σeqv (c) 172 174 184 180
σeqv (g) 165 148(1), 131(2) 163(1), 146(2) 160(1), 144(2)
Iron σeqv (b) 190 174 180 175
σeqv (k) 196 170 180 175
σI (c) 219 180 192 186
σI (g) 234 194(1), 192(2) 208(1), 205(2) 199(1), 197(2)
pblad (gap) (3) 42 (750,767um) 40 (740,760um) 40 (733,753um) 39 (730,750um)
Pcont -4, -12 -2, -5(1)
-24, -17(2)
-11, -14(1)
-33, -27(2)
-7, -11(1)
-28, -24(2)
Kp gap 0 um 0 um ~0 um 0 um(3) Pblad conservative due to model symmetry, otherwise around 10% lower -> lower coil and iron stresses during bladder operation
Straight vs Round Collars
Mariusz Juchno
QXF internal meeting19 February, 2013
Geometry
19/02/2013Mariusz Juchno 25
Coil contact
19/02/2013Mariusz Juchno 26
• Trapezoid collars– Possible to obtains
similar contact on the midnode
• Round collars– Layer 2 not evenly loaded
while layer 1 overloaded– Tension-compression
transition close to the midnode
-0.8 MPa
-7.7 MPa
-11MPa
-7 MPa
Collar contact
19/02/2013Mariusz Juchno 27
• Trapezoid collars– More uniform contact
distribution over the length similar to coils height
• Round collars– Force transfer close to
the midplane -> layer 2 gets not enough preload
Structure state
19/02/2013Mariusz Juchno 28
Reference RoundAl Al
15 mm 15 mm
Coil σeqv (b) 104 107
σeqv (k) 71 86
σeqv (c) 172 200
σeqv (g) 165 137
Iron σeqv (b) 190 190
σeqv (k) 196 191
σI (c) 219 214
σI (g) 234 231
pblad (gap) (3) 42 (750,767um) 42 (752,764um)
Pcont -4, -12 -14, -4
Kp gap 0 um 8 um
• Overloaded 1st layer (200 MPa in the coil after cooldown, and much higher contact pressure)
• 2nd layer not sufficiently loaded – keypole gap opened
• Round collars seem les sensitive to optimization due to force transfer close to the midplane