Mathcad - Mathcad - SpulOkayama‹…0.8,aj⋅0.8,la,d NL = Lteil()11, = Lteil()12 12, = Computation...
Transcript of Mathcad - Mathcad - SpulOkayama‹…0.8,aj⋅0.8,la,d NL = Lteil()11, = Lteil()12 12, = Computation...
Mini-Coil Design
Michael von Ortenberg
Fukui University, Okayama University,Humboldt University at Berlin
presented at
Mini Coil Workshop
IMR, Tohoku University, November 24,2006
PARAMETERS of the Mini-Coils constructed:
Coil A B C D E Bore (mm) 3/2.6 3/2.6 3/2.6 3/2.6 3/2.6 Wire (mm) 0.5 Cu/Ag 0.5 Cu/Ag 0.5 Cu/Ag 0.5 Cu/Ag 2x0.5
Cu/Ag Number of layers
12 16 14 12 12
Total height (mm)
20 20 20 15 20
R300 (Ohm) With steel flanges
1.017 1.76 1.304 0.747 0.285
R300 (Ohm) Without steel flanges
- - - 0.728 0.285
R77 (Ohm) with steel flanges
0.257 0.371 - - -
L300 (µH) without steel flanges
- - - 138,7 -
L300 (µH) with steel flanges
292 758 549.7 182.45 86
B/Ucoil (T/V)
2.98/100 1.30/102 2.82/101 3.64/101 3.20/103
Bmax (T)/Ucoil 47.78/1997 41.06/1997 46.29/1997 51.85/1806 51.72/1805 τ/2 =(tBmax-tB0) (msec) at 77 K
0.74 1.03 0.886 0.564 0.414
Operation crowbar crowbar crowbar crowbar crowbar
Schematic of Construction:
nn n+1n+1
BBnn
BBn+1n+1
aann
aan+1n+1=a=ann+d+d
aan+2n+2=a=an+1n+1+d+d
BBnn+B+Bn+1n+1
22
BBn+1n+1==BBnn--ΔΔBn(aBn(ann))
aann+0.5d+0.5d
σσnn=(B=(Bnn+B+Bn+1n+1)*I*(a)*I*(ann+0.5d)+0.5d)22 πd2
4
Schematic of Calculation:
Farad U1 5000:= Volt 90 kJC2 0.96 10 3−⋅:= Farad U2 2000:= Volt 4 kJ
Current density in wire at maximum current:
jI0
0.25 π⋅ d2⋅
:= j 9.065= [kA/mm^2] in wire
jeI0
d2:= je 7.12= :[kA/mm^2] effective homogeneous current density
Fieldfunctions in the ho mogeneous coil:Field [T] in the coil axis at height z [mm] above the center:a [mm] inner radius of the one-layer coil, la [mm] total height of the coil, Icurrent in kA
B z a, la, I,( )1.26 I⋅2 d⋅
0.5zla
−
0.5zla
−⎛⎜⎝
⎞⎟⎠
2 ala
⎛⎜⎝
⎞⎟⎠
2+
0.5zla
+
0.5zla
+⎛⎜⎝
⎞⎟⎠
2 ala
⎛⎜⎝
⎞⎟⎠
2+
+
⎡⎢⎢⎢⎢⎣
⎤⎥⎥⎥⎥⎦
⋅:= : for effective homogneous currentdensity I0/(d^2)
Field [T] at position y [mm] for z=0 inside or outside of the coil:
Math-CAD Program:
Program SpulOkayama-Sendai.mcd for the computation of MINI COILS inOKAYAMA for Cu/Ag-wire with breaking tension 0.78 G-PaVersion 26.10.06 based on SpulNeu.mcd of 7.11.02
i 1−:=
Parameters:a1 1.5:= [mm] inner radius of the coil
la 20:= [mm] total height of coil
I0 1.78:= [kA] current in the wireBm 48:= [T] maximum field of coild 0.5:= [mm] diameter of wireρd 9.96:= [gr/cm-3] density CuNL 12:= : number of windings
Capacitor banks #1 and #2:
C1 7.2 10 3−⋅:=
BE y a, la, I,( )1.26 I⋅2 π⋅ d⋅
0
π
φ
1ya
⎛⎜⎝
⎞⎟⎠
cos φ( )⋅−
1ya
⎛⎜⎝
⎞⎟⎠
2+ 2
ya
⎛⎜⎝
⎞⎟⎠
⋅ cos φ( )⋅−⎡⎢⎣
⎤⎥⎦
0.25yla
⎛⎜⎝
⎞⎟⎠
2+ 1
yla
⎛⎜⎝
⎞⎟⎠
⋅ cos φ( )⋅−ala
⎛⎜⎝
⎞⎟⎠
2+⋅
⌠⎮⎮⎮⎮⎮⎮⌡
d⋅:=
Remark: BE(y) has a pronounced step at the position y=a and reduces to 0 forlarger y.
Fieldstrength BF(0,y,z) inside and outside the coil:y-component:
BFy y z, a, la, I,( )1.26− I⋅
2 π⋅ d⋅
0
π
φcos φ( )
la2 a⋅
za
+⎛⎜⎝
⎞⎟⎠
21+ 2
ya
⋅ cos φ( )⋅−ya
⎛⎜⎝
⎞⎟⎠
2+
⎡⎢⎣
⎤⎥⎦
0.5
cos φ( )−
la2 a⋅
za
−⎛⎜⎝
⎞⎟⎠
21+ 2
ya
⋅ cos φ( )⋅−ya
⎛⎜⎝
⎞⎟⎠
2+
⎡⎢⎣
⎤⎥⎦
0.5+
...⌠⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⌡
d⋅:=
z-component:
BFz y z, a, la, I,( )1.26− I⋅
2 π⋅ d⋅
0
π
φ
ya
cos φ( )⋅ 1−⎛⎜⎝
⎞⎟⎠
la2 a⋅
za
−⎛⎜⎝
⎞⎟⎠
⋅
1 2ya
⋅ cos φ( )⋅−ya
⎛⎜⎝
⎞⎟⎠
2+
⎡⎢⎣
⎤⎥⎦
la2 a⋅
za
−⎛⎜⎝
⎞⎟⎠
21+ 2
ya
⋅ cos φ( )⋅−ya
⎛⎜⎝
⎞⎟⎠
2+
⎡⎢⎣
⎤⎥⎦
0.5
⋅
ya
cos φ( )⋅ 1−⎛⎜⎝
⎞⎟⎠
la2 a⋅
za
+⎛⎜⎝
⎞⎟⎠
⋅
1 2ya
⋅ cos φ( )⋅−ya
⎛⎜⎝
⎞⎟⎠
2+
⎡⎢⎣
⎤⎥⎦
la2 a⋅
za
+⎛⎜⎝
⎞⎟⎠
21+ 2
ya
⋅ cos φ( )⋅−ya
⎛⎜⎝
⎞⎟⎠
2+
⎡⎢⎣
⎤⎥⎦
0.5
⋅
+
...
⌠⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⌡
d⋅:=
Computation of the tension in element (d x d) in homogeneous coil enclosingwire of diameter d:
a: inner radius of the winding considered in [mm]d: wire diameter in [mm]B: Magnetfield produced by the winding abover the considered winding in [T], j: Current density within the considered winding in [kA/mm^2]
Bb1 Bm:= a1 1.5:=
n 2 NL 1+..:=
an an 1− d+:=
Bbn Bbn 1− B 0 an 1−, la, I0,( )−:=
σn 1−Bbn 1− Bbn+( )
2I0⋅
an 1−d2
+⎛⎜⎝
⎞⎟⎠
d2( )⋅ 10 3−⋅:= : G-Pa in square element (d x d) in contrast to wire
cross section πd^2/4
σ
0
0.57
0.663
0.724
0.756
0.758
0.732
0.679
0.599
0.493
0.363
0.21
0.033
0.033−
⎛⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎝
⎞⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎠
=Bb
0
48
43.564
39.166
34.814
30.517
26.284
22.119
18.028
14.016
10.086
6.24
2.479
1.196−
⎛⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎝
⎞⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎠
=a
0
1.5
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
⎛⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎜⎝
⎞⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎟⎠
=
σ; strain in wire of winding in G-Pa
a: inner winding radius in mm Bb: magnetic field inside winding in T
NL 12=Total number of windings:
Hence within +/- 2 mm around center field inhomgeneity very small!BB 2( )BB 0( )
0.991=
10 8 6 4 2 0 2 4 6 8 1020
30
40
50
60
BB z( )
z
BB z( )
1
NL 1+
n
B z an 0.5 d⋅+, la, I0,( )∑=
:=
Calculation of the z-dependence of the field in the center axis:
: [mm]Dtotal 15=Dtotal 2 aNL d+( )⋅:=Total diameter of coil:
σNL 0.033=σNL 1+BbNL 1+
2I0⋅
aNL 1+d2
+⎛⎜⎝
⎞⎟⎠
d2( )⋅ 10 3−⋅:=
’·‚³
1
NL
liter
aliterd2
+⎛⎜⎝
⎞⎟⎠
2⋅ π⋅lad
⋅∑=
:= ’·‚³ 1.357 104×= : mm total wire length
n 1 NL 1+..:=
0 2 4 6 8 10 120
50
100
150
Bbn
σn 100⋅
78
n
del 0.25001:= : [mm]
n 0 50..:=
m 0 50..:=
Myn m,
1
NL
liter
BFy del n⋅ del m⋅, aliter, la, I0,( )∑=
:=
Mzn m,
1
NL
liter
BFz del n⋅ del m⋅, aliter, la, I0,( )∑=
:=
Mn m, Myn m, i Mzn m,⋅+:=
Radial Component
My
Axial Component
Mz
M
Computation of the compressive force in k-Newton/meter=N/mm on the wire at the coil edge(radial unit: 0.25 mm):
liter 1 50..:= faradliter Myliter 40, I0⋅:=
faradliter
liter
Computation of the axial force acting on the different layers:m 1 20..:= r 6 8, 28..:=
—Ím1
12
r
My4 2 r⋅+ m 2⋅, I0⋅ ar⋅ 2⋅ π⋅∑=
:= : force produced by layer inposition z=0.25*m on thetotal of coil
—Ím =
: Newton
—Ím
m
Summation of all forces within one coil-half side:
sum
1
20
n
—Ín∑=
:= sum = :Newton , total force of one coil-half
MLm n, Lteil m n,( ):=
n 1 NL..:=m 1 NL..:=
:HenryInduk =Induk
1
NL
i 1
NL
j
Lteil i j,( )∑=
∑=
:=
Lteil 12 12,( ) =Lteil 1 1,( ) =NL =Lteil i j,( ) L ai 0.8⋅ aj 0.8⋅, la, d,( ):=
Computation of the inductance of the magnetic coil in consideration
Henry4 π⋅ 10 7−⋅ 106⋅ 25⋅ π⋅10 3−
1000⋅ =
classical result for "long" coil of one layerand 1000 windings
HenryL 5 5, 1000, 1,( ) =test result of present calculation:
L a b, L, d,( )2 π⋅ 10 10−⋅ a⋅ b⋅
d2
0
2 π⋅
cos φ( )−
L−
2
L
2
zln
L2
z−⎛⎜⎝
⎞⎟⎠
2 L2
z−⎛⎜⎝
⎞⎟⎠
2a2
+ b2+ 2 a⋅ b⋅ cos φ( )⋅+ 10 5−++
a2 b2+ 2 a⋅ b⋅ cos φ( )⋅+ 10 5−+
⎡⎢⎢⎢⎣
⎤⎥⎥⎥⎦
ln
L2
z+⎛⎜⎝
⎞⎟⎠
2 L2
z+⎛⎜⎝
⎞⎟⎠
2a2
+ b2+ 2 a⋅ b⋅ cos φ( )⋅+ 10 5−++
a2 b2+ 2 a⋅ b⋅ cos φ( )⋅+ 10 5−+
⎡⎢⎢⎢⎣
⎤⎥⎥⎥⎦
+
...
⎡⎢⎢⎢⎢⎢⎢⎢⎢⎣
⎤⎥⎥⎥⎥⎥⎥⎥⎥⎦
⌠⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⌡
d
⎡⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎣
⎤⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎦
⋅
⌠⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⎮⌡
⋅:=
Number of windings per layer: la/d
d =la =
Computation of the inductance of the coil constructed by several layers ofwindings:Length measurements in [mm], Induktivität in [Henry].Height of the coil la [mm]Radii of windings a1, a2 to be considered for cross inductanceDiameter of wire d [mm]
Calculation of temperature increase in coil after shot considering the "Action Integral" for ahalf-sinus current pulse of length τ and maximal current density j:0.5*τ*j^2=Integral from starting temperature Tex before the shot to final temperature Tf afterpulse with length τ of integrand [ρd*c(T)/ρCu]ρd: density of Cu (not temperature dependent)c(T) specific heat as function of temperature T
[mm] skin depth T=4.2 KδHe =δHe δRT 0.04⋅:=
[mm] skin depth at T=77 Kδ77 =δ77 δRT 0.13⋅:=
[kV] Voltage for maximum fieldUmax =Umax Rblind I0⋅:=
[Ohm] ratio of U/IRblind =RblindLgescoul
:=
[mm] Skin depth at T=273 KδRT =δRT 0.2 τ⋅ρ
μ0 π⋅⋅:=Skin depth in wire::
V*sec/A*cmμ0 4 π⋅ 10 9−⋅:=
[msec] pulse lengthτ =τ π 1000⋅ Lges coul⋅⋅:=
Lges Induk:=Ohm*cm at T=273 Kρ 1.95 10 6−⋅:=
[Farad] capacity of the bankcoul =coul C2:=
Fixing of the system parameters:
ML =
q12 0.42132:=
Tq11 98.169:=q11 0.33307:=Tq10 84.921:=q10 0.24208:=Tq9 73.68:=q9 0.16937:=
Tq8 66.449:=q8 0.12628:=Tq7 57.528:=q7 0.08050:=Tq6 49.032:=q6 0.04516:=
Tq5 43.162:=q5 0.02744:=Tq4 36.80:=q4 0.01380:=Tq3 30.972:=q3 0.00638:=
Tq2 23.278:=q2 0.00157:=Tq1 14.558:=
T0 4 5, 300..:=
[Ohm*cm]ρCu T0( ) interp vsq Tq, q, T0,( ) 10 6−⋅:=vsq cspline Tq q,( ):=
Tq20 297.855:=q20 1.7055:=Tq19 250.187:=q19 1.3865:=Tq18 208.061:=q18 1.1017:=
Tq17 188.174:=q17 0.9687:=Tq16 175.55:=q16 0.8784:=Tq15 152.720:=q15 0.7191:=
Tq14 133.033:=q14 0.58023:=Tq13 117.178:=q13 0.46746:=Tq12 110.620:=
r5 0.0063:=r6 0.0074:=r7 0.0097:=r8 0.014:=r9 0.0195:=
tesla10 60.0:=tesla11 80.0:=tesla12 100.0:=tesla13 150.0:=tesla14 200.0:=tesla15 250.0:=tesla16 273.0:=
r10 0.065:=r11 0.15:=r12 0.23:=r13 0.46:=r14 0.68:=r15 0.92:=r16 1.0:=
Resistance values:
Temperature dependence of the specific resistance of Cu:Data after HENNING (Fritz Herlach)
( )ρCu(T) specific resistance as function of temperature T
q1 0.00016:=Tq0 4:=q0 0.0005:=
Data after LANDOLT-BÖRNSTEIN:
[Ohm*cm]ρcu T0( ) interp vs tesla, r, T0,( ) ρ⋅:=vs cspline tesla r,( ):=
tesla0 4.0:=tesla1 6.0:=tesla2 8.0:=
r0 0.004:=r1 0.0044:=r2 0.0048:=
tesla3 10.0:=tesla4 15.0:=tesla5 20.0:=tesla6 25.0:=tesla7 30.0:=tesla8 35.0:=tesla9 40.0:=
r3 0.005:=r4 0.0057:=
0
ρcu T0( )
ρCu T0( )
0 300T0
: Ohm*cm
Specific Heat after Debye:
Specific heat after Debye:c T0( )T0343
⎛⎜⎝
⎞⎟⎠
3
0
343
T0x
x4 ex⋅
ex 1−( )2
⌠⎮⎮⎮⎮⌡
d
⎡⎢⎢⎢⎢⎣
⎤⎥⎥⎥⎥⎦
⋅ 1.2017⋅:= c 4.2( ) = c 77( ) =
T0 5 6, 250..:=
c T0( )
ρcu T0( ) 105⋅
ρCu T0( ) 105⋅
c T0( )5
ρcu T0( ) 108⋅⋅
c T0( )5
ρCu T0( ) 108⋅⋅
0 250T0
Tex 4.2:= [K] starting temperature of the coil before shot
j = :maximum current dernsity
w T0( )
Tex
T0
teslac tesla( )
ρCu tesla( )
⌠⎮⎮⌡
d:= f0.5 τ⋅ j2⋅ 1.6501⋅
ρd106⋅:= :Factor 1.6501 resulting from magneto
resistance of the form: (1+0.00766*B(t))[after Fritz HERLACH]
f =
T0 10 40, 500..:=
w T0( )
0 500T0
[K] final temperature of thecoil after shot
Tf =Tf rootw T0( )
f1− T0,
⎛⎜⎝
⎞⎟⎠
:=T0 100:=
w T0( )
0 500T0
T0 10 40, 500..:=
f =
:Factor 1.6501 resulting frommagneto resistance of the form:(1+0.00766*B(t))
f0.5 τ⋅ j2⋅ 1.6501⋅
ρd106⋅:=w T0( )
Tex
T0
teslac tesla( )
ρCu tesla( )
⌠⎮⎮⌡
d:=
:maximum current densityj =
[K] starting temperature of the coil before shotTex 77:=
[K] final temperature of the coilafter shot
Tf =Tf rootw T0( )
f1− T0,
⎛⎜⎝
⎞⎟⎠
:=T0 7:=
φd