TD-01-066

plantcalicobeansUrban and Civil

Nov 29, 2013 (3 years and 6 months ago)

58 views

1

f

Fermilab





Fermi National Accelerator
Laboratory








P.O. Box 500


Batavia, Illinois


60510



TD
-
01
-
066

September 27, 2001




Bolt Instrumentation for the First Nb
3
Sn Racetrack Model


S. Yadav

Fermi National Accelerator Laboratory

MS 316, PO Box
500
,

Batavia, IL, USA



Abstract

In this note we present some of the results from the instrumented bolts used to
measure the required pre
-
load for the first Nb
3
Sn racetrack model fabricated at
Fermilab.





Fermilab is involved in the development of a high

field accelerator magnet for future
hadron colliders using Nb
3
Sn superconductor and the react
-
and
-
wind technology. In
order to develop and optimize the fabrication techniques and to study the conductor
performance, a magnet (HFDB01) with flat racetrack ty
pe coils in a common coil
configuration was assembled and tested. The fabrication details for this racetrack model
along with the test results can be found in reference [
1
]. The mechanical assembly of the
racetrack is shown in Figure 1. The main components

of the mechanical structure are two
40 mm thick stainless steel plates (“main plates” in the following), which provide pre
-
stress and support of the main component of the magnetic force (in the direction normal
to the coil plane). 57 stainless steel bolts

(main bolts, L), with a 25 mm (1.0") diameter,
(pre
-
loaded at 2300 Kg after magnet impregnation) should restrain the coil separation
within 0.2 mm at maximum field. Side pushers provide vertical pre
-
stress and support by
means of 32 bolts (side bolts, S),

each with 12 mm (0.5") diameter. In the ends pre
-
stress
and support are given on each side by a 25 mm
-
thick plate and 8 bolts (end bolts, M),
each with a 20 mm (0.75") diameter.


Four of each of the three different kinds (main, side and end) of bolts were

instrumented
with strain gauges to measure the applied pre
-
load. Two strain gauges were mounted on
the diametrically opposite sides of each bolt on a 0.080" deep slot machined on the bolt.
The instrumented bolts were then calibrated, in compression and at

room temperature and
4.2 K, in the Engineering Laboratory by Tom Wokas. Note that although in application
the bolts are loaded in tension, we calibrated them in compression due to the available

2




Figure
1
: Racetrack assembly.




equipment and fixture. Further, stainless steel (the bolt material) has similar behavior in
tension and compression. The only difference in tension and compression behavior in an
instrumented bolt can occur due to the difference in the bond behavior in

the two modes
of loading. However, this difference is very small, less than 50 microstrains. Therefore,
our methodology of calibrating the bolts in compression but using them in tension is
adequate.


Figures 2 to 4 present the calibration data for the bo
lts. Any apparent bending in the bolt
was subtracted out by averaging of the data from the diametrically opposite gauges. Note
that bending is equal in magnitude but opposite in sign for the two gauges. Calibrations
were performed at both the room temperat
ure and at 4.2 K. The resistance value of the
gauge with no applied load (R
-
zero) was obtained from the calibration data. The slope of
the calibration curve (in pounds per microstrain) was also obtained from the calibration
curve. Figure 5 presents a summa
ry of the calibration data for all the gauges. We also
obtained theoretical estimates for the calibration constant for the three different types of
bolts at room temperature. These estimates are summarized in Figure 6. Note that the
difference in the calcu
lated and the measured values is due to the difference in the slot
Main Bolts (L)

End Bolts (M)

Side Bo
lts (S)

3

size used in the calculation and the actual machined slot size
1
. With the above caveat, we
observe that the computed values for the calibration constant are very close to the
experimentally

measured values.


Another observation that we would like to make is the fact that the ratio of the calibration
constants at 4.2 K and at room temperature should be essentially equal to the ratio of the
Young's modulus at these two temperatures
2
, which is

210 GPa/190 GPa = 1.1. The
measured data at 4.2 K does provide support to this observation. For the large bolts, we
could not calibrate them at 4.2 K due to the restrictions imposed by our equipment.
Therefore, the calibration constants at 4.2 K for the l
arge bolts were obtained by
multiplying the room temperature constant by a factor of 1.1.




Figure
2
: A typical calibration curve for the instrumented bolts in compression (compressive strain is
positive in the above graph). Thre
e calibration runs numbered from 1 to 3 were performed. 'A'
refers to data from one gauge and 'B' refers to data from the diametrically opposite gauge. The
bending strain was compensated by taking an average of the two strain gauge readings. As expected
a
linear load vs. deformation behavior is observed when the bending strain is averaged out. Initially
for gauge A (for loads less than 200 lbs), the tensile microstrain (
-
ive in our convention) due to
bending is larger than the imposed compressive microstrai
n.





1

We did not place tight tolerances on the machined slots on the two side
s of the bolt
.

2

Ignoring the effect of the change in bolt dimensions at 4.2 K.

0
2000
4000
6000
8000
10000
12000
-100
0
100
200
300
400
500
600
700
microstrain
Load (in lbs)
A-1
B-1
Avg-1
A-2
B-2
Avg-2
A-3
B-3
Avg-3
4

Figure
3
: Calibration curve at room temperature for the main bolt, L02. This calibration run shows
significantly less bending than that observed in Fig. 2.



Figures 7 to 9 show the data obtained when the bolts were tightene
d on the racetrack in
ICB3. Figure 10 shows the quench current for a series of quenches performed on the
racetrack on July 19, 2001. Figures 11 to 17 show the main bolt data. As expected, the
load in the main bolts versus the square of the current is a lin
ear behavior. We have also
presented the resistance changes for the compensator gauges in Figures 18 to 19 to
indicate that there is not a significant change in resistance due to the change in magnetic
field. This would suggest that the use of the compensa
tor gauges is not warranted for, at
least for current values up to 8 kA in the present case. A summary of the main bolts data
is presented in Table 1 on the next page. The average loss in prestress due to cooldown
was 2600 lbs. Whereas, the average gain in

bolt load due to energization up to 8.3 kA
was 1800lbs. We did not observe any significant change in the side and the end bolt
loads.

0
2000
4000
6000
8000
10000
12000
0
100
200
300
400
500
600
microstrain
Load (lbs)
A-1
B-1
Avg-1
A-2
B-2
Avg-2
A-3
B-3
Avg-3
5

Figure
4
: Calibration curve at room temperature for the main bolt, L04. This calibration run
e
ssentially shows no bending at large strains (compare it with Fig. 2 and 3).




Table
1
: Summary of the main bolt data.



MAIN
BOLT


273 K

(Lbs)


4.4 K

(Lbs)


Cool
down
loss

(lbs)

Energization
Gain for 8.3
kA

(lbs)

L1

6540

4140

2
400

1600

L2

3600

590

3000

2000

R3

4260

1800

2460

1800

R4

5950

2400

3550

1000

AVERAGE



2600

1800


0
2000
4000
6000
8000
10000
12000
0
100
200
300
400
500
600
microstrain
Load (lbs)
A-1
B-1
Avg-1
A-2
B-2
Avg-2
A-3
B-3
Avg-3
6

Figure
5
: Measured calibration data for the gauges.

Figure
6
: Theoretical calculations for the calibr
ation constant at room temperature.


7

Figure
7
: The side bolts were torqued to a final torque of 300 in
-
lbs using a torque wrench. This
figure shows the load readings for the different applied torque.


S-Bolts (0.5")
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0
50
100
150
200
250
300
350
Torque (in-lbs)
Load (lbs)
S01
S02
S03
8

Figure
8
: The end bolts were tightened using a regular wrench. This figure shows the load readings
after the final tightening.


M Bolts (0.75")
-500
0
500
1000
1500
2000
2500
3000
3500
4000
0
1
Data Steps
Load (lbs)
M01
M02
M03
M04
9

Figure
9
: The main bolts were torqued to a final torque of 110 ft
-
lbs using a torque wrench. This
figure

shows the load readings for the different applied torque.


1" Dia. Bolts
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
0
20
40
60
80
100
120
Torque (ft-lbs)
Load (lbs)
L01
L02
L03
L04
10

Figure
10
: Current vs. time data for the quench series conducted on July 19, 2001 for HFDB01.

0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
Time (second)
Current (A)
Fastscribe 010719: Racetrack Data Current
11

Figure
11
: Load (in lbs) versus the square of th
e current (A^2) for the main bolt L02 for the first quench in Fig. 10. As expected a linear behavior is
observed.

1400
1900
2400
2900
3400
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
80000000
Current Square (A^2)
Main L2 Load (lbs)
I^2
MainL2-FastScribe0719-Q1
12

Figure
12
: Load (in lbs) versus the square of the current (A^2) for the main bolt L03 for the first quench in Fig. 1
0. As expected a linear behavior is
observed.

1200
1700
2200
2700
3200
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
80000000
Current Square (A^2)
Main L3 Load (lbs)
MainL3-FastScribe0719-Q1
13

Figure
13
: Load (in lbs) versus the square of the current (A^2) for the main bolt L01 for the first quench in Fig. 10. As expected a l
inear behavior is
observed.

8000
8500
9000
9500
10000
10500
11000
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
80000000
Current Square (A^2)
Main L1 Load (lbs)
MainL1
14

Figure
14
: Load (in lbs) versus the square of the current (A^2) for the main bolt L04 for the first quench in Fig. 10. As expected a l
inear behavior is
observed.

6800
7000
7200
7400
7600
7800
8000
8200
0
10000000
20000000
30000000
40000000
50000000
60000000
70000000
80000000
Current Square (A^2)
Main L4A Load (lbs)
Series1
MainL4A-FastScribe0719-Q1
15

Figure
15
: The increase in load (in lbs) for the main bolt L02
as a function of a time parameter for the quench series shown in Fig. 10.

1600
2100
2600
3100
3600
0
200
400
600
800
1000
1200
Point No.
Main L2 Load (lbs)
Fastscribe 010719: Racetrack Data
MainL2-FastScribe0719-Q
16

Figure
16
: The increase in load (in lbs) for the main bolt L03 as a function of a time parameter for the quench series shown in Fig. 1
0.

1400
1600
1800
2000
2200
2400
2600
2800
3000
3200
3400
0
200
400
600
800
1000
1200
Point No.
Main L3 Load (lbs)
Fastscribe 010719: Racetrack Data
MainL3-FastScribe0719-Q
17

Figure
17
: The increase in load (in lbs) for the main bolt L01 as a function of a time parameter for the quench series shown in Fig. 1
0.

8200
8400
8600
8800
9000
9200
9400
9600
9800
10000
10200
0
200
400
600
800
1000
1200
Point No
Main L1 Load (lbs)
Fastscribe 010719: Racetrack Data
18

Figure
18
: The change is resistance of the compensator gauge R3C for the main bolts as

a function of time for the quench series shown in Fig. 10. The
data suggests a very small change in the gauge resistance due to the magnetic field effects.

Main Bolt Compensator Gauges-R3C
351.555
351.56
351.565
351.57
351.575
351.58
351.585
351.59
351.595
351.6
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
Time
Resistance (ohms)
19

Figure
19
: The change is resistance of the compensator gauge R4C for the

main bolts as a function of time for the quench series shown in Fig. 10. The
data suggests a very small change in the gauge resistance due to the magnetic field effects.

Main Bolt Compensator Gauges-R4C
351.43
351.435
351.44
351.445
351.45
351.455
351.46
351.465
351.47
351.475
351.48
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
Time
Resistance (ohms)
20


References




[
1
]

G. Ambrosio et al., “Development and Test of a Nb
3
Sn Racetrack Magnet Using The Reach and Wind Technology,” CEC
Conference, Madison, WI, (2001).