The Augustus Group Products Projects and Engineering Consulting
TORQUE PRELOAD CONTROL Jesse Meisterling
Copyright © The Augustus Group 2006
BOLT PRELOAD Definition: • Initial Preload: the tensile stress created in an individual fastener when first tightened. • Residual Preload: the tensile stress in an individual fastener when the joint tightening process is completed. • Operational Bolt Load or In Service Tension: resultant bolt load under operating conditions; temperature, pressure, external loads, etc. 3/1/2006
Copyright © The Augustus Group 2006
2
BOLT BASICS A bolt is a spring.
Angle Torque
∂F≅ ∂ The elastic behavior of our bolting materials keeps everything together.
Load
Stretch has been used for many years to control preload. As long as the initial ‘loading’ of the fastener is below the yield point, the ∂ from“zero load” to the target will be proportional to the tension force in the fastener.
Stretch
Load
If the fastener is taken beyond the yield point it is still the elastic “stretch” that produces the residual bolt load. An “unload” stretch measurement will produce an accurate load measurement in the fastener. 3/1/2006
Copyright © The Augustus Group 2006
3
TORQUE PRELOAD RELATIONSHIP T = Fp(P/2π + μtRt/cosΒ+ μnRn) Where: T=torque Fp=Preload P= Thread Pitch μt=Friction on thread flank Rt=Radius of thread flank Β=Flank angle/2 μn=Friction under head or nut Rn =Radius of nut bearing surface
3/1/2006
Copyright © The Augustus Group 2006
4
TORQUE PRELOAD RELATIONSHIP What are the forces?
μ
•Fp(P / 2π) - creates load in the fastener - lead screw •Fp(μtRt/cosΒ) - reaction torque in threads •Fp(μnRn) - reaction torque under nut
Lube Graphite, mineral oil Graphite, machine oil Molykote-G Fel-Pro C5A Crane 425A Grafo 360 Neolube
min 0.035 0.035 0.03 0.045 0.07 0.1 0.03
max 0.06 0.055 0.075 0.08 0.09 0.115 0.09
avg 0.048 0.045 0.053 0.063 0.08 0.108 0.06
The reaction forces work against us tightening the bolt and keep the screw together when installed
3/1/2006
Copyright © The Augustus Group 2006
5
TORQUE PRELOAD RELATIONSHIP Bolt Tension using Torque In the ideal world there would be a linear repeatable correlation.
B o l t P r e l o a d
Torque applied
3/1/2006
Copyright © The Augustus Group 2006
But:
6
TORQUE PRELOAD RELATIONSHIP Variables Affecting Bolt Torque/Tension VDI analysis UASF study came up w/ 71
3/1/2006
Copyright © The Augustus Group 2006
7
TORQUE PRELOAD RELATIONSHIP From our first equation we see the friction dependence. •Fp(P / 2π) •Fp(μtRt/cosΒ) •Fp(μnRn)
3/1/2006
Copyright © The Augustus Group 2006
8
TORQUE PRELOAD RELATIONSHIP 160
120000
Load as k changes (2% of base) 120
% Change in Load
100000
% K 2% 80000
%
100 80
60000
60
Load lbs
140
40000
40 20000
20 0 0.35
0 0.3
0.25
0.2
0.15
0.1
k
3/1/2006
Copyright © The Augustus Group 2006
9
TORQUE PRELOAD UNCERTAINTY Experiments show that the torque/preload relationship is uncertain. There is a family of torque/preload lines. The Air Force study identified 74 factors. Some are: •Lube and application method •Hardness •Surface finish •Plating •Material •Angularity of parts •Thread fit •Combination of above... 3/1/2006
Copyright © The Augustus Group 2006
10
MATERIAL HARDNESS AND TORQUE 50 each - 1.25” axle flange bolts, 4 strength grades, tightened to 2000 ft-lbs. Data taken after assembly and after 24 hrs. 4000XXX Engineering Test Report Ultrasonic Clamp Load M easurement In Axle Assembly - 13 Bolts per Sample Group Installed to 2,000 ft-lb 140,000 180 ksi Proof Load - 139,500 lbs
130,000 170 ksi Proof Load - 131,700 lbs 2nd Tightening On Right Side Of Axle 150 ksi
120,000
Ultrasonic Clamp Load (lbs)
110,000 100,000 90,000 80,000 70,000
1st Tightening On Left Side Of Axle 180 ksi
60,000 1st Tightening On Right Side Of Axle 170 ksi
50,000
2nd Tightening On Left Side Of Axle 170 ksi
40,000 30,000 20,000
Average + 3 Sigma Average - 3 Sigma Average 180 ksi - 10 min
180 ksi - 24 hrs
170 ks i - 10 min
170 ksi - 24 hr
170 ksi - 10 min
170 ksi - 24 hr
Gr 8 - 10 min
Gr 8 - 24 hr
Bolt Strength - M easurement Period
3/1/2006
Copyright © The Augustus Group 2006
11
SHORT FORM TORQUE TO PRELOAD T= kDFp Where: T = Torque k = Nut factor D = Nominal diameter Fp = Preload
The nut factor is an empirical factor that is obtained by tests. It is developed with the target lubricant. The “short form” k factor is not one number, it has a range and mean for each lubricant used. 3/1/2006
Copyright © The Augustus Group 2006
12
COMMON k FACTORS Nut Factors (K)
3/1/2006
Lube
Min. Reported
Mean
Max. Reported
As-Received
0.158
0.2
0.267
Alloy or Mild Steel Fasteners As-Received Stainless Steel Fasteners Cadmium Plate (Dry) Copper Based Anti-Seize Cadmium Plate (Waxed) Fel-Pro CS4 Fel-Pro C70 Fel-Pro N 5000 (paste) Machine Oil Moly Plate or Grease Never-Seeze (Paste) Neolube Phos-Oil Solid Film PTFE Zinc Plate (Waxed) Zinc PIate (Dry)
-
0.3
0.106 0.08 0.17 0.08 0.08 0.13 0.10 0.10 0.11 0.14 0.15 0.09 0.071 0.075
0.2 0.132 0.187 0.132 0.095 0.15 0.21 0.13 0.17 0.18 0.19 0.12 0.288 0.295
Copyright © The Augustus Group 2006
0.328 0.23 0.198 0.23 0.15 0.27 0.225 0.18 0.21 0.20 0.23 0.16 0.52 0.53
13
NON-PARAMETRIC k STUDY Fel Pro to N5000 Fel-Pro N5000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Fel Pro N5000
Molycote G 0.1113 0.1113 0.123 0.1346 0.1352 0.1352 0.1371 0.1397 0.1412 0.1447 0.1452
0.1463 0.1523 0.1533 0.1551 0.1577 0.1598 0.1605 0.1621 0.1631 0.1696 0.172 0.1757 0.1768 0.1852 0.189 0.1954 0.2043 0.2061 0.2267 0.2703 0.2703
Num ber of runs - 8 Rank sum of Y's - 23 Num ber of X's above m edian 13 Num ber of Y's above m edian -3 Median of com bined set - 16.5
3/1/2006
Fel Pro to Molycote G 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Never Seeze 0.1183 0.1183 0.1452
0.1463 0.1523 0.1537 0.1537 0.1551 0.1577 0.1577 0.1598 0.1604 0.1605 0.1621 0.172 0.1752 0.1768 0.178 0.1852 0.189 0.1892 0.1936 0.1954 0.1968 0.2013 0.2043 0.2061 0.2102 0.2267 0.2365 0.2703 0.2703
Number of runs - 18 Rank sum of Y's - 58 Number of X's above median - 9 Number of Y's above median - 7 Median of combined set - 16.5
Copyright © The Augustus Group 2006
14
K FACTOR: MULTIPLE TIGHTENINGS 2 1/4 - 8 x 12 B16 Studs A - Mo2 Grease b - Clean and Dry
3/1/2006
Copyright © The Augustus Group 2006
15
HIGH-SPEED TIGHTENING
3/1/2006
Copyright © The Augustus Group 2006
16
ASSEMBLY BOLT STRESS - Txx Tensile Stress (Txx) Txx = Fp/A = Fp/(π D2/4) Since we are applying torque (T) T = kD Fp Then: Fp = T/kD and Txx = 4T/(π D3)
3/1/2006
Copyright © The Augustus Group 2006
17
TORSIONAL STRESS - TXY Assume only 50% of the applied torque twists the bolt, the other 50% is lost in overcoming the friction at the bearing surface of the head and into loading the bolt.
TXY = (0.5T)C/J Where: J = Polar moment of inertia = π D2 /32 C = Max radius = D/2
TXY = (0.5T) 16/ π D3) 3/1/2006
Copyright © The Augustus Group 2006
18
COMBINED STRESS Calculate the maximum stress from the combined tensile and shear stress Txx max = Txx/2 + [Txy2 + (Txx/2)2] 1/2 Form a ratio Txy/Txx Txy/Txx = (0.5(16)T/ (n D3) )/ 4T / (n (k) D3) Txy/Txx = 2 (k)
3/1/2006
Copyright © The Augustus Group 2006
19
COMBINED STRESS Now pick a nut factor (k) and go back to Equation 1 Txx max = 1/2 + [(0.3)2 + (1.0/2)2] 1/2 Txx max = 1/2 + 0.58 Txx max = 1.08 The torsional stress increases the maximum stress by 8%. This is crude; haven't worried much about the diameter. The VDI calculation is a little more detailed. 3/1/2006
Copyright © The Augustus Group 2006
20
DISTORTION ENERGY THEORY PER ANSI B 1.1 Syp = [St2 + 3Ss2]1/2 Where: Syp = Yield stress St = Tensile stress Ss = Shear stress Ss = 0.3 St for k = 0.15
Syp = = [St2 + 3(0.3 St 2]1/2 St = 0.89 Syp 3/1/2006
Copyright © The Augustus Group 2006
21