ME 324: Mechanical engineering design I

Shafts

Problem 7-23: Shaft design

7.23
The shaft shown in the figure is driven by a gear at the right keyway, drives a fan at the left
keyway, and is supported by two deep-groove ball bearings. The shaft is made from AISI
1020 cold-drawn steel. At steady-state speed, the gear transmits a radial load of 1000 N and
a tangential load of 2500 N at a pitch diameter of 200 mm.

325
215 5 55
70 10 35 20
50 30 45 45 35
25 30

50

1.5
3 2.5
4
1 3 1

25 35 175 80 10

Determine fatigue factors of safety at any potentially critical locations using the DE-Gerber
failure criterion.

Solution:

Label the approximate locations of the effective centers of the bearings as A and B,
the fan as C, and the gear as D, with axial dimensions as shown. Since there is only
one gear, we can combine the radial and tangential gear forces into a single resultant
force with an accompanying torque, and handle the statics problem in a single plane.

From statics, the resultant reactions at the bearings can be found to be RA = 844.7 N
and RB = 1847.9 N. The bending moment and torque diagrams are shown, with the
maximum bending moment at D of MD = 844.7(175) = 147 822.5 Nmm and a torque
transmitted from D to C of T = 2500 (100) = 250 000 N mm.

Due to the shaft rotation, the bending stress on any stress element will be completely
reversed, while the torsional stress will be steady. Since we do not have any
information about the fan, we will ignore any axial load that it would introduce. It
would not likely contribute much compared to the bending anyway.
 325
215 5 55
70 10 35 20
50 30 45 45 35
25 30

50

1.5
3 2.5
4
1 3 1

25 35 175 80 10

147 822.5
Nmm 129 344.7
101 628.0

Nmm
250 000

Potentially critical locations are identified as follows:

• Keyway at C, where the torque is high, the diameter is small, and the keyway
creates a stress concentration.
• Keyway at D, where the bending moment is maximum, the torque is high, and the
keyway creates a stress concentration.
• Groove at E, where the diameter is smaller than at D, the bending moment is still
high, and the groove creates a stress concentration. There is no torque here.
• Shoulder at F, where the diameter is smaller than at D or E, the bending moment is
still moderate, and the shoulder creates a stress concentration. There is no torque
here.
• The shoulder to the left of D can be eliminated since the change in diameter is very
slight, so that the stress concentration will be much less than at D.

Table A-20: Sut = 470 MPa, Sy = 380 MPa

Eq. (6-8): S’e = 0.5(470)=235 MPa
Eq. (6-19): ka = 4.51(470)-0.265=0.883

Keyway at C
Since there is only steady torsion here, only a static check needs to be performed.
We’ll use the maximum shear stress theory.
 t = Tr/J = 250000(25/2)/(p(254)/32) = 81.5 MPa

Eq. (5-3): ny = (Sy/2)/ t = (380/2)/81.5 = 2.33

Keyway at D
Assuming r / d = 0.02 for typical end-milled keyway cutter (p. 365), with d = 45 mm,
r = 0.02d = 0.9.

Table 7-1: Kt = 2.14, Kts = 3.0

Fig. 6-20: q = 0.66
Fig. 6-21: qs = 0.72
Eq. (6-32): K f = 1 + q ( Kt - 1) = 1 + 0.66(2.14 -1) = 1.8
K fs = 1 + qs ( K ts - 1) = 1 + 0.72(3.0 - 1) = 2.4
-0.107
æ 1.75 ö
Eq. (6-20): kb = ç ÷ = 0.828
è 0.30 ø
Eq. (6-18): Se = 0.883(0.828)(235) = 171.8 MPa

We will choose the DE-Gerber criteria since this is an analysis problem in which we
would like to evaluate typical expectations.
Using Eq. (7-9) with Mm = Ta = 0,

A = 4 ( K f M a )2 = 4 éë(1.8)(1459 ) ùû 2 = 5252 lbf × in = 5.252 kip × in

2 2

A = 4(K f Ma ) = 4 éë(1.8)(1459 ) ùû = 5252 lbf × in = 5.252 kip × in

= 532 161 Nmm2
B = 3 ( K fsTm )2 = 3 éë( 2.4 )( 2532 ) ùû 2 = 10 525 lbf × in = 10.53 kip × in
2

B = 3 ( K fsTm ) = 3 éë( 2.4 )( 2532 ) ùû = 10 525 lbf × in = 10.53 kip × in

= 1 039 230.5 Nmm
ì é 2 1/2 ü
1 8A ï æ 2 BSe ù ï
ö
= í1 + ê1 + ç ÷ ú ý
n p d 3 Se ï ê è ASut ø ú ï
î ë û þ
ì é 2 1/2 ü
n = 4.2 8 ( 5.252Ans.
) ï ê æ 2 (10.53)( 24.9 ) ö ùú ï
= í1 + 1 + ç ÷ ý
( )
p 1.753 ( 24.9 ) ï ê çè ( 5.252 )( 68 ) ÷ø ú ï
î ë û þ
Groove at E
We will assume Figs. A-15-14 is applicable since the 50 mm diameter to the right of
the groove is relatively narrow and will likely not allow the stress flow to fully
develop.

r / d = 2.5 / 45 = 0.056, D / d = 50 / 45 = 1.1

Fig. A-15-14: Kt = 2.1
Fig. 6-20: q = 0.76
Eq. (6-32): K f = 1 + q ( Kt -1) = 1 + 0.76(2.1 -1) = 1.8
-0.107
æ 1.55 ö
Eq. (6-20): kb ==0.839
ç ÷ = 0.839
è 0.30 ø
Eq. (6-18): Se = 0.883(0.839)(34)
0.883(0.839)(235)= =25.2
174.1 MPa
kpsi

Using Eq. (7-9) with Mm = Ta = Tm = 0,

A = 4 ( K f M a ) == 465640.9
4 éë(1.8)(Nmm
1115)ùû = 4122 lbf × in = 4.122 kip × in
2 2

B=0
ì é 2 1/2 ü
ù ï
1 8A ï æ 2 BS ö
= í 1 + ê1 + ç
e
÷ ú ý
n p d 3 Se AS
ïî ë è ut ø úû ïþ
ê
= 0.14958 ( 4.122 )
=
( )
p 1.553 ( 25.2 )
n = 6.69
1 +
ë
Ans.
{
é1 + ( 0 )2 ù
û
1/2

}
Shoulder at F
r / d = 3 / 35 = 0.086, D / d = 50 / 35 = 1.43
Fig. A-15-9: Kt = 1.7
Fig. 6-20: q = 0.78
Eq. (6-32): K f = 1 + q ( Kt - 1) = 1 + 0.78(1.7 - 1) = 1.5
-0.107
æ 1.40 ö
Eq. (6-20): kb = 0.85
ç ÷ = 0.848
è 0.30 ø
Eq. (6-18): Se = 0.883(0.848)(34) = 25.5 MPa
0.883(0.85)(235)=176.4 kpsi

Using Eq. (7-9) with Mm = Ta = Tm = 0,

A = 4 ( K f M a ) = 304884
4 éë(1.5Nmm
)(845)ùû = 2535 lbf × in = 2.535 kip × in
2 2

B=0

ì é 2 1/2 ü
1 8A ï æ 2 BSe ù ï
ö
= í1 + ê1 + ç ÷ ú ý
n p d Se ï ê è ASut ø ú ï
3

î ë û þ
= 0.20538 ( 2.535 )
=
p (1.403 ) ( 25.5)
1 +{é1 + ( 0 )2 ù
ë û
1/2

}
n = 4.87 Ans.
Problem 7-3: Shaft design

The rotating solid steel shaft is simply supported by bearings at points B and C and is driven
by a gear (not shown) which meshes with the spur gear at D, which has a 150-mm pitch
diameter. The force F from the drive gear acts at a pressure angle of 20°. The shaft transmits
a torque to point A of T A 5 340 N ? m. The shaft is machined from steel with S y 5 420 MPa
and S ut 5 560 MPa. Using a factor of safety of 2.5, determine the minimum allowable
diameter of the 250-mm section of the shaft based on (a) a static yield analysis using the
distortion energy theory and (b) a fatigue-failure analysis. Assume sharp fillet radii at the
bearing shoulders for estimating stress-concentration factors.

Solution:

F cos 20°(d / 2) = TA, F = 2 TA / ( d cos 20°) = 2(340) / (0.150 cos 20°) = 4824 N.
The maximum bending moment will be at point C, with MC = 4824(0.100) = 482.4 N·m. Due
to the rotation, the bending is completely reversed, while the torsion is constant.
Thus, Ma = 482.4 N·m, Tm = 340 N·m, Mm = Ta = 0.

For sharp fillet radii at the shoulders, from Table 7-1, Kt = 2.7, and Kts = 2.2. Examining Figs.
6-20 and 6-21 (pp. 303 and 304 respectively) with Sut = 560 MPa, conservatively estimate q
= 0.8 and qs = 0.9. These estimates can be checked once a specific fillet radius is
determined.

Eq. (6-32): K f = 1 + 0.8(2.7 - 1) = 2.4

K fs = 1 + 0.9(2.2 - 1) = 2.1

(a) We will choose to include fatigue stress concentration factors even for the static
analysis to avoid localized yielding.
1/2
éæ 32 K f M a ö2 æ 16 K fsTm ö 2 ù
Eq. (7-15): ¢
s max = êç ÷ + 3ç ÷ ú
êëè p d è p d ø úû
3 3
ø
 Sy p d 3S y é 4 ( K M )2 + 3 ( K T )2 ù
-1/2
Eq. (7-16): n= =
¢
s max 16 êë f a fs m úû
Solving for d,
1/3
ìï 16n 1/2 ü
ï
d =í éë 4( K f M a ) 2 + 3( K fsTa ) 2 ùû ý
îï p S y þï
1/3
æ 16(2.5) ö

( ){ }
2 1/2
4 [ (2.4)(482.4) ] + 3 [(2.1)(340) ]
2
=ç ÷
ç p ( 420 ) 106 ÷
è ø

d = 0.0430 m = 43.0 mm Ans.

(b) ka = 4.51(560)-0.265 = 0.84

Assume kb = 0.85 for now. Check later once a diameter is known.

Se = 0.84(0.85)(0.5)(560) = 200 MPa

Selecting the DE-ASME Elliptic criteria, use Eq. (7-12) with M m = Ta = 0.

1/3
ì é æ 2 2 1/2 ü
ù
ï16(2.5) ê 2.4(482.4) ö æ ö
2.1(340) ú ï
d =í 4ç ÷ + 3ç ÷ ý
ï p ê çè 200 (10 ) ÷ø ç 420 (106 ) ÷ ú ï
ê 6

ë è ø úû
î þ
= 0.0534 m = 53.4 mm

With this diameter, we can refine our estimates for kb and q.

kb = 1.51d -0.157 = 1.51(53.4 )

-0.157
Eq. (6-20): = 0.81

Assuming a sharp fillet radius, from Table 7-1, r = 0.02d = 0.02 (53.4) = 1.07 mm.

Fig. (6-20): q = 0.72

Fig. (6-21): qs = 0.77

Iterating with these new estimates,

Eq. (6-32): Kf = 1 + 0.72 (2.7 – 1) = 2.2

Kfs = 1 + 0.77 (2.2 – 1) = 1.9
Eq. (6-18): Se = 0.84(0.81)(0.5)(560) = 191 MPa
Eq. (7-12): d = 53 mm Ans.

Further iteration does not change the results.

Problem 7-28: Shaft design

A 25-mm-diameter uniform steel shaft is 600 mm long and supported between two
bearings.

(a) Find the lowest critical speed of the shaft.

(b) If the goal is to double the critical speed, find the new diameter.
(c) A half-size model of the original shaft has what critical speed?

Solution:

In Eq. (7-22) set

pd4 pd2
I= , A=
64 4
to obtain
2
æ p ö æ d ö gE
w =ç ÷ ç ÷ (1)
è l ø è4ø g
or
4l 2w g
d= (2)
p 2
gE

(a) From Eq. (1) and Table A-5

2
æ p ö æ 0.025 ö 9.81(207)(10 )
9
w =ç ÷ ç ÷ = 883 rad/s Ans.
è 0.6 ø è 4 ø 76.5 103 ( )
(b) From Eq. (1), we observe that the critical speed is linearly proportional to the
diameter. Thus, to double the critical speed, we should double the diameter to
d = 50 mm. Ans.

(c) From Eq. (2),

p2 d gE
lw =
4 l g

Since d / l is the same regardless of the scale,

lw = constant = 0.6(883) = 529.8

529.8
w= = 1766 rad/s Ans.
0.3

=> The first critical speed doubles!