Universal Joint

A joint that transmits rotary motion between two shafts that aren't in a straight
line. Depending on its design, a universal joint can accommodate a large angular
variation between its inputs and outputs. The simplest kind of universal joint,
called a "Hooke joint," causes the output shaft to speed up and slow down twice
for every revolution of the input shaft. This speed fluctuation increases with the
angular difference between the shafts. Another, more common, name for a
Cardan joint.

It consists of two yokes attached to their respective shafts and connected by means of a
spider. The angle between the shafts may have any value up to approximately 35°, if angular
velocity is moderate when the angle is large. Although one shaft must make a single
revolution for each revolution of the second shaft, the instantaneous angular displacement
of the first shaft is the same as that of the second shaft only at the end of each 90° of shaft
rotation. Thus, only at four positions during each revolution is angular velocity of both shafts
the same.

he variation in angular displacement and angular velocity between driving and driven shafts,
which is objectionable in many mechanisms, can be eliminated by using two Hooke's joints,
with an intermediate shaft. This arrangement is conventional for an automobile drive shaft.
The axes of the driving and driven shafts need not intersect; however, it is necessary that the
axes of the two yokes attached to the intermediate shaft lie respectively in planes containing
the axes of adjoining shafts.

Equation of motion
Diagram of variables for the universal joint. Axle 1 is perpendicular to the red plane
and axle 2 is perpendicular to the blue plane at all times. These planes are at an angle
β with respect to each other. The angular displacement (rotational position) of each
axle is given by γ1 and γ2 respectively, which are the angles of the unit vectors and
with respect to their initial positions along the x and y axes. The and vectors
are fixed by the gimbal connecting the two axles and so are constrained to remain
perpendicular to each other at all times.

The Cardan joint suffers from one major problem: even when the drive shaft axle
rotates at a constant speed, the driven shaft axle rotates at a variable speed, thus
causing vibration and wear. The variation in the speed of the driven shaft depends on
the configuration of the joint, which is specified by three variables:

 γ1 The angle of rotation for axle 1

 γ2 The angle of rotation for axle 2
 β The bend angle of the joint, or angle of the axles with respect to each other,
with zero being parallel or straight through.

These variables are illustrated in the diagram on the right. Also shown are a set of
fixed coordinate axes with unit vectors and and the planes of rotation of each
axle. These planes of rotation are perpendicular to the axes of rotation and do not
move as the axles rotate. The two axles are joined by a gimbal which is not shown.
However, axle 1 attaches to the gimbal at the red points on the red plane of rotation in
the diagram, and axle 2 attaches at the blue points on the blue plane. Coordinate
systems fixed with respect to the rotating axles are defined as having their x-axis unit
vectors ( and ) pointing from the origin towards one of the connection points.
As shown in the diagram, is at angle γ1 with respect to its beginning position along
the x axis and is at angle γ2 with respect to its beginning position along the y axis.

is confined to the "red plane" in the diagram and is related to γ1 by:

 is confined to the "blue plane" in the diagram and is the result of the unit vector on
the x axis being rotated through Euler angles ]:

A constraint on the and vectors is that since they are fixed in the gimbal, they
must remain at right angles to each other:

Thus the equation of motion relating the two angular positions is given by:

The angles γ1 and γ2 in a rotating joint will be functions of time. Differentiating the
equation of motion with respect to time and using the equation of motion itself to
eliminate a variable yields the relationship between the angular velocities ω1 = dγ1 / dt

Angular (rotational) output shaft speed versus Output shaft rotation angle, , versus input shaft
rotation angle for different bend angles of rotation angle, , for different bend angles, , of
the joint the joint
and ω2 = dγ2 / dt:

As shown in the plots, the angular velocities are not linearly related, but rather are
periodic with a period twice that of the rotating shafts. The angular velocity equation
can again be differentiated to get the relation between the angular accelerations a1 and
a2:

Double Cardan Shaft

A configuration known as a double Cardan joint drive shaft partially overcomes the
problem of jerky rotation. This configuration uses two U-joints joined by an
intermediate shaft, with the second U-joint phased in relation to the first U-joint to
cancel the changing angular velocity. In this configuration, the assembly will result in
an almost constant velocity, provided both the driving and the driven shaft are parallel
and the two universal joints are correctly aligned with each other - usually 90°.
This assembly is commonly employed in rear wheel drive vehicles, where it is known
as a drive shaft or propeller (prop) shaft.

Even when the driving and driven shafts are parallel, if 0°, oscillating moments
are applied to the three shafts as they rotate. These tend to bend them in a direction
perpendicular to the common plane of the shafts. This applies forces to the support
bearings and can cause "launch shudder" in rear wheel drive vehicles. The
intermediate shaft will also maintain a sinusoidal angular velocity, which contributes

to vibration and stresses.

Thompson coupling
The Thompson constant velocity joint (TCVJ), also known as a Thompson coupling,
is a constant velocity universal joint that can be loaded axially and continue to
maintain constant velocity over a range of input and output shaft angles with low
friction and vibration. It consists of two cardan joints assembled within each other,
thus eliminating the intermediate shaft, along with a control yoke that geometrically
constrains their alignment. The control yoke maintains equal joint angles between the
input shafts and a relative phase angle of zero to ensure constant angular velocity at
all input and output shaft angles. While the geometric configuration does not maintain
constant velocity for the control yoke (aka intermediate coupling) that aligns the pair
of cardan joints, the control yoke has minimal inertia and generates virtually no
vibration. Eliminating the intermediate shaft and keeping the input shafts aligned in
the homokinetic plane virtually eliminates the induced shear stresses and vibration
inherent in traditional double cardan shafts.[6][7]

The use of cardan joints within the Thompson Coupling also reduces the wear, heat
and friction[8] when compared with Rzeppa type constant velocity joints. Cardan
joints, including Thompson couplings, utilise roller bearings running
circumferentially, whereas Rzeppa constant velocity joints use balls which roll and
slide axially along grooves.

The novel feature of the coupling is the method to geometrically constrain the pair of
cardan joints within the assembly by using, for example, a spherical four bar scissors
linkage and it is the first coupling to have this combination of properties.[9]
The coupling earned its inventor, Glenn Thompson, the Australian Society for
Engineering in Agriculture Engineering Award.

Both input axes and all

pivoting joint axes pass
through the coupling center
labeled 29. The blue input
yoke 14 is not directly
connected to the yellow
control yoke 21.

Removing two yokes reveals

the alignment mechanism. Pins
on the input shafts adjust the
extension and rotation of the
scissors attached to the control
yoke. This insures that both
Cardan joints remain aligned
in the homokinetic plane.
How to Change a Universal Joint
Instructions

1. Step 1

Using a floor jack, raise the Blazer high enough to allow easy access to the
driveshaft. Securely support the vehicle on jack stands before crawling under
it.

2. Step 2

Remove the four 11mm bolts holding the rear yoke of the driveshaft to the
differential.
3. Step 3

Place the pry bar between the universal joint and the differential yoke. Pry the
driveshaft out of the differential yoke if it does not easily come out.

4. Step 4

Pull the driveshaft toward the rear of the vehicle to take it out of the
transmission.

5. Step 5

Melt the nylon locks out of the joint if you are replacing an original equipment
universal joint. Any joint that has been previously replaced will have lock
rings installed so this step is not necessary. These nylon locks are located in
the ears of the yoke, just behind the area where the bearing caps are pressed in.
They are visible as small white plastic dots, one on each side of the ear.

6. Step 6

Clamp the universal joint in the universal joint press and push it out of the
yoke. The press pushes the cap out when you turn the driving screw in a
clockwise direction. Be sure to line up the end of the press which has the hole
to allow the bearing cap to enter it without obstruction.

7. Step 7
 Remove the bearing cap on the side of the joint which was pressed out of the
yoke. The bearing cap simply pulls off the joint post. Turn the the universal
joint press over and push the joint out of the other side of the yoke.

8. Step 8

Remove the second bearing cap when it is exposed. Next, remove the tool
from the driveshaft and take the universal joint out of the middle of the yoke.

9. Step 9

Take the bearing caps off one cross of the new universal joint. Being very
careful to avoid contacting the yoke with the universal joint, place one of the
capless posts of the joint through the yoke to the outside. Put the bearing cap
back on.

10. Step 10

Using the universal joint tool, press the bearing cap back into the yoke. Push it
far enough through to allow you to install the bearing cap back on the
opposing post.

11. Step 11

Install a lock ring on the bearing cap you pushed through the yoke. The locks
snap into the groove on the inner end of the bearing cap.

12. Step 12

Press the joint into the yoke until the lock contacts the inside of the opposing
ear and you can install the lock on the second bearing cap.

13. Step 13

Install the supplied grease fitting in the joint. Some replacement joints do not
have a grease fitting, so this step may not be necessary.

14. Step 14

Slide the driveshaft back into the transmission.

15. Step 15

Insert the rear of the driveshaft into the differential yoke. Be certain to center
the bearing caps in the yoke. This should be a fairly tight fit. You will likely
need to turn the differential yoke by hand until the yoke and the driveshaft line
up properly.

16. Step 16
 Use a drop of threadlocker, such as Locktite 242, on each of the 11mm bolts.
Thread the bolts into the differential yoke and tighten uniformly.

17. Step 17

Recheck that the bearing caps are centered in the differential yoke.

18. Step 18

Remove the jack stands and test drive the vehicle.

Multi-Objective Shape Optimization of an Automotive Universal

Joint, as an Assembly

Introduction:

A universal joint is a linkage used to connect

rotating shafts that are coplanar, but not
coinciding. A universal joint is made up of three
main components; two yokes and a cross
trunnion. In the automotive industry, yokes are
classified by the methods with which they
connect to other driveline components.

In the automotive industry, the cross trunnion in

a universal joint is designed to fail at a pre-
determined value to prevent further damage to
more expensive driveline components. Despite careful calculation of the cross
trunnion failure limit, little research has been performed on the optimization of the
remaining components of the assembly; the yokes. This may result in unnecessarily
heavy and costly parts.

When conducting design optimization of complex systems, two options exist: (1) to
optimize each individual component separately, considering approximations on both
boundary and loading conditions and (2) to optimize the entire assembly at once,
using measurable boundary and loading conditions. Optimization of only individual
components is computationally more efficient; however, approximations to boundary
conditions and the necessity to exclude additional constraint functions make
component-level optimization undesirable in many situations. This research
considered a Universal Joint as a case study to emphasize the differences in results
between component level optimization and assembly level optimization. The results
show that although component level optimization requires less computational time,
the Pareto frontier does not match that of the assembly level optimization, or true
behaviour of the system.

Modeling:

The primary objective of this research was to consider the optimization of an

automotive universal joint flange yoke and weld yoke separately, and in a complete
assembly. In both phases Volume, Joint Angle and Manufacturing Cost were
considered as objective functions, and von Mises stress and Strain Energy Density
were used as constraints.

Modeling and simulation of the universal joint was conducted using the commercial
Finite Element Analysis software, ANSYS v9.0. Since shape optimization requires
systematic modification of part geometry, a robust and parametric model was
constructed using a technique known as “bottom-up” modeling. For increased
accuracy, all parts were meshed using well-shaped quadratic, quadrilateral elements
(SOLID95). Gradient based optimization was conducted using MATLAB v7.0.

Boundary conditions applied during optimization were as follows: The Gencoz

distribution was used to simulate the static loading delivered by the cross trunnion
bearing to the yoke ears in this universal joint component level model. At the
assembly level, an approximation of the contact forces between the bearing and yoke
was developed to increase computational efficiency.

Three components are parametrically assembled to permit all joint and rotation angle
combinations.
Results:

During all optimization phases, an

improvement over the baseline design
was achieved. Since the current design
configurations of both the flange yoke
and weld yoke do not lie on the Pareto
frontier of optimal designs in either case,
the original designs may therefore be
considered over-designed. Furthermore,
it was shown that the Pareto frontiers
generated for component level
optimization do not exactly match those
generated for assembly level
optimization. While these tradeoff
curves indicate the same general trend -
to achieve an increase in maximum joint angle requires a substantial and simultaneous
increase in part volume, or mass – the absolute values of the anchor points and the overall
shape of the curve are significantly different. Therefore, feasible optimum designs cannot
be obtained from component level optimization alone; assembly level optimization is
required to consider the interaction between components and keep designs within the
feasible region.
Mathematical
Description: P01 -
Universal Joint
The universal joint P01 can be considered to be a spherical 4-bar linkage with
, the output crank b
variable in the range , and the fixed (grounded) link g
being the angle between the input and output shafts which can be adjusted in the
range . Note that the links a, h, and g are
not actually links, however, the constraints of the mechanism operate as if they were
spherical links. See 4-bar linkages for a discussion of the eight main types of spherical
4-bar linkages.

When this mechanism is used in the drive shaft of an automobile, then it is usual that
and in this case the mechanism is called Hooke’s joint.
See History of Hooke’s Joint. For the Hooke’s joint, using the notation from 4-bar
linkages, we have
. Thus (noting that on the sphere, |l|S = |l|, if |l| , and |l|S
= - |l|, if |l| )

From the chart in 4-bar linkages we have that this is a double crank, which means that
both the input and output shafts can rotate completely around, as can be seen in the
movie above. {The reader can check that as long as b is within of
and g is within of , then this universal joint
will still be a double crank. In fact, this is important in its automotive use because as
an automobile goes over bumps the angle g between the shafts will change.}

We now look at the relationship between the rotation of the input shaft to the rotation
of out shaft. One of the problems with Hooke’s joint is that, though one rotation of the
input shaft results in one rotation of the output shaft, the rotations are not in sync
during the revolution. We look at the standard case of ,
which we depict in Figure 1. Since
 , the corresponding arcs are of great circles, thus A must the pole for
the great circle (dashed in the figure) passing through O and B; and the great circle
arcs h and a must intersect this great circle at right angles (since A is the pole).
Likewise, B must be the pole for the great circle passing through A and C; and the
great circle arcs h and b must also intersect this great circle at right angles. Since the
arc h is perpendicular to both of the dashed great circles, then the arc must bisect lune
determined by these two great circles. Since h is great circle in length, then
angle of this lune (at P) must be , as marked. In addition, we use
to label the angle at A and to label the angle at B. Now, since A is the pole
for the great circle P-O-B then the arc OB subtends the angle at the center of
the sphere and so the radian measure of OB is . By the same reasoning the
radian measure of the arc AC is .

The angle measures the rotation of the Input Shaft and the angle
measures the rotation of the output shaft. Note that, when =0 then =0
also. As changes in the positive (CCW) direction, will be changing in
the negative (CW) direction; thus, when is positive, will be negative
and we will need to use when denoting the angle in the triangle OCA.

In order to find the relationship between g (in radian measure) and the angles
and , we use properties of spherical triangles that are familiar to anyone who
has studied spherical geometry but unfortunately few people study spherical
geometry these days. You can find the formulas we need at the end of this document.

If we apply the “Spherical Pythagorean Theorem” to the right triangle OPC, we

obtain:

If we apply the Law of Sines to triangle OCA, then we obtain:

From triangle OCB we obtain:

Putting these together we get:

Solving

We put the negative as indicated because g in practice is near and thus

is positive. Note that when then
and thus the two shafts turn in
unison.
Spherical Trigonometry Formulas

If we have a spherical triangle with sides r, s, t (in radian measure) whose opposite
angles are labeled , then the Law of Sines states:

The Law of Cosines states:

In the case that the angle opposite c is a right angle then we get the “Spherical
Pythagorean Theorem”: