This action might not be possible to undo. Are you sure you want to continue?
ME 301
Kinematics & Dynamics of Machines
Class Notes
Dr. Bob
Mechanical Engineering
Ohio University
© Dr. Bob Productions
williar4@ohio.edu
http://www.ent.ohiou.edu/~bobw/
2
Table of Contents
INTRODUCTION............................................................................................ 3
MATRIXVECTOR INTRODUCTION........................................................ 11
MATLAB INTRODUCTION........................................................................ 25
MOBILITY..................................................................................................... 29
FOURBAR MECHANISM POSITION ANALYSIS................................... 34
GRASHOF’S LAW........................................................................................ 49
SLIDERCRANK MECHANISM POSITION ANALYSIS.......................... 53
VELOCITY ANALYSIS INTRODUCTION ................................................ 62
FOURBAR MECHANISM VELOCITY ANALYSIS................................. 67
SLIDERCRANK MECHANISM VELOCITY ANALYSIS........................ 78
ACCELERATION ANALYSIS INTRODUCTION...................................... 85
FOURBAR MECHANISM ACCELERATION ANALYSIS ...................... 91
SLIDERCRANK MECHANISM ACCELERATION ANALYSIS ........... 100
INPUT MOTION SPECIFICATION........................................................... 106
DYNAMICS INTRODUCTION.................................................................. 110
MASS, CENTER OF GRAVITY, MASS MOMENT OF INERTIA.......... 115
SINGLE ROTATING LINK INVERSE DYNAMICS................................ 124
FOURBAR MECHANISM INVERSE DYNAMICS................................. 133
SLIDERCRANK MECHANISM INVERSE DYNAMICS........................ 147
CAM INTRODUCTION.............................................................................. 156
ANALYTICAL CAM SYNTHESIS............................................................ 168
GEAR INTRODUCTION............................................................................ 175
GEAR TRAINS AND GEAR STANDARDIZATION................................ 185
3
ME 301 Kinematics & Dynamics of Machines
Introduction
Kinematics:
Kinema  Greek for motion
Dynamics:
Rigid Body Mechanics Diagram:
Required Math: Geometry, trigonometry, vectors, matrices, calculus
Mechanisms: linkages, cams, gears, gear trains
Analysis vs. Synthesis
• Analysis – determination of position, velocity, acceleration, etc.
for a given mechanism
• Synthesis – design of mechanism to do a specific job
4
Mobility  number of degreesoffreedom (dof):
• Structure – static, no motion
• Mechanism – 1 dof device with rigid links connected with joints
• Machine – collection of mechanisms to transmit force (input /
output)
• Robot – an electromechanical device having greater than 1 dof,
programmable for a variety of tasks.
Motion – Translation and Rotation
Planar – all motion is 2D (projected onto a common plane)
Helical  rotation about fixed axis and translation along axis  screw
Spherical  3D motion; all points in a body moves about a fixed point
Spatial  3 independent translations and rotations
5
Joints – Pairing elements
Lower – surface contact
Revolute – pin joint, turning pair
Prismatic – sliding pair
Higher – point or line contact
ball bearing
gears
cam and follower
Link – rigid body
Kinematic chain – number of links connected by joints
open – serial robot
closed – mechanism, parallel robot
Kinematic Inversion – change which link is fixed – same relative
motion, different absolute motion.
Examples – in class; also see following Atlas
6
A Brief Atlas of Structures, Mechanisms, and Robots
Dr. Bob
Statically Determinate Structure Statically Indeterminate Structure
4Bar Mechanism Offset SliderCrank Mechanism
Inverted SliderCrank Mechanism ScotchYoke Mechanism
7
Geared 5Bar Mechanism Stephenson I 6Bar Mechanism
Stephenson II 6Bar Mechanism Stephenson III 6Bar Mechanism
Watt I 6Bar Mechanism Watt II 6Bar Mechanism
8
Spur Gear Mechanism CamandFollower Mechanism
Geneva Wheel Mechanism Planar 3dof Robot
Adept 4dof SCARA Robot Mitsubishi 5dof Robot
9
PUMA 6dof Robot NASA 8dof ARMII
2dof 5Bar Parallel Robot 3dof 3RRR Parallel Robot
3dof 3RPR Parallel Robot 3dof Carpal Wrist
10
Connection to Machine Design
In ME 301 we focus on kinematics & dynamics analysis, not
synthesis (design).
However, the skills gained in this course support general
(electro)mechanical design.
Before one can design a machine, the required motion must be
satisfied. All design candidates must be analyzed regarding the
motion each would provide (position, velocity, and acceleration, both
translational and rotational). This requires kinematics analysis.
Before one can size the links, joints, bearings, gear box, and actuators
(motors) in a machine, the worstcase force and moment loading
condition(s) must be known, for statics and dynamics. This requires
dynamics analysis.
Engineering design is iterative by nature: each candidate design must
be thoroughly analyzed to determine its performance relative to the
design specifications and relative to other design candidates.
This kinematics & dynamics analysis is facilitated using a computer.
Without the computer, it is difficult to determine the worstcase
loading cases, and overdesigned factors of safety may be
inefficiently applied.
The goal of ME 301 is to give the student general skills in general
matrix/vectorbased kinematics and dynamics analysis which may be
applied in later classes and later careers.
11
MatrixVector Introduction
Vectors
Arrow in the plane with magnitude and direction. Used to
represent position, velocity, acceleration, force. Also, arrow normal
to the plane to represent angular velocity, angular acceleration, and
torque (moment) vectors (see later in notes).
Cartesian representation:
Polar representation: Magnitude at angle: θ @ P
(atan2  quadrantspecific inverse tangent function)
12
Vector Addition
Vectors add tailtohead (subtract headtotail); express
components in same coordinate frame.
13
Vector Dot Product
Dot product is projection of one vector onto another. Scalar
result.
14
Vector Cross Product
Cross product of two vectors gives a third vector mutually
perpendicular to the original two vectors. Vector result.
Direction via righthandrule: Put right hand fingers along first
vector
1
P , rotate into second vector
2
P ; right thumb is direction of
2 1
P P × .
15
k
ˆ
Vectors
In planar kinematics, angular velocity, angular acceleration, and
torque (moment) vectors are arrows along about the k
ˆ
axis (the unit
direction for the Z axis, perpendicular to the plane). Still has
magnitude and direction, but simplifies to a single component with ±
sign. We will often represent these k
ˆ
vectors by curled arrows in the
XY plane.
Example:
k
ˆ
ω ω ± = ;
+ ccw (curling in the direction of the right hand fingers)
– cw (curling in the opposite direction of the right hand fingers)
16
Vector Examples
)
`
¹
¹
´
¦
=
2
1
1
P
)
`
¹
¹
´
¦
=
2
3
2
P
Addition: = +
2 1
P P
Dot Product: = •
2 1
P P
Cross Product: = ×
2 1
P P
17
Matrices
Matrix: m x n array of numbers, where m is the number of rows
and n in the number of columns.
 
11 12 1
21 22 2
1 2
n
n
m m mn
a a a
a a a
A
a a a
=
"
"
# # % #
"
Used to simplify and standardize the solution of n linear equations in
n unknowns (where m=n). Used in velocity, acceleration, and
dynamics analysis linear equations (not used in position which is a
nonlinear solution).
Special Matrices
Square (m=n=3)
 
11 12 13
21 22 23
31 32 33
a a a
A a a a
a a a
=
Diagonal
 
11
22
33
0 0
0 0
0 0
a
A a
a
=
Identity
 
1 0 0
0 1 0
0 0 1
I
=
18
Transpose
 
11 21 31
12 22 32
13 23 33
T
a a a
A a a a
a a a
=
Symmetric
   
11 12 13
12 22 23
13 23 33
T
a a a
A A a a a
a a a
= =
Column Vector (3x1 matrix)
{ }
1
2
3
x
X x
x
¦ ¹
¦ ¦
=
´ `
¦ ¦
¹ )
Row Vector (1x3 matrix)
{ } { }
1 2 3
T
X x x x =
Matrix Addition Just add up like terms
a b e f a e b f
c d g h c g d h
+ +
+ =
+ +
Matrix Multiplication with Scalar Just multiply each term
a b ka kb
k
c d kc kd
=
19
Matrix Multiplication
       
C A B B A = ≠
Row, Column indices have to line up as follows:
    
( ) ( )( )
C A B
mxn mxp pxn
=
≡
That is, the number of columns in the lefthand matrix must equal the
number of rows in the righthand matrix; if not, the multiplication is
undefined and cannot be done! Multiplication proceeds by
multiplying and adding terms along the rows of the lefthand matrix
and down the columns of the righthand matrix: (use your index
fingers from the left and right hands):
Example:
 
( ) ( )( )
2 1 2 3 3 1
g
a b c ag bh ci
C h
d e f dg eh fi
i
x x x
¦ ¹
+ +
¦ ¹
¦ ¦
= =
´ ` ´ `
+ +
¹ )
¦ ¦
¹ )
≡
note the inner indices (p=3) must match, as stated above and the
dimension of the result is the outer indices, i.e. 2x1.
20
Matrix Multiplication Examples
 
1 2 3
4 5 6
A
=
 
7 8
9 8
7 6
B
=
    
7 8
1 2 3
9 8
4 5 6
7 6
7 18 21 8 16 18 46 42
28 45 42 32 40 36 115 108
C A B =
=
+ + + +
= =
+ + + +
( ) ( )( )
2 2 2 3 3 2 x x x ≡
    
7 8
1 2 3
9 8
4 5 6
7 6
7 32 14 40 21 48 39 54 69
9 32 18 40 27 48 41 58 75
7 24 14 30 21 36 31 44 57
D B A =
=
+ + +
= + + + =
+ + +
( ) ( )( )
3 3 3 2 2 3 x x x ≡
21
Matrix Inversion
Matrix “division”: given
    
C A B = , solve for [B]
    
C A B = ⇒
        
  
 
1 1
A C A A B
I B
B
− −
=
=
=
     
1
B A C
−
⇒ =
Matrix [A] must be square to invert.
        
1 1
A A A A I
− −
= =
where [I] is the identity matrix, the matrix “1”. To calculate the
matrix inverse:
 
( )
1
Adjoint A
A
A
−
=
where: A Determinant of [A]
( ) ( )
Adjoint Cofactor
T
A A =
Cofactor(A)
( )
1
i j
ij ij
a M
+
= −
Minor
ij
M is the determinant of the submatrix with row i
and column j removed.
22
System of Linear Equations
We can solve n linear equations in n unknowns with the help of
a matrix. For n=3:
11 1 12 2 13 3 1
21 1 22 2 23 3 2
31 1 32 2 33 3 3
a x a x a x b
a x a x a x b
a x a x a x b
+ + =
+ + =
+ + =
Using matrix multiplication (backwards), this is written as:
 { } { }
A x b =
where:
 
11 12 13
21 22 23
31 32 33
a a a
A a a a
a a a
=
(known coefficients)
{ }
1
2
3
x
x x
x
¦ ¹
¦ ¦
=
´ `
¦ ¦
¹ )
(unknowns to be solved)
{ }
1
2
3
b
b b
b
¦ ¹
¦ ¦
=
´ `
¦ ¦
¹ )
(known righthand sides)
Unique solution { }   { }
1
x A b
−
= only if [A] has full rank. If not,
0 A = and the inverse of matrix [A] is undefined (dividing by zero).
23
Matrix Example
Solution of simultaneous linear equations.
1 2
1 2
2 5
6 4 14
x x
x x
+ =
+ =
⇒
)
`
¹
¹
´
¦
=
)
`
¹
¹
´
¦
14
5
4 6
2 1
2
1
x
x
 
=
4 6
2 1
A { }
)
`
¹
¹
´
¦
=
2
1
x
x
x { }
)
`
¹
¹
´
¦
=
14
5
b
{ }   { } b A x
1 −
=
( ) ( ) 8 6 2 4 1 − = − = A Determinant nonzero; unique solution!
 
−
−
=
−
−
=
−
8 / 1 4 / 3
4 / 1 2 / 1
1 6
2 4
1
1
A
A
check:         
= = =
− −
1 0
0 1
2
1 1
I A A A A
)
`
¹
¹
´
¦
=
)
`
¹
¹
´
¦
−
−
=
)
`
¹
¹
´
¦
2
1
14
5
8 / 1 4 / 3
4 / 1 2 / 1
2
1
x
x
Answer.
check: Plug answer into original equations and compare to the
{b} we need to get.
24
Vector and Matrix Matlab Examples
P1 = [1;2;0]; % Define two vectors
P2 = [3;2;0];
sum1 = P1+P2; % Vector addition
sum2 = P2+P1;
dot1 = dot(P1,P2); % Vector dot product
dot2 = dot(P2,P1);
cross1 = cross(P1,P2); % Vector cross product
cross2 = cross(P2,P1);
A = [1 2;6 4]; % Define a matrix and vector
b = [5;14];
dA = det(A); % Calculate determinant of A
invA = inv(A); % Calculate the inverse of A
x = invA*b; % Solve linear equations
x1 = x(1); % Extract answers
x2 = x(2);
A*x % Check answer – should be b
25
Matlab Introduction
Matrix laboratory
Control systems simulation and design software. Very
widespread in other fields. Introduction to basics, programming,
plots, animation, matrices, vectors. Based on C language,
programming is vaguely Clike, but much simpler to use. Sold by
Mathworks (http://www.mathworks.com).
Can buy student version software and manual for about the price
of one textbook (can use it for many classes!). ENT college has a
Matlab license; it is installed in most computer labs.
Doubleclick on Matlab icon to get started. Type
>>demo
to get a comprehensive overview of Matlab including builtin
functions. Try all the categories under Matlab first; you can ignore
Toolboxes, Simulink, and Stateflow for now. (Exception: there is
Symbolic Math under Toolboxes for the adventurous student!).
Type in commands (such as the Vector/Matrix examples given
earlier) at the Matlab prompt >>. Press <Enter> to see result or ;
<Enter> to suppress result.
Recommended operation mode: mfiles. Put your sequence of
Matlab statements in an ASCII file name.m (create a file with the
beautiful Matlab Editor/Debugger  this is colorcoordinated, tab
friendly, with parentheses alignment help and debugging
26
capabilities). A % indicates a comment. One basic way to run your
program is to hit the ‘save and run’ button on the editor toolbar.
Alternative: at the >> prompt type the MFile name name, without
the .m, assuming your file is in the search path. Matlab language is
interpretive and executes linebyline. Use the ; at the end of
statements to suppress intermediate results. If you use this
suppression, the variable name still holds the resulting value(s) – just
type the variable name at the prompt after the program runs to see the
value(s). If there is a syntax or programming logic error, it will give
a message at the bad line and then quit. Type:
>>who
to show you what variables you have defined;
>>whos
will show the variables, plus their matrix dimensions (scalar, vector
array, or matrix), very useful for debugging. Plus, after running a
file, place the cursor over different variables in the MFile inside the
Editor/Debugger to see the values! Online help is generally great:
>>help
Example mfiles (given on the following two pages)
1) MatEx1.m: Input, programming, plots, animation.
2) MatEx2.m: Matrix and vector definition, multiplication, transpose,
and solution of linear equations.
27
%
% Matlab Example Code 1: MatEx1.m
% Matrix, Vector examples
% Dr. Bob, ME 301
%
clc; clear; % Clear the cursor and clear any previously defined variables
%
% Matrix and Vector definition, multiplication, and transpose
%
A1 = [1 2 3; ... % Define 2x3 matrix [A1] (... is continuation line)
1 1 1];
x1 = [1;2;3]; % Define 3x1 vector {x1}
v = A1*x1; % 2x1 vector {v} is the product of [A1] times {x1}
A1T = A1'; % Transpose of matrix [A1]
vT = v'; % Transpose of vector {v}
%
% Solution of linear equations Ax=b
%
A2 = [1 2 3; ... % Define matrix [A2] to be a 3x3 coefficient matrix
1 1 1; ...
8 2 10];
b = [3;2;1]; % Define righthand side vector of knowns {b}
detA2 = det(A2); % First check to see if det(A) is near zero
x2 = inv(A2)*b; % Calculate {x2} to be the solution of Ax=b by inversion
check = A2*x2; % Check results;
z = b  check; % Better be zero!
%
% Display the usercreated variables (who), with dimensions (whos)
%
who
whos
%
% Display some of the results
%
v
x2
z
28
%
% Matlab Example Code 2: MatEx2.m
% Menu, Input, FOR loop, IF logic, Animation, and Plotting
% Dr. Bob, ME 301
%
clc; clear; % Clear the cursor and clear any previously defined variables
r = 1; L = 2; DR = pi/180; % Constants
%
% Input
%
anim = menu('Animate Single Link?','Yes','No') % Menu to screen
the = input('Enter [th0, dth, thf] (deg): ') % User types input
th0 = the(1)*DR; dth = the(2)*DR; thf = the(3)*DR; % Initial, delta, final thetas
th = [th0:dth:thf]; % Assign theta array
N = (thfth0)/dth + 1; % Number of iterations for loop
%
% Animate single link
%
if anim == 1 % Animate if user wants to
figure; % Give a blank graphics window
for i = 1:N; % For loop to animate
x2 = [0 L*cos(th(i))]; % Single link coordinates
y2 = [0 L*sin(th(i))];
plot(x2,y2); grid; % Animate to screen
set(gca,'FontSize',18);
xlabel('\itX (\itm)'); ylabel('\itY (\itm)');
axis('square'); axis([2 2 2 2]); % Define square plot limits
pause(1/4); % Pause to see animation
if i==1 % Pause to maximize window
pause; % User hits Enter to continue
end
end
end
%
% Calculate circle coordinates and cosine function
%
xc = r*cos(th); % Circle coordinates
yc = r*sin(th);
f1 = cos(th); % Cosine function of theta
f2 = sin(th); % Sine function of theta
%
% plots
%
figure; % Coplot cosine and sine functions
plot(th/DR,f1,'r',th/DR,f2,'g'); grid; set(gca,'FontSize',18);
legend('Cosine','Sine');
axis([0 360 1 1]); title('Functions of \it\theta');
xlabel('\it\theta (\itdeg)'); ylabel('Functions of \it\theta');
figure; % Plot circle
plot(xc,yc,'b'); grid; set(gca,'FontSize',18);
axis(['square']); axis([1.5 1.5 1.5 1.5]); title('Circle');
xlabel('\itX (\itm)'); ylabel('\itY (\itm)');
29
Mobility
Mobility:
Degreesoffreedom (dof):
How many dofs does an unconstrained planar link have?
What is the effect of constraining that link with a revolute joint?
Grubler's Criterion: Planar Jointed Devices
Where: M is the mobility
N is the total # of links, including ground
J
1
is the number of onedegreeoffreedom joints
J
2
is the number of twodegreeoffreedom joints
30
Onedegreeoffreedom joints:
Revolute
Prismatic
Twodegreeoffreedom joints (all have rolling and sliding):
Cam joint
Gear joint
Slottedpin joint
Caution: m links joining at one revolute location, must count m1
joints!
Caution: must count ground link (its freedom is subtracted in formula
with n1.
Planar mechanical device classification:
1 > M
1 = M
0 = M
0 < M
31
Planar Mobility Examples:
1) 3link serial robot
2) 4bar linkage
3) Slidercrank linkage
4) Scotch Yoke mechanism
5) Cam and follower
32
6) Gear pair
7) 4bar linkage with parallel link
8) Watt 6bar linkage
9) Staticallydeterminate structure
10) Staticallyindeterminate structure
33
11) 5bar linkage
12) Geared 5bar linkage
13) Cammodulated 4bar linkage
14) 3RRR parallel robot
34
FourBar Mechanism Position Analysis
Position (Displacement) Analysis: determination of relative
orientation/ position of links in a mechanism. Required for testing
motion of a synthesized mechanism. Also required for further
analysis: velocity, acceleration, dynamics, forces.
Generic Mechanism Position Analysis Statement: Given the
mechanism and one dof of position input, calculate the position
unknowns.
Fourbar Mechanism Position Analysis
Step 1. Draw the Kinematic Diagram:
r
1
– fixed ground link θ
1
– ground link angle
r
2
– input link θ
2
– input angle
r
3
– coupler link θ
3
– coupler angle
r
4
– output link θ
4
– output angle
All angles measured in righthand sense from horizontal to link.
35
Step 2. State the problem:
Step 3. Draw the Vector Diagram. Define all angles in positive
sense, measured from the right horizontal to the link vector (tailto
head). Don’t try to force acute angles; the relationships we can see so
easily in the first quadrant hold for all four quadrants:
)
`
¹
¹
´
¦
=
θ
θ
sin
cos
L
L
P ; good for all θ .
Vector Diagram:
Step 4. Derive the VectorLoopClosure Equation. Start at one
point, add vectors tailtohead until reach a second point. Write
equation by starting and ending at same points, but choosing a
different path.
36
Step 5. Write XY Components for VectorLoopClosure Equation.
Break one vector equation into its two scalar components (X and Y):
Step 6. Solve for the Unknowns from the XY Equations. Two
coupled nonlinear equations in the two unknowns
4 3
,θ θ . Isolate and
eliminate
3
θ and solve for
4
θ . Then go back to find
3
θ .
Square and add:
37
This equation has the form:
Solve using the tangent half angle substitution (Text Equation 4.9):

.

\

=
2
tan
4
θ
t
2
2
4
1
1
cos
t
t
+
−
= θ
2
4
1
2
sin
t
t
+
= θ
We converted a complicated coupled transcendental set of
equations into a quadratic polynomial. Much easier to solve (but we
doubled the order of the equation!).
Two solutions for
4
θ :
38
With factor two, no need to use the atan2 function.
Why two solutions? (Graphically demonstrate the two branches.)
What if 0
2 2 2
< − + G F E ? Imaginary solution, physically
means the mechanism cannot assemble for that input angle. See
section on Grashof's Law.
Go back to find
3
θ , one for each solution branch. Go back to original
two XY scalar equations.
Use ratio of Y to X equations:
Show graphical interpretation:


.

\

−
−
=
−
X X
Y Y
A B
A B
1
3
tan θ
39
The basic fourbar mechanism position analysis problem is now
solved. Now that we know the angular unknowns, we can find the
translational position of any point on the mechanism, e.g. coupler
point C:
Fourbar mechanism transmission angle: Transmission angle u:
relative angle between coupler and output links. Measure of
mechanical advantage of mechanism; 90
o
is ideal; 0,180
o
zero
transmission; as a rule of thumb, the absolute value of u should
remain in the range
D D
140 40 < < u for good transmission in a
mechanism. By geometry:
40
FourBar Mechanism Position Analysis: Term Example 1
Given
7
8
3
18 . 11
4
3
2
1
=
=
=
=
r
r
r
r
in
1
2
3
4
0.284
0.076
0.203
0.178
r
r
r
r
=
=
=
=
m
and
D
3 . 10
1
= θ (Ground link is 11" over and 2" up). Also given
5
/
=
A C
R (in) and
D
9 . 36
3
= δ for the coupler link point of interest.
Snapshot Analysis (one input angle)
Given this mechanism and
D
30
2
= θ , calculate u θ θ , ,
4 3
, and
C
P for
both branches. Results:
0.076
0.005
0.036
E
F
G
=
=
=
Branch t
3
θ
4
θ
u
C
P
Open 1.79
D
8 . 53
D
7 . 121
D
9 . 67
0.06, 0.16
Crossed 1.57
D
0 . 47 −
D
9 . 114 −
D
9 . 67
0.19, 0.02
These two branch solutions are demonstrated in the figures on the
following page. We use the SI system (m). Note u is identical for
both branches due to the conventions presented earlier.
41
0.1 0 0.1 0.2 0.3
0.1
0.05
0
0.05
0.1
0.15
0.2
0.25
X (m)
Y
(
m
)
4bar Example Snapshot, Open Branch
0.1 0 0.1 0.2 0.3
0.1
0.05
0
0.05
0.1
0.15
0.2
0.25
X (m)
Y
(
m
)
4bar Example Snapshot, Crossed Branch
42
Graphical Solution: The 4bar position analysis may be solved
graphically, merely by drawing the mechanism and determining the
mechanism closure. This is an excellent method to validate your
computer results at a given snapshot.
• Draw the known ground link (points O
2
and O
4
).
• Draw the given input link 2 length at the given angle (point A).
• Draw a circle of radius r
3
, centered at point A.
• Draw a circle of radius r
4
, centered at point O
4
.
• These circles intersect in general in two places.
• Connect the two branches and measure the unknown values.
Graphical Solution Figure:
43
4bar Snapshot Matlab code:
This program solves the 4bar position analysis problem for both
branches given a single
2
θ . The results are drawn to the screen.
%
% 4bar linkage snapshot position analysis  both branches
% Fbarplec.m, with graphical output, Dr. Bob, ME 301
%
clc; clear; % Clear cursor and clear previously defined variables
% Inputs
DR = pi/180;
R = input('Enter [r1, r2, r3, r4, rca, th1, th2, del3] (m and deg): ');
r1 = R(1); r2 = R(2); r3 = R(3); r4 = R(4); rca = R(5);
th1 = R(6)*DR; th2 = R(7)*DR; del3 = R(8)*DR; % Change degrees to radians
r1x = r1*cos(th1); r1y = r1*sin(th1);
% Position analysis: theta4
E = 2*r4*(r1*cos(th1)  r2*cos(th2));
F = 2*r4*(r1*sin(th1)  r2*sin(th2));
G = r1^2 + r2^2  r3^2 + r4^2  2*r1*r2*cos(th1th2);
t(1) = (F + sqrt(E^2 + F^2  G^2)) / (GE); % Crossed Branch
t(2) = (F  sqrt(E^2 + F^2  G^2)) / (GE); % Open Branch
th4(1) = 2*atan(t(1));
th4(2) = 2*atan(t(2));
% th3, coupler point, transmission angle; calculate for both branches
for i = 1:2,
ax = r2*cos(th2); % theta3
ay = r2*sin(th2);
bx = r4*cos(th4(i)) + r1x;
by = r4*sin(th4(i)) + r1y;
th3(i) = atan2(byay,bxax);
bet = th3(i) + del3; % coupler point
pcx(i) = r2*cos(th2) + rca*cos(bet);
pcy(i) = r2*sin(th2) + rca*sin(bet);
mu(i) = abs(th4(i)th3(i)); % transmission angle
end
% Plot 4bar position results
for i = 1:2,
x2 = [0 r2*cos(th2)]; % Coords of link 2
y2 = [0 r2*sin(th2)];
x3 = [r2*cos(th2) r1x+r4*cos(th4(i)) pcx(i)]; % Coords of link 3
y3 = [r2*sin(th2) r1y+r4*sin(th4(i)) pcy(i)];
x4 = [r1x r1x+r4*cos(th4(i))]; % Coords of link 4
y4 = [r1y r1y+r4*sin(th4(i))];
figure;
plot(x2,y2,'r',x4,y4,'r'); patch(x3,y3,'r');
axis('square'); set(gca,'FontSize',18);
xlabel('\itX (\itm)'); ylabel('\itY (\itm)');
axis([0.1 0.3 0.15 0.25]); grid;
end
44
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from position analysis is to report the
position analysis unknowns for the entire range of mechanism
motion. The first plot gives
3
θ (red),
4
θ (green), and u (blue), all
deg, for all
D D
360 0
2
≤ ≤θ , for Term Example 1, open branch only.
The second plot gives the coupler point location for this branch,
plotting
CY
P vs.
CX
P .
0 50 100 150 200 250 300 350
20
40
60
80
100
120
140
160
180
θ
2
(deg)
A
n
g
l
e
(
d
e
g
)
Thetas 3 (red) and 4 (green), Mu (blue)
45
0.1 0 0.1 0.2 0.3
0.1
0.05
0
0.05
0.1
0.15
0.2
0.25
X (m)
Y
(
m
)
4bar Example Snapshot, Open Branch Coupler Curve
46
Trigonometric Uncertainty
Return to
3
θ solution: XY scalar equations:
2 2 4 4 1 1 3 3
2 2 4 4 1 1 3 3
s r s r s r s r
c r c r c r c r
− + =
− + =
Since
4
θ has been solved, why not calculate
3
θ using Y equation?:
e.g. ( ) 5 . 0 sin
1
3
−
= θ ; figure:
Problem: inverse sine function is doublevalued; for each
4
θ there
are two possible solutions, only one of which is correct! Why not
calculate
3
θ using X equation? Inverse cosine has a similar problem;
e.g.

.

\

=
−
2
3
cos
1
3
θ ; figure:
Problem: inverse cosine function is doublevalued; for each
4
θ there
are two possible solutions, only one of which is correct!
47
So we must use information from both sine and cosine (i.e. both X
and Y equations)  this suggests using the tangent (as we did earlier in
the
3
θ solution):


.

\

− +
− +
=
−
2 2 4 4 1 1
2 2 4 4 1 1
1
3
tan
c r c r c r
s r s r s r
θ
e.g.

.

\

=
−
3
1
tan
1
3
θ ; figure:
Problem: the plain atan inverse tangent function is still double
valued!; for each
4
θ there are two possible solutions, only one of
which is correct! Solution: use the quadrantspecific inverse
tangent function atan2. Input to this function is both a numerator and
denominator; the function has builtin logic to determine the correct
quadrant for the angle answer, given the signs ± of the numerator
and denominator. The plain atan function takes a single quotient
input; hence this sign information is lost and the true quadrant is
unknown. No uncertainty with atan2:
e.g. =


.

\

+ + =
2
3
,
2
1
2 tan
3
a θ
=


.

\

− − =
2
3
,
2
1
2 tan
3
a θ
( )
2 2 4 4 1 1 2 2 4 4 1 1 3
, 2 tan c r c r c r s r s r s r A − + − + = θ
48
Now, having just cleared up this Trigonometric Uncertainty, we
already have an exception in the
4
θ tangent halfangle solution:
( ) t
1
4
tan 2
−
= θ
(there are two branches, one for each t value; only showing one here.)
With the 2 multiplying the inverse tangent result, it doesn't matter
whether we use atan or atan2 since the final answer will come to the
same angle. Example:
For

.

\

=
−
3
1
tan
2
1
4
θ
, from before, we don't know if the solution is
D
30
2
4
=
θ
or
D
210
2
4
=
θ
However, the multiple 2 takes care of this uncertainty:
D
60
4
= θ or
D D
60 420
4
= = θ
Now, for next time consider the following: Do the solutions for
4
θ
always exist? What if 0
2 2 2
= − + G F E ? What if
0
2 2 2
< − + G F E ? Stay tuned . . .
49
Grashof’s Law
Grashof was a German Engineer in the late 1800s. Grashof's
Law is used to determine the relative rotatability of the input and
output links in a 4bar mechanism:
Crank  full rotation, no limits
Rocker  not full rotation, rotates backandforth between limits
Mechanism types (input / output links):
Identify longest, shortest, intermediate 2 links: L, S, P, Q
1) If L + S < P + Q Then we call this a Grashof Mechanism and
there are four different mechanisms and rotation conditions:
Diagrams:
a)
b)
c)
50
2) If L + S > P + Q Then we call this a NonGrashof Mechanism
and the are four different mechanism inversions yield only one
rotation condition:
3) If L + S = P + Q Then we call this a Special Grashof
Mechanism and the four different mechanism inversions yield the
identical rotation conditions from 1) Grashof Mechanism. However,
there is the additional interesting and troublesome feature that the
mechanism may jump branches! Centerlines of links can become
collinear.
Examples
1) L = 10, S = 4, P = 8, Q = 7 – demonstrate the 4 possibilities
2) L = 10, S = 6, P = 8, Q = 7 – all Double Rockers
3) L = 10, S = 5, P = 8, Q = 7 – demonstrate branch jumping
Another interesting example: L = P = 10, S = Q = 4
parallel, locomotive linkage – subject to branch jumping unless
constrained. Also, very easy analysis:
u θ θ = =
4 2
0
3
= θ for all motion!
51
4Bar Joint Limits
If Grashof's Law predicts the input link is a rocker, there will be
rotation limits on the input link. These joint limits occur when links
3 and 4 are aligned. As shown in the figure, there will be two joint
limits, symmetric about the ground link.
To calculate the joint limits, we use the law of cosines:
( )
L
r r r r r r
2 2 1
2
2
2
1
2
4 3
cos 2 θ − + = +
( )
+ − +
± =
−
2 1
2
4 3
2
2
2
1 1
2
2
cos
r r
r r r r
L
θ
± by symmetry about
1
r
52
Example 1: Given 7 , 8 , 6 , 10
4 3 2 1
= = = = r r r r
L + S > P + Q (10 + 6 > 8 + 7)
so we predict only double rockers from this mechanism.
( )
( )( )
 
D
9 . 137 742 . 0 cos
6 10 2
7 8 6 10
cos
1
2 2 2
1
2
± = − ± =
+ − +
± =
− −
L
θ
Example 2: Given 7 , 8 , 4 , 10
4 3 2 1
= = = = r r r r
L + S < P + Q (10 + 4 < 8 + 7),
so we predict this mechanism is a crankrocker. Therefore, there are
no joint limits!
( )
( )( )
  3625 . 1 cos
4 10 2
7 8 4 10
cos
1
2 2 2
1
2
− ± =
+ − +
± =
− −
L
θ
which is undefined!
Caution: the figure on the previous page does not apply in all joint
limit cases. For certain mechanisms, the limiting conditions occur
when links 3 and 4 fold upon each other instead of stretching straight
out. The previous method can also be used to find angular limits on
link 4 when it is a rocker; here links 2 and 3 either stretch out in a
line or fold upon each other.
Example 3: (Term Example Fourbar)
Given 7 , 8 , 3 , 18 . 11
4 3 2 1
= = = = r r r r (in) and
D
3 . 10
1
= θ , limits are:
D
1 . 120
4
=
L
θ (links 2 and 3 stretched in a line)
D
5 . 172
4
=
L
θ (links 2 and 3 folded upon each other in a line)
There are no limits on
2
θ since it is a crank.
53
SliderCrank Mechanism Position Analysis
Converts linear motion to rotary or rotary motion to linear via
connecting rod. Internal Combustion Engine – explosion drives
piston (input), output is rotation of drive shaft. Air Compressor –
electric motor drives crank (input), piston (output) compresses air.
Two dead points where piston is at limits. Use flywheel on crank to
avoid locking. Unlike the fourbar mechanism, the four kinematic
inversions of the slidercrank mechanism yield radically different
types of motion. In class we will solve the Air Compressor case
where the crank is the input and the slider is the output.
Step 1. Draw the Kinematic Diagram:
r
2
– input link length θ
2
– input angle
r
3
– coupler link length θ
3
– coupler angle
h
– slider offset x – output displacement
Link 1 is the fixed ground link. All angles measured in righthand
sense from horizontal to link. x is measured horizontally from the
origin to the slider/coupler revolute joint location.
54
Step 2. State the problem:
Step 3. Draw the Vector Diagram. Define all angles in positive
sense, measured from the right horizontal to the link vector (tailto
head).
Vector Diagram:
Step 4. Derive the VectorLoopClosure Equation. Start at one
point, add vectors tailtohead until reach a second point. Write
equation by starting and ending at same points, but choosing a
different path.
55
Step 5. Write XY Components for VectorLoopClosure Equation.
Break one vector equation into its two scalar components (X and Y):
Step 6. Solve for the Unknowns from the XY Equations. Two
coupled nonlinear equations in the two unknowns
3
,θ x . We could
isolate on unknown, square & add, and solve as in the fourbar
approach. However, notice that the two XY equations are coupled
only in
3
θ but not in x. There a simpler method  solve
3
θ using the
Y equation only and then solve x from the X equation:
What about trigonometric uncertainty? The inverse sine function is
doublevalued and so there are two valid solution branches.
Graphically demonstrate the two branches.
56
Fullrotation condition
For solution to exist for entire motion range (r
2
is a crank),
absolute value of the inverse sine argument must be less than or equal
1:
1
3
2 2
≤
−
r
s r h
2 2 3
s r h r − ≥
which must hold for all motion. The worst case is
D
90
2
− = θ , which
yields
2 3
r h r + ≥
This condition was derived assuming positive h; allowing negative h:
2 3
r h r + ≥ .
57
SliderCrank Mechanism Position Analysis: Term Example 2
Given:
3
8
4
3
2
=
=
=
h
r
r
in
2
3
0.102
0.203
0.076
r
r
h
=
=
=
m
Snapshot Analysis (one input angle)
Given this mechanism and
D
30
2
= θ , calculate x and
3
θ for both
branches. Results:
Branch x (m)
3
θ
Open 0.290
D
2 . 7
Crossed 0.114
D
8 . 172
These two branch solutions are demonstrated in the figures on the
following page. We use the SI system (m).
58
0.1 0 0.1 0.2 0.3
0.2
0.1
0
0.1
0.2
X (m)
Y
(
m
)
SliderCrank Example Snapshot, Open Branch
0.1 0 0.1 0.2 0.3
0.2
0.1
0
0.1
0.2
X (m)
Y
(
m
)
SliderCrank Example Snapshot, Crossed Branch
59
Graphical Solution: The SliderCrank position analysis may be
solved graphically, merely by drawing the mechanism and
determining the mechanism closure. This is an excellent method to
validate your computer results at a given snapshot.
• Place the grounded revolute for the crank at the origin.
• Draw the line of the slider, offset vertically from the origin by h.
• Draw the given input link 2 length at the given angle (point A).
• Draw a circle of radius r
3
, centered at point A.
• This circle intersects the slider line in general in two places.
• Connect the two branches and measure the unknown values.
Graphical Solution Figure:
Slider Limits
The crank will rotate fully if the previouslyderived condition is
met. The slider reaches its maximum displacement when links 2 and
3 are aligned straight out and its maximum displacement when link 2
if folded onto link 3. We can draw two right triangles representing
these conditions and easily calculate the x limits to be
2951 . 0 0671 . 0 ≤ ≤ x , as seen in the full motion x plot, next page.
60
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from position analysis is to report the
position analysis unknowns for the entire range of mechanism
motion. The first plot gives x (m), for all
D D
360 0
2
≤ ≤θ , for Term
Example 2, right branch only. The second plot gives
3
θ (deg), for all
D D
360 0
2
≤ ≤θ , for the right branch only.
0 50 100 150 200 250 300 350
0.05
0.1
0.15
0.2
0.25
0.3
θ
2
(deg)
x
(
m
)
Slider Displacement
61
0 50 100 150 200 250 300 350
10
0
10
20
30
40
50
60
θ
2
(deg)
θ
3
(
d
e
g
)
Coupler Angle
62
Velocity Analysis Introduction
Velocity analysis is important for kinematic motion analysis.
Some tasks have timing, rates. Position analysis must be completed
first. Velocity analysis is also required for dynamics: position,
velocity, acceleration, dynamics, forces, machine design. Velocity
analysis is solution of coupled linear equations. Velocity is the first
time derivative of the position. Vector quantity:
Magnitude of velocity is speed; direction also crucial. Analytical
velocity analysis: write position vectors, take first time derivatives,
solve for unknowns. Units (translational and rotational):
Basic Velocity Derivation Figure:
Most general planar case: Translating and rotating rigid rod with a
slider on it. Find the total velocity of point P on the slider. Express
the position vector in Cartesian coordinates:
= + = L P P
O P
63
The angle is changing with angular velocity:
Only the planar case is this simple; the spatial rotation case is more
complicated. The length of the rod is changing with sliding velocity:
Product and Chain Rules of Differentiation
We’ll need to use the product and chain rules over and over in
velocity and acceleration analysis derivations.
Product rule:
( )
dt
dy
x y
dt
dx
xy
dt
d
+ = x, y both functions of time.
Chain rule:
( ) ( ) ( )
dt
dx
dx
df
t x f
dt
d
= f is a function of x, which is a function of t.
Example:
( ) ? cos = θ L
dt
d
64
Back to basic velocity derivation
First time derivative of position vector:
= =
dt
P d
V
P
P
65
We have just derived the ThreePart Velocity Equation:
L V V V
O P
× + + = ω
The terms for the ThreePart Velocity Equation can be
expressed in various ways, summarized below:
Vector
O
V
V
L × ω
Name
Point O
Velocity
Sliding
Velocity
Tangential
Velocity
XY
Components
Magnitude /
Direction
66
ThreePart Velocity Equation Example:
Given (instantaneously) L = 2 m,
D
30 = θ , ω = 1 rad/s, 3 V L = =
m/s
(outward), { }
3 2
T
O
V = m/s, calculate
P
V .
( )
( )
cos sin 3 3cos30 2 1 sin30
sin cos 2 3sin30 2 1 cos30
3 2.598 1 4.598
2 1.5 1.732 5.232
OX
P
OY
P
V V L
V
V V L
m
V
s
θ ω θ
θ ω θ
+ − ¦ ¹ + −
¦ ¹
= =
´ ` ´ `
+ + + +
¹ )
¹ )
+ −
¦ ¹ ¦ ¹
= =
´ ` ´ `
+ +
¹ ) ¹ )
D D
D D
or, 6.965@48.7
P
V =
D
m/s
Show magnitude and direction of each velocity component:
Vector
O
V
V
L × ω
Name
Point O
Velocity
Sliding
Velocity
Tangential
Velocity
XY
Components
Magnitude /
Direction
67
FourBar Mechanism Velocity Analysis
Velocity Analysis: determination of angular and linear velocities of
links in a mechanism. Required for complete motion analysis. Also
required for further analysis: acceleration, dynamics, forces, machine
design. Linear equations result from first time differentiation of
position equations. Unique solution for each mechanism branch.
Position analysis must be complete first. 1dof mechanism, so one
velocity input must be given.
Generic Mechanism Velocity Analysis Statement: Given the
mechanism, complete position analysis, and one dof of velocity input,
calculate the velocity unknowns.
Fourbar Mechanism Velocity Analysis
Step 1. Position Analysis must first be complete.
Step 2. Draw the Velocity Diagram:
where
i
ω , i = 2,3,4, is the absolute angular velocity of link i. 0
1
= ω
since the ground link is fixed.
68
Step 3. State the problem:
Step 4. Derive the velocity equations. Take the first time derivative
of the vector loop closure equations from position analysis, in XY
component form.
Fourbar mechanism position equations:
2 3 1 4
r r r r + = +
2 2 3 3 1 1 4 4
2 2 3 3 1 1 4 4
r c r c rc r c
r s r s r s r s
+ = +
+ = +
First time derivative for velocity equations: (use chain rule several
times) Chain rule:
( )
cos
cos
sin
sin
i i
i
i
i i
i i
d d d
dt d dt
θ θ
θ
θ
θθ
θ ω
=
= −
= −
( )
sin
sin
cos
cos
i i
i
i
i i
i i
d d d
dt d dt
θ θ
θ
θ
θθ
θ ω
=
=
=
Don’t have to use product rule because 0 =
i
r (rigid links).
69
The first time derivative of the position equations is:
Gathering unknowns on the LHS:
Substituting simpler terms:
Written in matrix form:
70
Step 5. Solve the velocity equations for the unknowns
4 3
, ω ω .
Algebra solution:
71
Alternate matrix solution (yields same solution):
72
FourBar mechanism singularity condition:
When does the solution fail? This is a mechanism singularity,
when the determinant of the coefficient matrix goes to zero. The
result is dividing by zero, for infinite answers
4 3
, ω ω . Let’s see what
this means physically.
Physically, this happens when links 3 and 4 are straight out or folded
on top of each other (what does this correspond to?):
73
The basic fourbar mechanism velocity analysis problem is now
solved. Now that we know the angular unknowns, we can find the
translational velocity of any point on the mechanism, e.g. coupler
point C:
74
Fourbar mechanism velocity example:
Given r
1
= 0.284, r
2
= 0.076, r
3
= 0.203, r
4
= 0.178, R
C/A
= 0.127
m, and
1
10.3 θ =
D
,
2
30 θ =
D
,
3
53.8 θ =
D
,
4
121.7 θ =
D
,
3
36.9 δ =
D
. This is
the open branch of the fourbar mechanism position example (Term
Example 1).
Snapshot Analysis (one input angle)
Given this mechanism position analysis plus
2
ω π = rad/s (+, so
ccw), calculate
4 3
,ω ω , and
C
V for this instant (snapshot).
3
4
0.164 0.151 0.120
0.120 0.093 0.207
ω
ω
− −
¦ ¹ ¦ ¹
=
´ ` ´ `
− −
¹ ) ¹ )
3
4
1.271
0.587
ω
ω
−
¦ ¹ ¦ ¹
=
´ ` ´ `
−
¹ ) ¹ )
Both are negative, so cw direction. These results are the absolute
angular velocities of links 3 and 4 with respect to the ground link.
Coupler point translational velocity:
0.042
0.209
C
V
¦ ¹
=
´ `
¹ )
(m/s)
75
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from velocity analysis is to report the
velocity analysis unknowns for the entire range of mechanism
motion. The plot below gives
3
ω (red) and
4
ω (green), (rad/s), for
all
D D
360 0
2
≤ ≤θ , for Term Example 1, open branch only. Since
2
ω is constant, we can plot the velocity results vs.
2
θ (since it is
related to time t via t
2 2
ω θ = ).
0 50 100 150 200 250 300 350
1.5
1
0.5
0
0.5
1
1.5
2
θ
2
(deg)
ω
3
a
n
d
ω
4
(
r
a
d
/
s
)
Omegas 3 (red) and 4 (green)
76
The plot below gives the translational coupler point velocity for all
D D
360 0
2
≤ ≤θ , for Term Example 1, open branch only.
0 50 100 150 200 250 300 350
0.2
0.1
0
0.1
0.2
0.3
θ
2
(deg)
V
C
X
a
n
d
V
C
Y
(
m
/
s
)
Coupler Point Velocities X (red) and Y (green)
77
Derivative/Integral Relationships
When one variable is the derivative of another, what are the
relationships? For example:
dt
d
3
3
θ
ω =
∫
+ = dt
3 30 3
ω θ θ
0 50 100 150 200 250 300 350
0.4
0.6
0.8
1
1.2
θ
3
(
r
a
d
)
0 50 100 150 200 250 300 350
1
0
1
θ
2
(deg)
ω
3
(
r
a
d
/
s
)
The value of
3
ω at any point is the slope of the
3
θ curve at that point.
The value of
3
θ at any point is the integral of the
3
ω curve up to that
point (the value of
3
θ at any point is the area under the
3
ω curve up
to that point).
This graph is vs.
2
θ , but the same type of relationships hold as
for time t since
2
ω is constant. This is the Term Example 1 result,
but we changed
3
θ from deg to rad for better comparison.
78
SliderCrank Mechanism Velocity Analysis
Again, we will solve the Air Compressor case where the crank is
the input and the slider is the output. The Internal Combustion
Engine case (slider input/crank output) is equally interesting.
Step 1. Position Analysis must first be complete.
Step 2. Draw the Velocity Diagram:
where
i
ω , i = 2,3 is the absolute angular velocity of link i. x is the
variable slider velocity. 0
4
= ω since the slider cannot rotate.
Step 3. State the problem:
79
Step 4. Derive the velocity equations. Take the first time derivative
of the vector loop closure equations from position analysis, in XY
component form.
Slidercrank mechanism position equations:
2 3
r r x h + = +
2 2 3 3
2 2 3 3
r c r c x
r s r s h
+ =
+ =
First time derivative for velocity equations:
Gathering unknowns on the LHS:
Written in matrix form:
80
Step 5. Solve the velocity equations for the unknowns x ,
3
ω .
Actually, these equations are decoupled so we don't need a
matrix solution. First, solve
3
ω from Y equation:
Then solve x from the X equation using the
3
ω result:
The alternate matrix solution:
will yield identical results.
81
Slidercrank mechanism singularity condition:
When does the solution fail? This is a slidercrank mechanism
singularity, when the determinant of the coefficient matrix goes to
zero. The result is dividing by zero, resulting in infinite answers
x ,
3
ω .
3 3
0 A r c = =
0 = A when 0 cos
3
= θ , or "
D D
, 270 , 90
3
= θ
Physically, this happens when link 3 is straight up or down
(
3
90 θ = ±
D
). Doesn’t happen for nominal fullrotation slidercrank
mechanisms, even with offsets.
Of course r
3
cannot go to zero, otherwise we have a degenerate
slidercrank mechanism.
82
Slidercrank mechanism velocity example:
Given r
2
= 0.102, r
3
= 0.203, h = 0.076 m, and
2
30 θ =
D
,
3
7.2 θ =
D
, x = 0.290 m. This is the right branch of the slidercrank
position example (Term Example 2).
Snapshot Analysis (one input angle)
Given this mechanism position analysis plus
2
/ 2 ω π = rad/s (+,
so ccw), calculate
3
,ω x for this instant (snapshot).
3
1 0.025 0.080
0 0.202 0.138
x
ω
−
¦ ¹ ¦ ¹
=
´ ` ´ `
−
¹ ) ¹ )
3
0.062
0.686
x
ω
−
¦ ¹ ¦ ¹
=
´ ` ´ `
−
¹ ) ¹ )
Both are negative, so the slider is currently traveling to the left and
the coupler link is currently rotating in the cw direction. These
results are the absolute linear and angular velocities of links 4 and 3
with respect to the fixed ground link.
83
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from velocity analysis is to report the
velocity analysis unknowns for the entire range of mechanism
motion. The plot below gives x (red, m/s) and
3
ω (green, rad/s), for
all
D D
360 0
2
≤ ≤θ , for Term Example 2, right branch only. Since
2
ω is constant, we can plot the velocity results vs.
2
θ (since it is
related to time t via t
2 2
ω θ = ).
0 50 100 150 200 250 300 350
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
0.8
θ
2
(deg)
x
d
(
m
/
s
)
a
n
d
ω
3
(
r
a
d
/
s
)
Xdot (red) and Omega3 (green)
84
Derivative/Integral Relationships
When one variable is the derivative of another, what are the
relationships? For example:
dt
dx
x =
∫
+ = dt x x x
0
0 50 100 150 200 250 300 350
0.1
0.2
0.3
x
(
m
)
0 50 100 150 200 250 300 350
0.2
0
0.2
θ
2
(deg)
x
d
(
m
/
s
)
The value of x at any point is the slope of the x curve at that point.
The value of x at any point is the integral of the x curve up to that
point (the value of x at any point is the area under the x curve up to
that point).
This graph is vs.
2
θ , but the same type of relationships hold as
for time t since
2
ω is constant. This is the Term Example 2 result.
85
Acceleration Analysis Introduction
Acceleration analysis is required for dynamics: position,
velocity, acceleration, dynamics, forces, machine design. Important
for kinematic motion analysis. Position and velocity analyses must be
completed first. Acceleration analysis is solution of linear equations.
Acceleration is the first time derivative of the velocity and second
time derivative of the position. Vector quantity:
Analytical acceleration analysis: write position vectors, take first
two time derivatives, solve for unknowns. Units (translational and
rotational):
Basic Acceleration Derivation Figure:
Rotating rigid rod with a slider on it. Find the total acceleration of
point P on the slider.
86
Recall the 2part position and 3part velocity results:
)
`
¹
¹
´
¦
+
+
= + =
θ
θ
sin
cos
L P
L P
L P P
OY
OX
O P
)
`
¹
¹
´
¦
+ +
− +
= × + + =
θ ω θ
θ ω θ
ω
cos sin
sin cos
L V V
L V V
L V V V
OY
OX
O P
The angle is changing with angular velocity and acceleration:
Only planar case is this simple; the spatial rotation case is more
complicated. The length of the rod is changing with sliding velocity
and acceleration:
87
Product and Chain Rules of Differentiation
Again, we’ll need to use the product and chain rules over and
over in acceleration analysis derivations.
Product rule:
( )
dt
dy
x y
dt
dx
xy
dt
d
+ = x, y both functions of time.
Chain rule:
( ) ( ) ( )
dt
dx
dx
df
t x f
dt
d
= f is a function of x, which is a function of t.
Example:
( )
2
2
cos ?
d
L
dt
θ =
88
Back to basic acceleration derivation
First time derivative of velocity vector (Second time derivative of
position vector):
= = =
2
2
dt
P d
dt
V d
A
P P
P
89
We have just derived the FivePart Acceleration Equation:
( ) L L V A A A
O P
× × + × + × + + = ω ω α ω 2
These terms can be expressed in various ways, summarized below:
Vector
O
A
A
V × ω 2
L × α
( ) L × × ω ω
Name
Point O
Acceleration
Sliding
Acceleration
Coriolis
Acceleration
Tangential
Acceleration
Centripetal
Acceleration
XY
Components
Magnitude /
Direction
90
FivePart Acceleration Equation Example:
Continuation of 3part velocity example.
Given (instantaneously) L=2 m,
D
30 = θ , 1 ω = rad/s, 2 α = rad/s
2
,
3 V L = =
m/s (outward),
{ }
3 2
T
O
V = , 4 A L = =
m/s
2
(outward),
{ }
1 2
T
O
A = , calculate
P
A .
2
2
2
cos 2 sin sin cos
sin 2 cos cos sin
1 3.464 3 2 1.732 2.268
2 2 5.196 3.464 1 11.660
OX
P
OY
A A V L L
A
A A V L L
m
s
θ ω θ α θ ω θ
θ ω θ α θ ω θ
¦ ¹ + − − −
=
´ `
+ + + −
¹ )
+ − − − −
¦ ¹ ¦ ¹
= =
´ ` ´ `
+ + + −
¹ ) ¹ )
or, 11.879@101.0
P
A =
D
m/s
2
Show magnitude and direction of each Acceleration component:
Vector
O
A
A
V × ω 2
L × α
( ) L × × ω ω
Name Point O
Acceleration
Sliding
Acceleration
Coriolis
Acceleration
Tangential
Acceleration
Centripetal
Acceleration
XY
Components
Magnitude /
Direction
91
FourBar Mechanism Acceleration Analysis
Acceleration Analysis  determination of angular and linear
accelerations of links in a mechanism. Required for complete motion
analysis. Also required for further analysis: dynamics, forces,
machine design. Linear equations result from second time
differentiation of position equations. Unique solution for each
mechanism branch. Position and velocity analyses must be complete
first. 1dof mechanism, so one acceleration input must be given.
Generic Mechanism Acceleration Analysis Statement: Given the
mechanism, complete position and velocity analyses, and one dof of
acceleration input, calculate the acceleration unknowns.
Fourbar Mechanism Acceleration Analysis
Step 1. Position and Velocity Analyses must first be complete.
Step 2. Draw the Acceleration Diagram:
where
i
α , i = 2,3,4 is the absolute angular acceleration of link i.
0
1
= α since the ground link is fixed.
92
Step 3. State the problem:
Step 4. Derive the acceleration equations. Take the first time
derivative of the fourbar mechanism velocity equations from
velocity analysis, in XY component form.
Fourbar mechanism velocity equations:
2 2 2 3 3 3 4 4 4
2 2 2 3 3 3 4 4 4
r s r s r s
r c r c r c
ω ω ω
ω ω ω
− − = −
+ =
The first time derivative of the velocity equations is:
Gathering unknowns on the LHS:
93
Substituting simpler terms:
Written in matrix form:
94
Step 5. Solve the acceleration equations for the unknowns
4 3
,α α .
Matrix solution (Algebra solution yields the same results):
95
FourBar mechanism singularity condition:
Same coefficient matrix A as velocity case, so singularity
condition is identical:
"
D D
, 180 , 0
3 4
= −θ θ
This condition is the same problem for position, velocity, and
acceleration. At this singularity, there is zero transmission angle u
and Link 2 is at a joint limit!
The basic fourbar mechanism acceleration analysis problem is now
solved. Now that we know the angular unknowns, we can find the
translational acceleration of any point on the mechanism, e.g.
coupler point C:
96
Fourbar mechanism acceleration example:
Given r
1
= 0.284, r
2
= 0.076, r
3
= 0.203, r
4
= 0.178, R
C/A
= 0.127
m, and
1
10.3 θ =
D
,
2
30 θ =
D
,
3
53.8 θ =
D
,
4
121.7 θ =
D
,
3
36.9 δ =
D
;
2
ω π = ,
3
1.271 ω = − ,
4
0.587 ω = − rad/s. This is the open branch of the
position and velocity example (Term Example 1).
Snapshot Analysis (one input angle)
Given this mechanism position and velocity analysis, plus
0
2
= α rad/s
2
, calculate
4 3
,α α for this instant (snapshot).
3
4
0.164 0.151 0.877
0.120 0.093 0.589
α
α
− −
¦ ¹ ¦ ¹
=
´ ` ´ `
− − −
¹ ) ¹ )
3
4
0.213
6.030
α
α
¦ ¹ ¦ ¹
=
´ ` ´ `
¹ ) ¹ )
Both are positive, so ccw direction. These results are the absolute
angular accelerations of links 3 and 4 with respect to the ground link.
Coupler point translational acceleration:
0.676
0.582
C
A
−
¦ ¹
=
´ `
−
¹ )
m/s
2
97
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from acceleration analysis is to report
the acceleration analysis unknowns for the entire range of mechanism
motion. The plot below gives
3
α (red) and
4
α (green), (rad/s
2
), for
all
D D
360 0
2
≤ ≤θ , for Term Example 1, open branch only. Since
2
ω is constant, we can plot the acceleration results vs.
2
θ (since it is
related to time t via t
2 2
ω θ = ).
0 50 100 150 200 250 300 350
6
4
2
0
2
4
6
θ
2
(deg)
α
3
a
n
d
α
4
(
r
a
d
/
s
2
)
Alphas 3 (red) and 4 (green)
98
The plot below gives the translational coupler point acceleration for
all
D D
360 0
2
≤ ≤θ , for Term Example 1, open branch only.
0 50 100 150 200 250 300 350
0.8
0.6
0.4
0.2
0
0.2
0.4
0.6
0.8
θ
2
(deg)
A
C
X
a
n
d
A
C
Y
(
m
/
s
2
)
Coupler Point Accelerations X (red) and Y (green)
99
Derivative/Integral Relationships
When one variable is the derivative of another, recall the
relationships from calculus. For example:
dt
d
4
4
θ
ω =
∫
+ = dt
4 40 4
ω θ θ
dt
d
4
4
ω
α =
∫
+ = dt
4 40 4
α ω ω
0 50 100 150 200 250 300 350
2
2.5
3
θ
4
(
r
a
d
)
0 50 100 150 200 250 300 350
1
0
1
ω
4
(
r
a
d
/
s
)
0 50 100 150 200 250 300 350
5
0
5
α
4
(
r
a
d
/
s
2
)
θ
2
(deg)
100
SliderCrank Mechanism Acceleration Analysis
Again, we will solve the Air Compressor case where the crank is
the input and the slider is the output.
Step 1. Position and Velocity Analyses must first be complete.
Step 2. Draw the Acceleration Diagram:
where 3 , 2 ; = i
i
α is the absolute angular acceleration of link i. 0
4
= α
since the slider cannot rotate.
Step 3. State the problem:
101
Step 4. Derive the acceleration equations. Take the first time
derivative of the velocity equations from velocity analysis, in XY
component form.
Slidercrank mechanism velocity equations:
2 2 2 3 3 3
2 2 2 3 3 3
0
r s r s x
r c r c
ω ω
ω ω
− − =
+ =
The first time derivative of the velocity equations is:
Gathering unknowns on the LHS:
Written in matrix form:
102
Step 5. Solve the acceleration equations for the unknowns x ,
3
α .
Actually, these equations are decoupled so we don't need a matrix
solution. First, solve
3
α from Y equation:
Then solve x from the X equation using the
3
α result:
103
Slidercrank mechanism singularity condition:
Same coefficient matrix as velocity case, so singularity
condition is identical (see the singularity discussion in the slider
crank velocity section).
Slidercrank mechanism acceleration example:
Given r
2
= 0.102, r
3
= 0.203, h = 0.076 m, and
2
30 θ =
D
,
3
7.2 θ =
D
, x = 0.290 m; and
2
/ 2 ω π = ,
3
0.686 ω = − rad/s, 0.062 x = −
m/s. This is the right branch of the position and velocity example
(Term Example 2).
Snapshot Analysis (one input angle)
Given this mechanism position and velocity analysis plus,
0
2
= α rad/s
2
, calculate
3
,α x for this instant (snapshot).
3
1 0.025 0.312
0 0.202 0.137
x
α
−
¦ ¹ ¦ ¹
=
´ ` ´ `
− −
¹ ) ¹ )
3
0.329
0.681
x
α
−
¦ ¹ ¦ ¹
=
´ ` ´ `
¹ ) ¹ )
These results are the absolute linear and angular accelerations of links
4 and 3 with respect to the fixed ground link.
104
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from acceleration analysis is to report
the acceleration analysis unknowns for the entire range of mechanism
motion. The plot below gives x (red, m/s
2
) and
3
α (green, rad/s
2
),
for all
D D
360 0
2
≤ ≤θ , for Term Example 2, right branch only. Since
2
ω is constant, we can plot the velocity results vs.
2
θ (since it is
related to time t via t
2 2
ω θ = ).
0 50 100 150 200 250 300 350
2.5
2
1.5
1
0.5
0
0.5
1
θ
2
(deg)
x
d
d
(
m
/
s
2
)
a
n
d
α
3
(
r
a
d
/
s
2
)
Xdotdot (red) and Alpha3 (green)
105
Derivative/Integral Relationships
When one variable is the derivative of another, recall the
relationships from calculus. For example:
dt
dx
x =
∫
+ = dt x x x
0
dt
x d
x
=
∫
+ = dt x x x
0
0 50 100 150 200 250 300 350
0.1
0.2
0.3
x
(
m
)
0 50 100 150 200 250 300 350
0.2
0
0.2
x
d
(
m
/
s
)
0 50 100 150 200 250 300 350
0.5
0
0.5
θ
2
(deg)
x
d
d
(
m
/
s
2
)
106
Input Motion Specification
Up to this point, for full range of motion we have assumed that
the input link rotates fully with a given constant input angular
velocity. Our input constraints have thus been
D D
360 0
2
≤ ≤θ , ω
2
constant, and α
2
= 0. This input motion specification is plotted like
this:
Note that we have been plotting calculated results vs. θ
2
. Since ω
2
is
constant, we have
2 2
t θ ω = , so we could just as well plot all results
vs. time t, since both θ
2
and t increase steadily (linearly).
107
This constant ω
2
input specification is fine for mechanisms whose
input rotates fully and considering steadystate motion only. Many
useful mechanisms have input links that do not rotate fully but travel
between joint limits, starting and stopping at zero angular velocity.
Why is the previous page’s plots unacceptable in this case?
Simplest change – linear angular velocity starting and stopping at
zero:
We cannot plot vs. θ
2
since it is not increasing linearly – plot vs. t.
What is the weakness of this approach? (Discontinuous acceleration
function yields infinite jerk at start, middle, and finish.)
108
We can fix this with a trapezoidal input acceleration profile:
This input motion specification should be fine (trapezoidal input
torque is often used for industrial robots), but there are many
different zones to handle – what acceleration profile is similar but
with a single function?
109
Fullcycloidal function input angle specification
( ) ( )
( )
( )
( )
( )
( )
( )
2 20 2 20
2 20
2
2 20
2
2
2
2 20
2
3
1 2
sin
2
2
1 cos
2
2
sin
4
2
cos
F
F F
F
F F
F
F F
F
F F
t t
t
t t
t
t
t t
t
t
t t
t
t
t t
π
θ θ θ θ
π
θ θ
π
ω
π θ θ
π
α
π θ θ
π
β
 
= + − −

\ .
−  
= −

\ .
−  
=

\ .
−  
=

\ .
0 0.5 1 1.5 2 2.5 3
60
80
100
θ
2
(
d
e
g
)
0 0.5 1 1.5 2 2.5 3
0
2
4
ω
2
(
r
a
d
/
s
)
0 0.5 1 1.5 2 2.5 3
1
0
1
α
2
(
r
a
d
/
s
2
)
0 0.5 1 1.5 2 2.5 3
2
0
2
time (sec)
β
2
(
r
a
d
/
s
3
)
Example with
20
60 θ =
D
,
2
120
F
θ =
D
, and t
F
= 3 sec.
110
Dynamics Introduction
Chart:
Kinematics:
translational
rotational
Kinetics:
translational Newton’s 2
nd
Law:
rotational Euler’s equation:
Dynamics of a single rigid body in the plane
Rigid body acted on by a system of forces and moments to
produce planar motion. What is the first step in analysis? Draw . . .
111
Free Body Diagram (FBD)
Isolate each rigid body and show the forces and moments acting.
This contains all the info needed to write Newton’s 2
nd
Law and
Euler’s equation.
FBD Simplified FBD
MAD (massacceleration diagram)
112
Internal and External Forces and Moments
All internal and external forces and moments must be included
on the FBD.
External forces/moments:
Internal forces/moments:
Write dynamics equations
Newton’s 2
nd
Law:
Euler’s equation:
G
A is the linear acceleration of center of gravity – same direction as
R. Different points in rigid body have different linear accelerations.
α angular acceleration of rigid body. The entire rigid body
experiences the same α .
113
D’Alembert’s Principle
Turn dynamics problem into a statics problem by the inclusion
of a fictitious “inertial force”
G
A m F − =
0
and a fictitious “inertial
moment” α
G
I M − =
0
. “Centrifugal force”
2
ω mr − is an example of
an inertial force; it’s not really a force but an effect of acceleration
and inertia. Subtract RHS of equations, then sum to zero as in statics.
We won’t use this method, just wanted you to know in case you ran
into it somewhere.
0
0
G
O
R mA
R F
− =
+ =
0
0
G
O
T r R I
T r R M
α + × − =
+ × + =
114
Two Types of Dynamics Problems
Forward Dynamics:
Given the mechanism, external forces and moments, and the
applied driving force (or torque), find the resulting mechanism
motion and internal joint forces.
Inverse Dynamics:
Given the mechanism, external forces and moments, and the
desired mechanism motion, find the required driving force (or torque)
and internal joint forces.
4Bar Linkage Forward Dynamics:
Given
2
τ and
EXT EXT
M F , , find the motion
4 3 2
, , θ θ θ ,
4 3 2
, , ω ω ω ,
4 3 2
, , α α α and internal forces
ij
F .
4Bar Linkage Inverse Dynamics:
Given the motion
4 3 2
, , θ θ θ ,
4 3 2
, , ω ω ω ,
4 3 2
, , α α α , and
EXT EXT
M F , , find
2
τ and internal forces
ij
F .
Next lecture: Newton's 2
nd
Law and Euler's equation require:
translational: mass center of gravity
rotational: center of gravity mass moment of inertia
m P
G
I
G
115
Mass, Center of Gravity, Mass Moment of Inertia
G
A m F =
∑
α
G
G
I M =
∑
Translational: mass center of gravity
Rotational: center of gravity mass moment of inertia
Mass
In Newton’s 2
nd
Law a m F =
∑
, mass m is the proportionality
constant. Mass is measure of translational inertia – resistance to
change in motion, Newton’s 1
st
Law. Mass is also measure of storage
of translational kinetic energy
2
2
1
mv KE
T
= .
116
Examples for m, CG, I
G
:
System of particles General rigid body
Rectangular rigid body
Mass calculation:
System of particles:
General rigid body:
Rectangular rigid body:
117
Center of Gravity (CG, G)
Also called center of mass, mass center, centroid
CG calculation:
System of particles:
General rigid body:
118
Rectangular rigid body:
Using an XY coordinate frame centered at the geometric center.
/ 2
/ 2
/ 2
/ 2
/ 2
2
/ 2
2 2
2
0
2 4 4
x
x
b
b
b
b
b
b
xdm
X
dm
xdV
m
xthdx
m
th
xdx
m
th x
m
th b b
m
ρ
ρ
ρ
ρ
ρ
−
−
−
=
=
=
=
=
 
= − =

\ .
∫
∫
∫
∫
∫
/ 2
/ 2
/ 2
/ 2
/ 2
2
/ 2
2 2
2
0
2 4 4
y
y
h
h
h
h
h
h
ydm
Y
dm
ydV
m
ytbdy
m
tb
ydy
m
tb y
m
tb h h
m
ρ
ρ
ρ
ρ
ρ
−
−
−
=
=
=
=
=
 
= − =

\ .
∫
∫
∫
∫
∫
For a homogeneous, regular geometric body, the CG is the geometric
center.
119
Mass Moment of Inertia (I
G
) is not the same as Area moment of
inertia (I
G
) for beam bending:
2
Ax
y
I y dA =
∫
2
Ay
x
I x dA =
∫
Units:
4
A
I m ≡
Mass Moment of Inertia (I
G
)
In Euler’s equation
GZ
G
M I α =
∑
, I is the proportionality constant. I
is measure of rotational inertia – resistance to change in motion,
Newton’s 1
st
Law. Also, it is a measure of how hard it is to
accelerate in rotation about certain axes. I is also measure of storage
of rotational kinetic energy
2
1
2
R G
KE I ω = .
Units:
2
G
I kgm ≡ .
120
Mass Moment of Inertia I
G
calculation:
System of particles:
where r
i
is the scalar perpendicular distance from the axis to the i
th
particle. With squaring, all terms will be positive, no there can be no
canceling like for CG. If first moment is balanced, second moment
will be doubled about the CG.
General rigid body:
What is the only term that matters for XY planar motion?
In the example shown above:
XXG YYG ZZG
I I I > > also
ZZG ZZ
I I >
121
Rectangular rigid body:
Using an XY coordinate frame centered at the CG.
( ) ( )
∫ ∫ ∫
− −
+ = + =
2 /
2 /
2 /
2 /
2 2 2 2
b
b
h
h
body
ZZG
tdxdy y x dm y x I ρ
∫
∫
−
−
−


.

\



.

\

−
− +

.

\

−
− =



.

\

+ =
2 /
2 /
3 3
2
2 /
2 /
2 /
2 /
3
2
8 8 3
1
2 2
3
b
b
b
b
h
h
ZZG
dx
h h h h
x t
dx
y
y x t I
ρ
ρ



.

\

+ =


.

\

+ =
−
−
∫
2 /
2 /
3 3
2 /
2 /
3
2
12 3 12
b
b
b
b
ZZG
x h hx
t dx
h
hx t I ρ ρ
( )
2 2
3 3
3 3 3
12 12 12
2 2 12 8 8 3
h b
tbh bh h b
t
b b h b b h
t I
ZZG
+ =


.

\

+ =


.

\


.

\

−
− +


.

\

−
− =
ρ
ρ
ρ
( )
2 2
12
h b
m
I
ZZG
+ = (because tbh V m ρ ρ = = )
122
Units: mass times distance squared,
2
kgm .
Checks with result given in the textbook.
How do we find mass moments of inertia in the realworld?
• look up in tables
• CAD package such as SolidEdge
123
Parallel Axis Theorem
The mass moment of inertia through the CG is related to mass
moments of inertia of parallel axes through different points as
follows:
where d is the scalar distance separating the axis of interest from the
axis through the CG. Notice
ZZG
I is a small as it can get; any
ZZ
I
must be greater, due to the term
2
md , which is always positive.
Parallel axis theorem example:
Rectangular rigid body:
( )
( )
2 2
2 2
2 2 2 2
2 2
2 2
12 4 4
12 4 12 4
3 3
3
ZZ
m b h
I b h m
b b h h
m
b h
m
m
b h
 
= + + +

\ .
 
= + + +

\ .
 
= +

\ .
= +
Agrees with result given in dynamics textbooks.
124
Single Rotating Link Inverse Dynamics
Generic Mechanism Inverse Dynamics Analysis Statement:
Given the mechanism, external forces and moments, and the
desired mechanism motion, find the required driving force (or torque)
and internal joint forces.
Single Rotating Link Inverse Dynamics Analysis
Step 1. Position, Velocity, and Acceleration Analyses must first be
complete.
Step 2. Draw the Diagrams:
Physical Dynamics Diagram:
Free Body Diagram (FBD):
125
Step 3. State the problem:
Step 4. Derive the NewtonEuler Dynamics Equations.
Newton's 2
nd
Law:
Euler's Equation:
Count # of unknowns and # of equations:
126
Step 5. Derive XYZ scalar equations from the vector equations.
Written in matrix form:
127
Step 6. Solve for the unknowns
Actually, we don’t need matrix solution; the first two equations
are decoupled and the solution is straightforward:
Step 7. Calculate Shaking Force and Moment
After the inverse dynamics problem is solved, we can calculate
the vector shaking force and moment, which is the force/moment
reaction on the ground link due to the mechanism, motion, and
external loads:
128
Terms for the inverse dynamics equations
The inverse dynamics problem has been solved analytically for
the single rotating link. Now, how do we calculate the various terms
that appear in the dynamics equations? These all must be derived
from given information.
=
)
`
¹
¹
´
¦
=
GY
GX
G
A
A
A
=
)
`
¹
¹
´
¦
=
EY
EX
E
F
F
F
=
)
`
¹
¹
´
¦
=
Y
X
r
r
r
12
12
12
=
)
`
¹
¹
´
¦
=
EY
EX
E
r
r
r
=
GZ
I
129
Single rotating link inverse dynamics example:
Given: L = 1 m, h = 0.1 m, m = 2 kg, ω = 100 rad/s, α = 0,
F
E
=150 N, 0 =
E
φ (constant relative to horizontal), M
E
=0 Nm.
Calculated terms: 5 . 0
12
= =
E
r r m
2
17 . 0 kgm I
GZ
=
2500
4330
− =
=
Gy
Gx
A
A
2
s
m
Snapshot Analysis (one input angle)
At
D
150 = θ , given this link, motion, and external force,
calculate τ , ,
12 12 Y X
F F and
S S
M F , for this instant (snapshot).
¦
)
¦
`
¹
¦
¹
¦
´
¦
− =
¦
)
¦
`
¹
¦
¹
¦
´
¦
5 . 37
4980
8510
1 433 . 0 250 . 0
0 1 0
0 0 1
12
12
τ
Y
X
F
F
¦
)
¦
`
¹
¦
¹
¦
´
¦
− =
¦
)
¦
`
¹
¦
¹
¦
´
¦
5 . 66
4980
8510
12
12
τ
Y
X
F
F
N, Nm
)
`
¹
¹
´
¦
−
= − = =
4980
8510
12 21
F F F
S
N
k M
S
ˆ
5 . 66 − = − = τ Nm
130
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from inverse dynamics analysis is to
report the unknowns for the entire range of mechanism motion. The
plot below gives the required driving torque τ (Nm, red) for all
D D
360 0 ≤ ≤θ , assuming the given ω is constant, for the same
example from the previous page. This shows the torque that must be
supplied by an external DC servomotor to cause the specified motion.
Also plotted is the average torque (green) τ
AVG
= 0 and the root
meansquare (RMS) torque value (blue) τ
RMS
= 106.1 Nm.
0 50 100 150 200 250 300 350
150
100
50
0
50
100
150
θ
(deg)
τ
(
N
m
)
Tau (red) with average (green) and rms (blue)
131
The plots below give the Shaking Force
S
F and CG translational
acceleration results, respectively, for all
D D
360 0 ≤ ≤θ . In both plots,
the X components are red and the Y green.
0 50 100 150 200 250 300 350
1
0.5
0
0.5
1
x 10
4
θ
(deg)
F
S
X
a
n
d
F
S
Y
(
N
)
Shaking Force, X (red) and Y (green)
132
0 50 100 150 200 250 300 350
5000
0
5000
θ
(deg)
A
G
X
a
n
d
A
G
Y
(
m
/
s
2
)
CG Acceleration, X (red) and Y (green)
The Shaking Moment
S
M is merely the negative of the driving
torque τ plot shown previously and hence is not shown separately. Is
the static loading (mg) significant?
133
FourBar Mechanism Inverse Dynamics
Generic Mechanism Inverse Dynamics Analysis Statement:
Given the mechanism, external forces and moments, and the
desired mechanism motion, find the required driving force (or torque)
and internal joint forces.
FourBar Mechanism Inverse Dynamics Analysis
First, can we simplify and solve the problem linkbylink, like the
single rotating link? Count # of unknowns and # of equations:
Step 1. Position, Velocity, and Acceleration Analyses must first be
complete.
Step 2. Draw the Diagrams:
Physical Dynamics Diagram:
134
Free Body Diagrams (FBDs):
ij
F :
ij
r :
Step 3. State the problem:
135
Step 4. Derive the NewtonEuler Dynamics Equations.
Newton's 2
nd
Law:
Euler's Equation:
Count # of unknowns and # of equations:
136
Step 5. Derive XYZ scalar equations from the vector equations.
137
Write these equations in matrix/vector form:
21
21
12 12 32 32 32
32
43
23 23 43 43 43
14
14
34 34 14 14 2
1 0 1 0 0 0 0 0 0
0 1 0 1 0 0 0 0 0
0 0 0 0 1
0 0 1 0 1 0 0 0 0
0 0 0 1 0 1 0 0 0
0 0 0 0 0
0 0 0 0 1 0 1 0 0
0 0 0 0 0 1 0 1 0
0 0 0 0 0
X
Y
Y X Y X X
Y
X
Y X Y X Y
X
Y
Y X Y X
F
F
r r r r F
F
F
r r r r F
F
F
r r r r τ
− ¦
¦
−
¦
¦ − −
¦
−
¦
¦
−
´
− −
−
−
− −
¹
( )
( )
( )
2 2
2 2
2 2
3 3 3
3 3 3
3 3 3 3 3 3 3
4 4 4
4 4 4
4 4 4 4 4 4 4
G X
G Y
G Z
G X E x
G Y E y
G Z E x E y E y E x E
G X E x
G Y E y
G Z E x E y E y E x E
m A
m A g
I
m A F
m A g F
I r F r F M
m A F
m A g F
I r F r F M
α
α
α
¦ ¹ ¹
¦ ¦ ¦
+
¦ ¦ ¦
¦ ¦ ¦
¦ ¦ ¦
−
¦ ¦ ¦
¦ ¦ ¦
+ − =
` ´ `
¦ ¦ ¦ ¦
− + −
¦ ¦ ¦ ¦
−
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
+ −
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦ − + −
) ¹ )
 { } { } b v A =
Coefficient matrix [A] dependent on geometry (kinematics solutions).
RHS {b} dependent on inertial terms, gravity, and given external
forces and moments.
138
Step 6. Solve for the unknowns
Simultaneous matrix solution: { }   { } b A v
1 −
=
Actually, using Gaussian elimination is more efficient and robust.
Solution to internal forces and input torque are contained in the
components of {v}.
Step 7. Calculate Shaking Force and Moment
After the basic inverse dynamics problem is solved, we can
calculate the vector shaking force and moment, which is the
force/moment reaction on the ground link due to the motion:
139
Details for the general fourbar mechanism model
The inverse dynamics problem has been derived analytically for
the fourbar mechanism. Now, how do we calculate the various
terms that appear in the dynamics equations? These all must be
derived from given information. See Fig. P112. Let us do link 3
terms (next page). Here is the general link 3 diagram for these
derivations:
140
Link 3 details:
=
)
`
¹
¹
´
¦
=
Y
X
r
r
r
23
23
23
=
)
`
¹
¹
´
¦
=
Y
X
r
r
r
43
43
43
=
)
`
¹
¹
´
¦
=
Y G
X G
G
A
A
A
3
3
3
=
)
`
¹
¹
´
¦
=
Y
X
P
P
P
F
F
F
3
3
3
=
)
`
¹
¹
´
¦
=
Y
X
P
P
P
R
R
R
3
3
3
3 E
M = given
141
Figure for Term Example 1 Inverse Dynamics Example starting on
the next page:
The coupler link 3 is a rectangle of dimensions 8” x 6” x 0.5”.
The triangle tip we have been using all along in Term Example 1 is
actually the CG; of the actual rectangular link for inverse dynamics.
0.1 0 0.1 0.2 0.3
0.1
0.05
0
0.05
0.1
0.15
0.2
0.25
0.3
X (m)
Y
(
m
)
142
Fourbar mechanism inverse dynamics example:
This is the mechanism from Term Example 1 (open branch),
with one crucial difference: the input angular velocity was too low
for interesting dynamics, so I changed it from
2
ω π = to
2
20 ω =
rad/s.
Given r
1
= 0.284, r
2
= 0.076, r
3
= 0.203, r
4
= 0.178, R
G3
= 0.127
m, and
1
10.3 θ =
D
,
2
30 θ =
D
,
3
53.8 θ =
D
,
4
121.7 θ =
D
,
3
36.9 δ =
D
;
2
20 ω = ,
3
8.09 ω = − ,
4
3.73 ω = − rad/s;
2
0 α = ,
3
8.65 α = ,
4
244.4 α =
rad/s
2
. This is the open branch of the position, velocity, and
acceleration example (Term Example 1).
All moving links are wood, with mass density 0.03 ρ = lb
m
/in
3
.
Links 2 and 4 have rectangular dimensions 0.75 by 0.50 by r
i
(in);
link 3 has rectangular dimensions 8 by 6 by 0.5 (in), as shown on the
previous page. The calculated inertia parameters are
2
0.015 m = ,
3
0.327 m = ,
4
0.036 m = kg and
6
2
7.9 10
G Z
I
−
= × ,
3
3
1.8 10
G Z
I
−
= × ,
5
4
9.5 10
G Z
I
−
= × kgm
2
. All external forces and moments are zero but
gravity is included.
Snapshot Analysis (one input angle)
At
2
30 θ =
D
, given this mechanism and motion, calculate the
four vector internal joint forces, the driving torque
2
τ , and the
shaking force and moment ,
S S
F M for this instant (snapshot).
143
¦
¦
¦
¦
¦
¦
)
¦
¦
¦
¦
¦
¦
`
¹
¦
¦
¦
¦
¦
¦
¹
¦
¦
¦
¦
¦
¦
´
¦
−
−
−
−
−
=
¦
¦
¦
¦
¦
¦
)
¦
¦
¦
¦
¦
¦
`
¹
¦
¦
¦
¦
¦
¦
¹
¦
¦
¦
¦
¦
¦
´
¦
−
−
− − −
−
−
− −
−
−
0233 . 0
095 . 0
638 . 0
015 . 0
497 . 4
955 . 8
0
034 . 0
202 . 0
0 047 . 0 076 . 0 047 . 0 076 . 0 0 0 0 0
0 1 0 1 0 0 0 0 0
0 0 1 0 1 0 0 0 0
0 0 0 122 . 0 037 . 0 002 . 0 127 . 0 0 0
0 0 0 1 0 1 0 0 0
0 0 0 0 1 0 1 0 0
1 0 0 0 0 033 . 0 019 . 0 033 . 0 019 . 0
0 0 0 0 0 1 0 1 0
0 0 0 0 0 0 1 0 1
2
14
14
43
43
32
32
21
21
τ
y
x
y
x
y
x
y
x
F
F
F
F
F
F
F
F
Solution by Gaussian elimination or: { }   { } b A v
1 −
=
Snapshot Answer:
{ }
¦
¦
¦
¦
¦
¦
)
¦
¦
¦
¦
¦
¦
`
¹
¦
¦
¦
¦
¦
¦
¹
¦
¦
¦
¦
¦
¦
´
¦
−
−
− =
¦
¦
¦
¦
¦
¦
)
¦
¦
¦
¦
¦
¦
`
¹
¦
¦
¦
¦
¦
¦
¹
¦
¦
¦
¦
¦
¦
´
¦
=
43 . 0
52 . 5
60 . 3
61 . 5
96 . 2
11 . 10
99 . 5
08 . 10
20 . 6
2
14
14
43
43
32
32
21
21
τ
y
x
y
x
y
x
y
x
F
F
F
F
F
F
F
F
v N, Nm
)
`
¹
¹
´
¦
=
56 . 4
80 . 9
S
F N k M
S
ˆ
68 . 1 − = Nm
144
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from inverse dynamics analysis is to
report the unknowns for the entire range of mechanism motion. The
plot below gives the required driving torque
2
τ (Nm) for all
D D
360 0
2
≤ ≤θ , for the Term Example 1 mechanism, assuming the
given 20
2
= ω rad/s is constant (Remember: this has been changed
from the π ω =
2
rad/s in the kinematics examples!). This plot shows
the torque (red) that must be supplied in all configurations by an
external DC servomotor to cause the specified motion. Also plotted
is the average torque (green) τ
2AVG
= 0 and the rootmeansquare
torque value (blue) τ
2RMS
= 0.36 Nm.
0 50 100 150 200 250 300 350
0.4
0.2
0
0.2
0.4
0.6
0.8
θ
2
(deg)
τ
2
(
N
m
)
Tau2 (red) with average (green) and RMS (blue)
145
The plots below give the shaking force
S
F (N) and shaking
moment
S
M (Nm) results, respectively, for all
D D
360 0
2
≤ ≤θ . In
the force plot, the X component is red and the Y green.
0 50 100 150 200 250 300 350
10
5
0
5
10
θ
2
(deg)
F
S
X
a
n
d
F
S
Y
(
N
)
Shaking Force, X (red) and Y (green)
146
In the shaking moment plot, there is only the Z component:
0 50 100 150 200 250 300 350
2
1.5
1
0.5
0
0.5
1
1.5
2
θ
2
(deg)
M
S
(
N
m
)
Shaking Moment
147
SliderCrank Mechanism Inverse Dynamics
This problem is very similar to the fourbar mechanism inverse
dynamics problem. In fact, links 2 and 3 are handled identically!
Step 1. Position, Velocity, and Acceleration Analyses must first be
complete.
Step 2. Draw the Diagrams:
Physical Dynamics Diagram:
Free Body Diagrams (FBDs):
ij
F : internal force of link i acting on link j
ij
r : moment arm pointing to link i from the CG of link j
148
Step 3. State the problem:
Step 4. Derive the NewtonEuler Dynamics Equations.
Again, links 2 and 3 are identical so let us focus on link 4, the
slider. There are two kinematic constraints on the slider:
Newton's 2
nd
Law:
Euler's Equation:
Count # of unknowns and # of equations: We need an additional
equation:
149
Step 5. Derive XYZ scalar equations from the vector equations and
beam these equations into matrix/vector form. Substitute the friction
constraint to eliminate one unknown (F
14X
); also eliminate one
equation (
4
4 4
G
G Z
M I α =
∑
).
( )
2 2 21
2 2 21
2 32 12 12 32 32
32
43
43 23 23 43 43
14
2
1 0 1 0 0 0 0 0
0 1 0 1 0 0 0 0
0 0 0 1
0 0 1 0 1 0 0 0
0 0 0 1 0 1 0 0
0 0 0 0
0 0 0 0 1 0 0
0 0 0 0 0 1 1 0
G X X
G Y Y
G Z X Y X Y X
Y
X
Y Y X Y X
Y
m A F
m A g F
I F r r r r
F
F
F r r r r
F u
τ
− ¦ ¹
¦ ¦
+ −
¦ ¦
¦ ¦ − −
¦ ¦
−
¦ ¦
=
´ `
−
¦ ¦
¦ ¦
− −
¦ ¦
− ±
¦ ¦
¦ ¦
−
¹ )
( )
2
3 3 3
3 3 3
3 3 3 3 3 3 3
4 4 4
4 4
G X E x
G Y E y
G Z E x E y E y E x E
G X E x
E y
m A F
m A g F
I r F r F M
m A F
m g F
α
α
¦ ¹
¦ ¦
¦ ¦
¦ ¦
¦ ¦
−
¦ ¦
´ `
+ −
¦ ¦
¦ ¦
− + −
¦ ¦
−
¦ ¦
¦ ¦
−
¹ )
 { } { } b v A =
Coefficient matrix [A] dependent on geometry (kinematics solutions).
Always choose proper sign of u to be opposite to the current x
direction. RHS {b} dependent on inertial and statics terms.
Step 6. Solve for the unknowns
Simultaneous matrix solution:
{ }   { } b A v
1 −
=
Actually, using Gaussian elimination is more efficient and robust.
Solution to internal forces and input torque contained in the
components of {v}.
150
Step 7. Calculate Shaking Force and Moment
After the basic inverse dynamics problem is solved, we can
calculate the vector shaking force and moment, which is the
force/moment reaction on the ground link due to the motion.
Figure for example starting on the next page: The slidercrank
mechanism is shown at the starting (and ending) position, with zero
input angle θ
2
.
0.1 0 0.1 0.2 0.3
0.25
0.2
0.15
0.1
0.05
0
0.05
0.1
0.15
0.2
X (m)
Y
(
m
)
151
Slidercrank mechanism inverse dynamics example:
This is the mechanism from Term Example 2 (right branch
only), in this case keeping the low input angular velocity
2
2 ω π =
rad/s so the previous snapshots and fullrangeofmotion results still
apply.
Given r
2
= 0.102, r
3
= 0.203, h = 0.076 m, and
2
30 θ =
D
,
3
7.2 θ =
D
, x = 0.290 m; and
2
/ 2 ω π = ,
3
0.686 ω = − rad/s, 0.062 x = −
m/s;
2
0 α = ,
3
0.681 α = rad/s
2
, 0.329 x = − m/s
2
. This is the right
branch of the position, velocity, and acceleration example (Term
Example 2).
All moving links are wood, with mass density 03 . 0 = ρ (lb
m
/in
3
).
Links 2 and 3 have rectangular dimensions 0.75 by 0.50 by r
i
(in);
link 4 has rectangular dimensions 0.75 by 0.50 by 3 (in). The
calculated inertia parameters are m
2
= 0.020, m
3
= 0.041, m
4
= 0.015,
(kg) and I
G2Z
= 1.819e005, I
G3Z
= 1.418e004 (kgm
2
). There is a
constant external force of 1 N acting at the center of the piston,
directed horizontally to the left; gravity is included but all other
external forces and moments are zero. We assume u = 0.2
(coefficient of friction between piston and wall);
Snapshot Analysis (one input angle)
At
2
30 θ =
D
, given this mechanism and motion, calculate the
four vector internal joint forces, the driving torque
2
τ , and the
shaking force and moment ,
S S
F M for this instant (snapshot).
152
21
21
32
32
43
43
14
2
1 0 1 0 0 0 0 0 0.002
0 1 0 1 0 0 0 0 0.1
0.025 0.044 0.025 0.044 0 0 0 1
0 0 1 0 1 0 0 0
0 0 0 1 0 1 0 0
0 0 0.013 0.101 0.013 0.101 0 0
0 0 0 0 1 0 0.2 0
0 0 0 0 0 1 1 0
X
Y
X
Y
X
Y
Y
F
F
F
F
F
F
F
τ
− ¦ ¹
¦ ¦
−
¦ ¦
¦ ¦
¦ ¦
−
¦ ¦
=
´ `
−
¦ ¦
¦ ¦
¦ ¦
−
¦ ¦
¦ ¦
−
¹ )
99
0
0.011
0.398
0.0001
0.995
0.150
¦ ¹
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
´ `
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¦ ¦
¹ )
Solution by Gaussian elimination or: { }   { } b A v
1 −
=
Snapshot Answer:
{ }
21
21
32
32
43
43
14
2
0.935
0.517
0.938
0.318
0.949
0.081
0.231
0.011
X
Y
X
Y
X
Y
Y
F
F
F
F
v
F
F
F
τ
¦ ¹ ¦ ¹
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
= =
´ ` ´ `
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¦ ¦ ¦ ¦
¹ ) ¹ )
(N, Nm)
0.982
0.748
S
F
− ¦ ¹
=
´ `
−
¹ )
(N)
ˆ
0.053
S
M k = − (Nm)
153
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from inverse dynamics analysis is to
report the unknowns for the entire range of mechanism motion. The
plot below gives the required driving torque
2
τ (Nm) for all
D D
360 0
2
≤ ≤θ , for the Term Example 2 slidercrank mechanism,
right branch only, assuming the given
2
2 ω π = rad/s is constant.
This plot shows the torque (red) that must be supplied in all
configurations by an external DC servomotor to cause the specified
motion. Also plotted is the average torque (green) τ
2AVG
= 0.002 and
the rootmeansquare torque value (blue) τ
2RMS
= 0.086 Nm.
0 50 100 150 200 250 300 350
0.1
0.05
0
0.05
0.1
0.15
0.2
θ
2
(deg)
τ
2
(
N
m
)
Actuator Torque
154
The plots below give the shaking force
S
F (N) and shaking
moment
S
M (Nm) results, respectively, for all
D D
360 0
2
≤ ≤θ . In
the force plot, the X component is red and the Y green.
0 50 100 150 200 250 300 350
1.05
1
0.95
0.9
0.85
0.8
0.75
0.7
θ
2
(deg)
F
S
X
a
n
d
F
S
Y
(
N
)
Shaking Force, X (red) and Y (green)
155
In the shaking moment plot, there is only the Z component:
0 50 100 150 200 250 300 350
0.06
0.04
0.02
0
0.02
0.04
0.06
0.08
θ
2
(deg)
M
S
(
N
m
)
Shaking Moment
156
Cam Introduction
Chapter 8
Applications
Compared to linkages, easier to design desired motion with cams, but
much more expensive and difficult to produce.
Cam Classification: Disk cams with followers
157
Degrees of Freedom Recall a cam joint has twodof; allows both
rolling and sliding.
Function Generation
The output parameter is a continuous function of the input
parameter. With linkages, we can only satisfy a function exactly at a
finite number of points: 3, 4, or 5, usually. For example, a 4bar
linkage:
( )
2 4
θ θ f =
With a cam and follower mechanism, we can satisfy function
generation at infinite points.
( ) θ f S = ( ) θ φ f =
Cam input angle is θ , output is S for reciprocating (translating) and
φ (rotating) for oscillating follower.
158
Cam Motion Profiles
Up to this point, we have been mostly concerned with
mechanism analysis: given a mechanism design and its input
parameters, determine the position, velocity, acceleration, and
dynamics behavior. With cams we must consider mechanism
synthesis for the first time: given the motion requirements (follower
motion and timing with input angle), design the cam. The first step is
to determine a “smooth” cam follower motion profile. Classification:
When the motion transitions between different motion functions, we
must ensure “smooth” motion.
159
Fundamental Law of Cam Design:
Which means:
If the Fundamental law of Cam Design is satisfied, the resulting
dynamic performance will be acceptable for highspeed cam/follower
operation. If not, there will be performance degradation due to
noise, vibrations, high wear, etc. Cyclical impulse hammering when
acceleration is not continuous.
160
S V A J Diagrams
In synthesis, we are only given total motion range and perhaps
some timing requirements. It is the engineer’s job to determine the
position curves and to match the velocity and acceleration across
junctions. Position is automatically matched by shifting axes. Draw
S V A J diagrams vs. time to graphically see if the Fundamental Law
of Cam Design is satisfied for candidate curves. We can plot vs. time
or vs. input cam angle θ (assuming constant angular velocity,
t ω θ = ).
Check out Examples 81 (terrible)
82 (bad)
83 (acceptable)
Slope of a function is the value of its derivative at a point. Therefore,
for continuous velocity and acceleration curves, the slopes of the
position and velocity curves must match across all junctions. The
slope of the acceleration can be discontinuous (leading to finite jumps
in jerk), but the acceleration itself must be continuous.
161
Generic Cam Follower Motion Profile figure:
Define each separate function so the value is zero at the initial angle,
which is zero. Then to put the whole thing together, just shift the θ
and S axes.
Match S: easy, just do it  shift S axes.
Match V: slope of S must match across junctions.
Match A: slope of V must match across junctions.
162
Cam Follower Motion Profile Examples Example 1
rise – dwell portion. Specify Parabolic (constant acceleration) to
Straight Line (constant velocity) rise, followed by a dwell.
S: ( )
2
1 0 1 1
2
1
θ θ A f = ( )
2 0 2 2
θ θ V f = ( ) 0
3 3
= θ f
V:
A:
J:
Match S at junction B: Just shift axis up.
Match V at junction B:
Try to match A at junction B:
Plot on next page.
163
Example 1 Plots
164
Cam Follower Motion Profile Examples Example 2
Fix rise portion only. Specify HalfCycloidal function
(sinusoidal in cam angle) to Straight Line (constant velocity) rise.
S: ( )


.

\

− =
1
1
1
1
1 1 1
sin
1
β
πθ
π β
θ
θ L f ( )
2 0 2 2
θ θ V f =
V:
A:
J:
Match S at junction B: Just shift axis up.
Match V at junction B:
Match A at junction B:
Plot on next page.
165
0 20 40 60 80
0
1
2
S
(
m
)
Half Cycloid
0 20 40 60 80
0
0.05
V
(
m
/
d
e
g
)
0 20 40 60 80
0
5
x 10
4
A
(
m
/
d
e
g
2
)
0 20 40 60 80
2
0
2
x 10
5
θ (deg)
J
(
m
/
d
e
g
3
)
166
Cam Follower Motion Profile Examples Example 3
Specify FullCycloidal function (sinusoidal in cam angle). This
will rise all the way to meet a dwell smoothly; it satisfies the
Fundamental Law of Cam Design.
S:
( )
1 1
1 1 1
1 1
1 2
sin
2
f L
θ πθ
θ
β π β
 
= −

\ .
( )
2 2
0 f θ =
V:
( )
1 1
1 1
1 1
2
1 cos
L
v
πθ
θ
β β
 
= −

\ .
( )
2 2
0 v θ =
A:
( )
1 1
1 1
2
1 1
2 2
sin
L
a
π πθ
θ
β β
 
=

\ .
( )
2 2
0 a θ =
J:
( )
2
1 1
1 1
3
1 1
4 2
cos
L
j
π πθ
θ
β β
 
=

\ .
( )
2 2
0 j θ =
Plot on next page.
167
0 20 40 60 80
0
1
2
S
(
m
)
Full Cycloid
0 20 40 60 80
0
0.05
V
(
m
/
d
e
g
)
0 20 40 60 80
2
0
2
x 10
3
A
(
m
/
d
e
g
2
)
0 20 40 60 80
2
0
2
x 10
4
θ (deg)
J
(
m
/
d
e
g
3
)
168
Analytical Cam Synthesis
Disk Cam with Radial FlatFaced Follower
Assume a valid cam motion profile has been designed according
to the Fundamental Law of Cam Design; i.e. we now have continuous
S, V, A curves. Given the motion profile, determine the cam contour.
Is it as simple as polar plotting of ( ) θ f S = vs. cam angle θ ?
We will use kinematic inversion to simplify the synthesis.
DCRFFF Figure:
169
As seen in the figure, the radius R out to the flatfaced follower (not
to the point of contact (x,y)) is:
where C is the minimum cam radius, a design variable, and ( ) θ f S =
is the given motion profile. The radius R and the flatface length L
can be related to the contact point x,y and the cam angle through
geometry:
Notice that:
To calculate the follower flatface length, double the maximum of L
from above. Doubled because by symmetry the contact point will
change to the other side at
D
180 = θ .
170
To summarize thus far:
This is sufficient to manufacture the cam; it is machined with θ, R, L
coordinates. If we want to know the cam contour in Cartesian
coordinates, we must solve the relationships for x, y. In matrix form:
The coefficient matrix [A] is orthonormal, which means    
T
A A =
−1
.
The solution is:
171
Minimum Cam radius to Avoid Cusps
A cusp is a point in the cam, or actually undercut; this is to be
avoided for good motion. The condition is that for a finite θ ∆ , there
is no change in x, y:
0 = =
θ θ d
dy
d
dx
will cause a cusp.
( ) ( )
( ) ( ) θ
θ
θ
θ
θ
θ
θ θ
θ
θ
θ
θ
θ
θ
θ
θ θ
θ
cos sin sin cos
sin cos cos sin
2
2
2
2
d
f d
d
df
d
df
f C
d
dy
d
f d
d
df
d
df
f C
d
dx
+ − + + =
− − + + − =
( )
( ) θ
θ
θ
θ
θ
θ
θ
θ
cos
sin
2
2
2
2


.

\

+ + =


.

\

+ + − =
d
f d
f C
d
dy
d
f d
f C
d
dx
0 = =
θ θ d
dy
d
dx
simultaneously only when:
( ) 0
2
2
= + +
θ
θ
d
f d
f C
Therefore, to avoid cusps on the entire cam contour,
( ) 0
2
2
> + +
θ
θ
d
f d
f C
Note C is always positive and ( ) θ f starts and ends at zero and never
goes negative.
172
Disk Cam with Radial FlatFaced Follower Design Example
Specify a fullcycloidal rise (total lift 50 mm), followed by a
high dwell, a fullcycloidal return (total fall 50 mm), and then a low
dwell. Each of these four motion steps occurs for 90 deg of cam
shaft rotation.
The cam motion profile associated with this specification is
shown below. Clearly, this satisfies the Fundamental Law of Cam
Design because the position, velocity, and acceleration curves are
continuous. The jerk is not continuous, but it remains finite over all
cam angles.
0 50 100 150 200 250 300 350
0
50
S
0 50 100 150 200 250 300 350
50
0
50
V
0 50 100 150 200 250 300 350
100
0
100
A
0 50 100 150 200 250 300 350
500
0
500
θ
J
173
Choosing a minimum cam radius of C = 100 mm), the resulting
cam contour is shown below.
150 100 50 0 50 100
100
50
0
50
100
150
X (mm)
Y
(
m
m
)
174
Let us check the cusp avoidance plot. To avoid cusps in this cam,
we require that:
( ) ( ) ( ) 0
2
2
> + + = + +
θ
θ θ θ
d
f d
f C A S C
As seen in the plot below, this inequality is satisfied for the entire
range of motion, so this cam design is acceptable.
0 50 100 150 200 250 300 350
0
50
100
150
200
250
θ
C
+
S
+
A
175
Gear Introduction
Transfer motion between rotating shafts in machinery, vehicles,
toys, etc. Gears used in electromechanical systems. Change in
angular velocity, torque, direction. Can openers to Aircraft carriers.
Related mechanisms  belt and chain drives.
Applications:
Gear Classification
External Spur Gears Internal Spur Gears
176
Rack & Pinion Bevel gears
Helical (Parallel Shaft) Helical (Crossed Shaft)
Herringbone Gears Gear Train
177
Automotive Differential Planetary Gear Train
Worm and Gear Harmonic Gearing
178
Harmonic Gearing
Taken from: http://www.roymech.co.uk/
“The harmonic gear allows high reduction ratios with concentric
shafts and with very low backlash and vibration. It is based on a very
simple construction utilising metals elastomechanical property.”
“Harmonic drive transmissions are noted for their ability to
reduce backlash in a motion control system. How they work is
through the use of a thinwalled flexible cup with external splines on
it lip, placed inside a circular thickwalled rigid ring machined with
internal splines. The external flexible spline has two fewer teeth
than the internal circular spline. An elliptical cam enclosed in an
antifriction ball bearing assembly is mounted inside the flexible cup
and forces the flexible cup splines to push deeply into the rigid ring at
two opposite points while rotating. The two contact points rotate at a
speed governed be the difference in the number of teeth on the two
splines This method basically preloads the teeth, which reduces
backlash.”
179
Gear Ratio
Common electric motors have high speed but low torque. A
robot joint needs lower rotation speed but high torque. A gear train
can accomplish both objectives – reduce speed and increase torque.
Gear ratio is a measure of the degree of reduction and increase.
Simple spur gear pair:
DOF:
A gear joint is like a cam joint; twodof, teeth in contact allow rolling
and sliding.
Gear 1 is input, gear 2 is output. Like two cylinders rolling without
sliding. Arc lengths are equal.
180
Define gear ratio n:
Radii inversely proportional to angular motion. For standard spur
gears, the radii are directly proportional to the number of teeth:
For relating angular velocities, tangential velocities are equal.
Most gear applications have constant angular velocities, for
accelerating up to (or down from) constant angular velocities:
181
For relating shaft torques, interface forces are equal.
Radii directly proportional to shaft torques
Summary:
The ratio of the number of teeth is directly proportional to the radii,
diameter, and shaft torques. The ratio of the number of teeth is
inverse proportional to the shaft angles, angular velocities, and
angular accelerations.
182
If 1 > n :
1 2
ω ω < Output has reduced speed
1 2
τ τ > Output has increased torque
This is the electric motor / robot joint case.
If 1 < n :
1 2
ω ω > Output has increased speed
1 2
τ τ < Output has reduced torque
If 1 = n :
1 2
ω ω = Output speed and torque unchanged
1 2
τ τ = direction reverses (external spur gears)
183
Gear ratio examples 1. Gear toy
2. Mountain Bike Transmission
Gear Ratios:
F
R
R
F
F
R
IN
OUT
N
N
N
N
n
τ
τ
ω
ω
= = = =
Schwinn Front
Sierra 48 38 28
14 0.29 0.37 0.50
16 0.33 0.42 0.57
Rear 18 0.38 0.47 0.64
22 0.46 0.58 0.78
26 0.54 0.68 0.93
30 0.62 0.79 1.07
Unlike electric motor example, mountain bike gearing generally:
• increases angular velocity
• decreases torque
184
Cannondale
Front
M400 44 32 22
11 0.25 0.34 0.50
12 0.27 0.38 0.54
Rear 14 0.32 0.44 0.64
16 0.36 0.50 0.73
18 0.41 0.56 0.82
21 0.48 0.66 0.95
24 0.54 0.75 1.09
28 0.64 0.88 1.27
32 0.73 1.00 1.45
BikeE
Front 34 Rear hub
recumbent 1.2913:1 1:1 0.7:1
11 0.25 0.32 0.46
13 0.30 0.38 0.55
Rear 15 0.34 0.44 0.63
18 0.41 0.53 0.76
21 0.48 0.62 0.88
24 0.55 0.70 1.01
28 0.64 0.82 1.18
But considering the difference in wheel sizes (26” Cannondale, 20”
BikeE), the effective BikeE high and low gear ratios are:
Stiff: 0.32 instead of 0.25
Granny 1.53 instead of 1.18
Original front was 46 teeth – changed for more granny gear.
185
Gear Trains and Gear Standardization
Simple Gear Trains
Mesh any number of spur gears. Leftmost is driving gear.
Rightmost is the output gear. All intermediate gears are first the
driven gear and then the driving gear as we proceed from left to right.
Let us calculate the overall gear ratio.
OUT
IN
GT
n
ω
ω
= Example:
We can find the overall gear ratio by canceling neighboring
intermediate angular velocities:
Each term in the above product may be replaced by its known
number of teeth ratio:
186
All intermediate ratios cancel, so:
We could have done the same with pitch radii instead of number of
teeth because they are in direct proportion:
So, the intermediate gears are idlers. Number of teeth effect cancels
out, but do change direction! We should have included sign:
So, for external spur gear trains:
Odd # of gears: Output same direction as input
Even # of gears: Output opposite direction as input
187
Different case:
Mesh any number of spur gears, where the driving and driven
gears are distinct, because each pair is rigidly attached to the same
shaft. See the figure. Again, let us calculate the overall gear ratio.
OUT
IN
GT
n
ω
ω
= Example:
Again, we use the equation:
But now the gears rigidly attached to the same shaft have the same
angular velocity ratio, so:
188
General formula:
Again, we must consider direction:
So, for external spur gear trains:
Odd # of pairs: Output opposite direction as input
Even # of pairs: Output same direction as input
189
Involute Spur Gear Details and Standardization
Rolling Cylinders
Mating spur gears are based on two pitch circles rolling without
slip. These are fictitious circles; you cannot look on a gear to see
them. The actual gear teeth both roll and slide (twodof joint).
Fundamental Law of Gearing:
From our study of linkage velocity, we know this is no easy feat.
Velocity ratios in a linkage vary wildly over the range of motion.
Velocity Ratio
Torque Ratio (Mechanical Advantage)
The author’s velocity ratio is the inverse of our gear ratio
definition and his torque ratio is the same as our gear ratio.
190
Involute Function
Standard spur gears have an involute tooth shape. If the gears’
center distance is not perfect (tolerances, thermal expansion, wear,
etc.), the angular velocity ratio will still be constant to satisfy the
Fundamental Law of Gearing. The involute is a curve generated by
unwrapping a taut sting from a circle:
191
Base Circle: Involute starts from this circle
Pitch Circle: Fictitious circle, pure rolling in contact
Pitch Point: Contact point between the two pitch circles
Pressure Angle: Angle between the common normal (also
called axis of transmission) of the two meshing teeth and the velocity
of the pitch point (tangent to both pitch circles). Point of contact
slides along this line. Similar angle is defined for cams and
followers.
Base circle, pitch circle, pressure angle relationship:
192
Length of contact along axis of transmission. Beginning of
contact is when tip of driven gear tooth intersects the axis of
transmission. End of contact is when tip of driving gear (pinion) tooth
intersects the axis of transmission. Only one or two teeth are in
contact at any one time. For harmonic gearing, many teeth are in
contact at any one time (higher gear ratio in a smaller package).
193
Increasing center distance increases the pressure angle, increases
the pitch circle radii, but doesn’t change the base circles (of course).
Thanks to the involute tooth shape, does not affect angular velocity
ratio.
How is this possible? Relationship from last page:
194
Backlash
Clearance. Distance between mating teeth measured along the
pitch circle circumference. All real gears must have some backlash
due to tolerances, thermal expansion, wear, etc. However, must
minimize backlash for smooth operation. Example: robot joints
which must be driven both directions. Changing direction, nothing
happens until the backlash is moved, and then impact  bad for
dynamics. Nonlinear effect in robots. On earth gravity tends to load
the backlash for predictable effects. In space however, the backlash
is less predictable! Figure:
195
Gear Standardization
To allow interchangeability in manufacturing and to allow
meshing of different size gears (radii and number of teeth) to achieve
desired gear ratios. For two spur gears to mesh, must have: 1) the
same pressure angle; 2) same diametral pitch; and 3) be made with
standard tooth proportions (Table 91, p. 441).
Diametral Pitch:
Module:
Module is the metric version of diametral pitch. Not
interchangeable with US gears because different tooth proportion
standards!
Circular Pitch:
196
Standard involute tooth proportions, see Table 91, p. 441.
Addendum is radial distance from pitch circle to Top Land of tooth.
Dedendum is radial distance from pitch circle to Bottom Land of
tooth (not to base circle). Clearance is radial distance from Bottom
Land to mating gear Top Land (kinda like radial backlash). Face
width is thickness of tooth and gear (mating widths needn’t be the
same). Tooth thickness is the circumferential length of each tooth..
Related to the circular pitch and backlash by:
2
Table of Contents
INTRODUCTION ............................................................................................ 3 MATRIXVECTOR INTRODUCTION ........................................................ 11 MATLAB INTRODUCTION ........................................................................ 25 MOBILITY..................................................................................................... 29 FOURBAR MECHANISM POSITION ANALYSIS................................... 34 GRASHOF’S LAW........................................................................................ 49 SLIDERCRANK MECHANISM POSITION ANALYSIS.......................... 53 VELOCITY ANALYSIS INTRODUCTION ................................................ 62 FOURBAR MECHANISM VELOCITY ANALYSIS ................................. 67 SLIDERCRANK MECHANISM VELOCITY ANALYSIS ........................ 78 ACCELERATION ANALYSIS INTRODUCTION...................................... 85 FOURBAR MECHANISM ACCELERATION ANALYSIS ...................... 91 SLIDERCRANK MECHANISM ACCELERATION ANALYSIS ........... 100 INPUT MOTION SPECIFICATION ........................................................... 106 DYNAMICS INTRODUCTION.................................................................. 110 MASS, CENTER OF GRAVITY, MASS MOMENT OF INERTIA .......... 115 SINGLE ROTATING LINK INVERSE DYNAMICS ................................ 124 FOURBAR MECHANISM INVERSE DYNAMICS................................. 133 SLIDERCRANK MECHANISM INVERSE DYNAMICS........................ 147 CAM INTRODUCTION .............................................................................. 156 ANALYTICAL CAM SYNTHESIS ............................................................ 168 GEAR INTRODUCTION ............................................................................ 175 GEAR TRAINS AND GEAR STANDARDIZATION................................ 185
3
ME 301 Kinematics & Dynamics of Machines Introduction Kinematics:
Kinema  Greek for motion Dynamics: Rigid Body Mechanics Diagram:
Required Math: Geometry, trigonometry, vectors, matrices, calculus Mechanisms: linkages, cams, gears, gear trains Analysis vs. Synthesis • Analysis – determination of position, velocity, acceleration, etc. for a given mechanism • Synthesis – design of mechanism to do a specific job
screw Spherical . no motion • Mechanism – 1 dof device with rigid links connected with joints • Machine – collection of mechanisms to transmit force (input / output) • Robot – an electromechanical device having greater than 1 dof. Motion – Translation and Rotation Planar – all motion is 2D (projected onto a common plane) Helical . programmable for a variety of tasks.number of degreesoffreedom (dof): • Structure – static.3 independent translations and rotations . all points in a body moves about a fixed point Spatial .4 Mobility .3D motion.rotation about fixed axis and translation along axis .
parallel robot Kinematic Inversion – change which link is fixed – same relative motion. Examples – in class. also see following Atlas .5 Joints – Pairing elements Lower – surface contact Revolute – pin joint. different absolute motion. turning pair Prismatic – sliding pair Higher – point or line contact ball bearing gears cam and follower Link – rigid body Kinematic chain – number of links connected by joints open – serial robot closed – mechanism.
Bob Statically Determinate Structure Statically Indeterminate Structure 4Bar Mechanism Offset SliderCrank Mechanism Inverted SliderCrank Mechanism ScotchYoke Mechanism . Mechanisms.6 A Brief Atlas of Structures. and Robots Dr.
7 Geared 5Bar Mechanism Stephenson I 6Bar Mechanism Stephenson II 6Bar Mechanism Stephenson III 6Bar Mechanism Watt I 6Bar Mechanism Watt II 6Bar Mechanism .
8 Spur Gear Mechanism CamandFollower Mechanism Geneva Wheel Mechanism Planar 3dof Robot Adept 4dof SCARA Robot Mitsubishi 5dof Robot .
9 PUMA 6dof Robot NASA 8dof ARMII 2dof 5Bar Parallel Robot 3dof 3RRR Parallel Robot 3dof 3RPR Parallel Robot 3dof Carpal Wrist .
and acceleration. not synthesis (design). This requires dynamics analysis. gear box. The goal of ME 301 is to give the student general skills in general matrix/vectorbased kinematics and dynamics analysis which may be applied in later classes and later careers. Before one can size the links. bearings. However. for statics and dynamics. and actuators (motors) in a machine. and overdesigned factors of safety may be inefficiently applied. Before one can design a machine.10 Connection to Machine Design In ME 301 we focus on kinematics & dynamics analysis. joints. This kinematics & dynamics analysis is facilitated using a computer. velocity. All design candidates must be analyzed regarding the motion each would provide (position. Without the computer. the required motion must be satisfied. Engineering design is iterative by nature: each candidate design must be thoroughly analyzed to determine its performance relative to the design specifications and relative to other design candidates. the skills gained in this course support general (electro)mechanical design. both translational and rotational). the worstcase force and moment loading condition(s) must be known. This requires kinematics analysis. it is difficult to determine the worstcase loading cases. .
velocity. acceleration. Also. force. and torque (moment) vectors (see later in notes). angular acceleration. Used to represent position. Cartesian representation: Polar representation: Magnitude at angle: P @ θ (atan2 . arrow normal to the plane to represent angular velocity.quadrantspecific inverse tangent function) .11 MatrixVector Introduction Vectors Arrow in the plane with magnitude and direction.
express .12 Vector Addition Vectors add tailtohead (subtract components in same coordinate frame. headtotail).
Scalar result.13 Vector Dot Product Dot product is projection of one vector onto another. .
Direction via righthandrule: Put right hand fingers along first vector P1 . right thumb is direction of P1 × P 2 . . rotate into second vector P 2 .14 Vector Cross Product Cross product of two vectors gives a third vector mutually perpendicular to the original two vectors. Vector result.
Still has magnitude and direction. Example: ˆ ω = ±ωk . We will often represent these k vectors by curled arrows in the XY plane. angular velocity. perpendicular to the plane). + ccw (curling in the direction of the right hand fingers) – cw (curling in the opposite direction of the right hand fingers) . and ˆ torque (moment) vectors are arrows along about the k axis (the unit direction for the Z axis. angular acceleration. but simplifies to a single component with ± ˆ sign.15 ˆ k Vectors In planar kinematics.
16 Vector Examples 1 P = 1 2 3 P2 = 2 Addition: P + P2 = 1 Dot Product: P • P2 = 1 Cross Product: P × P2 = 1 .
acceleration. Special Matrices a11 a12 [ A] = a21 a22 a31 a32 a11 0 [ A] = 0 a22 0 0 1 0 0 [ I ] = 0 1 0 0 0 1 a13 a23 a33 0 0 a33 Square (m=n=3) Diagonal Identity . a11 a12 a a22 A] = 21 [ am1 am 2 a1n a2 n amn Used to simplify and standardize the solution of n linear equations in n unknowns (where m=n). where m is the number of rows and n in the number of columns.17 Matrices Matrix: m x n array of numbers. Used in velocity. and dynamics analysis linear equations (not used in position which is a nonlinear solution).
18 Transpose a11 T [ A] = a12 a13 a21 a22 a23 a31 a32 a33 a12 a22 a23 a13 a23 a33 Symmetric a11 T [ A] = [ A] = a12 a13 x1 Column Vector (3x1 matrix) { X } = x2 x3 Row Vector (1x3 matrix) {X } T = { x1 x2 x3 } Matrix Addition Just add up like terms f a + e b + f = c + g d + h h Just multiply each term a b e c d + g Matrix Multiplication with Scalar a b ka kb k = c d kc kd .
e. the number of columns in the lefthand matrix must equal the number of rows in the righthand matrix. if not. the multiplication is undefined and cannot be done! Multiplication proceeds by multiplying and adding terms along the rows of the lefthand matrix and down the columns of the righthand matrix: (use your index fingers from the left and right hands): Example: [C ] = a b d e ( 2 x1) ≡ ( 2 x3)( 3x1) g c ag + bh + ci h = f dg + eh + fi i note the inner indices (p=3) must match.Matrix Multiplication [C ] = [ A][ B ] ≠ [ B ][ A] 19 Row. as stated above and the dimension of the result is the outer indices. Column indices have to line up as follows: [C ] = [ A][ B ] ( mxn ) ≡ ( mxp )( pxn ) That is. . i. 2x1.
20 Matrix Multiplication Examples 1 2 3 A] = [ 4 5 6 7 8 [ B ] = 9 8 7 6 [C ] = [ A][ B ] 7 1 2 3 = 9 4 5 6 7 7 + 18 + 21 = 28 + 45 + 42 8 8 6 8 + 16 + 18 46 42 = 115 108 32 + 40 + 36 ( 2 x 2 ) ≡ ( 2 x3)( 3x 2 ) [ D ] = [ B ][ A] 7 8 9 8 1 2 3 = 4 5 6 7 6 7 + 32 14 + 40 21 + 48 39 54 69 = 9 + 32 18 + 40 27 + 48 = 41 58 75 7 + 24 14 + 30 21 + 36 31 44 57 ( 3x3) ≡ ( 3x 2 )( 2 x3) .
the matrix “1”. solve for [B] [C ] = [ A][ B ] ⇒ [ A] [C ] = [ A] [ A][ B ] = [ I ][ B ] = [ B] −1 −1 ⇒ [ B ] = [ A] −1 [C ] Matrix [A] must be square to invert. .21 Matrix Inversion Matrix “division”: given [C ] = [ A][ B ] . To calculate the matrix inverse: [ A] where: A −1 Adjoint ( A ) = A Determinant of [A] T Adjoint ( A ) = Cofactor ( A ) Cofactor(A) aij = ( −1) i+ j M ij Minor M ij is the determinant of the submatrix with row i and column j removed. [ A][ A]−1 = [ A]−1 [ A] = [ I ] where [I] is the identity matrix.
22 System of Linear Equations We can solve n linear equations in n unknowns with the help of a matrix. If not. A = 0 and the inverse of matrix [A] is undefined (dividing by zero). this is written as: [ A]{ x} = {b} where: a11 a12 [ A] = a21 a22 a31 a32 a13 a23 a33 (known coefficients) x1 { x} = x2 x 3 b1 {b} = b2 b 3 −1 (unknowns to be solved) (known righthand sides) Unique solution { x} = [ A] {b} only if [A] has full rank. For n=3: a11 x1 + a12 x2 + a13 x3 = b1 a21 x1 + a22 x2 + a23 x3 = b2 a31 x1 + a32 x2 + a33 x3 = b3 Using matrix multiplication (backwards). .
. x1 + 2 x2 = 5 6 x1 + 4 x2 = 14 ⇒ 1 2 x1 5 6 4 x = 14 2 x1 x2 [ A] = 1 2 6 4 {x} = {b} = 5 14 {x} = [ A]−1{b} A = 1(4 ) − 2(6 ) = −8 Determinant nonzero. x1 − 1 / 2 1 / 4 5 1 = 14 = 2 x 2 3 / 4 − 1 / 8 check: Plug answer into original equations and compare to the {b} we need to get.23 Matrix Example Solution of simultaneous linear equations. unique solution! [ A]−1 = check: 1 4 − 2 − 1 / 2 1 / 4 = A − 6 1 3 / 4 − 1 / 8 [ A][ A]−1 = [ A]−1[ A] = [I 2 ] = 1 0 0 1 Answer.
invA = inv(A).2.P1).0]. A = [1 2.0].2.14]. % sum2 = P2+P1.P1). b = [5. x1 = x(1).P2). % dot2 = dot(P2. sum1 = P1+P2.P2). % P2 = [3. A*x Define two vectors Vector addition Vector dot product Vector cross product % Define a matrix and vector % % % % Calculate determinant of A Calculate the inverse of A Solve linear equations Extract answers % Check answer – should be b . x = invA*b.6 4]. cross1 = cross(P1. dA = det(A). dot1 = dot(P1.24 Vector and Matrix Matlab Examples P1 = [1. % cross2 = cross(P2. x2 = x(2).
m (create a file with the beautiful Matlab Editor/Debugger . Can buy student version software and manual for about the price of one textbook (can use it for many classes!). but much simpler to use.mathworks. ENT college has a Matlab license. with parentheses alignment help and debugging . and Stateflow for now.this is colorcoordinated. Press <Enter> to see result or . animation. programming is vaguely Clike. matrices.25 Matlab Introduction Matrix laboratory Control systems simulation and design software. Type in commands (such as the Vector/Matrix examples given earlier) at the Matlab prompt >>. vectors. you can ignore Toolboxes. Type >>demo to get a comprehensive overview of Matlab including builtin functions. (Exception: there is Symbolic Math under Toolboxes for the adventurous student!). <Enter> to suppress result. programming. Based on C language. Very widespread in other fields. plots. Put your sequence of Matlab statements in an ASCII file name.com). Sold by Mathworks (http://www. Introduction to basics. tabfriendly. it is installed in most computer labs. Doubleclick on Matlab icon to get started. Recommended operation mode: mfiles. Try all the categories under Matlab first. Simulink.
without the . Type: >>who to show you what variables you have defined. . very useful for debugging. vector array. One basic way to run your program is to hit the ‘save and run’ button on the editor toolbar.m: Matrix and vector definition. assuming your file is in the search path. A % indicates a comment. Alternative: at the >> prompt type the MFile name name. at the end of statements to suppress intermediate results. If there is a syntax or programming logic error.m: Input. programming. If you use this suppression. after running a file. it will give a message at the bad line and then quit. Matlab language is interpretive and executes linebyline. the variable name still holds the resulting value(s) – just type the variable name at the prompt after the program runs to see the value(s). place the cursor over different variables in the MFile inside the Editor/Debugger to see the values! Online help is generally great: >>help Example mfiles (given on the following two pages) 1) MatEx1. transpose. Use the .m.26 capabilities). Plus. animation. plots. plus their matrix dimensions (scalar. 2) MatEx2. or matrix). and solution of linear equations. >>whos will show the variables. multiplication.
. .m % Matrix. multiplication. is continuation line) Define 3x1 vector {x1} 2x1 vector {v} is the product of [A1] times {x1} Transpose of matrix [A1] Transpose of vector {v} A1T = A1'. = inv(A2)*b.1]. = [1... vT = v'. % % % A2 Solution of linear equations Ax=b = [1 2 3. Bob. 1 1 1]. Vector examples % Dr.check. = det(A2). clear. = [3. with dimensions (whos) who whos % % % v x2 z Display some of the results .2.. Better be zero! check = A2*x2. = A1*x1. ME 301 %clc.. % % % % % Define 2x3 matrix [A1] (. % % % % % % Define matrix [A2] to be a 3x3 coefficient matrix b detA2 x2 Define righthand side vector of knowns {b} First check to see if det(A) is near zero Calculate {x2} to be the solution of Ax=b by inversion Check results. z % % % = b ..2.27 %% Matlab Example Code 1: MatEx1. % % % A1 x1 v % Clear the cursor and clear any previously defined variables Matrix and Vector definition.. 8 2 10]. Display the usercreated variables (who).. .3]. . 1 1 1. and transpose = [1 2 3.
cos(th). xlabel('\it\theta (\itdeg)').yc. Animation.'r'. delta.f2. % x2 = [0 L*cos(th(i))]. dth = the(2)*DR. grid.'FontSize'. FOR loop. r clear. ME 301 %clc.y2).'FontSize'. sin(th).'g'). IF logic.'b').th/DR.18). grid.'No') the = input('Enter [th0. set(gca. r*sin(th). L % Clear the cursor and clear any previously defined variables DR = pi/180.f1. thf = the(3)*DR. legend('Cosine'. Bob. % % % % % Menu to screen User types input Initial. % % Input % anim = menu('Animate Single Link?'. thf] (deg): ') th0 = the(1)*DR. % % % Animate if user wants to Give a blank graphics window For loop to animate Single link coordinates Animate to screen Define square plot limits Pause to see animation Pause to maximize window User hits Enter to continue Circle coordinates Cosine function of theta Sine function of theta % % plots % figure. and Plotting % Dr. title('Circle'). axis([0 360 1 1]). % y2 = [0 L*sin(th(i))]. % Coplot cosine and sine functions plot(th/DR. xlabel('\itX (\itm)').5 1. ylabel('\itY (\itm)'). Input. .5]). = 2. figure.5 1. % pause(1/4). axis('square'). set(gca.18).'Sine'). plot(x2.m % Menu. axis([1. title('Functions of \it\theta').18). N = (thfth0)/dth + 1. xlabel('\itX (\itm)'). % set(gca. axis(['square']). % if i==1 % pause. % for i = 1:N. % Plot circle plot(xc. axis([2 2 2 2]).5 1. % end end end % % % xc yc f1 f2 Calculate circle coordinates and cosine function = = = = r*cos(th).28 %% Matlab Example Code 2: MatEx2. grid. final thetas Assign theta array Number of iterations for loop % % Animate single link % % if anim == 1 figure. dth.'Yes'. th = [th0:dth:thf]. ylabel('Functions of \it\theta').'FontSize'. ylabel('\itY (\itm)'). % Constants = 1.
including ground J1 is the number of onedegreeoffreedom joints J2 is the number of twodegreeoffreedom joints .29 Mobility Mobility: Degreesoffreedom (dof): How many dofs does an unconstrained planar link have? What is the effect of constraining that link with a revolute joint? Grubler's Criterion: Planar Jointed Devices Where: M is the mobility N is the total # of links.
Planar mechanical device classification: M >1 M =1 M =0 M <0 .30 Onedegreeoffreedom joints: Revolute Prismatic Twodegreeoffreedom joints (all have rolling and sliding): Cam joint Gear joint Slottedpin joint Caution: m links joining at one revolute location. must count m1 joints! Caution: must count ground link (its freedom is subtracted in formula with n1.
31 Planar Mobility Examples: 1) 3link serial robot 2) 4bar linkage 3) Slidercrank linkage 4) Scotch Yoke mechanism 5) Cam and follower .
32
6)
Gear pair
7)
4bar linkage with parallel link
8)
Watt 6bar linkage
9)
Staticallydeterminate structure
10) Staticallyindeterminate structure
33
11) 5bar linkage
12) Geared 5bar linkage
13) Cammodulated 4bar linkage
14) 3RRR parallel robot
34
FourBar Mechanism Position Analysis Position (Displacement) Analysis: determination of relative orientation/ position of links in a mechanism. Required for testing motion of a synthesized mechanism. Also required for further analysis: velocity, acceleration, dynamics, forces. Generic Mechanism Position Analysis Statement: Given the mechanism and one dof of position input, calculate the position unknowns. Fourbar Mechanism Position Analysis
Step 1. Draw the Kinematic Diagram:
θ1 – ground link angle r1 – fixed ground link r2 – input link θ2 – input angle θ3 – coupler angle r3 – coupler link r4 – output link θ4 – output angle All angles measured in righthand sense from horizontal to link.
Write equation by starting and ending at same points. Define all angles in positive sense. Start at one point. . measured from the right horizontal to the link vector (tailtohead). Draw the Vector Diagram. Don’t try to force acute angles. State the problem: Step 3. add vectors tailtohead until reach a second point. good for all θ . but choosing a different path.35 Step 2. the relationships we can see so easily in the first quadrant hold for all four quadrants: L cosθ P= . Derive the VectorLoopClosure Equation. L sin θ Vector Diagram: Step 4.
Square and add: . Isolate and eliminate θ 3 and solve for θ 4 . Write XY Components for VectorLoopClosure Equation.θ 4 . Then go back to find θ 3 . Two coupled nonlinear equations in the two unknowns θ 3 . Break one vector equation into its two scalar components (X and Y): Step 6. Solve for the Unknowns from the XY Equations.36 Step 5.
37 This equation has the form: Solve using the tangent half angle substitution (Text Equation 4.9): θ t = tan 4 2 cos θ 4 = 1 − t2 1 + t2 sin θ 4 = 2t 1 + t2 We converted a complicated coupled transcendental set of equations into a quadratic polynomial. Two solutions for θ 4 : . Much easier to solve (but we doubled the order of the equation!).
Go back to find θ 3 . one for each solution branch. See section on Grashof's Law. Go back to original two XY scalar equations.) What if E 2 + F 2 − G 2 < 0 ? Imaginary solution. Why two solutions? (Graphically demonstrate the two branches. physically means the mechanism cannot assemble for that input angle. no need to use the atan2 function. Use ratio of Y to X equations: Show graphical interpretation: θ 3 = tan −1 BY − AY B X − AX .38 With factor two.
g.180o zero transmission. we can find the translational position of any point on the mechanism. 0. the absolute value of µ should remain in the range 40 < µ < 140 mechanism. Measure of mechanical advantage of mechanism. 90o is ideal. as a rule of thumb. By geometry: for good transmission in a . Now that we know the angular unknowns.39 The basic fourbar mechanism position analysis problem is now solved. coupler point C: Fourbar mechanism transmission angle: Transmission angle µ : relative angle between coupler and output links. e.
203 r4 = 0.9 θ4 µ 67.9 for the coupler link point of interest.57 53. calculate θ 3 .0 θ3 121.076 F = 0.178 m r2 = 3 r3 = 8 r4 = 7 and θ1 = 10. We use the SI system (m). Note µ is identical for both branches due to the conventions presented earlier.076 r3 = 0. Results: E = 0.06. Also given RC / A = 5 (in) and δ 3 = 36.79 1.02 These two branch solutions are demonstrated in the figures on the following page. µ .8 − 47.036 Branch Open Crossed t 1. 0.7 − 114.18 Given r1 = 0.θ 4 . and P C for both branches.284 in r2 = 0.40 FourBar Mechanism Position Analysis: Term Example 1 r1 = 11.005 G = 0.19.9 67. . Snapshot Analysis (one input angle) Given this mechanism and θ 2 = 30 .3 (Ground link is 11" over and 2" up).16 0.9 PC 0. 0.
2 0. Open Branch 0.2 0.2 0.1 Y (m) 0.3 4bar Example Snapshot.1 0.05 0.15 0.1 0 0.05 0 0.1 0 0.05 0.1 X (m) 0. Crossed Branch .3 4bar Example Snapshot.25 0.25 0.1 Y (m) 0.05 0 0.1 X (m) 0.1 0.41 0.15 0.2 0.
These circles intersect in general in two places. Connect the two branches and measure the unknown values. centered at point A. Graphical Solution Figure: . This is an excellent method to validate your computer results at a given snapshot. merely by drawing the mechanism and determining the mechanism closure.42 Graphical Solution: The 4bar position analysis may be solved graphically. Draw a circle of radius r4. Draw a circle of radius r3. centered at point O4. Draw the given input link 2 length at the given angle (point A). • • • • • • Draw the known ground link (points O2 and O4).
43
4bar Snapshot Matlab code: This program solves the 4bar position analysis problem for both branches given a single θ 2 . The results are drawn to the screen.
%% 4bar linkage snapshot position analysis  both branches % Fbarplec.m, with graphical output, Dr. Bob, ME 301 %clc; clear; % Clear cursor and clear previously defined variables % Inputs DR = pi/180; R = input('Enter [r1, r2, r3, r4, rca, th1, th2, del3] (m and deg): '); r1 = R(1); r2 = R(2); r3 = R(3); r4 = R(4); rca = R(5); th1 = R(6)*DR; th2 = R(7)*DR; del3 = R(8)*DR; % Change degrees to radians r1x = r1*cos(th1); r1y = r1*sin(th1); % Position analysis: theta4 E = 2*r4*(r1*cos(th1)  r2*cos(th2)); F = 2*r4*(r1*sin(th1)  r2*sin(th2)); G = r1^2 + r2^2  r3^2 + r4^2  2*r1*r2*cos(th1th2); t(1) = (F + sqrt(E^2 + F^2  G^2)) / (GE); % Crossed Branch t(2) = (F  sqrt(E^2 + F^2  G^2)) / (GE); % Open Branch th4(1) = 2*atan(t(1)); th4(2) = 2*atan(t(2)); % th3, coupler point, transmission angle; calculate for both branches for i = 1:2, % theta3 ax = r2*cos(th2); ay = r2*sin(th2); bx = r4*cos(th4(i)) + r1x; by = r4*sin(th4(i)) + r1y; th3(i) = atan2(byay,bxax); bet = th3(i) + del3; % coupler point pcx(i) = r2*cos(th2) + rca*cos(bet); pcy(i) = r2*sin(th2) + rca*sin(bet); mu(i) = abs(th4(i)th3(i)); % transmission angle end % Plot 4bar position results for i = 1:2, x2 = [0 r2*cos(th2)]; y2 = [0 r2*sin(th2)]; x3 = [r2*cos(th2) r1x+r4*cos(th4(i)) pcx(i)]; y3 = [r2*sin(th2) r1y+r4*sin(th4(i)) pcy(i)]; x4 = [r1x r1x+r4*cos(th4(i))]; y4 = [r1y r1y+r4*sin(th4(i))]; figure; plot(x2,y2,'r',x4,y4,'r'); patch(x3,y3,'r'); axis('square'); set(gca,'FontSize',18); xlabel('\itX (\itm)'); ylabel('\itY (\itm)'); axis([0.1 0.3 0.15 0.25]); grid; end
% % %
Coords of link 2 Coords of link 3 Coords of link 4
44
FullRangeOfMotion (F.R.O.M.) Analysis
A more meaningful result from position analysis is to report the position analysis unknowns for the entire range of mechanism motion. The first plot gives θ 3 (red), θ 4 (green), and µ (blue), all
deg, for all 0 ≤ θ 2 ≤ 360 , for Term Example 1, open branch only. The second plot gives the coupler point location for this branch, plotting PCY vs. PCX .
Thetas 3 (red) and 4 (green), Mu (blue) 180 160 140 Angle (deg) 120 100 80 60 40 20 0 50 100 150 200 θ 2 (deg) 250 300 350
45
0.25 0.2 0.15 0.1 Y (m) 0.05 0 0.05 0.1 0.1 0 0.1 X (m) 0.2 0.3
4bar Example Snapshot, Open Branch Coupler Curve
θ 3 = sin −1 (0. e. why not calculate θ 3 using Y equation?: e.46 Trigonometric Uncertainty Return to θ 3 solution: XY scalar equations: r3c3 = r1c1 + r4 c4 − r2 c2 r3 s3 = r1s1 + r4 s 4 − r2 s 2 Since θ 4 has been solved. only one of which is correct! Why not calculate θ 3 using X equation? Inverse cosine has a similar problem. for each θ 4 there are two possible solutions. figure: Problem: inverse sine function is doublevalued. figure: 2 Problem: inverse cosine function is doublevalued. θ 3 = cos −1 3 .g. for each θ 4 there are two possible solutions.g.5). only one of which is correct! .
figure: 3 r1s1 + r4 s 4 − r2 s 2 r1c1 + r4 c4 − r2 c2 Problem: the plain atan inverse tangent function is still doublevalued!. θ 3 = tan −1 1 . Input to this function is both a numerator and denominator.g. No uncertainty with atan2: 1 3 = e. θ 3 = a tan 2 + . hence this sign information is lost and the true quadrant is unknown.+ 2 2 θ 3 = a tan 2 − . for each θ 4 there are two possible solutions. only one of which is correct! Solution: use the quadrantspecific inverse tangent function atan2. both X and Y equations) . the function has builtin logic to determine the correct quadrant for the angle answer.g.47 So we must use information from both sine and cosine (i.e.− 1 2 3 = 2 θ 3 = A tan 2(r1s1 + r4 s 4 − r2 s 2 .this suggests using the tangent (as we did earlier in the θ 3 solution): θ 3 = tan −1 e. The plain atan function takes a single quotient input. given the signs ± of the numerator and denominator. r1c1 + r4 c4 − r2 c2 ) .
only showing one here.) With the 2 multiplying the inverse tangent result. . it doesn't matter whether we use atan or atan2 since the final answer will come to the same angle. . one for each t value. the multiple 2 takes care of this uncertainty: θ 4 = 60 or θ 4 = 420 = 60 Now. we don't know if the solution is 3 2 θ4 θ4 2 2 = 30 or = 210 However. we already have an exception in the θ 4 tangent halfangle solution: θ 4 = 2 tan −1 (t ) (there are two branches. . from before.48 Now. for next time consider the following: Do the solutions for θ 4 always exist? What if E 2 + F 2 − G 2 = 0 ? What if E 2 + F 2 − G 2 < 0 ? Stay tuned . Example: θ For 4 = tan −1 1 . having just cleared up this Trigonometric Uncertainty.
P.full rotation.49 Grashof’s Law Grashof was a German Engineer in the late 1800s. intermediate 2 links: L. Q 1) If L + S < P + Q Then we call this a Grashof Mechanism and there are four different mechanisms and rotation conditions: Diagrams: a) b) c) . rotates backandforth between limits Mechanism types (input / output links): Identify longest. S.not full rotation. shortest. no limits Rocker . Grashof's Law is used to determine the relative rotatability of the input and output links in a 4bar mechanism: Crank .
locomotive linkage – subject to branch jumping unless constrained. there is the additional interesting and troublesome feature that the mechanism may jump branches! Centerlines of links can become collinear. Examples 1) L = 10. Q = 7 – all Double Rockers 3) L = 10. S = 4.50 2) If L + S > P + Q Then we call this a NonGrashof Mechanism and the are four different mechanism inversions yield only one rotation condition: 3) If L + S = P + Q Then we call this a Special Grashof Mechanism and the four different mechanism inversions yield the identical rotation conditions from 1) Grashof Mechanism. P = 8. Q = 7 – demonstrate the 4 possibilities 2) L = 10. S = Q = 4 parallel. S = 5. However. Also. P = 8. Q = 7 – demonstrate branch jumping Another interesting example: L = P = 10. very easy analysis: θ2 =θ4 = µ θ 3 = 0 for all motion! . P = 8. S = 6.
To calculate the joint limits. there will be two joint limits. As shown in the figure. we use the law of cosines: (r3 + r4 )2 2 = r12 + r2 − 2r1r2 cos θ 2 L θ 2 L = ± cos 2 −1 r1 2 + r2 − (r3 + r4 )2 2r1r2 ± by symmetry about r1 .51 4Bar Joint Limits If Grashof's Law predicts the input link is a rocker. symmetric about the ground link. These joint limits occur when links 3 and 4 are aligned. there will be rotation limits on the input link.
742] = ±137. r3 = 8.1 (links 2 and 3 stretched in a line) θ 4 L = 172.3625] Caution: the figure on the previous page does not apply in all joint limit cases. r3 = 8. there are no joint limits! θ 2 L = ± cos 2 2 2 −1 10 + 4 − (8 + 7 ) = ± cos 2(10 )(4 ) which is undefined! −1 [− 1.52 Example 1: Given r1 = 10. r4 = 7 L + S < P + Q (10 + 4 < 8 + 7). r2 = 3. r2 = 6. Therefore. r4 = 7 (in) and θ1 = 10.18. here links 2 and 3 either stretch out in a line or fold upon each other.9 Example 2: Given r1 = 10. For certain mechanisms.5 (links 2 and 3 folded upon each other in a line) There are no limits on θ 2 since it is a crank. r3 = 8. θ 2 L = ± cos 2 2 2 −1 10 + 6 − (8 + 7 ) 2(10 )(6 ) −1 = ± cos [− 0. . Example 3: (Term Example Fourbar) Given r1 = 11. so we predict this mechanism is a crankrocker. limits are: θ 4 L = 120.3 . r2 = 4. The previous method can also be used to find angular limits on link 4 when it is a rocker. r4 = 7 L + S > P + Q (10 + 6 > 8 + 7) so we predict only double rockers from this mechanism. the limiting conditions occur when links 3 and 4 fold upon each other instead of stretching straight out.
Unlike the fourbar mechanism.53 SliderCrank Mechanism Position Analysis Converts linear motion to rotary or rotary motion to linear via connecting rod. All angles measured in righthand sense from horizontal to link. the four kinematic inversions of the slidercrank mechanism yield radically different types of motion. Air Compressor – electric motor drives crank (input). Draw the Kinematic Diagram: r2 – input link length r3 – coupler link length h – slider offset θ2 – input angle θ3 – coupler angle x – output displacement Link 1 is the fixed ground link. Use flywheel on crank to avoid locking. piston (output) compresses air. Step 1. x is measured horizontally from the origin to the slider/coupler revolute joint location. . In class we will solve the Air Compressor case where the crank is the input and the slider is the output. output is rotation of drive shaft. Internal Combustion Engine – explosion drives piston (input). Two dead points where piston is at limits.
add vectors tailtohead until reach a second point. Define all angles in positive sense. State the problem: Step 3. Derive the VectorLoopClosure Equation. . Vector Diagram: Step 4. Write equation by starting and ending at same points. but choosing a different path.54 Step 2. Start at one point. measured from the right horizontal to the link vector (tailtohead). Draw the Vector Diagram.
There a simpler method . and solve as in the fourbar approach. . Graphically demonstrate the two branches. Solve for the Unknowns from the XY Equations. Two coupled nonlinear equations in the two unknowns x.θ 3 . However. Break one vector equation into its two scalar components (X and Y): Step 6. notice that the two XY equations are coupled only in θ 3 but not in x. Write XY Components for VectorLoopClosure Equation.solve θ 3 using the Y equation only and then solve x from the X equation: What about trigonometric uncertainty? The inverse sine function is doublevalued and so there are two valid solution branches. We could isolate on unknown. square & add.55 Step 5.
The worst case is θ 2 = −90 . which yields r3 ≥ h + r2 This condition was derived assuming positive h. allowing negative h: r3 ≥ h + r2 .56 Fullrotation condition For solution to exist for entire motion range (r2 is a crank). . absolute value of the inverse sine argument must be less than or equal 1: h − r2 s2 ≤1 r3 r3 ≥ h − r2 s2 which must hold for all motion.
102 r3 = 0. We use the SI system (m). .076 Snapshot Analysis (one input angle) Given this mechanism and θ 2 = 30 .203 m h = 0. calculate x and θ 3 for both branches.57 SliderCrank Mechanism Position Analysis: Term Example 2 Given: r2 = 4 r3 = 8 in h=3 r2 = 0.114 θ3 7.290 0.2 172. Results: Branch Open Crossed x (m) 0.8 These two branch solutions are demonstrated in the figures on the following page.
2 0.1 Y (m) 0 0.3 SliderCrank Example Snapshot.3 SliderCrank Example Snapshot.1 0 0.2 0.1 0.1 0. Crossed Branch .2 0.2 0.58 0.1 Y (m) 0 0.2 0.1 X (m) 0.1 X (m) 0.2 0. Open Branch 0.1 0 0.
Draw a circle of radius r3. as seen in the full motion x plot. centered at point A. Graphical Solution Figure: Slider Limits The crank will rotate fully if the previouslyderived condition is met. This is an excellent method to validate your computer results at a given snapshot.2951. next page. This circle intersects the slider line in general in two places.59 Graphical Solution: The SliderCrank position analysis may be solved graphically. • • • • • • Place the grounded revolute for the crank at the origin. merely by drawing the mechanism and determining the mechanism closure. The slider reaches its maximum displacement when links 2 and 3 are aligned straight out and its maximum displacement when link 2 if folded onto link 3. Draw the line of the slider. Connect the two branches and measure the unknown values. offset vertically from the origin by h. Draw the given input link 2 length at the given angle (point A). We can draw two right triangles representing these conditions and easily calculate the x limits to be 0. .0671 ≤ x ≤ 0.
05 0 50 100 150 200 θ 2 (deg) 250 300 350 .2 x (m) 0.R. for all 0 ≤ θ 2 ≤ 360 . right branch only.60 FullRangeOfMotion (F.3 0. for Term Example 2. Slider Displacement 0.25 0.1 0.) Analysis A more meaningful result from position analysis is to report the position analysis unknowns for the entire range of mechanism motion.15 0. for the right branch only. The first plot gives x (m).O. The second plot gives θ 3 (deg). for all 0 ≤ θ 2 ≤ 360 .M.
61 Coupler Angle 60 50 40 (deg) 30 20 10 0 10 0 50 100 150 200 θ 2 (deg) 250 300 350 θ 3 .
rates. Units (translational and rotational): Basic Velocity Derivation Figure: Most general planar case: Translating and rotating rigid rod with a slider on it. forces. acceleration. Analytical velocity analysis: write position vectors. direction also crucial. Vector quantity: Magnitude of velocity is speed. Express the position vector in Cartesian coordinates: P P = PO + L = .62 Velocity Analysis Introduction Velocity analysis is important for kinematic motion analysis. Find the total velocity of point P on the slider. take first time derivatives. Velocity analysis is also required for dynamics: position. velocity. Velocity is the first time derivative of the position. machine design. dynamics. Position analysis must be completed first. Some tasks have timing. Velocity analysis is solution of coupled linear equations. solve for unknowns.
63 The angle is changing with angular velocity: Only the planar case is this simple. dt dx dt Example: d (L cos θ ) = ? dt . Product rule: d dx dy ( xy ) = y + x dt dt dt x. Chain rule: d df dx ( f ( x(t ))) = f is a function of x. The length of the rod is changing with sliding velocity: Product and Chain Rules of Differentiation We’ll need to use the product and chain rules over and over in velocity and acceleration analysis derivations. the spatial rotation case is more complicated. y both functions of time. which is a function of t.
64 Back to basic velocity derivation First time derivative of position vector: VP = d PP = dt .
65 We have just derived the ThreePart Velocity Equation: V P = V O +V +ω × L The terms for the ThreePart Velocity Equation can be expressed in various ways. summarized below: Vector Name ω×L VO V Point O Velocity Sliding Velocity Tangential Velocity XY Components Magnitude / Direction .
598 − 1 4.232 s or. calculate V P . ω = 1 rad/s.66 ThreePart Velocity Equation Example: Given (instantaneously) L = 2 m. V O = {3 2} m/s.732 5. V = L = 3 m/s (outward).7 m/s Show magnitude and direction of each velocity component: Vector Name VO V ω×L Point O Velocity Sliding Velocity Tangential Velocity XY Components Magnitude / Direction .965@ 48.5 + 1. V P = 6. θ = 30 . T V + V cosθ − Lω sin θ 3 + 3cos30 − 2 (1) sin 30 V P = OX = VOY + V sin θ + Lω cosθ 2 + 3sin 30 + 2 (1) cos30 3 + 2.598 m VP = = 2 + 1.
Fourbar Mechanism Velocity Analysis Step 1.67 FourBar Mechanism Velocity Analysis Velocity Analysis: determination of angular and linear velocities of links in a mechanism. and one dof of velocity input.3. is the absolute angular velocity of link i. Linear equations result from first time differentiation of position equations. Generic Mechanism Velocity Analysis Statement: Given the mechanism. . Step 2. complete position analysis. Position analysis must be complete first.4. Position Analysis must first be complete. forces. Draw the Velocity Diagram: where ω i . i = 2. Required for complete motion analysis. dynamics. Unique solution for each mechanism branch. 1dof mechanism. ω 1 = 0 since the ground link is fixed. Also required for further analysis: acceleration. machine design. calculate the velocity unknowns. so one velocity input must be given.
Take the first time derivative of the vector loop closure equations from position analysis. Derive the velocity equations. in XY component form. .68 Step 3. Fourbar mechanism position equations: r 2 + r 3 = r1 + r 4 r2c2 + r3c3 = r1c1 + r4c4 r2 s2 + r3 s3 = r1s1 + r4 s4 First time derivative for velocity equations: (use chain rule several times) Chain rule: d cosθi dθi d ( cosθi ) = dt dθi dt = − sin θiθ i = − sin θiωi d sin θi dθi d ( sin θi ) = dt dθi dt = cosθiθi = cosθiωi Don’t have to use product rule because ri = 0 (rigid links). State the problem: Step 4.
69 The first time derivative of the position equations is: Gathering unknowns on the LHS: Substituting simpler terms: Written in matrix form: .
Algebra solution: . ω 4 . Solve the velocity equations for the unknowns ω 3 .70 Step 5.
71 Alternate matrix solution (yields same solution): .
this happens when links 3 and 4 are straight out or folded on top of each other (what does this correspond to?): . when the determinant of the coefficient matrix goes to zero.72 FourBar mechanism singularity condition: When does the solution fail? This is a mechanism singularity. Physically. The result is dividing by zero. ω 4 . for infinite answers ω 3 . Let’s see what this means physically.
e. coupler point C: . Now that we know the angular unknowns. we can find the translational velocity of any point on the mechanism.73 The basic fourbar mechanism velocity analysis problem is now solved.g.
042 Coupler point translational velocity: V C = (m/s) 0. θ3 = 53.7 .284.587 Both are negative.8 .3 .178.209 .151 ω3 −0.120 −0.120 −0. 0. and θ1 = 10. so cw direction. Snapshot Analysis (one input angle) Given this mechanism position analysis plus ω2 = π rad/s (+. so ccw).271 = ω4 −0.127 m. δ 3 = 36.74 Fourbar mechanism velocity example: Given r1 = 0.076.207 4 ω3 −1.093 ω = 0. calculate ω 3 .9 . RC/A = 0. This is the open branch of the fourbar mechanism position example (Term Example 1). r3 = 0. and V C for this instant (snapshot). r4 = 0. r2 = 0.164 −0. θ 4 = 121. θ 2 = 30 .203. 0. These results are the absolute angular velocities of links 3 and 4 with respect to the ground link. ω 4 .
) Analysis A more meaningful result from velocity analysis is to report the velocity analysis unknowns for the entire range of mechanism motion. θ 2 (since it is related to time t via θ 2 = ω 2 t ).5 0 ω 3 4 0. for all 0 ≤ θ 2 ≤ 360 .O. we can plot the velocity results vs.R. Since ω 2 is constant. (rad/s). for Term Example 1. open branch only.M. The plot below gives ω 3 (red) and ω 4 (green). Omegas 3 (red) and 4 (green) 2 1.5 0 50 100 150 200 θ 2 (deg) 250 300 350 .75 FullRangeOfMotion (F.5 1 1.5 (rad/s) and ω 1 0.
76
The plot below gives the translational coupler point velocity for all 0 ≤ θ 2 ≤ 360 , for Term Example 1, open branch only.
Coupler Point Velocities X (red) and Y (green) 0.3 0.2 VCX and VCY (m/s) 0.1 0
0.1 0.2 0 50 100 150 200 θ 2 (deg) 250 300 350
77
Derivative/Integral Relationships When one variable is the derivative of another, what are the relationships? For example: dθ ω3 = 3 θ 3 = θ 30 + ∫ ω3dt dt
1.2 (rad) 1 0.8 0.6 0.4 0 1 0 50 100 150 200 250 300 350
ω
(rad/s) 3
θ
3
1 0 50 100 150 200 θ 2 (deg) 250 300 350
The value of ω3 at any point is the slope of the θ 3 curve at that point. The value of θ 3 at any point is the integral of the ω3 curve up to that point (the value of θ 3 at any point is the area under the ω3 curve up to that point). This graph is vs. θ 2 , but the same type of relationships hold as for time t since ω 2 is constant. This is the Term Example 1 result, but we changed θ 3 from deg to rad for better comparison.
78
SliderCrank Mechanism Velocity Analysis
Again, we will solve the Air Compressor case where the crank is the input and the slider is the output. The Internal Combustion Engine case (slider input/crank output) is equally interesting. Step 1. Position Analysis must first be complete. Step 2. Draw the Velocity Diagram:
where ω i , i = 2,3 is the absolute angular velocity of link i. x is the variable slider velocity. ω 4 = 0 since the slider cannot rotate. Step 3. State the problem:
Derive the velocity equations. Take the first time derivative of the vector loop closure equations from position analysis. Slidercrank mechanism position equations: r2 + r3 = x + h First time derivative for velocity equations: r2c2 + r3c3 = x r2 s2 + r3 s3 = h Gathering unknowns on the LHS: Written in matrix form: . in XY component form.79 Step 4.
.80 Step 5. these equations are decoupled so we don't need a matrix solution. Actually. First. Solve the velocity equations for the unknowns ω3 . solve ω3 from Y equation: Then solve x from the X equation using the ω3 result: The alternate matrix solution: will yield identical results. x .
. resulting in infinite answers ω3 . when the determinant of the coefficient matrix goes to zero. even with offsets. A = r3c3 = 0 A = 0 when cos θ 3 = 0 . otherwise we have a degenerate slidercrank mechanism.270 . The result is dividing by zero. this happens when link 3 is straight up or down (θ3 = ±90 ).81 Slidercrank mechanism singularity condition: When does the solution fail? This is a slidercrank mechanism singularity. Of course r3 cannot go to zero. Physically. or θ 3 = 90 . x . Doesn’t happen for nominal fullrotation slidercrank mechanisms.
Snapshot Analysis (one input angle) Given this mechanism position analysis plus ω2 = π / 2 rad/s (+.ω3 for this instant (snapshot). h = 0. These results are the absolute linear and angular velocities of links 4 and 3 with respect to the fixed ground link.290 m.080 0 −0. x = 0. 1 0. r3 = 0.202 ω = 0. .686 Both are negative. calculate x.076 m.102. so the slider is currently traveling to the left and the coupler link is currently rotating in the cw direction.062 = ω3 −0. and θ 2 = 30 . so ccw).025 x −0.82 Slidercrank mechanism velocity example: Given r2 = 0.138 3 x −0. This is the right branch of the slidercrank position example (Term Example 2).2 .203. θ3 = 7.
4 0. for all 0 ≤ θ 2 ≤ 360 .2 0 3 0.6 (rad/s) xd (m/s) and ω 0. m/s) and ω 3 (green.8 0. θ 2 (since it is related to time t via θ 2 = ω 2 t ). rad/s). for Term Example 2.83 FullRangeOfMotion (F. right branch only.6 0.R.4 0.) Analysis A more meaningful result from velocity analysis is to report the velocity analysis unknowns for the entire range of mechanism motion.O.2 0.8 0 50 100 150 200 θ 2 (deg) 250 300 350 .M. Since ω 2 is constant. we can plot the velocity results vs. Xdot (red) and Omega3 (green) 0. The plot below gives x (red.
The value of x at any point is the integral of the x curve up to that point (the value of x at any point is the area under the x curve up to that point). what are the relationships? For example: dx x = x 0 + ∫ xdt x= dt 0. but the same type of relationships hold as for time t since ω 2 is constant.84 Derivative/Integral Relationships When one variable is the derivative of another.2 0 0.2 0. .2 0 50 100 150 200 θ 2 (deg) 250 300 350 The value of x at any point is the slope of the x curve at that point.3 x (m) 0. This is the Term Example 2 result. θ 2 .1 0 50 100 150 200 250 300 350 xd (m/s) 0. This graph is vs.
acceleration.85 Acceleration Analysis Introduction Acceleration analysis is required for dynamics: position. forces. dynamics. take first two time derivatives. Units (translational and rotational): Basic Acceleration Derivation Figure: Rotating rigid rod with a slider on it. Vector quantity: Analytical acceleration analysis: write position vectors. Acceleration is the first time derivative of the velocity and second time derivative of the position. velocity. Find the total acceleration of point P on the slider. . solve for unknowns. Position and velocity analyses must be completed first. Acceleration analysis is solution of linear equations. Important for kinematic motion analysis. machine design.
The length of the rod is changing with sliding velocity and acceleration: . the spatial rotation case is more complicated.86 Recall the 2part position and 3part velocity results: POX + L cos θ P P = PO + L = POY + L sin θ VOX + V cos θ − Lω sin θ V P = V O +V +ω × L = VOY + V sin θ + Lω cos θ The angle is changing with angular velocity and acceleration: Only planar case is this simple.
Product rule: d dx dy ( xy ) = y + x dt dt dt x. dt dx dt Example: d2 ( L cosθ ) = ? dt 2 . y both functions of time.87 Product and Chain Rules of Differentiation Again. we’ll need to use the product and chain rules over and over in acceleration analysis derivations. Chain rule: d df dx ( f ( x(t ))) = f is a function of x. which is a function of t.
88 Back to basic acceleration derivation First time derivative of velocity vector (Second time derivative of position vector): dV P d 2 P P = = AP = 2 dt dt .
89 We have just derived the FivePart Acceleration Equation: A P = AO + A + 2ω × V + α × L + ω × (ω × L ) These terms can be expressed in various ways. summarized below: Vector Name AO A 2ω × V α×L Tangential Acceleration ω × (ω × L ) Centripetal Acceleration Point O Acceleration Sliding Acceleration Coriolis Acceleration XY Components Magnitude / Direction .
268 m = = 2 + 2 + 5.879@101. AO = {1 2} . A = L = 4 m/s2 (outward).660 s 2 or.196 + 3.464 − 3 − 2 − 1. Given (instantaneously) L=2 m. α = 2 rad/s2.464 − 1 11. AP = 11. ω = 1 rad/s. T V = L = 3 m/s (outward).732 −2. calculate A P . V O = {3 2} . T AOX + A cosθ − 2V ω sin θ − Lα sin θ − Lω 2 cosθ AP = AOY + A sin θ + 2V ω cosθ + Lα cosθ − Lω 2 sin θ 1 + 3.0 m/s2 Show magnitude and direction of each Acceleration component: Vector Name AO Point O Acceleration A Sliding Acceleration 2ω × V Coriolis Acceleration α×L Tangential Acceleration ω × (ω × L ) Centripetal Acceleration XY Components Magnitude / Direction .90 FivePart Acceleration Equation Example: Continuation of 3part velocity example. θ = 30 .
91 FourBar Mechanism Acceleration Analysis Acceleration Analysis . Step 2. Also required for further analysis: dynamics.3.4 is the absolute angular acceleration of link i. Linear equations result from second time differentiation of position equations. Draw the Acceleration Diagram: where α i . forces. Position and velocity analyses must be complete first. i = 2. complete position and velocity analyses. Unique solution for each mechanism branch. 1dof mechanism. Generic Mechanism Acceleration Analysis Statement: Given the mechanism. Fourbar Mechanism Acceleration Analysis Step 1. Required for complete motion analysis. . calculate the acceleration unknowns. α 1 = 0 since the ground link is fixed. and one dof of acceleration input. Position and Velocity Analyses must first be complete. machine design. so one acceleration input must be given.determination of angular and linear accelerations of links in a mechanism.
Derive the acceleration equations. Fourbar mechanism velocity equations: − r2ω2 s2 − r3ω3 s3 = − r4ω4 s4 r2ω2c2 + r3ω3c3 = r4ω4c4 The first time derivative of the velocity equations is: Gathering unknowns on the LHS: . Take the first time derivative of the fourbar mechanism velocity equations from velocity analysis. in XY component form.92 Step 3. State the problem: Step 4.
93 Substituting simpler terms: Written in matrix form: .
94 Step 5.α 4 . Solve the acceleration equations for the unknowns α 3 . Matrix solution (Algebra solution yields the same results): .
so singularity condition is identical: θ 4 − θ 3 = 0 . there is zero transmission angle µ and Link 2 is at a joint limit! The basic fourbar mechanism acceleration analysis problem is now solved. Now that we know the angular unknowns.g. coupler point C: . velocity. we can find the translational acceleration of any point on the mechanism.180 . e. At this singularity. and acceleration. This condition is the same problem for position.95 FourBar mechanism singularity condition: Same coefficient matrix A as velocity case.
8 .076.96 Fourbar mechanism acceleration example: Given r1 = 0. plus α 2 = 0 rad/s2.120 −0.582 .587 rad/s.093 α = −0.877 −0.271. RC/A = 0.α 4 for this instant (snapshot).213 = α 4 6.164 −0. r3 = 0.9 . This is the open branch of the position and velocity example (Term Example 1). θ3 = 53. 0. These results are the absolute angular accelerations of links 3 and 4 with respect to the ground link. Snapshot Analysis (one input angle) Given this mechanism position and velocity analysis. so ccw direction. δ 3 = 36.151 α 3 −0.203. r2 = 0. and θ1 = 10. θ 4 = 121.030 Both are positive.3 . calculate α 3 . ω2 = π . ω3 = −1. −0.127 m.7 .284.589 4 α 3 0.178. r4 = 0. ω4 = −0. θ 2 = 30 .676 2 Coupler point translational acceleration: AC = m/s −0.
(rad/s2).97 FullRangeOfMotion (F.R. Since ω 2 is constant. for Term Example 1.O. θ 2 (since it is related to time t via θ 2 = ω 2 t ). for all 0 ≤ θ 2 ≤ 360 . open branch only.) Analysis A more meaningful result from acceleration analysis is to report the acceleration analysis unknowns for the entire range of mechanism motion. Alphas 3 (red) and 4 (green) 6 4 2 0 α and α 3 (rad/s ) 4 2 2 4 6 0 50 100 150 200 θ 2 (deg) 250 300 350 .M. The plot below gives α 3 (red) and α 4 (green). we can plot the acceleration results vs.
2 0 0.6 ACX and ACY (m/s ) 2 0.6 0.98 The plot below gives the translational coupler point acceleration for all 0 ≤ θ 2 ≤ 360 . Coupler Point Accelerations X (red) and Y (green) 0.4 0.2 0. for Term Example 1.4 0. open branch only.8 0 50 100 150 200 θ 2 (deg) 250 300 350 .8 0.
5 2 0 50 100 150 200 250 300 350 ω (rad/s) 4 θ 1 0 0 (rad/s ) 4 2 4 1 50 100 150 200 250 300 350 5 0 50 100 150 200 θ 2 (deg) 250 300 350 5 0 α . For example: ω4 = dθ 4 dt θ 4 = θ 40 + ∫ ω 4 dt ω 4 = ω 40 + ∫ α 4 dt α4 = dω 4 dt (rad) 3 2.99 Derivative/Integral Relationships When one variable is the derivative of another. recall the relationships from calculus.
Step 1.3 is the absolute angular acceleration of link i. Step 3. i = 2. α 4 = 0 since the slider cannot rotate. State the problem: . Position and Velocity Analyses must first be complete. Draw the Acceleration Diagram: where α i . we will solve the Air Compressor case where the crank is the input and the slider is the output. Step 2.100 SliderCrank Mechanism Acceleration Analysis Again.
Derive the acceleration equations.101 Step 4. Slidercrank mechanism velocity equations: − r2ω2 s2 − r3ω3 s3 = x r2ω2c2 + r3ω3c3 = 0 The first time derivative of the velocity equations is: Gathering unknowns on the LHS: Written in matrix form: . Take the first time derivative of the velocity equations from velocity analysis. in XY component form.
Actually. solve α 3 from Y equation: Then solve x from the X equation using the α 3 result: . x .102 Step 5. Solve the acceleration equations for the unknowns α 3 . these equations are decoupled so we don't need a matrix solution. First.
290 m.686 rad/s. and ω2 = π / 2 . ω3 = −0.681 These results are the absolute linear and angular accelerations of links 4 and 3 with respect to the fixed ground link.329 = α 3 0.2 . calculate x. 1 0. θ3 = 7.076 m.203.312 0 −0.103 Slidercrank mechanism singularity condition: Same coefficient matrix as velocity case. r3 = 0. This is the right branch of the position and velocity example (Term Example 2). x = 0. .062 m/s.137 3 x −0.α 3 for this instant (snapshot).102. h = 0.025 x −0. α 2 = 0 rad/s2. so singularity condition is identical (see the singularity discussion in the slidercrank velocity section). x = −0. Snapshot Analysis (one input angle) Given this mechanism position and velocity analysis plus.202 α = −0. Slidercrank mechanism acceleration example: Given r2 = 0. and θ 2 = 30 .
5 1 1.M. The plot below gives x (red.5 2 2. we can plot the velocity results vs. Xdotdot (red) and Alpha3 (green) 1 (rad/s ) 3 0. m/s2) and α 3 (green. for all 0 ≤ θ 2 ≤ 360 .O. for Term Example 2. Since ω 2 is constant.R.5 0 50 100 150 200 θ 2 (deg) 250 300 350 .104 FullRangeOfMotion (F.) Analysis A more meaningful result from acceleration analysis is to report the acceleration analysis unknowns for the entire range of mechanism motion. right branch only. rad/s2). θ 2 (since it is related to time t via θ 2 = ω 2 t ).5 0 2 2 xdd (m/s ) and α 0.
recall the relationships from calculus.3 x (m) 0.2 0.105 Derivative/Integral Relationships When one variable is the derivative of another.5 0 0.1 0 xd (m/s) 0.5 0 50 100 150 200 θ 2 (deg) 250 300 350 50 100 150 200 250 300 350 50 100 150 200 250 300 350 xdd (m/s ) 2 .2 0 0.2 0 0. For example: x= x= dx dt dx dt x = x 0 + ∫ xdt x = x 0 + ∫ xdt 0.
and α2 = 0. This input motion specification is plotted like this: Note that we have been plotting calculated results vs.106 Input Motion Specification Up to this point. so we could just as well plot all results vs. for full range of motion we have assumed that the input link rotates fully with a given constant input angular velocity. Since ω2 is constant. since both θ2 and t increase steadily (linearly). θ2. Our input constraints have thus been 0 ≤ θ 2 ≤ 360 . time t. we have θ 2 = ω2t . ω2 constant. .
Many useful mechanisms have input links that do not rotate fully but travel between joint limits. t.107 This constant ω2 input specification is fine for mechanisms whose input rotates fully and considering steadystate motion only. starting and stopping at zero angular velocity. What is the weakness of this approach? (Discontinuous acceleration function yields infinite jerk at start. Why is the previous page’s plots unacceptable in this case? Simplest change – linear angular velocity starting and stopping at zero: We cannot plot vs. θ2 since it is not increasing linearly – plot vs. and finish. middle.) .
108 We can fix this with a trapezoidal input acceleration profile: This input motion specification should be fine (trapezoidal input torque is often used for industrial robots). but there are many different zones to handle – what acceleration profile is similar but with a single function? .
and tF = 3 sec.5 2 2.5 2 2. .5 3 Example with θ 20 = 60 .5 3 2 0 0.109 Fullcycloidal function input angle specification θ 2 ( t ) = θ 20 + (θ 2 F − θ 20 ) ω2 ( t ) = tF t 1 2π t sin − t F 2π tF tF 2π (θ 2 F − θ 20 ) 2π t α2 (t ) = sin t 2 tF F 4π 2 (θ 2 F − θ 20 ) 2π t β2 (t ) = cos 3 tF tF (θ 2 F − θ 20 ) 1 − cos 2π t θ2 (deg) β 2 (rad/s3) α2 (rad/s2) ω (rad/s) 2 100 80 60 0 4 2 0 0 1 0 0.5 1 1.5 3 1 0 2 0 0.5 1 1.5 time (sec) 2 2.5 1 1.5 2 2.5 3 0.5 1 1. θ 2 F = 120 .
What is the first step in analysis? Draw . .110 Dynamics Introduction Chart: Kinematics: translational rotational Kinetics: translational Newton’s 2nd Law: rotational Euler’s equation: Dynamics of a single rigid body in the plane Rigid body acted on by a system of forces and moments to produce planar motion. . .
FBD Simplified FBD MAD (massacceleration diagram) . This contains all the info needed to write Newton’s 2nd Law and Euler’s equation.111 Free Body Diagram (FBD) Isolate each rigid body and show the forces and moments acting.
.112 Internal and External Forces and Moments All internal and external forces and moments must be included on the FBD. External forces/moments: Internal forces/moments: Write dynamics equations Newton’s 2nd Law: Euler’s equation: AG is the linear acceleration of center of gravity – same direction as R . α angular acceleration of rigid body. Different points in rigid body have different linear accelerations. The entire rigid body experiences the same α .
just wanted you to know in case you ran into it somewhere. then sum to zero as in statics. R − m AG = 0 R + FO = 0 T + r × R − IG α = 0 T +r×R+MO =0 . “Centrifugal force” − mrω 2 is an example of an inertial force. Subtract RHS of equations. it’s not really a force but an effect of acceleration and inertia.113 D’Alembert’s Principle Turn dynamics problem into a statics problem by the inclusion of a fictitious “inertial force” F 0 = − m AG and a fictitious “inertial moment” M 0 = − IG α . We won’t use this method.
ω 3 . external forces and moments. F EXT . α 3 . θ 4 . Inverse Dynamics: Given the mechanism. α 2 . find the motion θ 2 . α 2 . θ 3 . external forces and moments. and Next lecture: Newton's 2nd Law and Euler's equation require: translational: mass rotational: center of gravity center of gravity mass moment of inertia m PG IG . M EXT . ω 4 . α 4 . θ 4 . 4Bar Linkage Inverse Dynamics: Given the motion θ 2 . ω 2 . M EXT . 4Bar Linkage Forward Dynamics: Given τ 2 and F EXT . ω 2 . α 3 . θ 3 . ω 3 . find the resulting mechanism motion and internal joint forces. and the desired mechanism motion.114 Two Types of Dynamics Problems Forward Dynamics: Given the mechanism. find the required driving force (or torque) and internal joint forces. find τ 2 and internal forces F ij . α 4 and internal forces F ij . and the applied driving force (or torque). ω 4 .
Newton’s 1st Law. Center of Gravity. Mass is measure of translational inertia – resistance to change in motion. 2 .115 Mass. mass m is the proportionality constant. Mass Moment of Inertia ∑ F = m AG Translational: mass Rotational: ∑MG = IGα center of gravity center of gravity mass moment of inertia Mass In Newton’s 2nd Law ∑ F = ma . Mass is also measure of storage 1 of translational kinetic energy KET = mv 2 .
CG. IG: System of particles General rigid body Rectangular rigid body Mass calculation: System of particles: General rigid body: Rectangular rigid body: .116 Examples for m.
G) Also called center of mass. centroid CG calculation: System of particles: General rigid body: . mass center.117 Center of Gravity (CG.
the CG is the geometric center. X= = = ∫ xdm x ∫ dm ∫ xdV b/2 Y= = = ∫ ydm y ∫ dm m∫ y ρ ρ ρ mx ydV h/2 m ∫− b / 2 ρ th b / 2 = ∫−b / 2 xdx m = xthdx ytbdy m ∫− h / 2 ρ tb h / 2 ydy = m ∫− h / 2 = ρ ρ th x 2 m 2 b/2 ρ tb y 2 h/2 −b / 2 m 2 −h / 2 b2 = 4 − 4 =0 2m ρ th b 2 h2 = − =0 2m 4 4 ρ tb h 2 For a homogeneous. .118 Rectangular rigid body: Using an XY coordinate frame centered at the geometric center. regular geometric body.
Also.119 Mass Moment of Inertia (IG) is not the same as Area moment of inertia (IG) for beam bending: I Ax = ∫ y 2 dA y I Ay = ∫ x 2 dA x Units: I A ≡ m 4 Mass Moment of Inertia (IG) In Euler’s equation ∑ M G = I GZ α . . 2 Units: IG ≡ kgm2 . Newton’s 1st Law. I is measure of rotational inertia – resistance to change in motion. I is the proportionality constant. I is also measure of storage 1 of rotational kinetic energy KER = I Gω 2 . it is a measure of how hard it is to accelerate in rotation about certain axes.
120 Mass Moment of Inertia IG calculation: System of particles: where ri is the scalar perpendicular distance from the axis to the ith particle. With squaring. If first moment is balanced. General rigid body: What is the only term that matters for XY planar motion? In the example shown above: I ZZG > I YYG > I XXG also I ZZ > I ZZG . no there can be no canceling like for CG. all terms will be positive. second moment will be doubled about the CG.
I ZZG = body (x 2 + y 2 )dm = ∫−bb/ /22 ∫−hh/ /22 (x 2 + y 2 )ρtdxdy ∫ 3 h/2 y b/2 2 I ZZG = ρt ∫ x y+ dx −b / 2 3 −h / 2 3 − h3 b / 2 2 h − h 1 h x − dx = ρt ∫ + − −b / 2 2 2 3 8 8 b/2 3 h3 h3 x hx + dx = ρt + 12 12 3 −b / 2 I ZZG = ρt ∫ b/2 2 hx −b / 2 h b3 − b3 h3 b − b + − I ZZG = ρt − 3 8 8 12 2 2 b3h bh3 ρtbh 2 = = ρt + b + h2 12 12 12 m 2 b + h2 12 ( ) I ZZG = ( ) (because m = ρV = ρtbh ) .121 Rectangular rigid body: Using an XY coordinate frame centered at the CG.
kgm 2 . Checks with result given in the textbook. How do we find mass moments of inertia in the realworld? • look up in tables • CAD package such as SolidEdge .122 Units: mass times distance squared.
which is always positive. due to the term md 2 . . Notice I ZZG is a small as it can get. any I ZZ must be greater.123 Parallel Axis Theorem The mass moment of inertia through the CG is related to mass moments of inertia of parallel axes through different points as follows: where d is the scalar distance separating the axis of interest from the axis through the CG. Parallel axis theorem example: Rectangular rigid body: I ZZ b2 h2 m 2 2 = (b + h ) + m + 12 4 4 b2 b2 h2 h2 = m + + + 12 4 12 4 b2 h2 = m + 3 3 m = (b2 + h2 ) 3 Agrees with result given in dynamics textbooks.
and Acceleration Analyses must first be complete. Step 2. external forces and moments. and the desired mechanism motion. Single Rotating Link Inverse Dynamics Analysis Step 1.124 Single Rotating Link Inverse Dynamics Generic Mechanism Inverse Dynamics Analysis Statement: Given the mechanism. find the required driving force (or torque) and internal joint forces. Draw the Diagrams: Physical Dynamics Diagram: Free Body Diagram (FBD): . Position. Velocity.
125 Step 3. Derive the NewtonEuler Dynamics Equations. Newton's 2nd Law: Euler's Equation: Count # of unknowns and # of equations: . State the problem: Step 4.
Written in matrix form: .126 Step 5. Derive XYZ scalar equations from the vector equations.
we don’t need matrix solution. the first two equations are decoupled and the solution is straightforward: Step 7. motion. Calculate Shaking Force and Moment After the inverse dynamics problem is solved. and external loads: .127 Step 6. which is the force/moment reaction on the ground link due to the mechanism. Solve for the unknowns Actually. we can calculate the vector shaking force and moment.
AGX AG = = AGY FEX FE = = FEY r r12 = 12 X = r12Y r r E = EX = rEY I GZ = . Now. how do we calculate the various terms that appear in the dynamics equations? These all must be derived from given information.128 Terms for the inverse dynamics equations The inverse dynamics problem has been solved analytically for the single rotating link.
129 Single rotating link inverse dynamics example: Given: L = 1 m. 0 0 F12 X 8510 1 0 1 0 F12Y = − 4980 0.5k Nm . ω = 100 rad/s. h = 0. calculate F12 X .5 F12 X 8510 F12Y = − 4980 N.5 − 8510 F S = F 21 = − F 12 = N 4980 ˆ M S = −τ = −66. and external force. φ E = 0 (constant relative to horizontal). FE=150 N. Calculated terms: r12 = rE = 0.17 kgm 2 AGx = 4330 m AGy = −2500 s 2 Snapshot Analysis (one input angle) At θ = 150 .250 0. α = 0.5 m I GZ = 0.433 1 τ 37. M S for this instant (snapshot). m = 2 kg. Nm τ 66. τ and F S . ME=0 Nm. given this link.1 m. motion. F12Y .
red) for all 0 ≤ θ ≤ 360 . This shows the torque that must be supplied by an external DC servomotor to cause the specified motion.) Analysis A more meaningful result from inverse dynamics analysis is to report the unknowns for the entire range of mechanism motion. Tau (red) with average (green) and rms (blue) 150 100 50 τ (Nm) 0 50 100 150 0 50 100 150 200 θ (deg) 250 300 350 .M. Also plotted is the average torque (green) τAVG = 0 and the rootmeansquare (RMS) torque value (blue) τRMS = 106. assuming the given ω is constant.1 Nm. The plot below gives the required driving torque τ (Nm. for the same example from the previous page.O.130 FullRangeOfMotion (F.R.
131
The plots below give the Shaking Force F S and CG translational acceleration results, respectively, for all 0 ≤ θ ≤ 360 . In both plots, the X components are red and the Y green.
1
x 10
4
Shaking Force, X (red) and Y (green)
0.5 FSX and FSY (N)
0
0.5
1 0
50
100
150
200 θ (deg)
250
300
350
132
CG Acceleration, X (red) and Y (green) 5000
AGX and AGY (m/s )
2
0
5000 0
50
100
150
200 θ (deg)
250
300
350
The Shaking Moment M S is merely the negative of the driving torque τ plot shown previously and hence is not shown separately. Is the static loading (mg) significant?
133
FourBar Mechanism Inverse Dynamics Generic Mechanism Inverse Dynamics Analysis Statement: Given the mechanism, external forces and moments, and the desired mechanism motion, find the required driving force (or torque) and internal joint forces. FourBar Mechanism Inverse Dynamics Analysis
First, can we simplify and solve the problem linkbylink, like the single rotating link? Count # of unknowns and # of equations:
Step 1. Position, Velocity, and Acceleration Analyses must first be complete. Step 2. Draw the Diagrams: Physical Dynamics Diagram:
State the problem: .134 Free Body Diagrams (FBDs): F ij : r ij : Step 3.
Derive the NewtonEuler Dynamics Equations.135 Step 4. Newton's 2nd Law: Euler's Equation: Count # of unknowns and # of equations: .
Derive XYZ scalar equations from the vector equations. .136 Step 5.
RHS {b} dependent on inertial terms. and given external forces and moments. .137 Write these equations in matrix/vector form: −1 0 r12Y 0 0 0 0 0 0 0 −1 − r12 X 0 0 0 0 0 0 1 0 − r32Y −1 0 r23Y 0 0 0 0 1 r32 X 0 −1 − r23 X 0 0 0 0 0 0 1 0 − r43Y −1 0 r34Y 0 0 0 0 1 r43 X 0 −1 − r34 X 0 0 0 0 0 0 1 0 −r14Y 0 0 0 0 0 0 0 1 r14 X m2 AG 2 X 0 F21 X F m2 ( AG 2Y + g ) 0 21Y IG 2 Zα 2 F32 X 1 m3 AG 3 X − FE 3 x 0 F32Y F43 X = m3 ( AG 3Y + g ) − FE 3 y 0 I α −r F +r F −M 0 F43Y E3 G 3Z 3 E 3 x E 3 y E 3 y E 3 x F m4 AG 4 X − FE 4 x 0 14 X m4 ( AG 4Y + g ) − FE 4 y 0 F14Y τ I α − r F + r F − M 0 2 G 4 Z 4 E 4 x E 4 y E 4 y E 4 x E4 [ A]{v} = {b} Coefficient matrix [A] dependent on geometry (kinematics solutions). gravity.
we can calculate the vector shaking force and moment. Solve for the unknowns Simultaneous matrix solution: {v} = [ A]−1{b} Actually. using Gaussian elimination is more efficient and robust. which is the force/moment reaction on the ground link due to the motion: .138 Step 6. Calculate Shaking Force and Moment After the basic inverse dynamics problem is solved. Step 7. Solution to internal forces and input torque are contained in the components of {v}.
See Fig. Here is the general link 3 diagram for these derivations: . how do we calculate the various terms that appear in the dynamics equations? These all must be derived from given information.139 Details for the general fourbar mechanism model The inverse dynamics problem has been derived analytically for the fourbar mechanism. P112. Now. Let us do link 3 terms (next page).
140 Link 3 details: r23 X r 23 = = r23Y r r 43 = 43 X = r43Y AG 3 X A G3 = = AG 3Y FP3 X = F P3 = FP3Y R P3 X = R P3 = R P3Y M E 3 = given .
1 0. 0.15 Y (m) 0. of the actual rectangular link for inverse dynamics.2 0.05 0. The triangle tip we have been using all along in Term Example 1 is actually the CG.1 X (m) 0.5”.25 0.2 0.3 0.1 0.141 Figure for Term Example 1 Inverse Dynamics Example starting on the next page: The coupler link 3 is a rectangle of dimensions 8” x 6” x 0.3 .05 0 0.1 0 0.
All moving links are wood. and θ1 = 10. m3 = 0. m4 = 0. with one crucial difference: the input angular velocity was too low for interesting dynamics.9 . calculate the four vector internal joint forces.3 .142 Fourbar mechanism inverse dynamics example: This is the mechanism from Term Example 1 (open branch). θ 4 = 121. link 3 has rectangular dimensions 8 by 6 by 0. r4 = 0.7 .9 × 10−6 .09 . RG3 = 0. M S for this instant (snapshot). so I changed it from ω2 = π to ω2 = 20 rad/s. δ 3 = 36.076.8 × 10−3 . with mass density ρ = 0. given this mechanism and motion. α 2 = 0 . α 3 = 8. This is the open branch of the position. velocity.203. Links 2 and 4 have rectangular dimensions 0. and the shaking force and moment F S . θ3 = 53. r2 = 0.015 .127 m. ω3 = −8.73 rad/s.4 rad/s2.50 by ri (in).03 lbm/in3.284. ω4 = −3.036 kg and I G 2 Z = 7.5 (in).75 by 0. as shown on the previous page. and acceleration example (Term Example 1). . α 4 = 244. Given r1 = 0. ω2 = 20 . Snapshot Analysis (one input angle) At θ 2 = 30 .327 . All external forces and moments are zero but gravity is included. I G 3 Z = 1.8 .65 .178. I G 4 Z = 9. r3 = 0.5 × 10−5 kgm2. The calculated inertia parameters are m2 = 0. θ 2 = 30 . the driving torque τ 2 .
80 FS = N 4.019 0.20 F 10.52 τ 2 − 0. Nm F 5.034 1 F32 x 0 0 F32 y − 8.61 43 y F14 x − 3.037 0.076 0.047 0.019 0.033 − 0.015 0 F14 x − 0.095 0 τ 2 0.99 F32 y 10.033 0 0 0 0 0 −1 0 1 0 0 0 0 0 0 0 −1 0 1 0 0 − 0.002 − 0.076 0.60 F14 y 5.638 0 F14 y − 0.047 0 F21x − 0.43 9.122 0 0 0 0 0 −1 0 0 0 0 1 0 0 0 0 0 −1 0 1 0 0 0 0 0 0.143 0 1 0 0 0 0 0 −1 0 −1 0 1 0 0 0 0 − 0.56 ˆ M S = −1.96 N.68k Nm .497 0 F43 y 0.08 21 y F32 x 5.955 0 F43 x = − 4.11 {v} = F43 x = − 2.0233 Solution by Gaussian elimination or: {v} = [ A]−1{b} Snapshot Answer: F21x 6.127 − 0.202 0 F21 y 0.
Also plotted is the average torque (green) τ2AVG = 0 and the rootmeansquare torque value (blue) τ2RMS = 0.R.36 Nm.8 0.O.) Analysis A more meaningful result from inverse dynamics analysis is to report the unknowns for the entire range of mechanism motion.2 0. The plot below gives the required driving torque τ 2 (Nm) for all 0 ≤ θ 2 ≤ 360 .144 FullRangeOfMotion (F. This plot shows the torque (red) that must be supplied in all configurations by an external DC servomotor to cause the specified motion. Tau2 (red) with average (green) and RMS (blue) 0.M.4 (Nm) 0.2 0 τ 2 0. assuming the given ω 2 = 20 rad/s is constant (Remember: this has been changed from the ω 2 = π rad/s in the kinematics examples!).6 0. for the Term Example 1 mechanism.4 0 50 100 150 200 θ 2 (deg) 250 300 350 .
In the force plot. the X component is red and the Y green. respectively. Shaking Force. X (red) and Y (green) 10 5 FSX and FSY (N) 0 5 10 0 50 100 150 200 θ 2 (deg) 250 300 350 .145 The plots below give the shaking force F S (N) and shaking moment M S (Nm) results. for all 0 ≤ θ 2 ≤ 360 .
5 1 MS (Nm) 0.5 1 1.5 0 0.5 2 0 50 100 150 200 θ 2 (deg) 250 300 350 .146 In the shaking moment plot. there is only the Z component: Shaking Moment 2 1.
Step 2. Position. and Acceleration Analyses must first be complete. In fact. Draw the Diagrams: Physical Dynamics Diagram: Free Body Diagrams (FBDs): F ij : internal force of link i acting on link j r ij : moment arm pointing to link i from the CG of link j .147 SliderCrank Mechanism Inverse Dynamics This problem is very similar to the fourbar mechanism inverse dynamics problem. links 2 and 3 are handled identically! Step 1. Velocity.
There are two kinematic constraints on the slider: Newton's 2nd Law: Euler's Equation: Count # of unknowns and # of equations: We need an additional equation: .148 Step 3. Derive the NewtonEuler Dynamics Equations. links 2 and 3 are identical so let us focus on link 4. the slider. Again. State the problem: Step 4.
.149 Step 5. Solve for the unknowns Simultaneous matrix solution: {v} = [ A]−1{b} Actually. using Gaussian elimination is more efficient and robust. also eliminate one equation ( ∑ M G 4 = I G 4 Zα 4 ). Solution to internal forces and input torque contained in the components of {v}. RHS {b} dependent on inertial and statics terms. Derive XYZ scalar equations from the vector equations and beam these equations into matrix/vector form. Substitute the friction constraint to eliminate one unknown (F14X). −1 0 r12Y 0 0 0 0 0 0 −1 − r12 X 0 0 0 0 0 1 0 − r32Y −1 0 r23Y 0 0 0 1 r32 X 0 −1 −r23 X 0 0 0 0 0 1 0 − r43Y −1 0 0 0 0 0 1 r43 X 0 −1 m2 AG 2 X 0 F21 X F m2 ( AG 2Y + g ) 0 0 21Y IG 2 Zα 2 0 1 F32 X m3 AG 3 X − FE 3 x 0 0 F32Y = F43 X m3 ( AG 3Y + g ) − FE 3 y 0 0 0 0 F43Y I G 3 Z α 3 − rE 3 x FE 3 y + rE 3 y FE 3 x − M E 3 m4 AG 4 X − FE 4 x ± µ 0 F14Y m4 g − FE 4 y 1 0 τ 2 0 [ A]{v} = {b} Coefficient matrix [A] dependent on geometry (kinematics solutions). Always choose proper sign of µ to be opposite to the current x direction. Step 6.
Figure for example starting on the next page: The slidercrank mechanism is shown at the starting (and ending) position.1 0.2 0. 0. with zero input angle θ2.1 X (m) 0.05 0.3 . Calculate Shaking Force and Moment After the basic inverse dynamics problem is solved. we can calculate the vector shaking force and moment.2 0.1 0 0.15 0.15 0.2 0.05 Y (m) 0 0. which is the force/moment reaction on the ground link due to the motion.25 0.150 Step 7.1 0.
. calculate the four vector internal joint forces. IG3Z = 1.686 rad/s. velocity.2 . α 2 = 0 . and the shaking force and moment F S . link 4 has rectangular dimensions 0. the driving torque τ 2 . x = −0. Snapshot Analysis (one input angle) At θ 2 = 30 . All moving links are wood. m3 = 0.2 (coefficient of friction between piston and wall). and ω2 = π / 2 . We assume µ = 0. Links 2 and 3 have rectangular dimensions 0. ω3 = −0.50 by 3 (in).015. and θ 2 = 30 . with mass density ρ = 0.75 by 0.041. h = 0.076 m. and acceleration example (Term Example 2). θ3 = 7. gravity is included but all other external forces and moments are zero. m4 = 0.819e005. M S for this instant (snapshot). in this case keeping the low input angular velocity ω2 = π 2 rad/s so the previous snapshots and fullrangeofmotion results still apply.062 m/s.50 by ri (in). This is the right branch of the position.203.290 m. x = 0. x = −0.418e004 (kgm2). r3 = 0. given this mechanism and motion. α 3 = 0. The calculated inertia parameters are m2 = 0. directed horizontally to the left.681 rad/s2.03 (lbm/in3). There is a constant external force of 1 N acting at the center of the piston.102.151 Slidercrank mechanism inverse dynamics example: This is the mechanism from Term Example 2 (right branch only).75 by 0.020.329 m/s2. (kg) and IG2Z = 1. Given r2 = 0.
938 F32Y 0.013 0.025 0.081 F14Y 0.231 τ 2 0.011 −0.053k (Nm) .044 0.2 0 F14Y 0.935 F 0.002 −1 0 −1 0 1 0 0 0 0 F21Y 0.949 F43Y 0.152 0 1 0 0 0 0 0 F21 X 0.150 −1 0 Solution by Gaussian elimination or: {v} = [ A]−1{b} Snapshot Answer: F21 X 0.0001 0 0 0 0 0 0 0.044 0 0 0 1 F32 X 0 0 −1 0 1 0 0 0 F32Y 0.013 0.199 0. Nm) F43 X 0.101 0.025 0.398 0 0.101 0 0 F43Y 0.982 FS = (N) −0.318 {v} = = (N.517 21Y F32 X 0.748 ˆ M S = −0.995 −1 0 0 0 0 1 0 τ 2 0.011 0 = 0 0 0 −1 0 1 0 0 F43 X 0.
05 0. Also plotted is the average torque (green) τ2AVG = 0. right branch only.O. This plot shows the torque (red) that must be supplied in all configurations by an external DC servomotor to cause the specified motion.) Analysis A more meaningful result from inverse dynamics analysis is to report the unknowns for the entire range of mechanism motion. The plot below gives the required driving torque τ 2 (Nm) for all 0 ≤ θ 2 ≤ 360 .R.1 0 50 100 150 200 θ2 (deg) 250 300 350 .05 0 0.15 0. for the Term Example 2 slidercrank mechanism. assuming the given ω2 = π 2 rad/s is constant.1 τ2 (Nm) 0. Actuator Torque 0.086 Nm.M.002 and the rootmeansquare torque value (blue) τ2RMS = 0.153 FullRangeOfMotion (F.2 0.
the X component is red and the Y green.8 FSX and FSY (N) 0. for all 0 ≤ θ 2 ≤ 360 . X (red) and Y (green) 0.154 The plots below give the shaking force F S (N) and shaking moment M S (Nm) results. In the force plot.85 0.7 0.75 0. respectively.9 0.05 0 50 100 150 200 θ2 (deg) 250 300 350 . Shaking Force.95 1 1.
there is only the Z component: Shaking Moment 0.06 0.04 MS (Nm) 0.04 0.02 0.08 0.06 0 50 100 150 200 θ2 (deg) 250 300 350 .02 0 0.155 In the shaking moment plot.
but much more expensive and difficult to produce.156 Cam Introduction Chapter 8 Applications Compared to linkages. easier to design desired motion with cams. Cam Classification: Disk cams with followers .
we can satisfy function generation at infinite points. allows both Function Generation The output parameter is a continuous function of the input parameter. 4. For example. usually. we can only satisfy a function exactly at a finite number of points: 3. . With linkages.157 Degrees of Freedom rolling and sliding. or 5. S = f (θ ) φ = f (θ ) Cam input angle is θ . a 4bar linkage: θ 4 = f (θ 2 ) With a cam and follower mechanism. Recall a cam joint has twodof. output is S for reciprocating (translating) and φ (rotating) for oscillating follower.
acceleration. and dynamics behavior. design the cam. With cams we must consider mechanism synthesis for the first time: given the motion requirements (follower motion and timing with input angle). we have been mostly concerned with mechanism analysis: given a mechanism design and its input parameters.158 Cam Motion Profiles Up to this point. determine the position. Classification: When the motion transitions between different motion functions. velocity. we must ensure “smooth” motion. . The first step is to determine a “smooth” cam follower motion profile.
Cyclical impulse hammering when acceleration is not continuous. vibrations.159 Fundamental Law of Cam Design: Which means: If the Fundamental law of Cam Design is satisfied. . etc. the resulting dynamic performance will be acceptable for highspeed cam/follower operation. there will be performance degradation due to noise. If not. high wear.
we are only given total motion range and perhaps some timing requirements. the slopes of the position and velocity curves must match across all junctions. Check out Examples 81 (terrible) 82 (bad) 83 (acceptable) Slope of a function is the value of its derivative at a point. Position is automatically matched by shifting axes. The slope of the acceleration can be discontinuous (leading to finite jumps in jerk). Therefore. We can plot vs. for continuous velocity and acceleration curves. θ = ωt ). time or vs. time to graphically see if the Fundamental Law of Cam Design is satisfied for candidate curves. .160 S V A J Diagrams In synthesis. input cam angle θ (assuming constant angular velocity. It is the engineer’s job to determine the position curves and to match the velocity and acceleration across junctions. Draw S V A J diagrams vs. but the acceleration itself must be continuous.
just shift the θ and S axes.161 Generic Cam Follower Motion Profile figure: Define each separate function so the value is zero at the initial angle.shift S axes. Match S: Match V: easy. Then to put the whole thing together. Match A: slope of V must match across junctions. . just do it . slope of S must match across junctions. which is zero.
162 Cam Follower Motion Profile Examples Example 1 rise – dwell portion. . Specify Parabolic (constant acceleration) to Straight Line (constant velocity) rise. 1 2 S: f1 (θ1 ) = A0θ1 f 2 (θ 2 ) = V0θ 2 f 3 (θ 3 ) = 0 2 V: A: J: Match S at junction B: Just shift axis up. followed by a dwell. Match V at junction B: Try to match A at junction B: Plot on next page.
163 Example 1 Plots .
S: θ πθ 1 f1 (θ1 ) = L1 1 − sin 1 f 2 (θ 2 ) = V0θ 2 β1 β1 π V: A: J: Match S at junction B: Just shift axis up. Specify HalfCycloidal function (sinusoidal in cam angle) to Straight Line (constant velocity) rise. .164 Cam Follower Motion Profile Examples Example 2 Fix rise portion only. Match V at junction B: Match A at junction B: Plot on next page.
165 Half Cycloid 2 S (m) 1 20 40 60 80 V (m/deg) 0 0 0.05 J (m/deg ) A (m/deg2) 0 4 x 0 10 5 20 40 60 80 0 5 x 0 10 2 0 2 0 20 40 60 80 3 20 40 θ (deg) 60 80 .
. it satisfies the Fundamental Law of Cam Design.166 Cam Follower Motion Profile Examples Example 3 Specify FullCycloidal function (sinusoidal in cam angle). This will rise all the way to meet a dwell smoothly. θ 1 2πθ1 f1 (θ1 ) = L1 1 − sin β1 β1 2π L1 2πθ1 1 − cos β1 β1 S: f 2 (θ 2 ) = 0 V: v1 (θ1 ) = v2 (θ 2 ) = 0 A: a1 (θ1 ) = 2π L1 2πθ1 sin 2 β1 β1 a2 (θ 2 ) = 0 J: 4π 2 L1 2πθ1 j1 (θ1 ) = cos β13 β1 j2 (θ 2 ) = 0 Plot on next page.
167 Full Cycloid 2 S (m) 1 20 40 60 80 V (m/deg) 0 0 0.05 J (m/deg ) A (m/deg ) 0 3 x 0 10 2 0 2 4 x 0 10 2 0 2 0 2 20 40 60 80 20 40 60 80 3 20 40 θ (deg) 60 80 .
V. A curves. we now have continuous S.168 Analytical Cam Synthesis Disk Cam with Radial FlatFaced Follower Assume a valid cam motion profile has been designed according to the Fundamental Law of Cam Design. Given the motion profile. determine the cam contour.e. i. DCRFFF Figure: . cam angle θ ? We will use kinematic inversion to simplify the synthesis. Is it as simple as polar plotting of S = f (θ ) vs.
.y)) is: where C is the minimum cam radius.169 As seen in the figure.y and the cam angle through geometry: Notice that: To calculate the follower flatface length. double the maximum of L from above. Doubled because by symmetry the contact point will change to the other side at θ = 180 . The radius R and the flatface length L can be related to the contact point x. a design variable. and S = f (θ ) is the given motion profile. the radius R out to the flatfaced follower (not to the point of contact (x.
170 To summarize thus far: This is sufficient to manufacture the cam. If we want to know the cam contour in Cartesian coordinates. it is machined with θ. L coordinates. which means [ A]−1 = [ A]T . R. The solution is: . In matrix form: The coefficient matrix [A] is orthonormal. y. we must solve the relationships for x.
171 Minimum Cam radius to Avoid Cusps A cusp is a point in the cam. to avoid cusps on the entire cam contour. . The condition is that for a finite ∆θ . or actually undercut. there is no change in x. this is to be avoided for good motion. d2 f >0 C + f (θ ) + 2 dθ Note C is always positive and f (θ ) starts and ends at zero and never goes negative. y: dx dy = =0 will cause a cusp. dθ dθ dx df df d2 f = −(C + f (θ )) sin θ + cos θ − cos θ − sin θ 2 dθ dθ dθ dθ dy df df d2 f = (C + f (θ )) cos θ + sin θ − sin θ + cos θ 2 dθ dθ dθ dθ dx d2 f sin θ = − C + f (θ ) + dθ dθ 2 d2 f dy cos θ = C + f (θ ) + dθ dθ 2 dx dy = = 0 simultaneously only when: dθ dθ d2 f =0 C + f (θ ) + 2 dθ Therefore.
Each of these four motion steps occurs for 90 deg of cam shaft rotation. The jerk is not continuous. a fullcycloidal return (total fall 50 mm). this satisfies the Fundamental Law of Cam Design because the position. and acceleration curves are continuous. followed by a high dwell. and then a low dwell. velocity. but it remains finite over all cam angles. 50 S 0 0 50 0 V 50 100 150 200 250 300 350 50 0 100 0 100 0 500 0 500 0 J A 50 100 150 200 250 300 350 50 100 150 200 250 300 350 50 100 150 θ 200 250 300 350 . The cam motion profile associated with this specification is shown below.172 Disk Cam with Radial FlatFaced Follower Design Example Specify a fullcycloidal rise (total lift 50 mm). Clearly.
173 Choosing a minimum cam radius of C = 100 mm). the resulting cam contour is shown below. 150 100 Y (mm) 50 0 50 100 150 100 50 0 X (mm) 50 100 .
250 200 C+S+A 150 100 50 0 0 50 100 150 θ 200 250 300 350 . To avoid cusps in this cam. so this cam design is acceptable. we require that: C + S (θ ) + A(θ ) = C + f (θ ) + d2 f dθ 2 >0 As seen in the plot below. this inequality is satisfied for the entire range of motion.174 Let us check the cusp avoidance plot.
torque. Gears used in electromechanical systems. Can openers to Aircraft carriers.belt and chain drives. Change in angular velocity. Applications: Gear Classification External Spur Gears Internal Spur Gears . Related mechanisms . vehicles. etc. toys.175 Gear Introduction Transfer motion between rotating shafts in machinery. direction.
176 Rack & Pinion Bevel gears Helical (Parallel Shaft) Helical (Crossed Shaft) Herringbone Gears Gear Train .
177 Automotive Differential Planetary Gear Train Worm and Gear Harmonic Gearing .
uk/ “The harmonic gear allows high reduction ratios with concentric shafts and with very low backlash and vibration. placed inside a circular thickwalled rigid ring machined with internal splines.178 Harmonic Gearing Taken from: http://www.” . The external flexible spline has two fewer teeth than the internal circular spline.co. The two contact points rotate at a speed governed be the difference in the number of teeth on the two splines This method basically preloads the teeth. It is based on a very simple construction utilising metals elastomechanical property. which reduces backlash.” “Harmonic drive transmissions are noted for their ability to reduce backlash in a motion control system.roymech. An elliptical cam enclosed in an antifriction ball bearing assembly is mounted inside the flexible cup and forces the flexible cup splines to push deeply into the rigid ring at two opposite points while rotating. How they work is through the use of a thinwalled flexible cup with external splines on it lip.
teeth in contact allow rolling and sliding. twodof. Like two cylinders rolling without sliding. Gear ratio is a measure of the degree of reduction and increase. A robot joint needs lower rotation speed but high torque. .179 Gear Ratio Common electric motors have high speed but low torque. gear 2 is output. A gear train can accomplish both objectives – reduce speed and increase torque. Gear 1 is input. Arc lengths are equal. Simple spur gear pair: DOF: A gear joint is like a cam joint.
For standard spur gears. the radii are directly proportional to the number of teeth: For relating angular velocities.180 Define gear ratio n: Radii inversely proportional to angular motion. Most gear applications have constant angular velocities. for accelerating up to (or down from) constant angular velocities: . tangential velocities are equal.
and shaft torques. angular velocities. . and angular accelerations.181 For relating shaft torques. The ratio of the number of teeth is inverse proportional to the shaft angles. interface forces are equal. Radii directly proportional to shaft torques Summary: The ratio of the number of teeth is directly proportional to the radii. diameter.
182 If n > 1: ω 2 < ω1 τ 2 > τ1 Output has reduced speed Output has increased torque This is the electric motor / robot joint case. If n < 1 : ω 2 > ω1 τ 2 < τ1 Output has increased speed Output has reduced torque If n = 1: ω 2 = ω1 τ 2 = τ1 Output speed and torque unchanged direction reverses (external spur gears) .
79 28 0.78 0.29 0.58 0.38 0. Mountain Bike Transmission N OUT N R ω F τ R = = = N IN NF ωR τ F Gear Ratios: n= Schwinn Sierra Rear 14 16 18 22 26 30 48 0.57 0.42 0. Gear toy 2.62 Front 38 0.68 0.50 0.93 1.47 0.46 0.33 0. mountain bike gearing generally: • increases angular velocity • decreases torque .183 Gear ratio examples 1.37 0.54 0.64 0.07 Unlike electric motor example.
38 0.64 0.2913:1 0.88 1.82 0.44 0.34 0. .34 0.95 1.64 recumbent Rear Rear hub 1:1 0.82 0.36 0.25 0.18 But considering the difference in wheel sizes (26” Cannondale.27 0.62 0.38 0.53 instead of 1.75 0.27 1.09 1.00 22 0.50 0.54 0.45 BikeE Front 34 11 13 15 18 21 24 28 1.88 1.66 0.25 0.01 1.56 0.73 Front 32 0. the effective BikeE high and low gear ratios are: Stiff: 0. 20” BikeE).32 instead of 0.54 0.44 0.18 Original front was 46 teeth – changed for more granny gear.55 0.30 0.184 Cannondale M400 Rear 11 12 14 16 18 21 24 28 32 44 0.48 0.32 0.48 0.46 0.73 0.64 0.25 Granny 1.70 0.7:1 0.41 0.41 0.76 0.32 0.53 0.50 0.63 0.55 0.
Rightmost is the output gear. Let us calculate the overall gear ratio.185 Gear Trains and Gear Standardization Simple Gear Trains Mesh any number of spur gears. nGT ω IN = ω OUT Example: We can find the overall gear ratio by canceling neighboring intermediate angular velocities: Each term in the above product may be replaced by its known number of teeth ratio: . Leftmost is driving gear. All intermediate gears are first the driven gear and then the driving gear as we proceed from left to right.
Number of teeth effect cancels out. the intermediate gears are idlers. so: We could have done the same with pitch radii instead of number of teeth because they are in direct proportion: So.186 All intermediate ratios cancel. for external spur gear trains: Odd # of gears: Output same direction as input Even # of gears: Output opposite direction as input . but do change direction! We should have included sign: So.
where the driving and driven gears are distinct. so: . See the figure. because each pair is rigidly attached to the same shaft. we use the equation: But now the gears rigidly attached to the same shaft have the same angular velocity ratio. nGT = ω IN ω OUT Example: Again.187 Different case: Mesh any number of spur gears. let us calculate the overall gear ratio. Again.
for external spur gear trains: Odd # of pairs: Output opposite direction as input Even # of pairs: Output same direction as input .188 General formula: Again. we must consider direction: So.
The actual gear teeth both roll and slide (twodof joint). Velocity Ratio Torque Ratio (Mechanical Advantage) The author’s velocity ratio is the inverse of our gear ratio definition and his torque ratio is the same as our gear ratio. you cannot look on a gear to see them.189 Involute Spur Gear Details and Standardization Rolling Cylinders Mating spur gears are based on two pitch circles rolling without slip. . Velocity ratios in a linkage vary wildly over the range of motion. These are fictitious circles. Fundamental Law of Gearing: From our study of linkage velocity. we know this is no easy feat.
the angular velocity ratio will still be constant to satisfy the Fundamental Law of Gearing. The involute is a curve generated by unwrapping a taut sting from a circle: . etc.190 Involute Function Standard spur gears have an involute tooth shape. thermal expansion. wear.). If the gears’ center distance is not perfect (tolerances.
Point of contact slides along this line. pure rolling in contact Contact point between the two pitch circles Pressure Angle: Angle between the common normal (also called axis of transmission) of the two meshing teeth and the velocity of the pitch point (tangent to both pitch circles). pressure angle relationship: .191 Base Circle: Pitch Circle: Pitch Point: Involute starts from this circle Fictitious circle. pitch circle. Base circle. Similar angle is defined for cams and followers.
Beginning of contact is when tip of driven gear tooth intersects the axis of transmission. . End of contact is when tip of driving gear (pinion) tooth intersects the axis of transmission. For harmonic gearing. Only one or two teeth are in contact at any one time.192 Length of contact along axis of transmission. many teeth are in contact at any one time (higher gear ratio in a smaller package).
does not affect angular velocity ratio. increases the pitch circle radii. How is this possible? Relationship from last page: . but doesn’t change the base circles (of course).193 Increasing center distance increases the pressure angle. Thanks to the involute tooth shape.
However. All real gears must have some backlash due to tolerances. On earth gravity tends to load the backlash for predictable effects.194 Backlash Clearance. Changing direction. Nonlinear effect in robots. In space however. wear. and then impact . thermal expansion. etc. the backlash is less predictable! Figure: . must minimize backlash for smooth operation. nothing happens until the backlash is moved.bad for dynamics. Distance between mating teeth measured along the pitch circle circumference. Example: robot joints which must be driven both directions.
must have: 1) the same pressure angle. Not interchangeable with US gears because different tooth proportion standards! Circular Pitch: . and 3) be made with standard tooth proportions (Table 91. Diametral Pitch: Module: Module is the metric version of diametral pitch. 2) same diametral pitch. 441).195 Gear Standardization To allow interchangeability in manufacturing and to allow meshing of different size gears (radii and number of teeth) to achieve desired gear ratios. For two spur gears to mesh. p.
p. Addendum is radial distance from pitch circle to Top Land of tooth. see Table 91.196 Standard involute tooth proportions. Dedendum is radial distance from pitch circle to Bottom Land of tooth (not to base circle). Clearance is radial distance from Bottom Land to mating gear Top Land (kinda like radial backlash). Face width is thickness of tooth and gear (mating widths needn’t be the same). Related to the circular pitch and backlash by: .. 441. Tooth thickness is the circumferential length of each tooth.