o For every degree of freedom that a joint eliminates, a constraint equation must be
written. The following paragraphs outline the form of the constraint equations for some
of the common joints.
pG pG CB q2 CB r2
2 1 1 2
or
p p C
G2 G1 B1
q2 CB2 r2 0 (1)
1/5
o Eq. (1) represents a set of three scalar constraint equations that eliminate the three
translational degrees of freedom between the bodies.
o For incorporation into a set of equations of motion, the constraint equations may be
differentiated twice so they are in the form of second order differential equations.
Using inertial angular velocity components gives
0 vG vG CB q2 CB r2
2 1 1 2
vG2 vG1 B1 CB1 q2 B2 CB2 r2
vG vG B q2 B r2
2 1 1 2
vG vG q2 B r2 B
2 1 1 2
0 aG aG q2 B q2 B r2 B r2 B
2 1 1 1 2 2
(2)
0 vG vG CB q2 CB r2
2 1 1 2
vG2 vG1 CB1 B 1 q2 CB2 B 2 r2
vG vG CB q2 B CB r2 B
2 1 1 1 2 2
0 aG aG CB q2 B CB q2 B
2 1 1 1 1 1
CB2 r2 B2 CB2 r2 B2
or
0 aG aG CB q2 B CB B q2 B
2 1 1 1 1 1 1
(3)
CB2 r2 B2 CB2 B2 r2 B2
2/5
Spherical Joint: Relative Coordinates
s2 s2 s2 0 (4)
f B1 B1B2 f B1 RB2 RB1 0 or
f B2 B1B2 f B2 RB2 RB1 0 (5)
g B1 B1B2 g B1 R
B RB
2 1
0 or g B2 B1B2 g B2 B R
R
2
0
B1 (6)
o Initial conditions are used to ensure the alignment of hB2 with hB1 .
o Using inertial angular velocity components the first of Eqs. (5) and (6) may be written
B B CB f B
T T
0 f B1 2 1 1 1 B2 B1
f C
T T
B1
B1 B2 B1
3/5
and
0 g B1
T
B2 B1
g C
T T
B1 B1 B2 B1
0 f B1 CB1 f C
T T T T
B2 B1 B1 B1 B2 B1
f B1 CB1 f C
T T T T
B2 B1 B1 B1 B1 B2 B1
f B1 CB1 f C
T T T T T
B2 B1 B1 B1 B1 B2 B1
and
0 g B1 CB1 g C
T T T T
B2 B1 B1 B1 B2 B1
g B1 CB1 g C
T T T T
B2 B1 B1 B1 B1 B2 B1
g B1 CB1 g C
T T T T T
B2 B1 B1 B1 B1 B2 B1
0 f B1 CB1 f C
T T T T T
B2 B1 B1 B1 B1 B2 B1 (7)
0 g B1 CB1 g C
T T T T T
B2 B1 B1 B1 B1 B2 B1 (8)
0 f B1 CB2 / B1 B 2 B 1 f B1 CB1 CB2 B 2 B 1
T T T
0 g C g C
T T T
B1 B2 / B1 B2 B1 B1
B1 CB2 B2 B1
0 f B1 CB1 CB2 B 2 B1
T T
f C
(9)
CB2 B 2 B 1 CB1 CB2 B 2
T T T
B1 B1
4/5
0 g B1 CB1 CB2 B 2 B1
T T
g C
(10)
CB2 B 2 B1 CB1 CB2 B 2
T T T
B1 B1
o The translation constraints are the same as the spherical joint as given in Eq. (4).
o To derive the additional rotational constraint equations, we start with the same set-up
used for absolute coordinates. That is, let hBi be a vector fixed in Bi parallel to the
hinge joint, and let f Bi and g Bi be vectors fixed in Bi that are perpendicular to the
hinge joint (and, hence, hBi ). Then, the rotational constraint can be expressed directly in
terms of the angular velocities as follows
f B1 B1 B2 f B1 B2 0 or f B2 B1 B2 f B2 B2 0 (11)
g B1 B1 B2 g B1 B2 0 or g B2 B1 B2 g B2 B2 0 (12)
o As before, initial conditions are used to ensure the alignment of hB2 with hB1 .
f 0
T
B1 B2 (13)
g 0
T
B1 B2 (14)
f 0
T
B2 B2 (15)
g 0
T
B2 B2 (16)
5/5