Off Panhard bar

[Gary Derian]

Hi Joe, I would like to respond to your comments.

Low roll centers, 2 to 4 inches above the road, are preferred, yes.

The Rover DeDion does create a small amount of scrub.

My comment about the RX-7 offset watts linkage was not to claim it was the
same as the Alfa a-arm but to comment that offset locating devices, while
maybe not optimal from a kinematic standpoint, could still work well.

I don't know what you think I have not thought out.  A 4 link does have
wonderful flexibility for adjusting anti-squat.  Angling two of the links
creates a virtual a-arm (is this the Satchell Link ?) that laterally locates
the rear axle.  This is similar to the Alfa or Lotus but they used a real
a-arm, not a virtual one.

The problem with a Satchell Link is when the car leans in a corner, the
links bind.  On a road car with rubber bushings, it still all works.  On a
race car with solid joints, the 4 links, when viewed from above, must be
parallel to the longitudinal axis of the car.

A road car type 4 link, like GM, uses the links to locate the axle and
resist axle torque.  If rubber bushings are used to smooth the ride, wheel
hop occurs in high powered cars.  That is why it is better to use separate
mechanisms to locate and to resist axle torque.

Of the 6 degrees of freedom, moving along and rotating about the x,y and z
axes, a suspension needs to fully constrain 5.  For good ride, there should
be some longitudinal compliance (the tire moves back when it hits a bump)
all others constraints should be very rigid.  A 4 link cannot provide this
mix of rigid and compliant constraints.  That is fine for a race car but not
so good for a road car.

Gary Derian <gderian at oh.verio.com>

>
>I agree a solid axle has no jacking. However, a lower rear roll center will
>make a car handle more consistently even with a solid axle. Given a choice,
>I would opt for a low roll center.
>
>If the dead axle is allowed to change length, then the axle track dimension
>(centerline of tire to centerline of other tire) changes with it. If the
>axle track dimension changes then the tires scrub laterally.
>
>An a arm is not a Watts linkage. A Watts linkage resembles a Panhard bar
>with a bellcrank in the center. The bellcrank should be attached to the
>chassis and the two rods attached to either end of the axle. The rods are
>parallel to the axle in their normal position.
>
>You haven't thought this out. With a four link you can vary the angle of
the
>upper and lower links realitive to each other in side view to vary the
>location of the instant center. The distance from the instant center to
axle
>is the length of the effective arm the axle is attached to. This arm is
>infinitely long when the links are parallel. Your trailing arm length is
>12"? 18"?
>
>If I choose, I can place the instant center in side view on an imaginary
>line from the contact point of the rear tire to the intersection of the
>height of the center of gravity and the centerline of the front axle and
>have 100% anti-squat. I will have more rear wheel traction for leaving at a
>light or coming out of a turn and this means I can put more power to the
>road. The only wheel hop problem I have with the four link is brake hop if
I
>make the effective arm length too short chasing too much anti-squat.
>
>Come to think of it, you could reverse the angle of the arms
>
>By the way, looked it up, this is called a Satchell Link. According to Herb
>Adams in Chassis Engineering, the advantages of a Satchell are that it
>provides considerable anti-squat with roll understeer and a low roll
center.
>
>
>Regards,
>
>Joe