The Mark Ortiz Automotive


September 2013

Reproduction for free use permitted and encouraged.

Reproduction for sale subject to restrictions.  Please inquire for details.





Mark Ortiz Automotive is a chassis consulting service primarily serving oval track and road racers. This newsletter is a free service intended to benefit racers and enthusiasts by offering useful insights into chassis engineering and answers to questions.  Readers may mail questions to: 155 Wankel Dr., Kannapolis, NC 28083-8200; submit questions by phone at 704-933-8876; or submit questions by    e-mail to:  Readers are invited to subscribe to this newsletter by e-mail.  Just e-mail me and request to be added to the list.





When we run a roll test on the K&C rig, we roll the chassis through some pre-determined roll angle.  As we do so and weight transfers, the loading tires naturally compress and the unloading tires rebound.  Based on very accurate measurement of the wheel center heights in this test, we are able to calculate a “suspension roll angle” which subtracts the roll angle due to tire compression away from the total chassis roll angle.  Currently, we generate a plot of camber angle vs. suspension roll angle.  However, I had one customer suggest that the camber angle also needs to be corrected, so we need to report a “suspension camber angle” which would be the absolute camber angle minus the tire compression roll angle.  This may make sense because the chassis to which the control arms are attached is rolling through an excess roll angle due to tire compression.  It is further suggested that the “suspension camber angle” found in this way would match up with a rigid body kinematic model of the suspension.  I can see the logic of this but find it curious that no one has mentioned this before.  For example, it’s not something that Anthony Best Dynamics reports in their post-processing.  What do you think?  Does it make sense?


This question comes from my friends up the road at Morse Measurements in Salisbury, NC (  They offer kinematics and compliance (K&C) testing to the general public.  This is a very useful service that was previously only available to (or through) major manufacturers’ engineering departments or, more recently, very well funded race teams.  They occasionally refer me some work, so I’m happy to give them a plug.


The idea that there is a camber change component as well as a roll component due to tire compliance definitely does make sense.  And the tire compliance camber change should be identical to the tire compliance roll angle, since there is no camber recovery in tire compliance roll.


The camber change due to tire compliance, plus suspension roll, minus camber recovery due to geometry, will not necessarily equal measured camber change on the rig.  That’s not necessarily bad, however.  In fact, it’s a cloud with a definite silver lining.  The difference will be camber change due



to non-tire compliances, and that should be a very useful thing for a client to know.  Indeed, one of the main reasons for doing K&C testing is to evaluate compliances.


Since we generally buy our tires, and run them at whatever pressure gives best grip, we generally just live with whatever tire compliance we get.  Trouble is, without rig testing we don’t know what that tire compliance is.  Even with rig testing, we only have a decent approximation, because the tire will act a bit different at speed and temperature on the track than it does sitting still at room temperature on a rig.  But that decent approximation is still much better than no measurement at all.


For most forms of racing, we try to minimize other compliances, especially ones that cause camber change.  In passenger vehicles, we accept camber compliance as a necessary evil, in order to get the noise, vibration, harshness, and stiction reductions that come with rubber bushings.  In either case, we want to know what amount of camber compliance we have, from the pieces that we have some design control over.  By accurately measuring the component due to tire compliance, and subtracting, we can at least know the aggregate compliance from all the other parts.


We can then either use intuition and educated guessing to target areas where we think rigidity could be improved, and test again, or do further rig testing with additional measurement and instrumentation to pinpoint where the biggest or most easily remedied deflections are occurring.





If the drive-line is offset significantly, let's say to the left as in a super modified, is the moment applied to the axle distributed out to the contact patches using the asymmetric drive-line center to contact patch center line distance?

This would yield very dissimilar reaction forces at the left and right contact patches – is this line of thought correct? And if this is the case then what are the implications, if any, on calculating the sprung mass roll reaction and chassis roll moment distribution front to rear for calculating torque roll?


This may be obvious -- or I am just confusing the difference between a force couple and a moment applied asymmetrically to a beam.


The short answer is that the location of the driveshaft or the pinion does not matter at all.  The location of the torque arm is what matters.  Or, to broaden the statement a bit, it is the location and nature of whatever transmits axle forces (both torque and thrust) to the sprung structure that determines what roll moments the system creates.  Location of the drive shaft only matters to the extent that it relates to that.


If the driveshaft exerted a vertical force where it connected to the axle, its lateral location would matter.  But the driveshaft is designed so it can only transmit rotational forces.  If we try to push the


axle laterally or vertically with the driveshaft, the universal joints bend.  If we try to push the axle longitudinally with the driveshaft, the splines slip.


When we have a beam (such as the axle) supported on two points (such as the tires) and we apply a purely rotational force to the beam, it doesn’t matter where along the beam’s length the rotational force is applied.  The sum of the forces at the support points will not change when we apply the rotational force, and the change in their difference will be the rotational force times half their spacing.


The misconception is quite common that the pinion shaft somehow applies a jacking force to the suspension, and therefore its lateral location affects torque roll.  It is true that in a conventional rear end, the pinion shaft does exert an upward force against its bearings.  What people often miss is that there is an equal and opposite downward force at the differential carrier bearings, and the sum of the vertical forces is zero.  There is an offset in the forces’ lines of action, and therefore a couple, and a corresponding moment, or torque.  The torque tries to rotate the axle housing rearward, and there is a corresponding equal and opposite torque forward at the ring gear, axles, and wheels, which propels the car.  The rearward torque on the axle housing acts through the suspension, and the suspension can be arranged to produce roll moments in response.  However, these are not dependent on where the pinion shaft is.


The driveshaft also tries to roll the axle to the left, and roll the sprung structure to the right.  This creates torque wedge and torque roll.  The mechanism used to react axle torque can be arranged to counter the effect of driveshaft torque, or augment it.  But neither wedge change due to axle torque nor wedge change due to driveshaft torque are influenced by the lateral whereabouts of the driveshaft.


In a torque tube rear end, the location of the torque reacting mechanism is inextricably linked to the driveshaft location.  In a torque tube design, axle torque is transmitted to the sprung structure by a

tubular beam surrounding the driveshaft.  There is a large hollow ball, or spherically radiused formation, at the front of the torque tube, which sits in a cylindrical bore, usually at the rear of the transmission.  There is generally only one universal joint, arranged to be concentric with the ball.  


The shaft is then splined to the rear end input shaft, and also has splines at its forward end to absorb plunge.  Since the tube encloses the driveshaft, it has to be where the driveshaft is.


A beam that acts similarly but does not surround the driveshaft is called a torque arm, or sometimes (mainly in the US) a lift bar.  It can have a sliding connection to the sprung structure at its front end, as a torque tube does, but more commonly it has a drop link.  Compared to a ball or a rubber bushing, a drop link has the advantage that it only constrains the arm end vertically, and not horizontally.  A ball or bushing can create rear steer or bind, depending on the geometry of the axle locating linkage.  A rigid drop link is the usual choice in supermodifieds.  In dirt Late Models, a coilover is generally substituted for the drop link.  In IMCA-style modifieds, where coilovers are prohibited, a big coil on a slider is the most common choice.


Whether the drop link is deliberately made compliant or not, its plan view location is what determines the magnitude of the jacking force the arm creates, and where the jacking force acts.  Where the arm attaches to the axle doesn’t matter, nor does the height of the drop link or arm end.


The arm usually attaches to the rear end center section, and runs straight forward alongside the driveshaft.  This is convenient for strength and packaging, but there is no reason the arm can’t angle away from the driveshaft if there is room.  It can also be attached to an axle tube.


And of course we don’t have to use a torque arm at all.  We just have to react the axle torque somehow.


In supermodifieds, the rear axle is generally a closed-tube quick change with a spool and wide-five hubs (five lug studs, on a 10.25” bolt circle).  There are generally two disc brakes, with the calipers on birdcages.  Two parallel, equal-length trailing links run forward from each birdcage to the frame of the car.  The trailing links locate the axle longitudinally, and react braking torque.


The center section of the rear end is by no means in the center.  It is drastically offset to the left.  The driveshaft runs just inboard of the left rear tire.  The driver sits a bit to the right of that.  The driver’s right shoulder is left of the track midpoint.  The engine is alongside the driver’s legs, tilted to the left.


Between the driveshaft and the driver sits the torque arm.  Because the torque arm is so far left, it generates lots of rightward roll and wedge increase under power.  This is basically a good thing, because a car as left-heavy as a supermodified tends to be loose on exit, and a lot of torque wedge counters that.  However, in my limited experience with these cars, the torque arm overcompensates, and the car actually has a power push, which has to be compensated for with the rest of the setup.  The car then is freer than we would really like before the driver applies power.


If I were designing a supermodified, I would use a trailing link or pull bar rather than a torque arm, and provide multiple mounting points for its front pivot, to adjust the angle of the link.