The Mark Ortiz Automotive

CHASSIS NEWSLETTER

Presented free of charge as a service

to the Motorsports Community

April 2012

Reproduction for free use permitted and encouraged.

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

 

 

WELCOME

 

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: markortizauto@windstream.net.  Readers are invited to subscribe to this newsletter by e-mail.  Just e-mail me and request to be added to the list.

 

 

ANOTHER ANTI-DIVE IDEA

 

Geometric compliance arguably should improve tire adhesion.  I've not, however, seen any test results that support the concept, have you?

 

The system Lotus has developed seems overly complex and potential unreliable.

 

My thought is to use torsion bar front springs.  Place hydraulic cylinders at the fixed end of the bars rigged to rotate the bars appropriately in response to change in  brake line pressure, countering brake dive and  facilitating geometric compliance.  Your comments?

 

I don’t know of any experiment to test the hypothesis that it improves tire adhesion if the tire can move rearward when it encounters a bump, but it definitely reduces ride harshness.  That means it reduces acceleration and jerk of the sprung mass, and that should reduce load variation at the contact patch.

 

Reduced load variation at the contact patch helps because, to a substantial degree at least, the tire’s frictional performance is limited by those periods when it has the least normal force, and the value of that minimum normal force decreases when load variation increases.

 

It is not impossible to raise the front of the car, or raise the front and lower the rear, using fluid displaced by application of the brake pedal.  However, this requires the driver’s leg to do the work of displacing the system involved, against a load.  The driver’s leg will have to do this in addition to applying the brakes.

 

This means that there will be an increase in pedal travel, pedal force, or both.  If the slave cylinder is in series with the ride spring, and maybe also if it’s in parallel, changes in spring loading will be transmitted to the pedal.  When braking, the driver will feel bumps in the road through the brake pedal.

 

 

It would be possible to have acceptable force and travel at the pedal if the brakes are power-assisted.  However, any power assist for the brakes is necessarily powered in some manner by the engine.  If the brake system also is arranged to extend or compress the suspension, we then have engine power being used to do this, making the system a form of active rather than passive suspension, and therefore illegal in most forms of racing at present.

 

I have recently had another person contact me, wanting to know my opinion of a roll compensating system that operates off the steering system.  I won’t divulge how this person proposed to do this, but it amounts to a similar idea, applied to cornering and roll rather than braking and pitch.

 

Where it’s allowed, active ride height control can definitely be used to modify pitch, heave, roll, and/or warp behavior of a suspension system, with or without the use of additional springing.  This

does offer advantages.  However, in most cases we will want to keep such a system separate from the brakes, steering, or throttle, and have it controlled in response to accelerometer outputs instead.

 

 

EFFECTS OF UNIVERSAL JOINT ANGULARITY ON HANDLING

 

A topic I would like discussed relates to the dismal performance of the Williams F1 car last year.

The car had a low line gearbox to make room for air flow to the rear wing and as a result had severe angularity in the rear driveshafts.  The shaft misalignment is in the vertical plane; that is, the inner ends are about 6 inches lower than hub height.

 

All year the drivers complained about turn-in and mid corner instability.

 

My thoughts are the precession, especially at high speeds, will try pull the suspension into droop.

Turning in means the weight transfer has to fight the precession on the loaded wheel prior to the suspension actually being able to do something.

 

Despite the team claiming they had special c/v joints that would accept the angular drive, I wonder if this design was causing the fundamental problems the team were encountering.

 

What effect would gyroscopic precession in the shafts have on the performance of the suspension? Would it have inhibited suspension operation, giving the problems the drivers complained about?

 

What is notable, this year, is that with a return to a more standard transaxle layout, the Williams GP car is much more competitive.

 

I have my own thoughts about this, as you might have gathered from the questions posed, but thought it might be a good topic to discuss, especially for students involved in FSAE (where highly angled driveshafts are rife and defended by statements like ‘The maker says they are OK up to 15 degrees’).

 

 

I would be surprised if the driveshaft angularity would cause poor turn-in and mid-corner instability.  Reliability problems, maybe.

 

A driveshaft has such a small radius and mass that there are no significant precession forces – certainly not compared to the wheels and brakes, or the crankshaft and clutch.

 

As I understand it, a tripod joint running with angularity induces an axial force, but that’s not properly termed precession.  Precession is the induced force when a rotating object with significant rotational inertia is moved from its previous plane of rotation.

 

I looked for definitions of precession, and those I found didn’t all pertain to gyroscopic effects.  Anyway, what happens with a top’s axis describing a conical path, or a spinning bicycle wheel trying to yaw when it’s rolled, is that when the object’s axis of rotation is given an angular velocity about one axis perpendicular to it, a torque is induced about the other axis perpendicular to it.  For example, if one holds a bicycle wheel at arm’s length and spins it (about a transverse or y axis), then rolls it to the right (about a longitudinal or x axis), the spinning wheel generates a yaw torque to the right (about a vertical or z axis).  When a spinning top starts to lean from gravity, the velocity imparted by gravity induces a moment that makes the top move about the other axis perpendicular to its axis of rotation, and it keeps doing that and as a result, its axis describes a conical path.

 

I don’t know what they teach mechanics in other countries, but what I learned here was that the reason for trying to have equal and opposite angles (or really just equal ones) in the U-joints of a driveshaft – and also making sure the joints are correctly indexed – does not have to do with any gyroscopic effect, but rather has to do with the pulsation or sinusoidal variation in output shaft rotational velocity when a single Cardan joint runs at an angle, with a constant input speed.  If a second joint is in series with such a joint, it is possible to arrange things to get a cancelling pulsation at the second joint, so that the output after that does not pulsate.

 

I believe a traditional U-joint creates a small self-straightening moment, but not enough to materially affect suspension systems.  I read that a tripod joint does not have either of these effects, but does induce a small tension load in a driveshaft.  If the suspension’s front view instant center is below hub height – or more properly, if the driveshaft centerline passes above the suspension’s instant axis – that might produce a small pro-squat effect, but I don’t see how that would lead to poor turn-in or to mid-turn instability.

 

If there is a tension force induced in the driveshafts, that might jack the rear suspension down a little, but it would do that on both sides of the car, so I would not expect wheel loads to be much affected.

 

If a car is tight on entry, and then goes loose more than would reasonably be expected, I look for stiction under cornering load in the front suspension.  If a car is generally unstable under cornering loads and this doesn’t respond to adjustment, I look for something flexing, or binding, or bottoming.

 

 

 

Another thing I’d look at is the differential.  If Williams tried to get everything lower and more compact, what was the diff like?

 

As this relates to FSAE judging, I don’t think I’d penalize a team for having visible angularity in the driveshafts, either in top view or in front view, provided it didn’t look to be enough to create a reliability problem.

 

The 2006 UNC Charlotte FSAE car had very short driveshafts.  They didn’t have a lot of angularity at static condition, but they did when there was significant suspension displacement.  Not only was there significant shaft angularity (some of the time, at least), but there were big changes in shaft angularity.  The effect was sufficient to cause serious reliability problems.  The car had to be set up with just the right combination of track and camber, with the one dependent on the other.  If this was not done exactly right, either a tripod roller would come out of its slot on a bump, failing the joint instantly, or the shaft would bottom lengthwise against its buttons.

 

Despite this joint angularity (and plunge) problem being bad enough to cause serious reliability headaches, the car handled just fine.

 

If highly angled driveshafts are commonplace in a group of cars, and driveshaft angularity does in fact cause handling problems, then the expectation would be that all cars exhibiting this characteristic would have handling problems.  It would be difficult, if not impossible, for any car with pronounced driveshaft angularity to handle well.  I haven’t been to an FSAE event for a few years, but I don’t think serious handling problems are as rife as pronounced driveshaft angularity is, and indeed it would surprise me if there were any correlation at all.