Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

SAE TECHNICAL
PAPER SERIES 2004-01-3534

Un-sprung Weight, the Enemy

that Became a Friend
Peter E. Bryant
VP Engineering AMTECH Corporation

Reprinted From: Proceedings of the 2004 SAE Motorsports

Engineering Conference and Exhibition

Motorsports Engineering
Conference and Exhibition
Dearborn, Michigan
November 30-December 3, 2004

400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760 Web: www.sae.org
 Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

The Engineering Meetings Board has approved this paper for publication. It has successfully completed
SAE’s peer review process under the supervision of the session organizer. This process requires a
minimum of three (3) reviews by industry experts.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise,
without the prior written permission of SAE.

For permission and licensing requests contact:

SAE Permissions
400 Commonwealth Drive
Warrendale, PA 15096-0001-USA
Email: permissions@sae.org
Tel: 724-772-4028
Fax: 724-772-4891

For multiple print copies contact:

SAE Customer Service

Tel: 877-606-7323 (inside USA and Canada)
Tel: 724-776-4970 (outside USA)
Fax: 724-776-1615
Email: CustomerService@sae.org

ISSN 0148-7191
Copyright © 2004 SAE International

Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE.
The author is solely responsible for the content of the paper. A process is available by which discussions
will be printed with the paper if it is published in SAE Transactions.

Persons wishing to submit papers to be considered for presentation or publication by SAE should send the
manuscript or a 300 word abstract to Secretary, Engineering Meetings Board, SAE.

Printed in USA
 Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

2004-01-3534

Un-sprung Weight, the Enemy that Became a Friend

Peter E. Bryant
VP Engineering AMTECH Corporation

Copyright © 2004 SAE International

ABSTRACT transferable to coil spring suspensions for use in motor

racing. The dynamic stability benefits could be very
For years racecar designers and automotive suspension significant, especially in World Rally Championship
engineers have sought to keep the un-sprung wheel (WRC) type racing, where ground clearance is important
/axle component weight of a vehicle suspension and and suspension travel requirements are dictated by off
axle system to a minimum. In the early days, ultra light road conditions. In WRC cars aero ground effects
Magnesium wheels were considered a big break assisted handling is not as readily applicable at low
through. In the mean time, because we strived so hard speeds, so mechanical grip is at a premium.
to reduce un-sprung weight, we tended to consider it Some significant bi-products of this technology are that it
simply as an “Enemy” that had to be dealt with. So it helps control ride height and helps in preventing pitching
may be a little surprising that un-sprung mass could and it does not need any computer control to deploy.
actually become useful in controlling dynamic roll or
pitch, but it can. Tire footprint stability and body roll
recovery during negotiation of “S” bends and chicanes in
racing are the key to faster times through these turns. OPPOSING SPRING THEORY
Because the time element during a cornering maneuver
is so critical in racing, common sense dictates that If you When only the vehicle jounce travel is opposed by a
can control or shut down the initiation of roll by a few spring it is basically a one dimensional suspension
milliseconds, it will make a big difference to the system, because nothing apart from the shock
momentum and thus lessen the total amount of roll absorbers is opposing the rebound travel of the
generated and needed to be corrected by the stabilizer suspension. Opposing springs suspension stabilization
system. This says that good rebound travel control can means that the cars suspension system has changed
seriously enhance recovery. from a single dimensional spring system to a two
This Paper outlines a newly patented rebound control dimensional system.
method intended both for racing and street use that
utilizing the un-sprung weight. With opposing springs, when the vehicle begins to roll in
a corner or pitch up at the rear as it would during heavy
INTRODUCTION braking applications, a rebound control spring is added
to oppose the upward movement of the sprung mass
In 1998 the Author; a consulting automotive engineer, from normal ride height. Basically the rebound spring is
was retained by an automotive aftermarket product tethered to the un-sprung weight and is pulling against
company to design and develop a new aftermarket the sprung mass as it tries to react to the lateral G being
stabilizer that originally was intended for sale to SUV generated. Historically, trying to control the rebound
and pickup truck owners. The purpose of the product travel of the sprung mass in these instances has been
was to improve the on-highway handling and stability of the role taken by the shock absorbers, assisted
SUV, pickups and high CG vehicles and reduce the generally during cornering by anti-roll or sway bars.
propensity of those vehicles to rollover. In the process of
developing the new product, which was strictly intended UNDERSTANDING AND APPLYING REBOUND
for use on street vehicles with rear leaf spring type CONTROL
suspension, the need to address similar problems for
coil spring suspension equipped vehicles arose. The In racing now we commonly refer to two different types
simple technology developed and patented for the leaf of grip or tire adhesion. Mechanical grip and Aero
spring application was based around controlling the enhanced grip. Racing has simultaneously also evolved
sprung mass of the vehicle during roll or pitch events into two basic types of cars. Those with lots of aero grip
using the un-sprung weight as an opposing force to and Go-Cart type stiff suspension, with minimal up or
delay the inception of roll. Being a racing car designer, it down movement of the sprung mass as found in
occurred to me that the use of this technology is easily
 Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

Formula One and Indy type cars, and those that still is mounted around the shock control rod, between the
need ground clearance and suspension travel for shock absorber piston and the top of the shock reservoir
uneven surfaces and ground clearances similar to street where the rod exits the shock body as it is being
cars. As we all know, the main purpose of the extended in rebound. But there is a packaging problem
suspension springs are to support the sprung weight at with doing this. Generally the space above the piston in
the correct ride height with sufficient spring rate to a shock is needed for rebound travel of the piston. If we
control suspension movement to suit our designed put a spring in this space that is stiff enough to help
vehicle dynamics. With the exception of cars like GP control the sprung mass it will probably have a “Solid”
cars with “Flexural” type suspension mountings. Gravity length that takes up too much space. One alternative is
or down force of some kind returns the car to its normal to put the spring on the outside of the shock.
ride height after the suspension system has moved up or
down. (Jounce or Rebound). Flexural suspension Fig. 1. Below shows how a shock absorber can be used
mountings with no swiveling capability between chassis to induce rebound control with a coil spring mounted
and suspension “A” arms, naturally induce rebound outside the body of the shock. As the shock is extended
control simply because the same inherent resistance to it tries to compress the rebound spring.
bending of the solid mounted suspension arms during
jounce happens in rebound. The flexural suspension
wants to always return the sprung mass to a “set” ride
height. But when a car with swiveling type suspension (A
WRC CAR for example) experiences rebound travel, it
generally relies on the rebound control that is induced or
inherent in the shocks to help return it to normal design
height following rebound movement of the suspension.
During cornering or pitching of these cars, the unloading
suspension springs are trying to return to their basic free
length and in fact are “encouraging,” rebound
movement. In racing cars with very stiff suspension
spring rates this often means that during excessive
rebound travel, the suspension springs can become
totally unloaded and sometimes reach their free lengths.
At that point they, and the sprung mass are then actually
“out of control!” In racing it is important that no dynamic
handling component of the car ever be “out of control”
because the time delay experienced in returning the FIG.1
sprung mass to a condition of full control is critical.
Of course, there are many other places that rebound
Rebound control can be applied in many ways. When it control springs can be applied in a racing car
is done through the use of shock absorber technology, suspension system. The important thing about using
unless computer controlled it often has a “Fixed” or separate springs is that they can easily be tailored to suit
constant value and is applied through the design of the the rebound control characteristics needed.
valves and pistons that control the movement of the
hydraulic fluid from one chamber in the shock to Rebound control is very critical to ground effect racing
another. One problem with this is that when a shock cars, because it controls the air gap under the cars and
absorber is traveling from the full jounce position of the prevents it from increasing and causing a dramatic loss
suspension to the full rebound position, any rebound in grip.
control capability is virtually “wasted” until the
suspension reaches the design height of the car. This is Fig 1A below, shows how the rebound control springs
due to the suspension spring being the primary force could be deployed in a formula type single seat racing
opposing any rebound control being exerted by the car. The sketch shows a push rod type suspension on a
shocks. Without some kind of computer controlled partial tub. The inboard coil-over shocks are pushing
damping, the suspension spring rate is usually much against the rocker arm bell crank and the rebound
higher in poundage than any force that the shock could springs are behind the bell crank pushing back. The
typically impose. Thus the rebound control exerted by steering system is not shown and this is not an actual
the shock during jounce to design height suspension tub for a car.
movement, besides being negligible, is really not even
necessary. Rebound control is really only needed during
rebound travel.

Besides using shock valves, one alternative method to

apply rebound control is through the use of a spring that
 Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

Plots of variables were made for each vehicle A and B

as follows.

1. Lateral acceleration (of the vehicles cg in g’s) versus

time.
2. Speed versus time.
3.Tire normal force (Newton’s) versus time. Tire normal
force is the force exerted downward by the tire on the
surface. There are four individual plots, one for each
tire. When the tire force plot goes to zero force, it means
that the tire has lifted of the ground. Simultaneous lift off
of both inside tires indicates incipient rollover.
4. Effective tire friction versus time. There are again four
individual plots, one for each tire. Each plot is the result
FIG 1A.
of dividing the lateral force exerted by the tire on the
surface, by the normal force (previous plot) exerted
vertically downward on the surface. Sideslip occurs
whenever the effective tire friction exceeds the
measured lateral tire force characteristic multiplied by
the surface friction ratio. There were more plots made
but because we are primarily interested in mechanical
grip, we will show just these 4 sets of the data taken.

The title of each run is at the top of each graph. (See Fig
2A) The word “Fishhook” refers to the maneuver used.
(The “fish hook” maneuver is the one selected by the
USA National Highway and Traffic Administration
(NHTSA) as one of the tests they use to determine
rollover propensity of high CG vehicles in order to give
them safety ratings.) In the maneuver the car is
computer driven through 2 steering maneuvers at
speeds starting at 32 to 35 mph, and increasing in 10
mph increments up to 55 mph or until rollover is
experienced. In a steady state at 35 mph the steering
wheel is turned 360 degrees to the right, then
immediately 460 degrees to the left. This type of
maneuver is very similar to driving through a chicane
FIG. 1B. and is often experienced in World Rally type events.

Fig 1B. This sectioned sketch shows how a rebound The first simulation run was done without rebound
control spring could easily be applied to a strut type control. We called this “Normal “ or vehicle A in the data.
spring shock suspension member. Note that the springs The second run with rebound control springs installed
could be replaced by air bags. was entitled “Spring Stiff Extension,” or vehicle B by the
Simulation Consultants. We used a coefficient of friction
VALIDATION of 1.1:1 for the tire/road interaction. In vehicle B we used
Prior to testing we needed a way to evaluate how the same rate for the rebound control springs as was
effective rebound control could be in improving handling. used for the vehicle suspension springs.
We started off by having an independent consultant do a
simulation. “ADAMS CAR“ is a 3D computer simulation Please note: The following graphs were taken from the
program that can be used to put an actual vehicle original report by scanning them and creating digital
through a set maneuver and measure the various pictures at the highest possible resolution. Unfortunately
dynamic responses. We selected a standard SUV type they do not have the quality of resolution of the originals
vehicle for the simulation for 2 reasons. 1. They but the important contrast data is still discernable.
already had it modeled and had made a base run
through an extreme cornering maneuver. In Fig 2A and 2B below we show the simulation data for
2. We figured that using a high CG type vehicle would Lateral Acceleration versus Time.
show better contrast in the data. In the data the base
vehicle is A and the modified vehicle is B.
 Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

FIG. 2A SUV With normal suspension. FIG. 3B . Speed versus Time. Vehicle B

Compare Fig 2 A with Fig 2B below and note the The Speed versus Time graphs above indicated that the
significance of the amount of time that passes after the vehicle in the baseline configuration had stopped
second turn ends and before the two vehicles achieve steering at 4.7 seconds because it had rolled over.
zero G lateral forces. (Rollover). Vehicle B went a further 2 seconds at around 25 mph
before it also rolled over and the data stopped. A speed
of 25 mph equates to a distance of around 73 feet.

Note: In this amount of time and distance, the driver

would have had more than adequate time to apply the
brakes and stop prior to rollover.

FIG 2B SUV With rebound control springs installed.

FIG. 4A Normal tire force versus time. Vehicle A.

Note the continuing activity of the tire friction in FIG. 4B.

Compared to 4A.

FIG. 3A . Speed versus Time Vehicle A.

 Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

FIG. 5B clearly show Effective Tire Friction occurring

long after it has ended in FIG. 5A. Note the time
differences at the bottom edge of the graphs.

The simulation data was obviously very encouraging so

we then conducted some dynamic testing with the
rebound control shocks mounted on a high CG SUV
vehicle at Texas Proving Ground Services LLC in Hanis,
Texas.

Because the test service had no provision for outriggers

for protection in case of rollover, the tests conducted
were limited in speed and severity. Following is
comparison data from the base line vehicle (A) to the
same vehicle (B) equipped with rebound control shocks
with a rebound spring rate that was 50% of the base
vehicle suspension spring rate. The rebound springs
FIG. 4B Normal Tire Force Vehicle B. were activated at a shock length that was .75 inches into
the jounce movement, starting from normal design
. height. They were inactive during normal jounce
movement from that point. They were fully engaged for
the entire amount of rebound travel from that point. The
following data was recorded.

SLALOM EVENT. With cones @ 60 foot apart. Entry

speed 31.7 mph.
Average Vehicle Roll (Left and Right Motion)
Baseline Vehicle A: 1.576 degrees.
Vehicle B: 1.537 degrees (Reduction of 2.5%)
Average Vehicle Yaw (Twist around vertical axis)
Vehicle A. 1.042 degrees.
Vehicle B. 0.945 degrees (Reduction of 9.3%)

L-TURN EVENT
Average Vehicle Yaw (Twist around vertical axis)
Vehicle A. .810 degrees
Vehicle B. 0.667 degrees (Reduction of 17.6%)
FIG. 5A Tire Friction Vehicle A.
LANE CHANGE EVENT.
Average Vehicle Lateral Acceleration (L to R)
Vehicle A: 0.792 G
Vehicle B: .90 G (12% increase indicates reduced roll)

S-CURVE EVENT.
Vehicle Yaw (Twist around vertical axis
During S-Curve Event.
Vehicle A: 1.128 degrees.
Vehicle B: 0.785 degrees. (Yaw Reduction of 30.4%)

MAXIMUM BRAKING EVENT (Best run)

Vehicle A @30.80 Mph entry speed: 41.7 feet.
Vehicle B @30.73 Mph entry speed: 39.1 feet.
Stopping distance improvement: 2.6 feet or 6.2%.
Note: Average of all runs was .59 feet or 1.5%.
Average total pitch improvement during all runs with
Vehicle B was .484 degrees or 5.8%.

FIG. 5B Tire Friction Vehicle B. CHATTERBUMP EVENT

Average Vehicle Fore/Aft Acceleration.
 Downloaded from SAE International by Univ of Nottingham - Kings Meadow Campus, Friday, August 10, 2018

Vehicle A: 1.435 G passive in one respect it is proactive in another, which

Vehicle B: 1.21 G (15% reduction) could easily help offset that.
Note: in the test vehicles the rebound control spring was
preloaded so that it was inducing some resistance to
rebound travel of the suspension. We did this by setting CONTACTS
the rebound spring so that the free length was at a ride Peter Bryant. VP Eng. Amtech Corporation.
position of half an inch of travel into jounce. This meant Tom Neavitt. Pres. Amtech Corporation.
that the rebound control was being introduced pro- 3120 Venture Drive, Las Vegas, NV 89101.
actively, thus causing a slight delay in rebound (702) 233-3953 or (702) 312-6124
movement. This also caused a very slight lowering of the E-mail: pebryant@aol.com
ride height. The rate of the rebound control springs was
generally around half that of the suspension springs. We
reasoned that if the rebound springs were too stiff, they REFERENCES
would just delay the body roll momentarily until the
lateral G experienced by the car overcame the un- SAE Paper 02ADSC-73.
sprung mass of the wheel/axle system and the wheels Opposing or Counter Spring (Bi-Linear) Suspension
lifted. Technology for Optimum Vehicle Dynamic Roll Control
Without Computers.
VALIDATION CONCLUSIONS Author Peter E. Bryant. Published May 2002

The validation data indicates that controlling rebound TEXAS PROVING GROUNDS SERVICES LLC.
movement can increase cornering speed significantly in Hanis Texas.
slow corners and can also reduce roll in fast corners Report to Amtech Corporation on Rebound Control
depending on suspension configuration, aero package Shock Test.
and suspension travel. But it can definitely improve
mechanical grip in cars with real suspension movement. United States Patent 6,761,372
Opposing Spring Resilient Tension Suspension System.
The reduction in the braking distance demonstrated that Inventor Peter E. Bryant.
the rear brakes were working better because it had
helped to reduce the forward pitching. The low speed
chatterbump acceleration reduction is important
regarding wheel oscillation; this could be a very
significant factor in smoothing out tire chatter when
racing over and around curbs and chicanes, when there
is not sufficient travel on the shocks for them to exert
control.

The overall conclusion must be that rebound control

springs work simply because they are attached to the
un-sprung component, thus making un-sprung weight “A
friend” in improving dynamic handling; rather than just
being a nuisance and “An enemy.”

Notes:
To a certain extent we regret that the validation data in
this paper was generated using an SUV instead of a
racing car. Unfortunately, we were unable to secure a
suitable race test vehicle and initially were just looking to
get comparison data. But the fact that we are able to
show good data indicating that the technology works at
slow speeds on any kind of car is a clear indication that
it could quite possibly work very well on racing cars.

Perhaps the most significant fact that emerged is that

vehicle ride height can be controlled to a certain extent
without using complicated and expensive computer
technology. Active Suspension is banned in most
Professional Racing Series. Finding ways around the
rules is part of the sport. Although this technology is