Solar Cells, 31 (1991) 425-442

425

Sunrunner: the engineering report

Justin Beres
University of Michigan, A n n Arbor, MI 48109-2116 (U.S.A.)

Abstract
In November 1987, the first World Solar Challenge took place, a 3000 km (1860 miles)
transcontinental solar powered vehicle race from Darwin to Adelaide across the Australian
Outback. The race, held every three years, featured entries from various countries,
including Switzerland, Japan, Australia, and the United States. The winning car, General
Motors' Sunraycer, finished 970 km (600 miles) in front of its nearest competitor. Based
on this outstanding performance, General Motors decided not to return to Australia in
1990, but to instead sponsor a solar car race of its own and send the top three finishers
to the international competition. GM Sunrayce USA featured 32 cars from top engineering
colleges throughout North America and took place from Florida to Michigan, covering
2660 km (1650 miles) during July 1990.

1. I n t r o d u c t i o n
1.1. T h e p r o j e c t
T h e University o f Michigan Solar Car Project a l m o s t wholly c o m p r i s e s
u n d e r g r a d u a t e students. Organized into seven teams, s t u d e n t s have r e s p o n sibilities r a n g i n g f r o m e n g i n e e r i n g design t o b u s i n e s s administration. The
p r o j e c t h a s followed a strict timeline of dates and goals, as the c a r w e n t
f r o m c o n c e p t t o p a p e r to reality in only one y e a r f o r the GM S u n r a y c e USA.
During the p e r i o d b e t w e e n the S u n r a y c e a n d the 1 9 9 0 W o r l d Solar Challenge,
S u n r u n n e r u n d e r w e n t a 27.3 k g (60 lb) weight reduction, along with small
p e r c e n t a g e i n c r e a s e s in s u b s y s t e m efficiencies.
1.2. T h e c a r
R e g u l a t i o n s f o r t h e a f o r e m e n t i o n e d r a c e s allow only sun-derived prop u l s i o n e n e r g y , with b a t t e r y s t o r a g e o f this energy. In satisfying t h e s e rules
t h r o u g h design, S u n r u n n e r ' s solar cells, singularly referred to as " t h e a r r a y " ,
collect e n e r g y w h i c h is d i r e c t e d to an electric m o t o r o r b a t t e r y pack. The
m o t o r t u r n s a single r e a r wheel by direct drive r e d u c t i o n . A c o m p o s i t e b o d y
shell h o u s e s the driver, s e a t e d in a r e c u m b e n t n y l o n seat w h i c h is a t t a c h e d
t o a s p a e e f r a m e c o n s t r u c t e d of thin-walled a l u m i n u m tubing. The c a r f e a t u r e s
a M a c P h e r s o n strut f r o n t s u s p e n s i o n , a trailing a r m rear s u s p e n s i o n , a n d
f r o n t w h e e l steer, with the steering c o l u m n a l o n g the vehicle centerline.
S u n r u n n e r is the m a x i m u m l e n g t h a n d width, but is only 1.27 m (4.2 ft)
tall. It has a w h e e l b a s e o f 2.43 m (95.5 inch), a n d front and r e a r t r a c k

Elsevier Sequoia, Lausanne

426

Faculty Advisor [

I
I

Executive Conanittee

I ~j~'M~ager I I ~oty~ager

I
Body

] I Chassis I I Sol~,

IFu~.~isin~/~b,,o Relations I

I [ ~ow~Electronicsll Race S~ate~ I

I L~gisti./Fina~ce I

Body
a e r o d y n a m i c testing
shell fabrication
c a n o p y fabrication
canopy wiper
seat
instrument panel
ventilation
paint / decals

Solar A r r a y
cell layout
cell a s s e m b l y
cell attachment
ceLl wiring
Race Strategy
meteorology
c o u r s e study
vehicle m o n i t o r i n g

Chassis
spaceframe
suspensions
w h e e l s / tires / a r i e s
driveline
brakes
steering

P~lic Relations / Fundraisin~,

sponsor m a i n t e n a n c e
fundraising
LoL,istics / F i n a n c e
accounting
logistical a r r a n g e m e n t s

P o w e r Electronics
motor
batteries
instrumentation
telemetry
Scheme

1. T e a m

structure

and

organization.

widths of 1.60 m (63 inch) and 1.83 (72 inch) respectively, and weighs 232
kg (510 lb).

1.3. The design

In conceptualizing Stmrunner, the objective was to create a vehicle which
would obtain as much of the available energy as possible, while expending
as little as possible for propulsion. This led to vehicle shape design strategy
of maximizing energy accumulation while minimizing resistance forces, considering the e x p e c t e d race conditions of the GM Sunrayce USA. The primary
retarding force is aerodynamic drag, which is greater than the rolling resistance.
Therefore, the action plan became to achieve maximum solar collection
capacity and minimal aerodynamic drag.
This plan, however, spawns a dilemma in solar car design. A car with
solar cells laid out on a flat plate, which tilts to be perpendicular to the
sun, has the greatest solar gain. However, aerodynamics in this case are
quite poor, as flow over a large wetted surface area causes high drag, namely
skin friction. In contrast, a car with an aerodynamic t eardrop shape complicates

427
solar cell arrangement. Cells must be carefully positioned and electronically
monitored to insure homogeneous solar loading. The electronic controlling,
or "power tracking", inherently creates inefficiency.
The final Sunrunner configuration combined and compromised solar
collection and aerodynamics. The shape consisted of three airfoils: the main
body and two sidepods. Teardrop shapes maximize aerodynamic efficiency,
allowing for attached flow, which reduces form drag, typically the primary
contributor of total drag. Placed on the aft two-thirds of the car, 14 057
solar cells cover the largest amount of area practical, crudely configuring
an upside-down U (Fig. 1). The 8.3 m 2 cell area on top is divided into five
"facets", or flat lengthwise strips. Facets prevent fracture of rigid cells, by
restricting placement on three-dimensional compound curves. The upsidedown U is completed, with 2.4 m 2 of area on each sidepod. This arrangement
is optimal for sunlight collection, along the south-north route of the Sunrayce
and World Solar Challenge, for several reasons. The position of the sun
constantly changes, at no time shining directly overhead. Cells on the side
of the car are useful during morning and evening hours when the sun radiates
at minimal angles to the earth. During cloudy days, sunlight diffuses to the
horizon, also coming in at slight angles. Sidepod cells also collect light that
has been reflected off the ground.
Once the vehicle shape was finalized a quarter-scale clay model was
made. At first, sidepods were symmetric airfoils, in the plan view. Testing
in the 5 foot)< 7 foot wind tunnel at the University of Michigan offered
encouraging results. However, the sidepods were soon made asymmetrical,
allowing for more solar cell area in the rear of the car. Also, the pod regions
housing the front wheels were widened because they inhibited turning clearance
and suspension packaging. Unexpectedly, follow-up wind tunnel testing yielded
high drag numbers. Analysis indicated that a downforce was created in the
"underbody tunnel" owing to the pressure distribution of the new configuration.
The consequence was high negative lift which is quadratically related to
drag. The solution to reducing this drag, found through continued wind
tunnel testing, was raising the entire car, then pitching the nose of the car
upward. This allowed for a ground clearance of 102 mm (4.0 inch) at the

Fig. 1. Configuration of array facets.

428
lowest p o i n t o f the car. G r o u n d c l e a r a n c e s at the front and rear axles were
178 nun (7.0 inch) and 114 nun (4.5 inch) respectively (Fig.2).
The d r a g D w a s m e a s u r e d a c c o r d i n g to
D = pv2C,t
w h e r e D is the total drag, p the fluid m a s s density, v the flow velocity, Cd
the coefficient of drag, and A the frontal area. It is regulated b y only two
factors, the coefficient of drag Ca indicating shape, a n d the frontal area A
indicating size. The p r o d u c t CdA r e p r e s e n t s the total drag D b e c a u s e the
d y n a m i c p r e s s u r e f a c t o r of the e q u a t i o n is c o m p o s e d of variables. Quarterscale wind tunnel results indicated CaA = 0.15 m 2 ( 1 . 6 6 ftz), with C d = 0 . 0 9 1 ,
b a s e d on an area A = 1.70 m 2 ( 1 8 . 2 9 ft2). A frontal area scan d o n e on the
full-scale car at the General Motors A u t o m o t i v e Wind Tunnel in W a r r e n , MI,
verified this area; however, it yielded C a = 0 . 1 2 0 with a b l o c k a g e c o r r e c t i o n
f a c t o r o f 6% considered. Flow visualization with s m o k e s h o w e d a t t a c h e d

0.757
0.457
0.102

0.000
l

0.000

2.000 --

0.305

:=:::::>

1.695

2.000

....................

~:::::::>

.......................

0.000 --

1,270--

0.757-0,645 -0,457__
0.330--~
0,102__
0.000

..........

..........................

....

D. . . . . . . . . . . . . . . . .

..................... f : : : : : ............................... : .... .

"l

.,t"

0.000

[
1.13

1.70 1.85

I
3.43

Fig. 2. Drawing of vehicle outline, dimensions in meters.

5.85

6.00

429
flow, y e t v o r t i c e s did exit the r e a r c o r n e r s . After m i n o r a d j u s t m e n t s w e r e
m a d e t o wheel openings, C d = 0 . 1 0 8 was o b t a i n e d at zero yaw, p r o d u c i n g
C d A = 0 . 1 8 m 2 (1.98 ft2). This is a b o u t one-third the drag o f a c o n v e n t i o n a l
automobile.

2. Body
2.1. E x t e r i o r
Once the car was s h a p e d and d i m e n s i o n e d it was divided into sections
(Fig. 3). Here, the g r e a t e s t c o n c e r n c e n t e r e d a r o u n d the solar a r r a y piece.
The car had to b e split so that the a r r a y c o u l d b e tilted or r e m o v e d f r o m
the chassis for b a t t e r y charging and for m a i n t e n a n c e b e f o r e and after e a c h
race day. B e c a u s e o f f o r e s e e n h u m a n handling, the design had to limit the
risk o f d a m a g i n g the fragile and costly solar cells. Also, the a r r a y had to
satisfy the size constraints of the " h y p o t h e t i c a l b o x " rule o f GM S u n r a y c e
USA. O t h e r c o n s i d e r a t i o n s w e r e quick driver exit in o r d e r to p a s s a " 1 5
s e c o n d e s c a p e " t e s t o f the Sunrayce, and accessibility o f c o m p o n e n t s for
repair.
The b o d y was c o n s t r u c t e d o f c o m p o s i t e materials kevlar and N o m e x ,
c h o s e n on the criteria of light w e i g h t and ability to a b s o r b large i m p a c t
energy. Kevlar, a cloth similar to fiberglass, weighs a p p r o x i m a t e l y 0.93 kg
m -2 (0.19 lb ft -2) and a b s o r b s 15% m o r e e n e r g y t h a n c h r o m o l y steel. Two
t y p e s were used: 120, a s m o o t h fine w e a v e u s e d to r e d u c e skin friction drag
along the o u t e r surface of the car, and 285, a s t r o n g c o a r s e w e a v e for the
inner layer. N o m e x is a high s t r e n g t h h o n e y c o m b , w h o s e density is 28.9 kg
m -a (1.8 lb f t - 3 ) . Its t h i c k n e s s r a n g e d f r o m 9.5 m m ( 0 . 3 7 5 inch) to 12.7
m m (0.50 inch), in b e t w e e n two layers o f kevlar, f o r m i n g a sandwich.

Fig. 3. Drawing of body sectioning.

430
O n c e s e c t i o n e d , the m u l t i - s t e p p r o c e s s of b o d y f a b r i c a t i o n c o m m e n c e d .
T h e q u a r t e r - s c a l e clay wind tunnel m o d e l w a s digitized, for which a c o m p u t e r
s c a n n e d the clay s u r f a c e r e a d i n g t h o u s a n d s o f t h r e e - d i m e n s i o n a l d a t a points.
T h e m o d e l w a s t h e n s m o o t h e d a n d m a d e s y m m e t r i c a l f r o m side to side. In
the m e a n t i m e , full-scale a r m a t u r e s w e r e c o n s t r u c t e d of w o o d a n d f o a m and
carefully c o v e r e d with clay, to e n s u r e a d h e s i o n to the f o a m . W h e n c o m p l e t e d ,
t h e digitized d a t a w e r e d o w n l o a d e d to a n u m e r i c a l c o n t r o l milling m a c h i n e
w h e r e a five-axis c u t t e r c a r v e d the clay into t h e vehicle s h a p e . A f a c e c o a t
w a s a p p l i e d to t h e clay t o p r o d u c e a s m o o t h s u r f a c e finish a n d to seal it.
S h e e t s of fiberglass w e r e laid down, c a s t i n g f e m a l e m o l d s . T h e y w e r e r e m o v e d
f r o m a r m a t u r e s a n d p l a c e d into " e g g - c r a t e s " , f r e e - s t a n d i n g s u p p o r t i v e w o o d e n
s t r u c t u r e s , allowing t h e m to b e level to the g r o u n d . T h e s e f o u n d a t i o n s w e r e
p a i n s t a k i n g l y cut to fit e x a c t l y the o u t e r c u r v e s of the molds. S e a m s w e r e
c o v e r e d with fiberglass for stability.
W h e r e n e c e s s a r y , s e c t i o n l i n e s o r p a r t l i n e s w e r e s c r i b e d into the m o l d s
a n d multiple c o a t s of t h r e e different w a x e s -- f o u n d b y a trial a n d e r r o r
p r o c e s s -- w e r e applied to g u a r a n t e e r e l e a s e of the p i e c e o n c e it w a s laid
up. Kevlar a n d N o m e x w e r e cut to the partlines, the latter f o r m i n g to c u r v e s
w h e n p e r s u a d e d with a h e a t gun. Materials w e r e t h e n laid down: k e v l a r /
N o m e x / k e v l a r . W a x w a s p l a c e d at t h e e d g e s of p a r t s requiring a ridge to
j o i n a n o t h e r part. O v e r it, lips c o u l d f o r m .
To e n s u r e p r o p e r b o n d i n g a n d to eliminate e x t r a resin f r o m the kevlar,
s a n d w i c h e d p a r t s w e r e v a c u u m b a g g e d . The t h r e e layers w e r e sealed in
plastic a n d a t t a c h e d to a s t e a d y suction of 0.5 a t m . T h e p a r t w a s t h e n h e a t e d
to 93 C ( 2 0 0 F). After 1 h t h e p a r t w a s r e m o v e d a n d p r i e d f r o m the m o l d
with f o r c e d air or a knife if n e c e s s a r y .
A s t r u c t u r a l a d h e s i v e w a s u s e d to b o n d s o m e p i e c e s p e r m a n e n t l y . Mounting
the c h a s s i s s p a c e f r a m e to the b o d y followed. During t h e b o d y f a b r i c a t i o n
p r o c e s s , c o o r d i n a t e s o f f r a m e a t t a c h m e n t p o i n t s w e r e e t c h e d into the molds.
In t h e s e places, 152.4 m i n x 152.4 m m (6 i n c h x 6 inch) s q u a r e s of N o m e x
w e r e cut out a n d r e p l a c e d with six r e i n f o r c e m e n t l a y e r s o f kevlar. A l u m i n u m
plates, 3.2 m m thick ( 0 . 1 2 5 inch), w e r e g l u e d t o t h e kevlar, t h e n c o v e r e d
with fiberglass. H o l e s w e r e drilled u p t h r o u g h the b o d y into similar plates
w e l d e d t o the f r a m e , a n d the t w o w e r e s e c u r e d with e l e v a t o r bolts. Quartert u r n f a s t e n e r s w e r e u s e d to a t t a c h r e m o v a b l e b o d y p i e c e s t o g e t h e r .
The c a n o p y o f the solar c a r is likened to a w i n d s h i e l d o f a c o n v e n t i o n a l
auto. Its r e q u i r e m e n t s w e r e m i n i m i z e d weight, g o o d optical quality, a b r a s i o n
r e s i s t a n c e , a n d light reflection capability f o r d r i v e r c o m f o r t . In d e v e l o p i n g
S u n r u n n e r ' s c a n o p y t h e r e w e r e five c o n s i d e r a t i o n s : m a t e r i a l , t h i c k n e s s , f o r m ing, tinting, a n d attaching.
S t u d y of m a t e r i a l s led to t h e r m o p l a s t s , n a m e l y a c r y l i c s a n d p o l y c a r b o n a t e s ,
w h i c h c o u l d b e d r a p e or blow m o l d e d . Several tinting o p t i o n s w e r e r e s e a r c h e d ,
s u c h a s s e e - t h r o u g h mirror, Scotchtint, metallizing, coloring, a n d e v a p o r a t e d
m e t a l film covering. The final design w a s a d r a p e - m o l d e d , 3.2 m m ( 0 . 1 2 5
inch) thick, s c r a t c h - r e s i s t a n t acrylic, u r e t h a n e d to the k e v l a r - N o m e x c a n o p y
frame. S p u t t e r c o a t e d with titanium nitride, it a l l o w e d 17% light t r a n s m i s s i o n .

431

2.2. I n t e r i o r
E r g o n o m i c s w e r e t h o r o u g h l y analyzed in the design o f S u n r u n n e r ' s
interior with r e s p e c t to driver packaging, c o m f o r t , and safety. More specifically,
t o p i c s of seating, vision, ventilation, and c o n t r o l s w e r e given attention.
P a c k a g i n g s p a c e was r e s t r i c t e d as a result o f minimizing frontal area
to lessen drag. Initially, gliders and M1 tanks w e r e r e s e a r c h e d b e c a u s e o f
t h e i r similar seating conditions. The r e q u i r e m e n t for light weight led to a
h a m m o c k - s t y l e seat design e m p l o y i n g a s t r o n g s t r e t c h - r e s i s t a n t w o v e n m e s h
material. W e a v e p a t t e r n s are s t r o n g e r than knit b e c a u s e o f the bi-directional
interlacing o f strands. The anti-stretch characteristic p r e v e n t e d sag; a n imp o r t a n t consideration, b e c a u s e the allotted distance b e t w e e n the b o t t o m of
the seat and the inner b o d y surface was a m e r e 25.4 m m (1 inch).
After the seat was a t t a c h e d t o the s p a c e f r a m e the c a n o p y was designed.
Fulfilling vision r e q u i r e m e n t s o f GM S u n r a y c e USA, driver e y e s had to be
no less than 700 m m ( 2 7 . 6 inch) off the ground. The driver had to be able
to see 10 a b o v e the flat horizon, 90 f r o m e i t h e r side o f the center, and a
s p o t 8 m (26.2 ft) in f r o n t o f the car.
In o r d e r to d e t e r m i n e the c a n o p y daylight opening, a m o c k - u p o f the
c o c k p i t was c o n s t r u c t e d . Here, several drivers sat in the seat while five
w o o d e n t e m p l a t e s simulating the b o d y shell w e r e p l a c e d o v e r t h e m , s p a c e d
2 0 0 m m (7.9 inch) apart, starting f r o m the c a r ' s nose. F o r w a r d vision was
t e s t e d by increasing o r d e c r e a s i n g the first t e m p l a t e height. Side-to-side
testing was m o r e arbitrary, as n o distance or " p o i n t on the g r o u n d " rules
w e r e specified. Despite this, S u n r u n n e r ' s c a n o p y offered 180 of sight latitude.
Rear vision utilized a fiber optics b u n d l e r a t h e r t h a n m i r r o r s m o u n t e d
to the car's e x t e r i o r w h i c h would c a u s e i n c r e a s e d drag. On o n e end, the
b u n d l e a t t a c h e d to an e y e p i e c e t h r o u g h which the driver could view images
in a 40 range. An objective lens c o n n e c t e d at t h e o t h e r end and was m o u n t e d
on t o p of the car, j u s t fore o f the solar array, h o u s e d in a small a e r o d y n a m i c
fin. The bundle, as thick as a g a r d e n hose, was c o m p o s e d of t h o u s a n d s o f
s m o o t h glass fibers, e a c h i n d e p e n d e n t l y c o n d u c t i n g light b y m e a n s o f multiple
internal reflections. The length was limited to 1.2 m (4 ft) b e c a u s e r e s o l u t i o n
is inversely p r o p o r t i o n a l t o length.
C o m f o r t e n c o m p a s s e d two considerations: driver o v e r h e a t i n g and c o n t r o l s
accessibility. To investigate fully h o w h e a t affects drivers, a test was c o n d u c t e d
simulating p r e d i c t e d interior t e m p e r a t u r e s . Subjects sat in the seat s u r r o u n d e d
b y two s p a c e h e a t e r s p l a c e d o n w o o d e n p l a t f o r m s to the right and left.
Blankets w e r e d r a p e d o v e r t h e t e s t t e m p l a t e s a n d c a r d b o a r d was t a p e d
a r o u n d t h e back of the s p a c e f r a m e . All this was d o n e in an a t t e m p t to retain
as m u c h heat as possible. W e t a n d d r y bulb t h e r m o m e t e r s w e r e p l a c e d inside
the cockpit. The steering w h e e l was wired to a c o m p u t e r o n which a driving
simulation p r o g r a m was r u n to o c c u p y drivers during a 9 0 min test session.
Six times drivers w e r e asked to rate c o m f o r t o f the neck, back, arms, and
thighs, as t e m p e r a t u r e s r e a c h i n g 35 C (95 F) w e r e s i m u l t a n e o u s l y noted.
Results b a s e d o n o b s e r v e d r e a c t i o n s as well as c o m f o r t ratings and c o m m e n t s
f r o m subjects f o r c e d a d j u s t m e n t s t o the c o c k p i t design.

432
Regarding the seat, the chosen Lino Mesh allowed for airflow, keeping
the driver cool. Search for a nylon mesh material with adequate gaps between
the weave for ventilation and high denier, or fineness of thread, for strength,
led to consideration of materials such as Cordura, Dymetrol, and Textilene.
Upon final positioning, test results indicated that discomfort from body fatigue
o c cu r r ed after about 5 h: the maximum expected driving time for any person.
In conjunction with the seat and tinted canopy, the ventilation system
aided in driver cooling. Mylar ducts running from two inlets on the vehicle
nose branched to the driver, motor, m ot or controller, and batteries. The
inlets, 23.2 cm 2 (3.6 inch 2) in cross-sectional area, located in regions of
highest pressure as determined by wind tunnel pressure distribution experiments on the quarter-scale model, allowed for an airflow of 140 ft a min -1.
Initially, inlets were placed at the stagnation point of the car. However, fear
of destroying the laminar boundary layer led to the final configuration. A
fan supplying 35 ft a min-~ of forced ventilation was placed in the 102 mm
(4 inch) diameter tube ducted to the driver. Forced air was also provided
to the motor, controller, and battery pack using 40 ft 3 min-~ fans.
Aspects of the car which required monitoring directed design of the
instrument panel. Instrumentation included a speedometer, tachometer, battery
charge display, and solar array ammeter gauge. The cruise control knob,
turn signal knob, for w ar d- r e ve r s e switch, fan switch, horn button, and indicator
light lamps were controlled manually. Braking and acceleration were operated
by foot pedals. Arrangement of the controls was critical, with ease of
accessibility the highest priority. Final c o m p o n e n t placement was achieved
after experimentation with various positions.
With regard to safety, the distance from the driver's feet to the nose
of the car was over 305 mm (1 ft), providing crush area in the case of a
head-on collision. Head clearance was measured at 102 m m (4 inch), as the
spaceframe extended over the seat, protecting against roll-over conditions.
The driver also wore a six-point climbing harness for restraint. The harness's
male plugs buckled into female receptacles sewn t o the seat.

3. S o l a r a r r a y

3.1. Solar cells

Four types of cells were considered (listed in increasing order of
p e r f o r m a n c e and cost): am or phous silicon, polycrystalline silicon, monocrystalline silicon, and gallium arsenide. Efficiencies are approximately 5%,
12% 17% and 22% respectively. The cost of space grade gallium arsenide
cells is four~ t'nnes that of space grade monocrystalline cells, which, in turn,
cost ten times more than terrestrial and amorphous cells.
3.2. A r r a y layout
GM Sunrayce USA regulations dictated that the array fit into a hypothetical
volume 4 m (13.1 ft) by 2 m (6.6 ft) by 1.6 m (5.3 ft). Sunrunner's space

433
grade monocrystalline cells, supplied by Spectrolab, are 17% efficient, producing a peak p o w e r output of 1400 W.
The array was arranged two ways: into facets, conforming to the vehicle
configuration, and into "sections", pertaining to energy collection. The form er
allows for an airfoil shape, while the latter allows for power maximization
using " t r a c k e r s " . Cells connected in series form snake-like lengths called
"strings", which, also placed in series, form "modules". Modules connect ed
in parallel were segregated into sections. Sectioning linked the modules on
the car with similar three-dimensional orientation, since they experience
c o mp ar ab le solar intensity. Theoretically, modules in a section must receive
uniform radiation in or der to p r o d u c e the highest output. However, in the
case of Sunrunner, owing to time, geometry, and wiring constraints, the
sections were exactly facets.
Cells are manuf act ur e d to.a voltage specification at a certain t e m p e r a t u r e
as the ambient t e m p e r a t u r e increases, the voltage decreases linearly. Tests
showed voltages at m a xi m um power points of 0.39 V, 0.44 V, and 0.49 V,
at 70 C (158 F), 50 C (122 F), and 28 C (82 oF) respectively, p e r 20
m m 40 m m (0.8 inch 1.6 inch) cell. At their most efficient point, trackers
required 190 V. The array also had to p r o d u c e a potential above the maximum
battery voltage of 136 V, allowing the batteries to charge. Designing for
worst case t e m p e r a t u r e conditions, an average of 380 cells were organized
into 15--20 strings which formed modules, offering an operating potential
of 190 V and 280 mA at 28 C (82 oF). Five modules per facet were placed
on top o f the car, with six on each sidepod.
The lengths of the 676 strings were limited by facet width, cell width + 0.50
m m (0.020 inch), and a physical gap between cells of 1 m m (0.040 inch).
Cells in strings maintain the same current characteristics. The cell with the
lowest cu r r en t limits the output of the entire module as a result of the series
connection. Likewise, modules with the lowest voltage output limit the potential
going to a tracker. Because horizontal surfaces collect the most sunlight
t h r o u g h o u t the day, cells with highest currents were placed on top of the
car.
-

3.3. A r r a y assembly
The first step in assembling the array was attaching two silver interconnects
to the back of each cell. Next, cells were covered with a magnesium fluoride
coated antireflective glass, glued with two drops of a silicone adhesive. This
not only p r o t e c t e d the cells, but also diminished reflected power loss. Cells
were cured for 40 rain at 38 C (100 F) and 100 min at 54 C (130 F).
They were th en cleaned and soldered in strings, held steady by aluminum
CNC-milled jigs. "U-tabs" were added to string ends, over which silver bus
strips were soldered, forming modules.

3.4. Cell m a t c h i n g
Once soldering was completed, strings were examined on a light bench
to m atch like voltages and t o check for defective cells. Modules with equal

434

v o l t a g e s w e r e united to a t r a c k e r . T r a c k e r s electronically r e g u l a t e the a r r a y

o u t p u t current/array to p r o d u c e m a x i m u m o u t p u t p o w e r . The " b u s v o l t a g e "
v a r i e s constantly. A r r a y p o w e r o u t p u t Po,t is t h e p r o d u c t o f the a r r a y c u r r e n t
and bus voltage,

Solar insolation is d e p e n d e n t on t i m e o f d a y a n d weather. F o r e x a m p l e ,

u n d e r t r e e s h a d o w or d u r i n g c l o u d y late a f t e r n o o n , c u r r e n t o u t p u t s a r e l o w e r
t h a n at n o o n on a s u n n y day. T r a c k e r s c o n t r o l the a r r a y c u r r e n t and b u s
voltage, s e n s i n g t h o s e at " m a x i m u m p o i n t s " , offering m a x i m u m c h a r g i n g
p o w e r to the b a t t e r i e s while p r e v e n t i n g loss of p o w e r t h r o u g h d i s s i p a t e d
heat.
The light b e n c h s i m u l a t e d 1 Sun intensity conditions, equalling 100 m
W c m - ' -- the a v e r a g e a m o u n t of light r e a c h i n g t h e e a r t h at s o l a r n o o n
on a cloudless, 25 C ( 7 7 F) day. H e r e , eight 3 0 0 W light b u l b s w e r e
a t t a c h e d to c h e m i s t r y stands, f u s e d t o p r e v e n t h e a t i n g u p a n d self-destructing,
a n d c o n t i n u o u s l y c o o l e d by fans. O r i e n t a t i o n s w e r e a d j u s t e d to a c h i e v e
u n i f o r m illumination a n d t e m p e r a t u r e . A deviation of 7% w a s verified w h e n
a r e f e r e n c e cell with a k n o w n c u r r e n t w a s m o v e d a l o n g the r e c t a n g u l a r , 102
m m 660 m m (4 inch 26 inch) t e s t area, a n d an u n c h a n g i n g c u r r e n t o u t p u t
w a s m e a s u r e d . E a c h string of cells w a s p l a c e d on the bench, a n d its
c h a r a c t e r i s t i c s p l o t t e d on a t y p e - 5 7 5 t r a n s i s t o r c u r v e tracer, w h i c h is similar
to an oscilloscope. F r o m t h e g r a p h , m a x i m u m p o w e r p o i n t s w e r e t a k e n (Fig.
4).
A digital m u i t i m e t e r f o u n d o p e n - c i r c u i t v o l t a g e s a n d short-circuit c u r r e n t s
for e a c h string. At air m a s s (AM) 1.5, p e r cell, Vo = 5 8 0 mV a n d I ~ = 2 9 2
mA. F r o m t h e s e n u m b e r s , efliciencies w e r e calculated:
r/en = [(I~Vo)/Areace, ]/(sun intensity)

3.5. A t t a c h m e n t
After m a t c h i n g , high b o n d i n g d o u b l e - s i d e d t r a n s f e r film w a s u s e d to
a t t a c h m o d u l e s to t h e c a r ' s k e v l a r surface. G a p s b e t w e e n cells w e r e filled
CURRE]
I:
I

) x Vmp
179 A) x (0.478 V)
33 W

~TAGE
Fig. 4. C u r r e n t - v o l t a g e c u r v e for a s o l a r cell.

435
POS
NEG
BypassDicxic
Blocking
Diode

Btw 10.&15
o1"iumgs

OneString
20 - 30 Cells

SilverBus Strip

\
/
An individualcell

SilverBus Suip

Fig. 5. Diagram showing cell wiring.

with a silicone grout to prevent shorting caused by water seepage and

breakage caused by cell expansion at high temperatures. After complications
involving controlled fluidity, homogeneous viscosity and curing were resolved,
the grout was successfully applied.

3.6. Wiring
Modules were joined to power trackers via a "harness" employing 18
gage copper coated silver wire, chosen for its light weight and low resistance
per unit length. Its Teflon covering offered advantageous durability and
thermal and electrical insulation properties. Maximum operating ratings of
600 V and 105 C (221 F) were well beyond the application range. An
aircraft-type circular plastic pin connector was used to plug the array wiring
into the trackers.
In order to scrutinize array performance and troubleshoot difficulties,
a power and a ground wire emanated from each module. Limited line lengths
were achieved using a computer aided design program. Bypass and blocking
diodes were then wired into the array. One blocking, or isolation, diode
separated each module preventing a defective or unilluminated module from
acting in reverse bias and drawing current from an adjacent module. Bypass,
shunt, or shadow, diodes, connected in parallel with strings, are reverse
biased when all cells operate properly. If a cell was defective or shadowed,
the diode opened alternate routes through which current could flow, avoiding
the open circuit (Fig. 5).
4. P o w e r

electronics

4.1. Batteries
Solar energy not used to propel the car directly was reversed in batteries.
GM Sunrayce USA rules limited storage capacity to 5 k W h at a 10 h discharge

436
rate, with ventilation of 10 ft 3 m i n - 1 taken from and exhausted to the outside
of the car.

4.2. B a t t e r y s e l e c t i o n a n d u s e
"Specific energy", life expectancy, and cost were considered in the
battery selection process. The silver zinc (AgZn) type was chosen over nickel
cadmiun (NiCd) and lead acid (Pb-acid) types. AgZn maintain 62 W h lb -1,
which is four times the specific energy of the other two types, and also
feature the smallest voltage drop during discharge. Each of the 6840 A h
cells weighed 0.60 kg (1.33 lb), and p r o d u c e d a mean operating voltage of
1.55 V, with maximum voltage of 2.05 V.
Battery lifetime is a combination of "cyclic life" and "shelf-life". AgZn
batteries are rated for ten " d e e p discharge cycles" and over 700 "trickle
charge cycles". The "wet shelf-life" of the AgZn cells is three to six months,
a life span which fit well within Sunrunner requirements.
"Activation" consisted of adding a liquid potassium hydroxide electrolyte
to "dry cells" which were then allowed to stand for 48 h, permitting proper
bathing of the " s e par a t or " . Next, cells were fully charged to 2 V, discharged,
then charged again.
Typically, charging offers two options: constant potential and constant
current. A charging source held at constant potential Vs, delivers current,
until the battery voltage Vbus appr oa c he s Vs. When Vbus= Vs, the current stops
flowing, and charging terminates. During constant current charging the voltage
can float. In this case, overcharging is probable because the current continues
to flow even when Vb~s= Vmax. As a result of constantly fluctuating solar array
output currents, Sunrunner's batteries were variable current trickle charged,
a variation of the second method. The variable current charging was done
by the trackers, which in a sense acted as ideal current sources, maximizing
power to the batteries by regulating changing array currents and bus voltages
(Fig. 6).

4.3. M o t o r s e l e c t i o n
The goal in selecting Sunrunner's m o t o r was achieving the highest possible
efficiency, with a spatial volume constraint of 178 mm (7 inch) in length
by 152 mm (6 inch) in width by 152 mm (6 inch) in height.
Sunrunner's final m o t o r was an a.c. three-phase brushless inductance
motor, constructed with Magnaquench III per manent magnets. It was controlled
by a pulse-width-modulated (PWM) static inverter, changing direct current
voltages from the solar array or batteries to the required alternating current
voltages. The PWM frequency was such that it allowed the m o t o r and controller
to operate as a system at its highest point of efficiency, with an input power
of 9 0 0 - 1 1 0 0 W ( 1 . 2 1 - 1 . 4 7 hp), offering a peak pow er output capability of
4.47 kW (6 hp), with 126.5 N mm (3 foot lb) continuous and 337.3 N mm
(8 foot lb) peak torque.

437

Voltage

t
Charge and Discharge Characteristics
of AgZn Batteries

Time

Fig. 6. Charge and discharge characteristics of AgZn batteries.

4.4. D r i v e l i n e s y s t e m

During the selection p r o c e s s several m o t o r s w e r e t e s t e d on a d y n a m o m e t e r ,

including two d.c. b r u s h l e s s designs. Input v o l t a g e s w e r e k e p t c o n s t a n t at
m o t o r s p e e d s ~0 ranging f r o m 3 0 0 0 to 4 5 0 0 rev min - I , in i n c r e m e n t s o f
5 0 0 rev rain -~. I n p u t c u r r e n t s w e r e m e a s u r e d t o find the total input p o w e r
Pin- Also m e a s u r e d was the m o t o r t o r q u e Vmo~. The p o w e r o u t p u t Po~t was
calculated by multiplying the m o t o r s p e e d b y the t o r q u e and by a constant,
k(~'motor0)) =Pout

T h e m o t o r efficiency was f o u n d b y dividing the two p o w e r values,

Vmotor=Pout/Pin
A family o f efficiency c u r v e s was p l o t t e d b a s e d on this s p e e d vs. t o r q u e
d a t a (Fig. 7).
In designing t h e driveline system, the a m o u n t o f t o r q u e n e e d e d t o m o v e
t h e c a r Vwh,~ Was f o u n d as the dot p r o d u c t o f quantified resistance f o r c e s
and wheel radius. This multiplied b y a g e a r r e d u c t i o n r, yielded the m o t o r
torque
Tmoto r ~ Twheelr

F r o m the efficiency c u r v e s (Fig. 8) an o p t i m u m m o t o r s p e e d was found,

offering a r e l a t e d m o t o r t o r q u e . T h e a b o v e e q u a t i o n w a s b a l a n c e d b y varying
r . S u n r u n n e r ' s m o t o r o p e r a t e d with a m a x i m u m efficiency ~%~_~= 90%, w h e r e
w = 2 2 0 0 rev min - I , r e q u i r i n g a p p r o x i m a t e l y 1 0 0 0 W (1.34 hp) o f input
p o w e r . A direct drive g e a r ratio o f 4.28:1 g e n e r a t e d a wheel s p e e d vwh~ = 59.5
k m h - 1 (37 miles p e r h o u r ) .

438

65q,

75%

80%

Motor
Torque

Motor Speed
Fig. 7. Family of efficiency curves.

1800
II
II

1600

iI
IIIIIIlllllllll
IIII
I11 111 III
111111111111111

tlI I I I I |l l1l l1I1l l

III

III
III

12{]0

III
III

ooo.,,,,IIIIIIIII,

l..[..illlllllll

800 i i

iiii

iii

IJ
II

I|lilllllllJl
Illl
Ill[
II,#ll
I 11111111,,"lll
I Ill
Illllll
I

II
I Ilw'Pmllll
II
L~I~il l l l l l l ]
I I I,ii~l

III

,.,,,-,,IIIIIIIII,,I

L,I~IIIIIIII

II

I
II

llllllili

L~,,I~III iIii
iii
iiii
iii
i I
6 o o 1 1 1 i ~-,.,-,~
i ~ I - u 'ir n1...i 11 I iIii i i1 i li iIl l I i li [i i ii i[ 1i Li Ii . Ii 1

'lO0 111111111"t"21111111iilill
:zoo 11
20

I I .I l .l l .l l .l
25

. . . . .

III

I| I .I.I. I. I I

IIIIIlirlil
III1

i ....

III

i ....

30
35
40
45
Vehicle Speed (mph)
Fig. 8. Plot o f power v s . speed.

II
50

5. C h a s s i s
5.1. F r a m e
Sunrunner's tubular spaceframe bore static loads of subsystem comp o n en ts and incurred dynamic loads when in motion. Design requirements
included safety considerations, light weight, ease of fabrication, reliability,
and repairability.
The frame structure consisted of 67 thin-walled aluminum tubes welded
at 23 nodes, weighing 15 kg (33 lb) with bracketry. Composite materials
were ruled out because of the lack of weldability and availability of an epoxy
for nodal adhesion with satisfactory shear strength qualities.
The frame was modelled on a finite element analysis c o m p u t e r program,
where masses were " l u m p e d " at different nodes representing masses of
various subsystems. The total static loading equalled 224.4 kg (494 lb) and
was broken down as follows.

439

Subsystem

Mass (kg (lb))

Nodal
distribution

Batteries and power trackers

Body shell
Driver, seat and restraints
Motor and driveline
Solar cells

54.4
56.7
88.2
7.0
18.1

2
9
4
2
2

(120)
(125)
(194)
(15)
(40)

Dynamic driving and safety conditions were simulated, under nine loadcases, by placing forces axially on frame members. Each m e m b e r was checked
for buckling or yielding. The loadcases included still static loading, cornering
loading, braking force, combined cornering and braking, twist bump force,
jacking torsion force, towing force, simulated roll analysis, and side impact
analysis.
Under still static loading, the frame was constrained at the wheels,
suspension links were modelled as rigid elements, and one force of graviW
(1 G) was ex er ted on the lumped masses. This case ensured that the frame
would s u p p o r t its own weight, although the solar car would never experience
a 1 G acceleration. For the twist bump force, the front right and rear left
wheels were constrained, as a vertical downward acceleration of 4 G was
placed on lumped masses, simulating a torsion bump. Under simulated roll
analysis, the frame was constrained at the t o p two nodes and an upward 4
G acceleration excited the lumped masses. This case rated the stability of
the frame for rollover o c c u p a n c y protection. In side impact analysis, the
frame was sandwiched, as 454.5 kg (1000 lb) nodal forces were applied to
each side of the frame, testing occupant protection from the sides.
F r o m the analysis, frame m e m b e r forces were extracted and used to
calculate m a x i m u m stresses, which were c o m p a r e d with yield strength and
critical buckling stresses, offering the following results:
Loadcase

1 G
4 G
4 G
Side

still static loading

twist b u m p force
roll analysis
impact

5.2. Front s u . ~ . s ~

Element

Nodal

stress ( M P a )

displacement
(mm)

13.23
62.44
131.43
61.01

0.38
2.39
2.89
1.41

Buckling

No
No
No
No

( ~ . 9)

Concerns regarding packaging a front suspension within body sidepods

led to considerations of a double A-arm suspension and MacPherson strut.
Because A-arms would have been e x p o s e d and the aerodynamics affected,
the final design was a modified MacPherson strut, where a 70 lb inch -1
spring fit over a gas-charged damper. The d a m p e r was inserted into a tubular
housing, to the b ot t om of which a bearing housing was welded. The lower
spring seat, o u t e r steering arm, transverse link ring, and brake mounts were

440

Link Ring

Steering Tie Rod

I
Transverse Link - - ~

Body Shell
Bearing Housing
Wheel

YL
Fig. 9. Drawing of the front suspension.

c o n n e c t e d to the s t r u t housing. T h e o u t e r s t e e r i n g a r m a t t a c h e d to t h e
s t e e r i n g tie r o d b y m e a n s of a f e m a l e r o d end. Toe-in and t o e - o u t w e r e
c o n t r o l l e d b y adjusting a h e x a g o n a l stud, the e n d s o f w h i c h w e r e t h r e a d e d
in o p p o s i t e directions. T h e b e n t tie r o d s w e r e h o o k e d to inner s t e e r i n g a r m s ,
w h i c h b o l t e d into t h e r a c k a n d pinion. T h e split t r a n s v e r s e link ring r o t a t e d
a b o u t t h e s t r u t t u b e and w a s a t t a c h e d to a U - s h a p e d yoke, t h r o u g h w h i c h
a stud t h r e a d e d into t h e t r a n s v e r s e link. The t r a n s v e r s e link c o n n e c t e d to
a f r a m e n o d e in the c e n t e r of the c a r directly b e l o w t h e s t e e r i n g wheel.
B r a k e m o u n t s w e r e w e l d e d to the strut housing, h o l d i n g a h y d r a u l i c caliper.
T h e b e a r i n g h o u s i n g c o n t a i n e d p l a n e b e a r i n g s t h r o u g h w h i c h the axle w a s
inserted. At t h e b o t t o m of t h e b e a r i n g h o u s i n g s a stud w a s welded, o v e r
w h i c h a h i g h - d i s p l a c e m e n t m a l e r o d end w a s a t t a c h e d , t h r e a d e d into the
leading link, r e g u l a t i n g 5 of caster. The leading link a t t a c h e d to b r a c k e t s
in t h e f r a m e ' s rear.
Initially the s t r u t a n g l e ~b w a s d e s i g n e d at 15 f r o m t h e vertical. This
allowed 6.35 m m (0.25 inch) o f o u t b o a r d s t e e r i n g offset, e n s u r i n g t h a t the
f r o n t w h e e l s w o u l d self c e n t e r during braking. H o w e v e r , owing to b o d y
p i t c h i n g f o r a e r o d y n a m i c s and p a c k a g i n g c o n s t r a i n t s , the s t r u t a n g l e w a s

441

changed to 8 , forcing a 0.75 inch inboard steering offset. Extensive testing

revealed that this ge om et r y change did not affect overall vehicle handling,
although the wheels did not self c e nt e r during braking.
The transverse links, leading links, and strut tubes, were designed to
withstand max imum axial loads and bending moments, derived from most
e x t r eme accelerations and loadings.

5.J. R e a r suspension
Rear suspension design constraints included packaging, light weight, and
reliability u n d er 1 G cornering and 4 G twist bump conditions. Initially, to
avoid suspension links protruding from the body shell parallel to the ground,
a fork-type suspension was considered. However, inability to withstand bending
m o m e n t s generated during e x p e c t e d lateral motion ruled out this option.
The end choice was a modified trailing arm system where at one end the
arms were attached to the chassis with spherical rod ends, and at the other
they came t o g e t h e r and were welded to a bearing housing. A coil-over shock
a b s or b er providing springing and damping was m ount ed on the top of the
bearing housing. The final spring rate was 50 lb inch -~ with an average
motion ratio (AwheJAsp~g) of 1.01.
5.4. Axles, wheels a n d tires
Axles were designed based on vehicle weight and projected road inputs.
On one side of the bearing housing a double spline system was used; a
cross flange, with inner bore female splines fastened with a friction fit over
a section of the axle with extruding splines. The wheel hub was then coupled
to the outside o f the cross flange. On the other side of the bearing housing
a brake flange slid over intruding axle splines and bolted to a rotating brake
disk.
Axles were designed under worst case conditions that included a 2.6 G
vertical acceleration for road bumps, a 0.7 G horizontal acceleration for
e x t r e m e turning conditions, and a full caliper brake t orque on the disk. A
c o m p u t e r p r o g r a m was written to determine principal axle stresses with
these acceleration inputs.
The axles fit through the wheel hub which was strong enough to withstand
stresses o f spokes, yet soft enough to allow t h e m to seat properly. A com prom i se
was found in dimensioning the hub, as its length was directly relative to
lateral strength and inversely relative to its normal strength. All terrain bicycle
wheels are attached with straight spokes, which can withstand higher strain
p e r unit force than the double-butted type. Ability to withstand strain minimized
lateral wheel deflection, which prevented the wheels from coming out of
alignment.
Bald mountain bike tires with low rolling resistance fit over the rims,
and the entire assembly was tested. With the axle perpendicular to the
ground, 29.9 N (216 lb) were hung from the rim end, and a plastic deformation
of 1.5 mm (0.060 inch) was noted. 62.3 N (450 lb) was then placed on a

442

wheel, with the axle parallel to the ground. Here, no elastic deformation
was observed.

5.5. Brakes
GM Sunrayce USA regulated that brakes must be able to achieve a 0.43
G stopping deceleration. Initially, cable actuated calipers were installed;
however, unsatisfactory pe r f or m a nc e led to a hydraulic system, offering 0.51
G. The 5 inch bolts went through the floating calipers into unthreaded holes
in the brake mounts.

5.5. Steering
The steering system consisted of a rack and pinion, attaching to an
inner steering arm which connect ed to the outer steering arm by means of
the tie rod. A sprocket was threaded to the rack encasem ent and was
c o n n ected to another sprocket threaded to the steering column. The steering
wheel rotated the column sprocket, allowing for vehicle manipulation.
In designing the steering system, the objectives were to incorporate
"Ackerman steering", and to avoid " b u m p steer". Ackerman steering minimizes
frictional energy loss, or "tire scrub", when turning. Here, the outside wheel
actually turns less than the inside wheel, allowing lines perpendicular to the
center of each wheel to extend to a point called the "Ackerman center".
Also passing through this point is a line which intersects the rear wheel
axles perpendicularly. It was designed into the system when lines from the
outer steering arms were brought to a point which fell exactly in between
the rear wheels.
To avoid b u m p steer, suspension pivot points must move exactly like
steering pivot points. By designing both sets of pivot points to fall in the
same horizontal plane, bum py roads would not induce Sunrunner self-steering.
Part

Material

O.D.

W.T.

Function

Spaceframe
Strut housing
Leading link
Transverse link
T. I. y o k e
S t e e r i n g tie r o d
Outer steering arm
Inner steering arm
Trailing arms
Rack and pinion
Brake disk
Wheel hubs
Axles
Wheels

6061-T6Al
4130 steel
3A1-2.5V-Ti
3AI-2.5V-Ti
7075-T6Al
3AI-2.5V-Ti
4130 steel
6061-T6A1
3A1-2.5V-Ti
6AI-4V-Ti
7075-T6A1
2124A1
6A1-4V-Ti
6061A1

1
2
1
0.625
0.625
0.125
0.125
1
6 inch plate
0.669
26.0

0.049
0.125
0.051
0.042
0.042
plate
plate
0.051

Bear dynamic and static loads of car

Houses strut insert
Caster adjustment, bears braking forces
Camber adjustment
Connect strut housing to transverse link
Toe-in-toe-out adjustment,
Transfer turning force to wheel
T r a n s f e r t u r n i n g f o r c e t o tie r o d
Rear suspension
Transfer turning force to inner arm
Rotate at wheel rev rainSupport wheels
Support wheels
Support vehicle

0.125
-