My PhD research

1. "Slippery When Dry“
GripTester – UK Roads User Group
Meeting
Wednesday 9th August 2006,
National Motorcycle Museum,
Solihull
John C Bullas BSc MSc MIAT MIHT FGS
PhD Researcher
Transportation Research Group
University of Southampton
w.e.f. September 2006
Asst. Group Engineer
Transport Solutions
Atkins Highways & Transportation

2. Basics &
Background

3. Tyre/Road Friction
The Friction Force on a road is the sum of three components
Fad = Component due to (interfacial shear) adhesion, rubber vs substrate
Fdef = Component due to bulk rubber deformation and hysteresis loss at the
interface
Fab = Component due to abrasion and/or tearing of the rubber
For rough and or textured hard surfaces, the Fdef term is usually dominant since such texture
promotes a high level of interface deformation with attendant hysteresis losses. For smooth
surfaces the Fad term usually plays a dominant role. Both Fad and Fdef are usually present in any
frictional contact; it is their relative contributions that are important.
For any surface-rubber combination the magnitude of all three components is influenced by
temperature, sliding velocity and the presence of a lubricant.
VEITH, A. G. (1998) Rubber vs Countersurface Frictional Behaviour: How this Influences the Traction of Pneumatic Tyres.
Progress in Rubber and Plastics Technology, 14, 1.

4. Macro- and Microtexture
Microtexture:
An Indirectly
measured
property related to
the provision of wet
skidding resistance
Macrotexture:
A Directly measured property
influencing the Tyre/Road
contact improving skid
Resistance especially in the
wet at speed
HOSKING, R. (1992)
Road aggregates and
skidding, HMSO.
VINER, H. E., ROE, P.
G., PARRY, A. R. &
SIHAL, R. (2000) High
and low-speed
skidding resistance:
the influence of
texture on smooth
and ribbed tyre
friction.
Caracteristiques de
surface des
chaussees routieres
et aeronautiques;
Fourth international
symposium on
pavement surface
characteristics of
roads and airfields.
Nantes, France,
AIPCR/PIARC.

5. NTS: The New Generation
of Road Surfaces
OLD: roadwork delays, perceived rutting
problem, noise and spray issues. Two Part
Installation
~Positive Textured Hot Rolled Asphalt (HRA)
Asphalt + Rolled in Coated Chippings
NEW: Lower noise, lower spray, use
for thin resurfacing/regulating, Less
rutting(?), Single Stage Installation =
less delay
~ Stone Mastic/Matrix Asphalt (SMA),
a Negative Textured Surface (NTS)
http://www.highwaysmaintenance.com/

6. We (the highway engineers)
have assumed ..
NTS Negative Textured
Road Surfaces would
behave similarly to
Positive Textured
ones like HRA
….. Especially with
respect to their DRY
frictional properties
during emergency
braking…
WALSH, I. (2000) Out of the skid pan…. Surveyor.

7. Highway Engineers do not
measure DRY road friction!
BUT the UK we first
tried measuring the
dry friction of roads
in 1936 with our
motorbike wet skid
tester and the dry
friction was so good
we didn’t really try
again…
…until the 1970’s
when it was still
very good so we
stopped looking
BIRD, G. & SCOTT, W. J. O. (1936) Studies in Road Friction: I. Road Surface
Resistance to Skidding Department of Scientific and Industrial Research /
Ministry of Transport.
HOSKING, J. R. & WOODFORD, G. C. (1976) Measurement of skidding
resistance part i. guide to the use of SCRIM, Transport Research Laboratory
Report LR737. Transport Research Laboratory.

8. 1944: Bitumen & Maximum Friction
ZIPKES, E. (1944) Die Reibungskennziffer als Kriterium zur Beurteilung von Strassenbelägen Eidgenössissche.
Verlag A.G. Gebr. Leeman and Co.

9. 1944 E.Zipkes (Switzerland)
“…..On some road surfaces, the melting point of the binder may be
reached before that of the tread rubber, in which case the slipping co-
efficient will have a different value from that on which the rubber melts
first”
Remember this quote for later!
ZIPKES, E. (1944) Die Reibungskennziffer als
Kriterium zur Beurteilung von Strassenbelägen
Eidgenössissche. Verlag A.G. Gebr. Leeman and Co.

10. 1985: The M4 Motorway (UK), 13
are dead…
13 dead owing to a
combination of:
•Poor driving
•Substandard central
barrier height
•Extended DRY braking
distances [longer than in
the WET] on a coated
stone chippings with a
thick layer of bitumen
“…..some of the worst
DRY skid resistance
seen in 10 years” were
measured by the Police
Crash Investigator
ANON (1986) Low barriers conceded at M4 accident site. New Civil Engineer.
BYRD, T. & DADSON, J. (1986) Safety barrier review follows M4 carnage New Civil Engineer.
SHELSHEAR, G. (1986) Poor Skid Resistance cited at Crash Inquest. New Civil Engineer.
SHELSHEAR, G. (1986) Resume regarding "Poor Skid Resistance cited at Crash Inquest".
SHELSHEAR, G. (1986?) Statement by Mr D Simpson Concerning Safety Fences - M4 Berkshire
Crossover Accident 23.6.86
SHELSHEAR, G. (1986?) Statement on Skidding resistance by Mr P E Nutt - M4 Berkshire
Crossover Accident 23.6.86

11. …On a road more Slippery in
the Dry than in the Wet
MANDERSON, J. & RUDRAM, D. (1993) The variation in tyre/road friction with time. IN JACOB, B.
& BONTE, W. (Eds.) Vol 3; Forensic criminalistics 1. Dusseldorf; Germany, Berlin.

12. The Netherlands: 1990’s
Crash Investigation: ABS v NOABS
Porous Asphalt (P.A.) Negative Texture,
low noise, low spray
FAFIÉ, J. (2004) Early Life Dry Skid Resistance. Paper
presented at “Surf 2004”, 5th Symposium on Pavement
Surface Characteristics. Toronto, Ontario, Canada.
RICHARDSON, J. T. G. (1999) Stone Mastic asphalt
in the UK. SCI Lecture Series. Society of Chemical
Industry.

13. NOABS on P.A.
ABS on P.A.
ABS on ASPHALT
Elapsed Test Time 
Deceleration
JUTTE, R. H. & SISKENS, C. A. M. (1997)
A Material Technological Approach to the Low
Initial Skid Resistance of Porous Asphalt Roads.
European Conference on Porous Asphalt,
Madrid. Ministerio de Fomento, Direccion
General de Carreteras, Spain.

14. “The low skid resistance
(was) caused by melting of
the mortar” (the bitumen)”
The Porous Asphalt had a
thick layer of bitumen on the
chippings when new.. this
doubled the stopping time
when ABS was turned off!
Research published by D.W.W.
(NL) in 1997**
**JUTTE, R. H. & SISKENS, C. A. M. (1997) A Material Technological
Approach to the Low Initial Skid Resistance of Porous Asphalt
Roads. European Conference on Porous Asphalt, Madrid. Ministerio
de Fomento, Direccion General de Carreteras, Spain.

15. UK 2001: Two fatal crashes
on new(ish) SMAs
DRY =0.52, 0.51 Downhill 1:100
DRY =0.53, 0.55 Uphill 1:25
 Fatal Crash: Police NOABS Skid Testing
NB: Low values of Dry  were NOTHING to do
with the circumstances of the crashes
DRY =0.56, 0.48, 0.48, 0.49, 0.51
Fatal Crash: Police NOABS Skid Testing 
ALLEN, J. (2001) “Stone Mastic Asphalt”:
Presentation to the ACPO Senior Collision
Investigators Conference, Hull.

16. dry≈0.5.. IS NOT a typical
value for dry road friction
GOUDIE, D. W., BOWLER, J. J.,
BROWN, C. A., HEINRICHS, B. E.
& SIEGMUND, G. P. (2000) Tire
Friction During Locked Wheel
Braking. Accident
reconstruction: simulation and
animation. Detroit, MI, Society of
Automotive Engineers.

17. Any references to all
that Past Research?
• Little (if any) reference has
been made to any of the
research from the 1940’s,
1980’s and 1990’s by those
commenting on recent findings
in the UK … other than by the
surfacing industry [and
someone else]

18. And who even bothers to
measure the dry friction of
roads now?

19. ……to estimate the
braking speed and/or
speed at inception of loss
of control from skid marks
at the scene of fatal or
near-fatal crashes
Police
Collision
Investigators
do

20. Dry Friction Tests using Decelerometers
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
0 1 2 3 4 5
SkidMan/Vericom
decelerometers and data
output from a skid test
(hard copy and
downloaded)

21. Skid tests to estimate 
& critical speed
DERBYSHIRE
CONSTABULARY
ROAD POLICING
SUPPORT UNIT
(2005) Equations
Sheet.

22. Used in reconstruction of the
crash scene to combine survey
data from the scene and data
from the skid test to estimate
critical speeds from tyre marks
and curve radii

23. Today: NO ABS v ABS on a DRY road: extended
stopping time & distance for a NEW binder rich
surface and greater ABS/NOABS  ratios
Skidman Decellerometer Tests on DRY NEW HRA
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
0 1 2 3 4 5
Time (seconds)
(G)
NOABS 13/12/2004
17:07:00 35mph
ABS ON 13/12/2004
17:11:00 36mph

24. But Dry Friction improves over (some)
time

25. ABS vs. NOABS in
the DRY…
So what is happening during emergency braking
to give the low dry

26. The thermal/dynamic regime
causing the “bituplaning” event

27. Typical Bitumen Properties
AFTER: READ, J. & WHITEOAK, D. (2003) The Shell Bitumen
handbook, London, Thomas Telford.
Engine Oil Viscosity 0.5-0.2
-- Softening Point >

28. But is it actually melting?
Heat generated due to (interfacial shear) adhesion, rubber vs substrate
melts surface and then friction drops
OR
•Something else happens to reduce the friction and this leads to the heat
generation dropping with less friction

29. The (statistical) devil is in the
detail…..
• Tabulation and visual
classification of key phases
within each simulated
emergency braking event for
a 300+ SkidMan event
database
• To identify (and statistically
prove the significance of)
any differences between the
emergency braking
deceleration characteristics
delivered by the combination
of NEGATIVE & POSITIVE
surface texture types,
ABS/NOABS braking systems
and DRY and DAMP and WET
surfaces

30. Findings to date:
• The Low Dry Friction (‘bituplaning’)
phenomenon is easy to reproduce on
‘new’ bituminous surfaces of both
texture types with thick binder films
• The ‘bituplaning’ deceleration
characteristics observed are almost
identical to those reported on in NL on
PA.
• The ‘bituplaning’ phenomenon is NOT
just a high speed event ( just need
30mph)
• ABS braking can lead to momentary low
dry friction with “dash like” skidmarks
now more commonly left by ABS braked
vehicles
• Temperatures likely to be high enough to
melt bitumen have been measured and
the dynamic temperature generation
mechanism visualised for the first time.
Skidman Decellerometer Tests on NEW HRA
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
0 1 2 3 4 5
Time (seconds)
(G)
NOABS 13/12/2004
17:07:00 35mph
ABS ON 13/12/2004
17:11:00 36mph

31. Other Effects?
• Effect of ‘negative’ texture on the
transition between in “in-control” and
“out of control” states (which value of
mu to use in ABS cases?) and changes
to the assumed deceleration/slip
relationship
Slip versus Deceleration - Tack Coat
-20
0
20
40
60
80
100
120
25 26 27 28 29 30 31 32
Time (Seconds)
Slip(%Uncorrected)
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
Decellertion(G)
SLIP%
L AccG

32. BUT are Cars without ABS also equipped with fewer
secondary safety features and driven by the youngest
and oldest drivers least able to “handle the situation”?
REMEMBER: Extended
‘bituplaning’ only seems
to occurs with NOABS
Emergency braking on
DRY roads… AND only 4%
of ALL crashes DO NOT
have human contributory
factors associated with
them**
**BROUGHTON, J., MARKEY, K. A. & ROWE, D. (1998) A new system for recording
contributory factors in road accidents. Crowthorne, TRL.
“Toxicology reports on Mr Taylor
showed recent use of heroin and
cocaine, the inquest heard”

33. johnb@soton.ac.uk
john.bullas@gmail.com
Thank you for listening!