Download - Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
1/70
Report No. COOT
OTO R 93 3
INVESTIG TION OF THE
MODIFIED
LOTTM N TEST
TO PREDICT THE
PERFORM NCE OF
IN
COLOR DO
Timothy Aschenbrener
Colorado Department of Transportation
4340
East
Louisiana
Denver. Colorado 80222
Robert
B. McGennis
The
Asphalt Institute
Lexington
Kentucky
Final Report
April 1993
Prepared
in cooperation
with the
U S Department
of
Transportation
Federal
Highway of Administration
STRIPPING
P VEMENTS
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
2/70
The
contents of
t h i s repor t r e f l ec t
the
views of
authors
who are
respons ible
for the fac t s and
the
accuracy
of the data
presented
herein .
The
contents do not necessar i ly
r e f l ec t
the of f ic ia l
views
of
the
Colorado
Department
of
Transporta t ion
or
the
Federal Highway Adminis trat ion. This
repor t
does not cons t i tu te
s tandard
spec i f ica t ion o r
regula t ion .
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
3/70
CKNOWLEDGEMENTS
Larry
Lamar,
Dave
Gallegos, Brad Black,
Ken
Medina, Ju l ie
Le
and
Kim
Gilbert of the Colorado Department of
Transportat ion
CDOT) performed a l l of
the te s t ing
for t is study.
I
am also
grateful for the numerous Region Materials personnel
who
sampled
and delivered the
3
tonnes of aggregate used in the study.
The
CDOT Research Panel provided
many
excel lent
comments and
suggestions
for
the
study.
t included:
Gerry
Peterson
and
Jay
Goldbaum CDOT-Region 1 Materials) , Bob
LaForce CDOT-Staff
Materials) ,
Donna
Harmelink
CDOT-Research),
and
Mark
Swanlund
FHWA-Colorado
Division .
Special Thanks to
the
expert panel of Colorado
asphal t
paving
experts
who
provided
numerous
ideas
and
suggestions which
made
t i s
study more informational:
Bud
Brakey Brakey Consult ing
Engineers),
Jim Fife
Western
Colorado Testing),
Darrel
Holmquist
CTL/Thompson),
Joe
Proctor Morton
Internat ional) , and Eric
west
Western Mobile .
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
4/70
Technical Report Documentation Page
1.
Report No.
2. Government Accession No.
3
Recipient s Catalog No.
CDOT-DTD-R-93-3
4. Title and Subtitle
Investigation of the Modified Lottman Test to Predict the
Stripping Performance of Pavements in Colorado
5 Report
Date
Aj)ril1993
6.
Performing Organization
Code
File No 13.07
7. Author(s)
Timothv Aschenbrener and Robert
B
McGennis
9. PCI·forming
Organization Name
and Address
Colorado Department of Transportation
4201 East Arkansas Avenue
Denver Colorado 80222
12. Sponsoring Ageney
Name
and Address
Colorado
Department
of Transportation
4201 East Arkansas Avenue
Denver, Colorado 80222
15. Supplementary Notes
8. Performing
Organization
Rpt.No.
CDOT-DTD-R-93-3
10.
Work Unit
No. (TRAIS)
11. Contract
or
Grant No.
13. Type of
Rpt
and
Period
Covered
Final
Renort
14. Sponsoring Agency Code
Prepared in Cooperation with the U.S. Department of Transportation Federal
Highway Administration
16.
Abstract
Moisture damage to
hot
mix asphalt pavements has been a sporadic
but
persistent
problem in Colorado even though laboratory testing is performed to identify moisture
susceptible mixtures. The laboratory conditioning was often less severe than the conditioning
the hot mix asphalt pavement encountered in the field.
Twenty sites of known field performance with respect to moisture susceptibility, both
acceptable and unacceptable, were identified. Material from these sites were tested using
seven versions of the modified Lottman test (AASHTO T 283) and the boiling water test
(ASTM D 3624).
For modified Lottman testing, two levels of severity for conditioning laboratory samples
were identified that correlated well with conditions in the field.
For
mixtures placed under
high traffic, high temperatures, high moisture, and possible freeze, the severe laboratory
conditioning defined in the report should be used.
The
milder laboratory conditioning
defined in this report
is
appropriate for low traffic sites.
17. Key Words
Stripping
Lottman
Case Histories
19.5ccm·ity ClassiC. (rcpOl·t)
Unclassified
Boiling Water Test
Moisture Susceptibility
20.Seew·ity ClassiC. (page)
Unclassified
18.
Distribution Statement
No Restrictions: This report is
available to the public through
the National Technical Info.
Service.
Sprindield VA
22161
21. No. of Pages
7
22.
Price
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
5/70
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
6/70
VI.
TEST RESULTS
ND
DISCUSSION
• • • • • 21
VII.
VIII
Analysis
of
Untreated
Aggregate Quali ty
• • •
22
Inf luence-Freeze/No
Freeze/30-Minute
Satura t ion 22
Comparison
of
TSR s with Actual Performance • • • • • 25
Analys is
of
Swell
29
Analysis
of
Short
Term
Aging
• •••
31
Analysis
of
Lime Addit ion
• • • •• 34
Analysis
of the
Boil ing Water Test ••
35
MODIFIED LOTTM N REPEATABILITY
• • • • • • • • • • • •
36
C SE
HISTORIES
OF
OTHER INVESTIGATIONS •
37
Stuart
1986 37
Kennedy
1983
38
Parker
1988
39
Summary
39
IX
CONCLUS IONS 40
X IMPLEMENT TION
• •••• • ••••••••
43
XI.
RECOMMEND TIONS FOR
FUTURE
RESE RCH
• • • • • • • • • • • • 44
XII REFERENCES 45
v
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
7/70
LIST OF
T BLES
Table Page
Number Number
1 Si tes Used in This Study . • • • • • • • • . . . • . . . . . . . . . . . . • • 2
2
Aggregate
Gradations
and
Optimum
Asphalt
Contents
for HM Mixtures
Used in
This Study• • •
10
3 Comparison of the Original Lottman
to
Currently
Recommended
Procedures
. • • • • • . . . • . . . • • • • • • • • • . • • • • • 11
Experimental
Grid
for the Stripping Study
• • • . • . • • . 19
5 Comparison of Tensile
Strength Ratios
• • 23
6
Comparison of
Pavements
of Known
Field Performance
with AASHTO
T
283 Severi ty Level
2B)
••••
25
7 Comparison of Pavements of Known Field Performance
with AASHTO T 283
with No Freeze
Severi ty Level
2C) • . . • • • • • . . . . . . . • . . • • . • • • • • • • • • • 25
8 Comparison of Pavements of Known Field Performance
with
AASHTO T
283 with
a
30-Minute
Saturat ion
Severi ty Leve l l ) • . • . . . . • • . • • • . . . . . • • . • . • . • • • • • . .
25
9 Summary of Swell After Saturat ion
. • • • • • • • • • • • • • • • •
28
10
Summary of
Swell
After Conditioning • . • • • • • • • • . . • • •
31
11 Summary of Dry Tensile
Strengths
and Tensi le
Strength
Ratios
Using
Various
Lengths
of
Short-
Term Aging STA) • ••••••• ••••••••••••••••
32
12
Boil ing Water Test Results • . . • • • • • • • . . • . . . . • • • • • . .
36
13 Comparison of Pavements of Known Field Performance
with
the Boiling
Water Test
ASTM D 3625)
• • • • . • • . •
36
14 Comparison
of
the Lottman Test Level 2A) to
Actual Performance •• • • •• ••• 37
15 Comparison of the Lottman Test
Level
2C) to
Actual Performance
8
16 Comparison of the Lottman Test Level l )
to
Actual Performance • . . • . . . . • • • • ·
• •
38
17 Comparison
of
the Lottman Test Level
2B) to
Act
ual Performance 9
18 Comparison of the Lottman Test
Level
2C) to
Actual Performance 9
v i
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
8/70
LIST
O
FIGURES
Figure
Page
Number Number
1 A High
Maintenance
Mix
Experiencing
Raveling
f te r 15 Months •
2 A
Disintegrator After
6
Months
• • • • • • • • •
4
3
The
Surface
of
a Pavement
Requiring
Complete
Rehabi l i ta t ion
in 12
Months • • • • • • • • • • • • • •
6
4 A Core Showing
Str ipping
Below
the
Surface from
the
Pavement n igure
6
5 Environmental Conditions
for the
Pavement a t
Agate
Requiring
Complete
Rehabil i tat ion
f ter 12
Months
• 7
6 Environmental Conditions for the Pavement a t Cedar
Point
Requiring Complete
Rehabi l i ta t ion
f ter 11 Months 7
7 Environmental Conditions for the Pavement a t Arriba
Requiring Complete
Rehabil i tat ion
f te r 1 Months • • • • • 8
8 Environmental Conditions for the Pavement a t
Limon
Requiring
Complete
Rehabil i tat ion
f ter 1 Months • 8
9
Ranked
Order
of
Tensile Strength Ratios
from
SHTO
T 283 with a 30-Minute Saturat ion • • • • • • • • • •
27
1
Relat ionship
of
Swell After
Conditioning Versus
Tensi le Strength Ratio from
Samples Tested According
to
SHTO
T 283 with 30-Minute Satura t ion • • • • •
30
11 Influence of Short-Term
Aging on
ry Tensi le Strength • ••••••• 33
12 Influence of
Short-Term
Aging on
Tensi le Strength
Ratio • • • • • • • • • • • • • • • • • • • •
33
v i i
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
9/70
PPENDICES
ppendix • • Summary of Laboratory Test Data
ppendix
B
• • • •
Comparisons
of
Different
Versions
of the
Modified Lottman
Test
from
Other
Invest igators
vi i i
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
10/70
I INTRODU TION
Moisture damage,
otherwise
known as s tr ipping , to hot mix
asphal t HMA) pavements has been
a
sporadic but pers is tent
problem on
projec ts
in Colorado. After
premature
moisture
damage was observed on a
specif ic
project in
July
of
1991, the
Asphalt Ins t i tu t e was requested
to
perform a jo in t study with
the Colorado Department of
Transportation
CDOT). The
resul ts
of the jo in t study
were
reported by
McGennis (1)
in october of
1992. Recommendations included:
-
evaluating aggregates of
known
f ield performance with
several versions of the
moisture
suscep t ib i l i ty t es t
used by Colorado,
-
evaluating aggregates of
known
f ie ld performance
without
lime
or l iquid anti-s tr ipping
addit ives,
-
evaluating the sand
equivalent t e s t
-
implementing
a
bet ter
P200 management
st ra tegy during
construction,
l imi t ing the quantity
of
P200
in
HMA,
and
-
mill ing ruts instead of using level ing
courses.
This report
presents resul ts
from
the
f i r s t
two
recommendations;
a l l
of
the
other recommendations have been invest igated and
implemented. The
purpose of
th is
report s
to
compare HM
pavements of known f ie ld performance with resu l ts from
various
laboratory
moisture
susceptibi l i ty
t e s t s .
The laboratory
t e s t s
invest igated
were
the modified
Lottman
AASHTO
T
283)
and
various versions, and
the
boil ing water t e s t ASTM D
3625).
1
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
11/70
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
12/70
Good Some aggregate
sources in
Colorado have a
good
his tory of providing pavements
tha t
re s i s t moisture
damage.
Seven dif fe ren t aggregate sources with a history of excel lent
performance were selected for invest igat ion.
A
specif ic
project using
each
aggregate
source was then
studied
in deta i l
for
th is
invest igat ion.
High
Maintenance
These pavements have
received
an
exceptionally high level
of
maintenance.
Although
pavements
in th i s category are s t i l l in service
a f te r
two to f ive
years,
the i r
performance i s
considered
unacceptable
when
compared
to
the i r design l i f e .
The
maintenance required to
address
problems
from
moisture
damage to
the
HM
pavements
included
overlays and s ignif icant patching
of
potholes . A 15-month old
pavement
tha t required an overlay
on some
sections i s
shown
in Fig.
1 .
Dis integrators
There
are several
aggregate sources used
in
HM
pavements
tha t
have a notorious
history of
severe
moisture
damage.
A 6-month
old
pavement
tha t
disintegrated i s
shown
in Fig. 2. Since
contractors
have not used
these
aggregate
sources
on
CDOT
projects for many years ,
specif ic
mix
designs for
the
dis in tegrators
were di f f i cu l t
to
obtain.
The
mix designs with
the
aggregate sources thought
to
be
disintegrators
were reproduced
as
closely
as
possible
with
the
help
of
experienced, long-term employees of
the CDOT
Complete Rehabil i tat ion
Several pavements in
Colorado
required
complete
rehabil i ta t ion
when le ss than
two years old,
and
often when
le ss
than
one
year
old.
The
moisture
damage
was
re la ted to
a
unique
pavement design
feature, ru t res i s tant
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
13/70
Fig . I .
A
"High Maintenance" Mix R t p ~ r ; . i . p . n c i n g Rave l ing
Af t e r
I '
~ ' : o . n t l 1 . n .
Fig . 2 .
A
Dis in t eg ra to r
Af t e r
Months .
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
14/70
composite pavement,
tha t
ut i l ized a plant mixed seal
coat
PMSC) as described and evaluated by Harmelink
(4).
HM
pavements
di rec t ly
below the PMSC exhibited severe moisture
damage.
The
pavement
surface (Fig.
3)
and a core showing
the
moisture
damage
tha t occurred jus t below the surface
(Fig.
4)
are
shown
for
a
pavement requir ing
complete
rehabi l i ta t ion
a f t e r 12 months.
Even
though
the
PMSC was a
contribut ing
£actor to
the
dis t ress in the underlying
HMA,
the HM was s t i l l
considered to
be
susceptible to
moisture damage
since t fa i led
so
quickly.
The weather
condit ions
tha t contributed to the fa i lure of the
pavements
requiring
complete
rehabil i ta t ion
examined
in
th i s
study are shown in Figures 5 through
8.
The temperature i s
the
monthly mean maximum
temperature,
i e the average of the daily
high temperatures.
The precipi ta t ion i s
the
to ta l
accumulation
for
the month. The f i r s t month and
year
in
each
f igure
represents the
end of construction, and
the
f ina l
month
and
year in
each f igure
represents the time of
fa i lu re
Pavements
requiring complete rehabil i ta t ion a l l fa i led when
high
levels of precipi ta t ion
occurred in
the
hot tes t
part
of
the summer
Even
though
a l l
pavements in Colorado
are
subjected to freeze cycles, the
severe
moisture damage did
not correspond
with
freezing conditions.
The instantaneous
fa i lu res were direc t ly re la ted to a
simultaneous
combination
of
high temperature,
high
moisture, and
high
t r a f f i c
The
environmental data
used in
th is report was obtained from
the
weather
s ta t ion
located
closest
to
each
project
and
reported
by
the
National Oceanic
and
Atmospheric
Adm inistrat ion s
National Climatic
Data
Center.
5
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
15/70
Fig . 3 . The Surface o f a Pavement Requir ing
Comolete Rehab i l i t a t i on in 12 Months
F ig
4 A Core Showing St r ipp ing Below
the
Surface
from · t h r ~
P q N ~ m e n t
in Figure 3.
6
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
16/70
Environmental Conditions
Agate
100
Temperature
recipitation
~ - : : : : - - - - - - - - - - - , 4 . 5
90
80
70
60
50
40
30
20
10
8 9
10
12 2
3
4
88
89
Month and Year
_ Temperature
~
recipitation
5 6
4
2
1 5
1
Fig .
5 . Environmenttd
Condi t ions f o r t h e Pavement a t
Agate Requir..' ; ng Complete Rehab i l i t a t ion
A f t e r
12 : 1 ( o I · : h ~ .
Environmental Conditions
Cedar Point
Temperature recipitation
100 , - - - - - - - - - - - _ _ 4 5
90
4
80 3 5
70 3
60
2 5
50
2
40 1 5
30
20
10
9 10 11
2
2 3 4 5 6 7
89 90
Month and Year
_ Temperature
~
recipitation
0.5
o
Fig .
6
Environmenta l Condi t ions
fo r t he Pavement a t
Cedar Po in t Requir ing nComplete
R e h a b i l i t a t i o n
n
A f t e r
11
Months.
7
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
17/70
Environmental onditions
100
90
80
70
60
50
40
30
20
10
Arriba
Temperature recipitation
,
i i i
'
,
•
8 9
10 2 1 2
3
4
5 6 7 8 9
10 2 1 2
3
4 5
6 7
I
89
I
90
I
91
I
Month and Year
_
Temperature recipitation
4.5
4
3.5
3
2_5
2
1.5
0.5
o
Fig . 7 . Environmenta l Condit ions fo r
t h e Pavement a t
Arr iba Requi:o::ing "Complete R eh ab i l i t a t i o n
n
Afte r
24.
Mon-ths
.
Environmental onditions
Limon
Temperature recipitation
100 ---
--------------------
------------.4.5
90 4
80
3.5
70 3
60 2.5
50 2
40 1.5
30
20
10
10
2
90
34
9
Month and Year
5
_ Temperature recipitation
6 7
0.5
o
Fig . 8 . n v i r o n ~ q n t a l Condi t ions fo r t h e
Pavement
a t
Limon Ren"; .:: "-:-·;_:'1q "Complete R eh ab i l i t a t i o n
n
Afte r 10
on-'.:hn_
8
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
18/70
I I I . T ST
METHO OLOGY
The
original
mix design used
a t
each s i t e was
ident i f ied
Information
retr ieved
included
the aggregate sources
percentage of each
component
aggregate stockpile
component
and combined aggregate gradat ions optimum asphalt content
asphal t cement source and grade
and
anti s t r ipping treatment.
I t
was
not
possible to
use
the exact
aggregates and
asphal t
cements
from
the
original projects placed two to ten
years ago.
So
virgin aggregates
from the original
sources
used a t each
s i t e
were
sampled.
Addit ional ly
recent ly
produced
asphal t
cements
and anti s t r ipping treatments were
obtained
from the
original suppl iers
of materials
to the s i t e s .
The aggregates
from each s i te were then
blended
to
match
the
gradation used
on
the
project
as closely as possible.
A
mix
design
was then performed
to val idate the
optimum
asphal t
content
from
each
s i t e When
the optimum
asphalt content
of
the
new
mix
design
matched the
optimum
asphal t
content of the
original
mix
design
the moisture susceptibi l i ty t es t ing
proceeded. When the
optimum asphalt content of the
new
mix
design did
not
match
the
optimum
asphal t
content of the
original
mix
design t
was assumed the
aggregates had
changed
and the
new optimum asphalt
content was used.
o optimum
asphal t
contents used
in
th i s study varied
by more than 0.2
from the
original
designs.
The
aggregate
gradations
and
optimum asphal t
contents
of
the
HM mixtures used for th i s study are
shown
in
Table
2.
9
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
19/70
Table 2 Aggregate
Gradations and Optimum Asphalt
Contents
for RN Mixtures Used in This Study
Gradation mm and inches)
s i t e AC 19.0
12.5
9.50 4.75 2.36 0.60 0.30 0.15
0.08
3/4 1 /2
3/8
#4 #8 #30 #50 #100 #200
1
5.5 100
87
72
51
45 26
18
10
7.0
2 4.5
100
87
74
53
42
24
15 10
6.6
3
5.3
100
93
77
53
37 21
14
9
5 .9
4
4.9
100
88
66
50
40
21
14
8
5 .1
5
5 .0
100
94
80
52 41 31
18
10
7.1
6
6.0
100
88
51
37
22 14
10
5.9
7 5.7
100
91 74 49 3 7
18
12
8 4.7
8 4.8 100 94
77 49 38
24
18
12
8.1
9 5.9
100
96
62
41
25 13 10
6.1
10
5.0
100
86
77
55
43
26 18
13
8.6
11 4.9
100
97
57
40
21
15
11
7.8
12
5.0
100 86
76
54
42 25
18
13
8.4
13
5 .7 100
86
78
60 45 22 15
9 5.7
14 5 .3
100 86
78
63
47
25 16 10
7.7
15
5.6 100 85 76
62
49
27
18
13
8.3
16
5.4
100
88 79
61
50
30
20 13
8.3
17 5.6 100
95 72
44 24
17
12
7.3
18
5.6 100
95
70
39
21
15 11
7.2
19 5.5 100
96
93
83 69
32
20
14 11.
7
20 6.5
100
96
80
50
42 26 18 12
8 .3
IV. DEVELOPMENT
OF
THE
MO IFIE
LOTTM N
PROCEDURE
A modif ied Lottman
procedure i s
most commonly used
by
the CDOT
for mois ture
suscep t ib i l i t y t e s t ing of HMA.
Since nat ion-wide
exper ts have made numerous
modif ica t ions
to the procedure ,
understanding ts h i s to r i ca l development
i s
c r i t i c a l . In Table
3, the
or ig ina l procedure developed by Lottman
5,6 ,7) i s
compared
to
the two
most
commonly used vers ions today
AASHTO
T 283 and ASTM D
4867).
10
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
20/70
TABLE
3.
COMPARISON OF THE ORIGINAL LOTTMAN
5)
TO CURRENTLY
RECOMMENDED PROCEDURES
I
Or ig ina l
Modified
Lottman
Lottman
Ref.
S)
AASHTO T 283
ASTM
D
4867
Short-Term None
Loose mix:
None
Aging 16
hrs @
60·C
Compacted mix:
72-96 hrs
@
2S·C
Air Voids 3 to 5
6 to 8 6 to
8
Sample
Random
,Average
a i r Average
a i r
Grouping voids of
two
voids
o f
two
subse t s
should
subse t s should
.
be equal
be equa l
Satu ra t ion
100
55 to 80 55 to 80
Freez
e
15
hrs
@
-18·C
Min.
16
hrs
Optiona l :
@
-18·C
15
hrs @ -18·C
Hot water
Soak
24
hrs
@
60·e
24
hrs @
60·C 24
hrs
@
60
G
C
Strength
Ind i r e c t
In d i r e c t
Ind i r e c t
Proper ty
t ens ion o r
t ens ion t ens ion
diamentra l
modulu s
Loading Rate
1 .6
mm/min.
51 . rom/min.
'
51
mm/min.
@
13
G
C
@ 25·C
@
25·C
Prec i s ion
10
fo
r TSR
None
8 ps i
fo r
Statement i nd i r e c t
for
a Sing le
t e n s i l e
Operator s t r en g th
wet o r
dry)
Not spec i f i ed , bu t
r ep resen ta t ive
o f
a t yp ica l value
encountered .
11
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
21/70
n i t i a l Study
From
1968 through
1982, Lottman
(5,6,7,8,9,10,11) developed
a new
t es t procedure
to
identify
the moisture susceptibi l i ty of an HMA
The
procedure was
ver i f ied with aggregates
of
known
f ie ld performance and through
the construction of t e s t sections.
In 1970, Lottman (10) reported tha t
moisture damage
could
resul t from
excess
pore pressures tha t developed
in
the HM
pavement
from t r a f f i c and thermal
expansion.
Therefore,
the
moisture suscep t ib i l i ty t es t procedure included conditioning
phases
tha t created pressure
within
the
a i r
voids of the HM
sample.
The
condit ioning
included high
levels
of
sa tura t ion
and a freeze cycle to create pore pressure.
In
1974, Lottman (5) calculated indirect t ens i le s trength and
modulus
ra t ios as
the
value from
the
conditioned sample divided
by
the value from the unconditioned sample.
Conditioning
included vacuum sa tura t ion
followed
by e i ther single or
multiple cycles
of freezing
and hot-water soaking. Loading
ra tes
for
the indirect
tens i le
s trengths
were examined
a t 1.6
mm/min. (0.065 in . /min.)
a t
13
0
C
(55
0
F and 3.8 rom/min.
(0.15 in . /min.) a t 23
0
C
(73
0
F).
Lottman
(6)
reported the procedure in
1978,
and Lottman (7)
f inal ized
the
procedure in 1982. Testing parameters for
the
or ig inal Lottman procedure are
shown
in Table 3. In 1982,
f ie ld
evaluations on
eight t e s t
sections in seven s ta tes
including one
in
Colorado, provided validat ion of
the Lottman
moisture
suscep t ib i l i ty
t es t
(7).
12
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
22/70
Loading
Rate
In 1979,
Maupin
(12)
performed a study
to
implement the
1978
Lottman t e s t
procedure
for
Virginia.
For convenience
of using exis t ing
equipment, he
recommended a
loading ra te of 51 mm/min. 2 in . /min.)
a t
25
0
C 77
0
F) instead
of
Lottman s recommendation of 1.6
mm/min. (0.065 in. /min) a t
13
0
C (55
0
F).
No
s t a t i s t i ca l ly
s ignif icant difference
was
measured in the tens i le strength ra t ios between the
two
loading
ra tes
a t
the corresponding temperatures.
Testing in Colorado
was
performed
to
determine the difference
in the 51 mm/min. and 5.1 mm/min. loading ra tes Tensile
strength
ra t ios
prduced
from
the
two
different ra tes
were
iden t ica l
Dry strengths
using
the fas ter
ra te were 2.5 to 3
t imes
higher than those produced from the slow loading ra tes
The
two
modified Lottman procedures most commonly
used
today
recommend a loading ra te of 51 mm/min. (2 in . /min.)
a t
25
0
C
(77
0
F), as
shown in
Table 3.
Air
Voids Lottman (6) recommended compacting
the
laboratory
sample
to
match the projected a i r
voids
tha t would be in
the
HMA
pavement
a f te r approximately 6 years (3 to 5 a i r voids).
In order
to
improve
the
Lottman procedure,
Tunnicliff
(13,14,15) recommended modification of the ta rget a i r voids.
The sample should
be
compacted between 6
and 8 a i r voids
to
simulate
the
in-place voids of the HMA pavement
soon.
The two modified
Lottman
procedures
most
commonly used today
recommend l imit ing
a i r
voids
between
6
and
8 ,
as
shown
in
Table
3.
13
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
23/70
Saturation sample can
be damaged
i f
t swells
during
the
vacuum
sa tura t ion
process.
Based on
t es t ing of
aggregates of
known f ie ld performance,
Jimenez
16) indicated
tha t
swelling
during
vacuum
sa tura t ion
was
re la ted to s t r ipping
suscep t ib i l i ty Coplantz 17)
vacuum
saturated
samples
for
30
minutes a t a pressure of 610
of
mercury
to
provide
100
saturat ion. Vacuum
sa tura t ion
alone did not appear
to
in i t i a te
a st r ipping mechanism. Kennedy
18) found
excessive vacuum
saturat ion alone did
not
create
str ipping
unless the
aggregate
had shown poor st r ipping performance.
Stuar t
19)
performed
te s t ing
on
mixtures
of
known st r ipping
performance, both acceptable
and unacceptable. Based
on
t e s t
resul t s
from
the study, there was no conclusive evidence tha t
high
saturat ion
or
over-saturation
adversely effected
the t e s t
resul t s
The Lottman procedure 7) with high
sa tura t ion
and
the
modified Lottman procedure
developed
by Tunnicliff
15)
with
par t ia l
sa tura t ion were
comparable.
Dukatz
20) performed tes t ing that indicated no conclusions
could
be made
on
the effec t of
saturat ion
and
swell on
the
t ens i le
strength
of the
conditioned
sample. Various samples
saturated to high and low levels had
high
and low t ens i le
strengths.
In
order to determine i f par t ia l saturat ion could
predic t
pavement
performance, Tunnicliff
15) tes ted
the
eight samples
Lottman
7)
used for f ie ld verif ica t ion 6 years a f te r
Lottman).
He
concluded
tha t
l imi t ing sa tura t ion levels
correlated well with the
Lottman
procedure 7) tha t
allows
over-sa tura t ion
and
swell.
14
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
24/70
Tunnicl i f f 13) contended tha t damage from excessive
saturat ion could resul t in
low t ens i le
strength
ra t ios
even i f
the HM
was
not
moisture
susceptible. Although no t e s t
resul ts
were presented, past published
unreferenced)
and unpublished
l i t e ra tu re allegedly
supported
his claim. Over-saturat ion
of
one sample by
Tunnicliff
14)
indicated
tha t over-saturat ion
may be too severe.
The swell of
the
HM should
be
measured. The
two
modified
Lottman
procedures
most commonly used today recommend
l imit ing
the
level of saturat ion between
55
and 80 , as shown
in
Table 3.
Freeze Cycle
Tunnicliff
14,15)
also
recommended eliminating
the freeze cycle so
the
t e s t would be quicker and
easier
to
perform
for
f ield verif ica t ion
Lottman
5)
indicated tha t
the
freeze
cycle
with
high
sa tura t ion
predicted
the
str ipping
susceptibi l i ty of HM
pavements, even when the HM pavements in the f ie ld were
not
exposed
to freezing condit ions. Stuart 19)
indicated tha t
e i ther
the high sa tura t ion of
low
a i r voids with a freeze cycle
recommended
by Lottman
7)
or
the par t ia l
sa tura t ion of high
a i r voids
with no freeze
cycle
in
the modified
Lottman
procedure 15) would be
comparable.
I t was l ike ly tha t the
freeze was required
to
apply a s t ress in the sample as
discussed
by Professor
B.M. Gallaway
with Graf
21).
The
modified
Lottman
procedure
recommended
by
AASHTO
and
ASTM
allows
the freeze
cycle
to be optional,
as shown in
Table 3.
15
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
25/70
S h o r t ~ e r m Aging
he
time
of
exposure of the HM
sample
to
a
high
temperature
had
a s ignif icant
effec t on the
freeze-thaw
pedestal
moisture
susceptibi l i ty
t e s t
resul t s
he longer
the
HM
sample had exposure to a
high
temperature, the more
res i s tant t was to moisture
damage
(Graf,
21).
he
high
temperature
exposure increased the aging of the asphal t
cement
and provided
bet ter
coating, or
wetting .
Testing to
i sola te
the
two
variables indicated the increased resistance
to
moisture
damage was
primarily
re la ted
to the bet ter coating of
the
aggregate.
The
modified
Lottman
procedure
recommended
by AASHTO
requires
short- term aging whereas
the ASTM version has no short- term
aging,
as
shown
in Table
3.
Although
Lottman
(6) original ly
recommended the short-term aging speci f ied in AASHTO Lottman's
(7) final ized procedure
did not
mention short-term
aging.
Sample Grouping In 1987, Dukatz (20) reported the potent ial
for
tremendous
var iabi l i ty
in tens i le
s t rength ra t ios i f
the
average a i r
voids
in the conditioned and unconditioned
samples
were not equal. Ranges
in
the t ens i le s t rength
ra t ios were
as
high as 0.40.
When the
the
average
a i r voids
of
the
samples
in
the
conditioned and
unconditioned
groups were equal,
the
t ens i l e s t rength ra t ios were within
a
range
of 0.08.
he
two modified
Lottman procedures most commonly used today
recommend
the average
a i r voids of the conditioned and
unconditioned sample groups be equal, as
shown in
Table 3.
16
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
26/70
Mult iple Freeze Thaw Cycles . Several
researchers
have
invest igated
the
use
of multiple
freeze-thaw
condit ioning
cycles
and
have had varied conclusions. hen Lottman 5)
original ly invest igated the
procedure,
multiple
freeze-thaw
cycles
were
examined but
over-predicted
damage and were
considered too time
consuming
for pract ica l i ty
Lottman
concluded one freeze-thaw cycle
was adequate.
Scherocman 22) reported tha t multiple freeze-thaw cycles would
provide a
greater different ia t ion between the t ens i le strength
ra t ios
for
mater ia ls
with various levels
of
moisture
suscep t ib i l i ty
by Coplantz
17).
This conclusion i s supported
by
data
presented
To
the contrary, Kennedy 23) reported
tha t
ra tes of deter iorat ion in tens i le s trength
ra t ios
using
multiple freeze-thaw cycles were
not
s t a t i s t i ca l ly different
for
various
types of
materials . Materials with
higher
t ens i le
strength ra t ios a f te r
one
cycle would have
higher
t ens i le
strength
ra t ios a f te r
multiple cycles . The slopes or ra tes of
deter iorat ion of the
tensi le
s trength ra t ios were constant .
There i s tremendous var iabi l i ty in
conclusions drawn from
data
analysis of samples
tes ted
with multiple
freeze-thaw
cycles.
Additionally, t es t ing
samples with
multiple
freeze-thaw
cycles
requires an
addit ional
amount
of laboratory
t es t ing
time
tha t
may
not be
readi ly
available prior to paving.
Therefore,
multiple freeze-thaw
cycles were not
examined in
th i s
study.
~ h
Speci f i cat ion
The minimum
specif ied
t ens i le
s trength
ra t io
to
ensure
an
HM
pavement
will
perform
acceptably with
regard to
moisture susceptibi l i ty has varied.
Part of
the
reason
accounting for
the
specif icat ion to
vary
has been the
changes
in
the t es t
procedure. Based on
t es t ing of samples of
known f ie ld
performance,
tens i le
s trength ra t ios have been
17
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
27/70
recommended
a t
0.70 by
Lottman
(5)
and
Maupin
(12),
and a t 0.80
by
Lottman (7),
O Connor
(24),
and
Stuart
(19). A
survey of
s ta te
speci f ica t ions by
Tunnicliff
and Root (13) and
Hicks
(4)
revealed most s ta tes used 0.70 to
0.75
but
speci f ica t ions
ranged from 0.60 to 0.80. Computer
simulations
have resulted
in a t ens i le
strength
ra t io
recommendation of
0.85
by
Lottman (25).
Dukatz (20)
recommended a minimum t ens i le s trength a f te r
conditioning. A
very
weak
sample would not be
accepted.
This
also
prevented
acceptance of a mixture tha t had a low
unconditioned
t ens i le s trength ,
sometimes
caused
by
the
addition
of
l iquid ant i s t r ipping
addi t ives . Tunnicliff
(15)
recommended addi t ional research
be
performed to ident i fy a
minimum t ens i le
strength requirement.
v.
TEST PROCEDURES
Two different
t e s t procedures were investigated for
th is
study:
the AASHTO T 283 and the boi l ing
water
tes ts . Five different
var ia t ions
of
the
procedure were
used. Two
additional
t es t s
were performed: without any treatment for moisture
suscep t ib i l i ty and with hydrated lime. The procedures
used
to
inves t iga te
the
moisture
susceptibi l i ty of the HM
pavements
of
known f ie ld performance are described below. The experimental
grid of t e s t s performed on
samples
from the various s i tes i s
shown
in
Table 4.
Standard SHTO 283
The
materials
from
a l l
of the
s i t e s
in th i s
study
were tes ted
with
the
standard
procedure
(AASHTO T 283).
I t
includes
short- term aging, freezing, and
l imits on a i r
voids
(6
to
8 )
and
saturat ion
(55
to 80 ).
18
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
28/70
\
I
1
SHTO
T
283
X
SHTO
T
283
X
No f reeze
SHTO
T
283
X
30 min
sa t
SHTO T 283
X
No STA
SHTO
T
283 X
Double
STA
SHTO
T
283
X
No
addi t ive
SHTO
T 283
X
Lime
STM D 3625 X
Boi l ing Water
ST
A =Short Term Aging
T BLE 4
EXPERIMENTAL GRID
FOR THE STRIPPING
STUDY
GOOD PERFORMERS
HIGH
COMPLETE
M INTEN NCE
REHABILITATION
2
3
4
5 6
·7
8 9
10
11
12
13
14
15
16
X
X X X
X
X
X ·
X X X X X
X
X
X
X
X X X X
X
X
X
X
X X X X X
X
X
X
X X X
X X X
X
X
X
X X X X
X X X X X
X X X
X
X
X
X X X
X X X X X
X X X X X X X X X
X
X X X X X X X X X X X X X X X
X
X X X X
X X
X X
X
X
X X X X
X
X
X
X
X
X
X
X X X X
X X
X
X
DISINTEGRATORS
17
18
19
20
X
X
X
X
X
X
X
X
X
X X
X
X X X X
X X X X
X X X X
X X X X
X
X
X
X
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
29/70
S H ~
8l
No
Freeze) .
In
order to determine
i f
the
actual
pavement
performance could be predicted
without
the
freeze
cycle, the materials from a l l
of
the s i tes
in
th i s study
were tes ted without the freeze cycle.
SHTO
8l
lO-Minute Vacuum Saturat ion) . Some
inves t iga tors (17,18,20,26,27)
performed the modified Lottman
t e s t by
vacuum sa tura t ing a
sample
with
7 a i r voids
for 3
minutes. The procedure was performed with a vacuum sa tura t ion
of 3
minutes under
a
pressure
of 61
mm
of
mercury.
Consequently, the degree of
saturat ion was not controlled.
SHTO 283 No Short-Term Aging). The materials from
a l l
of
the
s i t e s in
th i s
study were tes ted without short- term
aging.
SHTO
283 Extra
Short-Term
Aging).
When HM i s produced
for a projec t
in Colorado,
a
sample i s
obtained
and delivered
to the Central Materials Laboratory for tes t ing.
After
delivery, the sample i s reheated for sp l i t t ing into the correct
sample size and reheated a
second
time for compaction.
In
to ta l the sample i s reheated approximately 4
to
8 additional
hours.
The ef fec t of the additional short- term
aging
was
invest igated. The
materials
from
a l l
of the s i t e s in
th i s
study
were
tes ted with
an additional
short- term aging of 5
hours
a t
121
C
(250
F
on
the loose
mix.
SHTO
283
No
Anti-Str ipping
Treatment). Numerous
aggregates
in
Colorado have
moisture
susceptibi l i ty
problems.
Anti-str ipping treatments in the form of lime and l iquid
anti s t r ipping additives have
commonly
been used
in Colorado.
The use of
lime
began
on
a
regional
basis in
the
early 1960 s
(28). The
DOT
specif ied
the use of l iquid anti s t r ipping
2
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
30/70
addi t ives
in
a l l
mixtures in
approximately
1983.
Even HM
with
l iquid
anti s t r ipping additives had
continued
problems
with
moisture susceptibi l i ty .
The
CDOT
then began
requiring
hydrated
lime in a l l
mixtures by
1 of the weight of the
aggregate in
1990.
Some
aggregate sources have modified Lottman t es t
resul t s tha t
are
very
low
when no ant i s t r ipping treatment
i s
provided.
McGennis (1) has
indicated
tha t
ant i s t r ipping
t reatments
should be able
to improve a
marginal HM
mixture,
but should
not
be expected
to
overcome severe
deficiencies. The
mater ia ls
from
a l l
of the s i tes in
th is
study
were tes ted
with
no
anti s t r ipping
treatment
to
determine
the
baseline
moisture
suscep t ib i l i ty potent ia l of the untreated HMA.
SHTO
T
283
With Lime). Some of
the
HM pavements
tha t
exhibited moisture dis t ress
were not
t rea ted with
hydrated
lime. The potent ia l moisture
susceptibi l i ty of
these
mater ia ls
with 1 hydrated lime by weight of the aggregate
was
investigated. I f an HM of
known
f ie ld performance did not
contain
hydrated
lime. When
constructed,
the procedure
was
performed
on
material from the s i t e with
hydrated lime.
oi l ing Water Test . Several studies have indicated the
boil ing
water
t es t (ASTM
D
3625) has
accurately
indicated the moisture
susceptibi l i ty performance of HM
pavements.
The t e s t
used
for
th i s
study involved
immersion of the sample in boil ing water
for
1 minutes. A
retained
coated area
over
95
i s usually
required.
VI.
TEST
RESULTS
ND DIS USSION
Results
from each
var ia t ion in
the AASHTO
T 283
t es t are
21
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
31/70
tabulated
in
Appendix A The
data includes a i r
voids,
saturat ion,
swell
a f te r
saturat ion and
conditioning,
conditioned and unconditioned
t ens i le
strengths, and
tens i le
s t rength
rat ios .
Analys is
o Untreated Aggregate Ouality .
Many
of the
aggregates t es ted
in
th i s study are
moisture
susceptible based
on known
f ie ld performance. Modified Lottman
t e s t
AASHTO T 283) resul t s were very
poor
when no l iquid
ant i - s t r ipping addi t ives or lime were used.
The
t ens i le
strength
ra t ios
of
samples t es ted
without treatment
are
shown
in Table 5.
Just enough treatment with l iquid an t i - s tr ipping
addit ives
or
hydrated lime
was
used
on
each
projec t
so
the
HM
samples would pass
the
t es t ; unfortunately condit ioning in the
f ie ld was more
severe than
conditioning in the laboratory,
and
the pavements fai led.
I t i s cr i t i ca l tha t the conditioning in the
laboratory
vacuum
saturat ion,
freeze,
hot-water
soak) be
equal to or greater
than
the
sever i ty
of condit ioning expected in the f ie ld. This i s
especial ly
important
when
marginal aggregates are used which
require
treatment with
l iquid
ant i - s t r ipping
addi t ives or
hydrated
lime. I f conditioning in the
laboratory i s less
severe
than condit ioning in the
f ie ld
an
engineer could
erroneously
assume
an
HM mixture would have good
f ie ld
performance for 10 or 20 years .
Field condit ions
re la t ing to
moisture damage
are
high t r a f f ic
high temperature,
high
moisture,
and possibly
freeze.
Inf luence
o
Freeze/No Freeze/30 Minute
Saturation.
Tensile
strength ra t ios for
samples
tested according to
AASHTO
T 283
freeze) , AASHTO
T 283
no freeze), and AASHTO
T 283
30-minute
sa tura t ion
and freeze) are
shown
in Table 5 .
No
s ta t i s t i ca l ly
22
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
32/70
s ign i f i c a n t d i f fe rence was found between AASHTO T 283 performed
with and without the f reeze
cycle .
By
using
the
30-minute sa tu ra t ion
and
f reeze , the
t e ns i l e
s t r e n g t h r a t i o s were s i g n i f i c a n t l y lower than t he samples
t e s t e d with p a r t i a l sa tu ra t ion AASHTO T
283
f reeze
and
no f reeze .
Table
5. Comparison o f Tensi le S t rength Ra t ios .
Tens i le
St reng th Rat ios
AASHTO
T 283)
S i t e Freeze No Freeze
30-Minute
Freeze
Sa tu ra t ion Note
1)
and Freeze
1.02
0.85 0.98
0.37
2
1.20
1.25 1.05 0.70
3 1.11 1.22 1.20
0.37
4
1.06 0.97 1.05
0.49
5 1.10
1.07
0.97 0.92
6 0.83
0.91
0.74
0.40
7
0.97
0.89
0.86
0.90
8
0.94
0.91
0.69
0.21
9
0.95 0.90
0.72 0.40
10 0.84 0.93
0.68
0.70
11
1.11
0.96 1.09
0.38
12
1.01
1.07 0.81
0.60
13 0.69 0.64 0.56 0.45
14 0.32 0.34 0.21
0.30
15
0.53 0.46 0.32
0.35
16
0.82
0.70 0.76 0.44
17
0.65
0.51 0.30 0.55
18
0.89 0.92 0.86
0.49
19
0.22
0.20
0.28
0.24
20
0.59 0.49 0.26 0.37
Avg.
0.84 0.81 0.72 0.48
S.D.
0.27 0.29 0.31 0.20
Note 1: The HMA sample was t e s t e d without t he use o f
l i q u i d
a n t i - s t r i pp ing
add i t ives
o r
hydrated
l ime.
23
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
33/70
In
separate
s tudies , Coplantz 17) and Dukatz 20) invest igated
samples compacted
to
approximately 7 a i r voids, Appendix B.
They
compared resu l ts from samples
tha t
were vacuum saturated
for
30 minutes
with
a freeze and par t i a l ly saturated
with
no
freeze.
They
found
tha t
samples
par t i a l ly
saturated
and
not
subjected
to
a freeze cycle
had tensi le strength ra t ios tha t
were two
to
three
t imes
higher than samples ful ly saturated and
subjected
to
a freeze cycle.
A modified Lottman t es t can be performed in
several
different
manners. Based
on
resu l ts from th i s study and others
15,17,19,20,27) shown
in
Appendix B, a
ranking of
t e s t s
in
decreasing
order
of
sever i ty
i s
l i s t ed
as
follows:
1) 30-minute saturat ion, 7 a i r voids,
freeze,
2A 30-minute
saturat ion,
4 a i r voids, freeze,
2B) 55-80
saturat ion, 7
a i r voids,
freeze,
2C) 55-80
saturat ion,
7
a i r
voids, no freeze.
All levels of sever i ty include hot-water soaking. Severi ty
Levels 2A, 2B, and 2C a l l
appear
to
provide approximately equal
resul ts . Severi ty Level 1 produces s ignif icantly lower t ens i le
strength
ra t ios .
The condit ioning
in
the laboratory should be
equal to
or
greater than
the conditioning in the f ield.
I t i s l ike ly tha t
different levels of f ie ld conditioning exis t throughout
Colorado. The
most
severe f ie ld conditions
are hypothesized
to be a
function of:
1) high t r a f f i c
2)
high
temperature,
3)
high levels of
moisture,
and
4)
possibly
very
low
temperatures.
High
levels of moisture
should
not
be determined on an annual
basis .
Based upon the performance of
pavements
requiring
complete
rehabil i ta t ion the amount of precipi ta t ion received in
a
high temperature
period i s more relevant.
24
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
34/70
The sever i ty level
of the modified
Lottman
t e s t
performed
in the
laboratory
should correlate with
the
sever i ty of the
f ie ld
condit ions. I f dif ferent
levels
of f ie ld condit ioning exis t
throughout
Colorado, then different
levels
of laboratory
condit ioning
should be
used.
Comparison
of Tensile Strength
Ratios with
Actual
Performance.
A
comparison of the various
severi ty levels of
the
modified
Lottman
t e s t
with pavements of known f ie ld performance i s shown
in
Tables 6, 7, and 8. A minimum
t ens i le
s trength
ra t io
of
0.80
was used.
Table
6.
Comparison
of
Pavements
of
Known
Field
Performance
with
SHTO
T 283 Severity
Level
28).
Good High
Complete Disint .
Maint.
Rehab.
Pass
7 5 1
1
Fail 3 3
Table 7.
Comparison of Pavements of Known
Field Performance
with SHTO
T 283
with
No
Freeze
Severi ty
Level
2C).
Good High
Complete Disint .
Maint.
Rehab.
Pass
7
5 1
ai l 4
3
Table
8. Comparison
of
Pavements of Known
Field Performance
with
SHTO
T 283 with a
30-Minute
Vacuum Saturation
Severi ty Level
1).
Good
High Complete Disint .
Maint. Rehab.
Pass
6
2
1
Fail
1
3
4 3
25
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
35/70
Based
on
these resul t s the
best predictor of
the
pavements
studied in
th i s
invest igat ion i s the modified Lottman t e s t
AASHTO T 283) with a 30-minute vacuum saturat ion. A plo t of
the t ens i le strength ra t ios in ranked order i s shown in
Fig.
9.
Although
a minimum
t ens i le
s trength
ra t io
of
0.80
was
used
to
develop Tables 6-8 and
Figure
9, considerat ion should
be
given
to
using a minimum t ens i le
strength
ra t io of 0.85,
as
recommended
by Lottman 25).
I t
i s known tha t many HM
mixtures
in Colorado
with
good
performance
would fa i l
the
sever i ty Level t e s t .
The
laboratory condit ioning specif ied for a par t icu la r
projec t
should
re la te
to the anticipated
f ie ld
condit ions.
Two
levels
of
laboratory
condit ioning should
exist . Severity
Level
should
be used
for high t r a f f i c s i t e s with severe environmental
conditions. Severi ty Level 2C should be used
for
low
t r a f f i c
s i t e s without
extreme environmental
condit ions.
Some out l ie rs
exis t
and
are
worth noting. Si te 11 i s a
quarried
source and very
f ine par t ic les conglomerated
onto
the
large par t ic les during processing of the aggregate. In order
to
minimize
the
generation of
dust,
a
large quantity of
water
i s sprayed on the aggregates. The dust
stays
out of the a i r
but remains
on
the aggregate.
After
the
water
dries the f ine
par t ic les can
be
removed by scraping an aggregate with
your
f inger
nai l .
I t
i s l ike ly tha t the
one cycle
of loading
applied by the modified
Lottman
i s not severe enough
to
penetra te the conglomeration. A t es t with multiple
cycles
would bet ter
identify
th i s
problem.
s i t e 6
has been
previously studied
and reported
by
McGennis 1). Identical t e s t s performed by the CDOT using
mater ia ls sampled
one
year apart were
substant ia l ly dif fe ren t .
26
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
36/70
o
a
I
CJ
Z
w
a
m
l l
J
m
z
w
f
0.4
0.3
0 21
1
3 4 2 1 5 18 7 12 16 6
9
8 10 3
5
7
19
20 14
SITE NUMBER
FIG.9 Ranked Order of Tensile Strength Ratios from AASHTO T 283 with a 30 Minute Saturation.
27
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
37/70
The di f fe rence
i s
not bel ieved to be
caused by r epea t ab i l i t y .
t i s wel l known t h a t the quarr ied source i s highly var i ab le
and conta ins seams of silt
and sha le .
When l a rge quan t i t i e s of
shale
are quar r ied , the ma ter ia l i s
l i ke ly to
be
highly
suscept ible
to
moisture, as
ind ica ted
by CDOT t e s t i n g performed
in 1992 Ref. 1 . When no shale
i s
quar r ied ,
the ma te r i a l
i s
l i k e l y to be marginal ly acceptab le as
ind ica ted
by
the
r e s u l t s
in t h i s
s tudy. Si tes
10 and 12 a l so had the same quar r ied
mater ia l .
Table 9. Summary of Swell Afte r
Sa tu ra t ion .
Swell
(
After Satura t ion ,
AASHTO
T
283
Si t e
Freeze
No 30-Minute No STA Ext ra
Freeze Sa tu ra t ion
STA
1
-0 .2
0.0
+0.1
-0 .2 0.0
2
-0 .1
-0 .2
0.0 -0 .2
-0 .1
3
-0 .4 -0 .2 -0 .4 +0.1 -0 . 4
4 -0 .3
-0 .3
-0 . 1
-0 .3
0.0
5 -0 .2 -0 .2 0.0
-0 .2
-0 .2
6
0.0
0.0
+0.1
-0 .1
-0 .2
7 -0 .6 -0 .3
-0 .4 -0 .2
-0 .3
8
-0 .5
-0 .4
+0.3
-0 .3 -0 .3
9
-0 .6
-0 .1 0.0 -0 .5 -0 .5
10
+0.2
-0 .6 -0 .3
0.0
0.0
11
-0 .7
-0 .7 -0 .2 -0 .2 -0 .7
12
-0 .5 -0 .3
-0 .1
-0 .6
-0 .8
13 -0 . 2
-0 .4
+0.4
-0 .6 -0 .3
14 -0 .3
+0.4
+0.8
0.0
+0.3
15
-0 .1
-0 .3 +0.3 0.0
0.0
16
0.0
-0 .1
+0.1
-0 .2 -0 .2
17 -0 . 2
-0 .2
0.0
-0 .2
-0 .2
18
+0.1
-0 .8 -0 .2 -0 .2
0.0
19
+0.3
-0 .1
+1.0 +0.2 +0.2
20
-0 .3
+0.2
-0 .2
-0 .1 -0 .5
STA
= Short-Term Aging
28
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
38/70
I f a
more judicious
aggregate
processing scheme
were
pract iced a t the quarries tha t provided
the
aggregates
for s i t e s
10,
11,
12,
and 16, a higher qual i ty HM
pavement could be produced
with
these aggregates.
Analysis o Swell Swell
was
measured
and
calculated a f te r
satura t ion
Table 9) and
a f te r
conditioning Table 10)
according to
the formulas
in
ASTM
D
4867.
For
samples saturated between
55
and 80 , there was no swell
more
than 0.5 a f te r
satura t ion.
For samples saturated for 30
minutes, swell more
than
0.5
af ter
satura t ion only occurred
on
two
samples
Sites
14
and
19): both
had
very
low
t ens i le
s t rength ra t ios and poor f ie ld
performance.
Swell often occurred a f te r conditioning. The amount of
swell
was
direc t ly
re la ted to the t ens i le s t rength ra t io
and known
f ie ld performance.
The higher
the
swell
a f te r conditioning;
the
lower
the t ens i le
s t rength
ra t io
and the
more
moisture
damage in the f ield A plot of
swell
a f te r conditioning and
t ens i le s t rength ra t io
i s shown in
Fig.
10 for
samples
with
30-minute
satura t ion.
The sample
can
be damaged by swell i f
allowed
to saturate for
30 minutes, but those
materials also
fa i l in
the
f ie ld based on
data in th i s study and data
reported
by
Jimenez 16) and
Kennedy 18).
Data
from Coplantz 17), Stuart 19) and Dukatz
20) indicated
tha t
excessive vacuum satura t ion
alone
did not
appear to in i t i a te
a s t r ipping
mechanism.
29
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
39/70
1.3
1.2
1 1
0
1
«
a:
I
0.9
CJ
0.8
z
w
a:
0.7
j)
w
0.6
J
j)
Z
0;5
0.4
0.3
o
O . 2 - - - - - - - - - - ~ - - ~ ~ - - - 4 - - ~ - - - ~ - ~ ~ ~
1
o
1 2
4 5
6
SWELL ( ) AFTER CONDITIONING
FIG.10 Relationship
of
Swell After Conditiorung Versus Tensile Strength Ratio from Samples Tested According
to AASHTO T 283 with 30-Minutes of Saturation.
3
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
40/70
Table
10
ummary
o
Swell After Conditioning
Swell
( After Condit ioning,
AASHTO
T
283
Si t e
Freeze
No 30-Minute No
STA
Ext ra
Freeze
Satu ra t ion
STA
1
+1.1 +1.3 +1.4
+1.7
+1.3
2
+0.2 +0.2
+0.4 +0.5
+0.2
3
+0.2
+0.3 +0.2
+0.6
0.0
4
+0.6 +0.6 +0.6 +0.7 +0.7
5
+0.5 +0.6 +0.8 +0.5 +0.7
6 +0.8 +0.6
+1.1
+1.0 +0.8
7 -0 .3
+0.4
+0.3 +0.4 +0.3
8 +1.3 +1.8 +1.8 +1.1 +1.2
9
+0.9 +1.0 +1.0
10
+0.3
+0.4 +1.0 +0.6
+0.2
11
0.0
+0.4
-0 .1
+0.3
+0.3
12
+0.5
+0.6 +0.1 +0.6
+0.3
13
+0.9
+1.7
+1.9 +1.4
+1.4
14
+3.4
+3.9
+4.6
+4.0
+3.8
15 +1.5
+1.9
+2.4
+1.6 +2.0
16 +0.8
+0.5
+1.1 +0.5 +1.3
17
+3.3
+3.9
+4.8
+3.6
+3.4
18
+1.3 +1.0 +1.3
+1.5 +1.8
19
+8.2 +8.1
+3.3 +9.4 +10.4
20
+3.8
+4.2 +5.9
+4.3 +4.3
STA = Short-Term Aging
Analysis o S h o r t T e ~ Aging The t ens i l e
s t r eng ths o f
uncondit ioned
samples
dry
t ens i l e
s t r eng ths
and t e n s i l
e
s t r eng th r a t io s us ing
th ree di f fe ren t
l eve l s of shor t - term
aging are shown in Table
11.
These
va lues
were
normal ized
to
the
values ob ta ined
from
the shor t - te rm
aging
spec i f i ed in
AASHTO T
283
and are
plo t ted
in
Figs . 11 and 12.
31
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
41/70
Figures
11
and 12 a r
e
box p lo t s . The l i ne i n the cen te r ~
the
box i s
the average .
The box encloses
p lus
and minus
one
s t anda rd dev ia t ion o f da ta .
The
whiskers
t ha t
extend
out
of
each
end of t he box are
the
range of
the da ta .
Table
11. ummary o f Dry Tens i le Strengths and ensi le Strength
Ratios
Using Various Lengths
o f
Short-Term Aging
STA).
Dry
St reng th kPa
Tens i le
Strength Ratio
Si t e
No
STA
Standard Extra
No STA
Standard Extra
ST
STA
ST
STA
440
500 620 1.00
1.02
0.87
2
460
490
7
00
1.13
1 .
20
1 .16
3
540
610 660
1.02 1.11 1.22
4
500
530 560
1.19 1.06
1.04
5
460
590
670
1.04 1.10
1.03
6
520
570
700
0.85
0.83
0.83
7
540
640
700
1.08 0.97
1.01
8
670 690 710
1.00
0.94 1.02
570 660 760 0.94
0.95 0.87
10
470
550
750
0.88
0.84
0.93
11
630 720
830
1.27
1.11
0.98
12
490
480
600
1.02 1. 01
0.90
13
650 660
880
0.72 0.69 0.74
14
680
700
900
0.35
0.32 0.29
15
700
680
800 0.55
0.53
0.53
16
610 680
770
0.90 0.82
0.61
17 400
460
490
0.74
0.65 0.65
18 420 460
540 0.93 0.89
0.84
19
500
550
590
0.15 0.22 0.16
20 550 550 700
0.51 0.59
0.49
Avg.
540
589
697 0.86 0.84
0.81
S.D.
91
86
110 0.29 0.27
0.28
32
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
42/70
0
....
"001
«
§ E
o
Zl l l
I-
.£:1
....
...
01'"
mE
..
/)
....
(/)"0
QI a
:="0
f/) C
C «
QI'"'
I-(/)
)o,QI
.... :
0 '
9
"0
N .-
: :
01
«1
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
43/70
The
HM samples
with longer aging have higher dry
t ens i le
strengths, Fig
11.
I f a dry
t ens i le
strength
i s
specif ied,
the
length
of
short-term
aging
must
also be specif ied. The dry
tens i le
strength
did
not
discriminate amoung any s i tes . The
wet t ens i le
strength
was
only
able
to
discriminate
between
good
and dis int igra tor s i t e s
a t
about 45 kPa
65 ps i .
In
most
cases,
the t ens i le
strength
ra t io remained constant
with increases in
aging,
Fig. 12.
However,
in one case
Si te
16), the t ens i le strength ra t io dropped because the dry
s t rength increased dramatical ly, and
the
wet strength did not
change. The t ens i le s trength ra t io i s
generally
insensi t ive
to the
length of
aging.
Kennedy
27) also determined
the
effect of short- term aging
on
the t ens i le
strength
ra t io was not s ignif icant .
By
eliminating
short-term
aging,
the time required
for
t es t ing
could be
shortened s ignif icantly .
Specifying a t ens i le strength ra t io appears to be superior
to
an absolute requirement
on
a t ens i le strength of a conditioned
sample.
The
inf luence of short- term
aging i s negated
when a
ra t io
i s used. In
the
f ie ld , condit ioning i s a function of
plant type, s torage time
in
the s i lo ,
haul time,
etc . with
a l l
of
the
f ie ld variables,
t
i s di f f icul t
to
quantify
the
amount of
short-term
aging
an
HM mixture
receives.
Analys is o
Lime Addit ion
Using SHTO
T
283, a l l
HM
samples
except one of
the
disintegrators , Si te 19) te s ted in th i s
study
had
acceptable
tens i le
s trength
ra t ios
when
lime
was
used. I t i s not
clear i f
the
addition
of
lime
would have
provided
good f ie ld performance
because the
sever i ty
Level 1
t e s t was
not
used.
4
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
44/70
Analysis o f
the
oi l ing Water
Test
The
resul t s
of the
boi l ing
water t e s t
are shown in Table 12.
Five
people
rated th
e
samples
and
the
t e s t resu l ts
are
very subject ive . A
large
sca t ter
in the
resul ts
was obtained
between each evaluator.
The boi l ing
water t e s t i s not an
idea l
t e s t because the resul t s
are subject ive. Addit ionally
the resul ts do
not consider
the
void s tructure permeabil i ty
or
gradation of the HM mixture.
In some
mixtures the
t r a f f i c loads are carr ied by
the
f ine
str ipping-susceptible
aggregates
and f ie ld performance
s
poor.
In
other mixtures with the same str ipping-susceptible
aggregates
performance
may
be
good
i f the
gradation
allows
the
t r a f f i c loads
to
be
carr ied
by the coarse nonstr ipping
susceptible aggregates.
The permeabil i ty of the sample
determined
by the void st ructure
should also be a factor in determining
the
suscep t ib i l i ty to
moisture
damage.
The boil ing water
t es t does
not
consider the
void s t ructure
since
the t e s t i s performed
on
a loose
mixture.
Results are summarized in Table 13 using a cutoff of 95 . The
boil ing t e s t i s a very severe
t e s t .
Most a l l of the samples
fa i led
regardless
of
known f ie ld performance.
35
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
45/70
Table
12. Boiling Water Test Results .
Si te
Category Coated Aggregate
1
Good 35
2
65
3
40
4
95
5
45
6
80
7
90
8 High Maintenance
55
9
95
10
80
11
65
12
95
13
Complete Rehab.
90
14
80
15
55
16
95
17
Disin t igra tors
50
18
50
19
40
20
75
Table
13. Comparison of Pavements of Known Field Performance
with the Boiling Water
Test
ASTM
D 3625).
Good High Complete Disint .
Maint. Rehab.
Pass
1
2
1 0
ail 6
3 3
VII. MODIFIED LOTTMAN REPEATABILITY
The
Colorado DOT
performed
an
invest igat ion to
determine
the
amount
of var iabi l i ty in the
ind i rec t
tens i le s t r ipping
t e s t
within the CDOT Central Materials
Laboratory
29).
A
s ingle operator
standard
deviat ion
in the
t ens i le s trength
36
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
46/70
ra t io
was
0.04. Maupin 30) performed
a
study in
Virginia
on
the var i ab i l i t y of the indirect t ens i le s trength r a t io One
standard deviation was 0.035. The indirect t ens i le
strength
ra t io i s
very repeatable within one
laboratory.
VIII . C SB H I S ~ R I S OF OTHER IRVESTIGATORS
The Lottman moisture suscep t ib i l i ty t e s t
can
be
performed
a t
various
levels
of severi ty.
The ColoradQ s i tes
with high
t ra f f ic in th i s study
are best predicted
with severi ty Leve l l
I t
was of i n t e res t to determine the
level
of severi ty tha t
predicted actual
performance
of o ther
pavements
reported
in
the
l i t e ra tu re Four case his tor ies in the l i t e ra tu re were
analyzed
to determine the level of
severi ty
of the Lottman t e s t
tha t predicted actual
pavement
performance.
Stuart
1986.
Stuart 19)
tes ted
materials from 14 s i tes with
good,
s l ight and
severe
f ie ld
performance with
respect to
moisture damage. The s i t e s were from
Georgia,
Maryland,
Mississippi
and
Utah and were
tes ted
with the
original
Lottman
t e s t Level
2A
and
the
Lottman
t es t
modified by Tunnicl i f f
Level 2C). Both t e s t s worked acceptably as shown in Tables 14
and
15. A minimum t ens i le strength ra t io of 0.80 was used.
Table 14. Comparison o the Lottman Test Level 2A
to Actual
Performance.
Actual Pavement Performance
Good
Slight Severe
Lottman
Pass
6
2
0
Test
Level 2A
Fai l
0 2 4
37
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
47/70
Table 15. Comparison
of
the Lottman
Test Level
2C)
to
Actual
Performance.
Actual Pavement Performance
Good Sl igh t Severe
Lottman
Pass
5 1
0
Tes t
Level 2C) Fa i l
1
3
Kennedy
1983. Kennedy
18)
analyzed e igh t s i t e s in Texas
with good
and
bad performance from moisture damage. Although
the
Lubbock
s i t e was or ig ina l ly c l a s s i f i ed as
good,
t was
changed
to
bad
when
a
low
area revealed s igns of
moisture
damage. The most severe vers ion L e v e l l ) of the Lottman t e s t
had good cor re la t ion as shown in Table 16.
A
minimum t e n s i l e
s t reng th
r a t i o o f 0.80 was used.
Table 16. Comparison of the
Lottman Test L e v e l l )
to Actual Performance.
Actual
Pavement Performance
Good Bad
Lottman
Pass 2
0
Tes t
L e v e l l )
Fai l
1
5
Kennedy 1983. One s i t e in Texas with poor performance was
analyzed
by Kennedy 26) . The
most severe ve rs ion L e v e l l )
o f
the Lottman t e s t showed good cor re l a t ion .
38
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
48/70
Parker
1988.
Parker
31) tes ted
five
aggregates
from Alabama
with good,
moderate, and
poor
performance his tor ies using the
Lottman t e s t
as
modified
for Levels
B and 2C. Correlat ion was
poor as shown in Tables 17 and 18. A minimum t ens i le st rength
ra t io
of 0.80 was used.
Table 17. Comparison
o the
Lottman Test Level 28)
to Actual Performance.
Actual
Pavement
Performance
Good Moderate
Poor
Lottman
Pass
0 1
0
Test
Level
2B Fai l 2 0 2
Table
18. Comparison
o the Lottman Test Level 2C)
to Actual
Performance.
Actual Pavement
Performance
Good
Moderate
Poor
Lottman
Pass 1 1 1
Test
Level
2C
Fai l
1
0
1
Summary. The modified Lottman and numerous versions
do
appear
to
have a
reasonably
good
corre la t ion with mixtures
of known
f ie ld performance.
However, the re la t ionship i s not ideal
When using a t e s t tha t does
not
ideal ly re la te
to
actual f ie ld
performance, t
i s
reasonable
tha t a large
factor
of
safe ty
be
applied
in
establ ishing the severi ty level of
the
t e s t
procedure
and
the specif ica t ion value.
39
-
8/18/2019 Investigation of the Modified Lottman Test to Predict the Stripping Performance of Pavements
49/70
IX. ON LUSIONS
The
aggregates
and
asphal t
cements used
for
th i s
study were
from the
same sources
but
were not the exact material
tha t was
used on each
project .
The proper
performance of
hot
mix
asphalt HMA) pavements i s
not
sole ly dependent on the material propert ies; improper
construction
or s t ruc tura l design
could also cause problems
with
the
HM
pavement.
For
the
fa i lures
studied
in
th i s
investigation, t i s not clear
how much
of the fa i lure could be
at tr ibuted
to
materia ls , construction, or s t ru