Determination of mechanical properties of hydraulic asphaltic concrete by means of a three-point bending test

Determination of mechanical

properties of hydraulic asphaltic concrete by means of a three-point bending test.

I .i. , ^ ^ ^ ^ Paper ' -r : eurobitume centres.. j I A

15^^'--~~ C'^ -e-uTSG "3fore SeoteTiCar '5. '952 • •: c^r -ai-onai fcca: 'üirt

Sheet 1 Feuillel 1

DEEEamiMünf OF MBCHAMICM. HCTEREIRS OF HYTKWIT.TC AfyHATfTTC

cnacs^mf. Ry MEMB CF A imEB-roiwr YSEJXSSX^ TEST

C.C. Hantauiban, H.F.C. van de Ven

Research - The Netherlands

1. mraxocTECN.

Asphalt is used as a protective revetment in nany sea walls. It is therefore an important material for protecting the Netherlands against flooding. Tliere are raany different applications such as asphaltic concrete, <3pen. stone asphalt and asphalt grouting mortars.

Asphaltic ccncrete is the most widely laiown of these because it is OGnmcnly used in the most severe attacked zone during ejctreane storm ccnditions. It forms an iapervious and flexible slope protection.

Asphaltic concrete exhibits plastic and viscous as well as elastic behaviour. One advantage of this is that the revetment can follow irregular settlements in the subsoil. Cracking is rare owing to the high bitumen ocnitent. Uhder rapid loads like wave isfjacts, the asphaltic concrete behaves as an elastic, relatively stiff material viiic^ ensures good spreading of the load.

The fact that asphalt is ooranonly used has led to the material for many years being given special attarticn in titó researdi conducted in the oontext of the Tedinical Advisory Ooramittee on Waterdefences in the area of slope protection. This research is concerned for exanple with:

Do net 'jse ciue nboc^. - .;,- ^as utiiiser :e •••joan ce ccLieur Dieue. Soacing: "I' - iines - inteniane: 1 ^ 2

Paper Autnoris) Surnarr.e -, Title or t n a c a c e r ^ Detenination Bechanical sheei .

1 C.c. Ibntauban \ properties of hydraulic l

V C-.-.i..^ •'H.F.C, van de Ven -rrrc ^ - • asiÈaltic concrete Poi.'p ƒ =^the mechanical böiaviour of asphalt revetanaits - the service life of ttie revetments

The design method for asphaltic concanete revetmerrts is described in the "Guidelines for tiie use of asphalt in hydraulic engineering" [ref. 1 ] .

The mechanical properties of hydraulic asphaltic concrete play an inportant role in tiie determination of the required thickness for the asphalt layer. This paper describes a method for assessing the required/present thickness of the asphalt stnxrture both for the design of sea walls and for existing sea walls.

The aim of this study was to develop a standard method for determining the bending tensile strength urxier dynamic stresses. The test nust be suitable for testing new asphalt mixtures and existing sea wall structures. An inportant factor was the possibility to take sanples in a sitcple and representative way.

2. MBCHANICKL TESHRG OF ASEHAIiT.

Building materials are generally characterised by ttieir "strength" and "stiffness". Strength can be established in nany different ways, depending en the nature of the material and the application. Bending tensile strength is determined as the most relevant property for tinber, and oonp-essive strength in the case of concrete. Stiffness is ^lerally expressed as the E modulus (Young's modulus), which indicates the linear relationship between stress arxi strain.

The situation is more complicated in öie case of asphalt. In this material, strength and stiffness are not only d^jaident on the conposition and nature of the basic materials but also depend heavily on the temperature and time of loading.

Vfe therefore cannot refer simply to the "strengtii" and "stiffness" of asphalt. These properties are determined as a

Ma^K on eacn page. OL.tsiGe:he'r^rfr at t h e : : ; ; ' t u i " • -arneis;.':i:o •:: •;V:ID0:'.-sresi "LimDer. ~hes6 notes wiil r.ct ce •jnntsc;. ;r,?v -r^re I D a v c r j errnr.T ."• ".aca-ieiX.'-z.

inaiquer sur chaque reuiHe a i'exTsr'e^r au cacre en '^ai.;: roiv, des aute;.::e, ::tre j u rapcort. n" de ia feuiiie. iCes inaicatcns qw r e .-.e'eiii pas reo; oc'-;,[-as. ont pcur GD;ei •-• ; v ; e r ies 'erreurs iors de !a mise en pages.;

?3Der

^uinonsiSurnare "=.:--:"-••' . Detemlnationlechanlcal .-.,.^,.,. C.C. Kontauban '= properties of hydraulic - - ^^ ^ -uT9ür-r>!':rr H.F.C, van de Ven - , ,,^. asphaltic concrete -.,, , •'

function o f tetrperatxtre a n d t i m e ó f loading. Strength"~arfl^ stiffness w i l l g e n e r a l l y d e c r e a s e a s t h e temperature a n d p e r i o d o f loading increase. S o i t i s inpartarrt t o k r x w t h e loading c o n d i t i o n s o f t h e s t r u c t u r e . I n D u t c h r o a d engineering, t h i s h a d l e d t o asphalt m i x t u r e s b e i n g s t u d i e d urvier specific c o n d i t i o n s : F o r e x a m p l e c r e e p a n d w h e e l - t r a c k i n g s t e s t s a t 4 0 * 0 aixi f a t i g u e t e s t s a t o'c a n d 2 0 ° C . T h e m e t h o d o f t e s t i n g i s t h u s g o v e r n e d b y t h e desired furKtioneil p r o p e r t i e s . W h e n a s p h a l t i c c o n c r e t e i s u s e d a s a s e a w a l l r e v e t m a t t , t h e functional p r o p e r t i e s and therefore t h e t e s t c o n d i t i o n s a r e d i f f e r e n t . U n d a r n o r m a l circumstances, t h e a s p h a l t layer m u s t b e s t i f f , s o i t d o e s n o t rx3ticeably creep on the slc^». In addition, the material must be flexible, so irregular settleaaents in the subsoil can be followed without cracking. Under extrerae conditions (st^ier^ storm), the asphalt must resist a particular number of wave inpacts and water pressures. Resistarxse to wave inpacts is increasingly regarded as the most relevant criterion. This has also resulted in extsisive analysis of this aspect in the "Guidelines" [ref. 1 ] .

3. MBCHAMICKL BQlAVIGaR OF HÏERAIILEC ASE«AEinC GCNCREEE. IWLike in road engineering, a sea wall is designed to withstand failure with a limited nunber of load repetitions due to wave forces. In a si^jer-storm there are a maximum of a few thousand load r^jetitions in a period of 36 hours. Si?)er-storms statistically occur very rarely. The svqper-storm will occur between 0 and lO'c, the load frequency of the wave inpact being between 1 aivi 10 Hz. [ref. 2 ] .

There is little data in the literature en this specific fatigue behaviour (N<10*). In order to simulate this behaviour to some extent in practice, foroe-oontrolled repetitive

i.Litncriö; Surnaire ^.uteijr^s; ••;om C.C. Hontauban H.F.C, van de Ven Detemination uechanical ; properties of hydraulic asphaltic concrete - a c e r

l <

leeding can be adopted in OT^ acMeve'genuine failure of the testspecimen. In addition, it is possible to obtain a good

impression of the deformatian of the various asphaltmixes. The (^^namic strength of hydraulic asphaltic concrete (as a measure of the resistance to failure) is an essential uuu^xment in the design of asphaltic concrete revetments as described in the "Guidelines" [ref. 1]. A formula is derived for calculating the layer thickness on the basis of wave inpact value and aphaltprcperties:

•—iS^Wl"

(1)

The dominant influence of the bending tensile strength is evident. This strength d^jends on the nuntoer of times the material is loaded. The relationship is linear on doubLe-logari1±ttnic scales and is written as:

log(Ne«=tu») = Ijog(k) - a*LDg(a) (2)

The relationship between N and a is thus determined by k and a.

Par calculating a layer thickness under more than one wave inpact, the wave-impact formula has been re-written as:

/r»

1

In^PT */• 275

16(1 - o')c (3)

«here the stress is represented by N, k and a.

No standard method for determining the baiding tensile strength as stated in this formula is given. In the absence of better informa1d.on, use has been xaade of results of fatigue testing as obtained in road engineering (for example the design nomographs from Shell). These fatigue results have

'aper Authorssi 2'jrnar-e ,^uieur:s! "-icm C.C. Hontauban H.F.C, van de Ven Detenination uechanical properties of hydraulic as{*altic concrete ^et

I

mostly been obtained t h r o u ^ standard four-point bending fatigue test. In the Netherlands, this test is carried out with long beams (450 ran), strain-controlled alternating stress being imposed. However, the long beams make sisple sampling difficult, and a different method has therefore being sou^it. A three-point bending test was cdiosen as a starting point for research, as performed by PaiiLmann and Gratz at the Technical Institute of Darmstadt (Germany) [ref. 3]. The method consists in dynamic loading of a short beam in a simple i±ree-point bending eguipnait.

4. DEVQIXMQIT GF A THREB-FOINr EEBIDINS TEST (THB-test). 4.1. Itaree-poixit bending test versus fam>point bending test

(EH&-test).

There is a preference for FPB-tests to stucty mechanical properties of asphalt, because of the constant momsit and the zero transverse forcelevel in the middle-section of the beam, see figure 1.

i-^

t

Fig. 1 Transverse force and moment lines

It has been found in studies on cement-bound materials [ref. 5] that the TEB-test results in a hi^ier strength. The reason for this is tiiou^it to be that in the FPBr-test approximately one third of the beam is available for the weakest point.

P a a e r C.C. Hontauban H.F.C, van de Ven Detemination uechanical properties of hydraulic as{^altic concrete i. 6

However, the TPB-test has been chosen because of simple and representative sanpling. This can only be adiieved with core drilling. Because the dimensions of drilled cores are limited, a TP&-test, with short beams is a solution.

Because of the dcGDiinant inf Ivience of bending tensile Strength in the design method, the TPB^test is carried out in a force-controlled manner with a nan zero mean stress level (figure 3, 4 ) . It is then possible to cause the beam to fail within a limited iruntoar of load r^jetitions (10^ to 10*), so that the relationship Utrmsta» - o can be determined. It is not possible to make a direct comparison with the standard FPB-test.

4.2 Es^erjjnental phase

The following were to be determined:

- the fatigue properties under repeated loading up to a number of N=10* at fedlure

- the stiffness modulus ejqjressed as E^yn - the deformatian during the fatigue test

It was decided to start with the same dimensions as in ref. 3. The dimensions of the test^aecimen are 160*50*50 ranf arxi can be sawn out of drilled cores of dia. 200 mm (see fig. 2 ) . The test was initally kept as simple as possible (see fig. 3 and 4 ) . So i6o

f=^.

t«

ISb t-'50 i6o 1<Xi

+

.-3D8r C.C. Hontauban H.F.C, van de Ven Deteraination uechanical properties of hydraulic asphaltic concrete

Fig.3 TPB-test load signal Fig.4 TPB-test, bendingcurve In a f i r s t series of t e s t s the teaiperature was varied between O and 20°C and the frequency between 1 and 10 Hz. These t e s t s (Figure 5) shewed that i t i s possible to achieve a good relationship between bending stress and the number of load repetitions [ref. 6 ] .

THREE-POINT BENDING TEST

Nf-(TnuDc lOOOOOOi z 3

f

I I I I I I 10 100

Maximum Bending stress (Tmax [MPs]

• T « 0 l 3 . ; f - 9 . 8 Hz D T « 0 ' C . ; f = l Hz x T - 1 0 ' C . ; f - 9 . 8 H • T - 1 0 * C . r f - 1 Hz A T - 2 0 t . : f - 9 . 8 H x T-20*C.;f-1 Hz

Fig. 5 N-tT relationship of short beams of hydraulic asphaltic concrete

The stiffness modulus was also calculated from the automatically recorded foroe and deflection signals. The theory of the FPB-test [ref. 7] was used for this purpose, the derived formulae having been adapted to a three-point bending system L^^ief. 8 ] . The stiffnesses c±tained in this way were

Paper

-:,,^,..,... .,„.-^, Detemination uechanical 3|,gg^ " ^ è /

C.C. Hontauban '" ' —"- •) properties of hydraulic 1 * • • • 4 ^ ^ H.F.C, van de Ven ----:. - - ; asphaltic concrete p^M.jie / .^'^

found t o d i f f e r greatly from stiffnesses determined b y t h e F P B - t e s t o n similar m a t e r i a l . T h e r e a s o n m a y h a v e b e e n t h e dimoisions o f t h e testbeam, b u t a l s o t h e fact t h a t t h e d e f l e c t i o n o f t h e b e a m i s a combination o f increasing p e r m a n e n t deflection a n d a deflection amplitude (Figure 4 ) . B e c a u s e b o t h signals a r e m e a s u r e d w i t h o n e L V D I (set t o a w i d e range for the total deflection), it is found that the deflection amplitude cannot be measured accurately.

4.3 Calibration testing

Because of the probleaoas associated wildi stiffness, a limited calibration study was carried out [ref. 9 ] . This study consists of comparative measurements with FPB-tests and TPB^ tests, examining in particular the effect of the test equipment, the beam loigth and the LVDT. Alimoinium beams were used in the stufy and the stiffness «ras determined and compared with E^. ~ 70,000 MPa.

The mean conclusions are:

the accuracy of the LVDT plays a major role.

- t h e w a y i n \i4iich t h e b e a m i s si^sported a n d h o w t h e f o r c e i s a p p l i e d a r e v e r y influential.

- i n t h e c a s e o f s h o r t b e a n e (effective length = 1 0 0 ran) E ^ i s considerably lower liian i n t h e FPB-beams

(effective length = 4 0 0 m m ) .

I t s h o u l d b e n o t e d t h a t E ^ i s f a r h i ^ i e r t h a n ELpbmt a n d "that t h e deflections w e r e v e r y s m a l l . I n 1993 a calibration stucfy w i l l b e carried o u t w i t h e p o x y b e a m s w i t h a stiffness com-p a r a b l e t o ascom-phaltic c o n c r e t e . I n a d d i t i o n , i t w i l l b e examined through a theoretical study \idiether t h e u s e o f t h e a d a p t e d FPBr-formulae i s c o r r e c t o r n e e d s t o b e adjusted.

4.4 Final method

A final method was provisionally developed from the

'.'la,-': n eacr; :a-e.-JL.iSice ;^--r;;—> . . ; t h ' ; - : " . ..^;.-:v.-^arri;s;. ;;;:•- . ...._ir. sheet i^jmDer.

'•:•.::: rem ces at^'ei.-. .ura du rapcort, n-au.Tss. ont DCL.r cc.e' :- èv:ier 'es erreurs Iors

Vs'^f'-'i^ _'T^rn'mw. Paper Author(s) Surname Auteur(s) Nom \C.C. ƒ H.F. Hontauban C. van de Ven

I itie cf the paper litre du -acpcrt Detenination aechanical \ properties of hydraulic ƒ asphaltic concrete Sheet ^ , ï

-} ' ^^

Feuille

eoqierimentcd phase and the calibration study.

Because the effective laigth of 100 mm in relation to the uniniTtum dimension (50 mm) was very small, it was decided to take longer testspecimens. It was found possible to drill cores with a diameter of 250 mm. From one layer (= slice 50 ran in hei^it), see fig. 2, it is possible to saw two testbeams

220 ran in length. In this way reasonably slender testbeams were obtained (220*50*50 mnr*), which are tested with an effective length of 200 ran. The effect of the transverse foroe then will be «wwH [ref. 4 ] . It was decided that stiffness would only be measured at tine start of the test with an accurate LVDT (small range). A small foroe amplitude is set v<hich is maintained over a limited nunber of load repetitions. The fatigue is measured with a wide range LVDT. It will soon be possible to combine two LVDT's, so the variation in stiffness can also be measured during fatigue.

4.5 Scovisional test ^sdeications

Extensive test specifications were compiled to ensure that in future the tests are performed according to agreed procedures

[ref. 10].

The specifications inclixie:

- preparation of the testbeams - the test equipment

- the test procedure

- data processing (with the formulae used) - a detailed example

In this way an attempt is made to contribute towards a clear interpretation of test results.

5. TEST HRS[ir.T3 5.1 locaticnB

A number of hydraulic asphaltic revetments coverings were

Mark on each page, outside ;he trance, at the loo: autr.cr narneis). title c: paper, sheet nunnper.

These notes will not be annted. they are to avoid eri-crs in cage-setting.

Indiquer sur chaque feuiiie a l'exténeur du cacre °n haul: rem des auteurs. ;itre du rapport, n' de la feuiiie. (Ces indicatcns qui ne seront pas •'sc'cauites. ont pour cDjet e sviter ;es erreurs Iors

de la mise en pages.)

•f;-^mW^

Autr::ns: -,.

C.C. Hontauban H.F.C, van de Ven

'itie o- Th Detemination uechanical

\ properties of hydraulic

• asphaltic concrete

neet

Paper

tested in order to examine their quality. The results of two revetments are presented here.

First of all a newly constructed revetment was studied. Tliis was a test slope in the Delta Flimje of the Voorst Hydraulics Laboratory (Figure 6 ) . The test slope was constructed in 1991 to examine the effect of heavy wave impact on a sea wall body covered with asphalt [ref. 2 ] .

A layer of asphaltic concrete was e^lied to a slope of 1:4 and ooopacted with vibrating rollers. In order to interpret the measured results of the construction (stresses, strains and displacements) it is important to know the mechanicêü. prcf)erties of the asphalt.

J£i4

w.Wu«?i«^w

,drolnoqil»iflinq

Fig. 6 Asphalt revetment of test slope in Delta Flume

The sea wall of the Boulevard in Vlissingen was protected in 1957 with an asphalt revetment consisting of two layers of asphaltic concrete (Figure 7 ) . After more than 30 years service the mechanical quality needs to be established. An assessment can then be made on the safety, remaining service life and durability of the revetment.

de :a ^.=

d e • .1 :T

c r r : 3 c

:M.e. -COS nc\;atc se en pages.)

Paper iutnoris) burnair -uce'.in;; 'Icrr C.C. Hontaito H.F.C, van de Ven Detenination uechanical properties of hydraulic as[^altic concrete .r.~~l

}

11 ASFALTBEKLEDING KRAAGSTUKKEN

Fig. 7 Asphalt revetment on Boulevard de Ruyter in Vlissingen Cores of 250 mm diameter were drilled from the two revetments. After the beams were wawn, the remaining parts of the slices were used to determine the standard properties:

Location

Uo-kition: Density: Edtaen DBlity: stone fraction sandfnction filler fEacticn bitaai specific density voids ratio penetration softening poijit 1 1 1 1

i

1 ) I

I

1 f penstration i n ^ l Delta FliBe av. 49.1 42.3 8.6 6.6 2345 3.1 74 50.0 -0.3 (s) 0.7 0.5 0.3 0.1 7 0.3 6 1.1 0.1 1 1 1 ( 1 1

J

Vlissingai toplayér av. 46.5 46.7 6.8 7.5 2287 4.3 74 49.5 •0.4 (s) 3.6 2.4 2.0 0.7 84 3.7 16 2.0 0.4

These results show that there are wide differences between the new and old revetment. Althou^ the asphalt of Vlissingen is more than 30 years old, the mean binder quality is found to agree with new asphalt. More detailed research is needed to examine vihether the method of recovery for these old "^pes of

?j'.an<--Thes-i ' : : inaicu^r- -; de la fs-;: delan";ise ^^. j.r i . en ';aqe. '.•I'i n e t . " a c u 3 : . . s n c •jaaes. lUl-. • ' . o '-• ; - . : • • c a i : .".f i r ^ouroc;^; -"•'-.':•. ••^mcor. : -"coc^t. r--i rrrsurs ics

Paper jn:cr";; ourr3.Tfi C.C. Hontauban H.F.C, van de Ven Detemination uechanical • properties of hydraulic asphaltic concrete

)

12

r

asphalt is adequate and \(*iat the chemical/physical changes have been.

5.2 Stiffness modulus and phase angle

Four test conditions were dKisen for testing the beams in the FPB-test: Temperature: 0 5 5 10 ("C)

Frequency: 9.8 9.8 1.0 1.0 (Hz)

First of all the initial stiffness moduli arri associated phase angles were measured by imposing a small load signal. The force anplitude was limited to a maximum of 400 N. The force and deflection signal was sampled five times over a toted maximum roDober of load repetitions of 250, from v ^ c h the average stiffness ncdulus and phase angle was calculated using Fourier transformation.

In order to exandne v*iether the results of the Delta Flxrnne are realistic, a comparison was made with similar results obtained with the standard FEB-test (Figures 8 and 9). This related to new hydraulic asphaltic concrete of a test segment of the West-R^)elle sea wall on which the E-dyn and 0 were determined as a function of temperature and frequency [ref. 13].

Edyn-FREQUENCY

WEST KAPEUE (WK) «id OELTAFLUME (OF)

PHASE ANOLE • FREQUBMrY WEST KAPEOE (WK) and OELTAFLUME (DF)

laooo' 14000' laooO' & tOOOO' ^ « 0 ^ 0000' lU 4000-3Q0O' flH»! »t»J - ^ T-20*C(WK) -•- T-IO'C. P M O - ^ T-O'C. (WK) - ^ T - « * C . ( O F ) * T-0'C.(OF) « T - K T C I O F ) Fig. 8 Fig. 9

The measurements show that both test methods simply comparable results on new hydraulic asphaltic concrete. The data quoted above can nam be used as reference values to test the results

de :a "^ü;::e. C--s v-r; de 'a n i s e en caaes. :.ui' oci-ii ; ;•.. ..."0^.^ ^:;cor*. n-; rrr^L-rs :crs

-aper

ircnsi Sur-a.T^

C.C. Hontauban H.F.C, van de Ven

^ace.' Detenination nechanical _{properties of hydraulic}

asphaltic concrete

l 13

r

of older structures. Only the results of the top layer are | given for the Vlissingen revetment because this mixture, like | hydraulic asphaltic concrete, is a daise asphaltic ccncrete with crushed stone (Figures 10 and 11).

E d y n • Frequiney

DELTAFUJME (OF) and VUSSNOB^ (VU OELTAFLUME (DF) ind VUSSINOEN (VL) PHASE ANGLE - FREQUENCY

RvqiMn^lWt n«4MnaV|Hl)

A TmO'CJPF) - • - T - S t . ( D F ) I T - i e ( t ( D F ) K T-O'C. (VU - ^ T - s t . ( V D " T - 1 0 t . ( V U

F i g . 10 F i g . 11

The results show that the old asphalt has a lower stiffness. It is not clear v*ether this is solely attributable to the

hitler bitumen content or ZLLSO to a certain degree of

stripping. The expected hi^ier phase angle was not found tut

appears to be sli^itly lower. Detailed stucty of recovered binder (nechanical arxi chemical) is required in order to find

an explanation. Another way of comparing the data is given in Figure 12, where the phase angle is plotted against the E modulxjs, so that eill observations can be contained in one curve. The difference between the old and new asphalt, is very

clear. gdyn - PHASE ANGLE

DELTAFUJME (DF) and VUSSINQ0I (VU

DELTACSOOT VUSSINGBI

i:

I:

Figure 1 2 . rJmttjiimtmi r»wfci>*w Nv M WMmm u G S i l » » Biyn(QPi4 -ct -^eorin'ec. v.ccs de 'a ~i:se en oaqes. ;i -T-^urS :ors

Paper . i n o r i s i L.urna.'^'v:-C.C. Hontauban H.F.C. van de Ven Detenination nechanical properties of hydraulic asphaltic concrete jnee:

i

5.3 Dynamic sLtnaigUi

The same test conditions are retained for determining the <fynamic strength. The foroe level (anplitude) is set so that the beams preferably fail after approx. 10*, Itf" and 10* load repetitions. In reality this proves difficult to manage, particularly in the case of old material.

Tests on the Delta Flume naterial provided the follcMing results: OELTAFLUME Nf-Ohiax 1(XXXX> _ 10000; Z

I

1000 10 100

Maximum Banding ttran <r max (MPa]

" • " Ö t i S S A SiCSAta iïi~^TÖHÏ Z lOtSIAte

Fig. 13: N-a relationship of hydraulic asphaltic concrete of Delta Oiannel

The regression and correlation coefficients were calculated in each test condition;

Temp. Freq. Log(k) a c.c.

0 5 5 10 9 . 8 9 . 8 1 . 0 1 . 0 5 . 4 6 . 4 4 . 4 3 . 7 2 . 3 4 . 7 3 . 5 3 . 0 0.80 0.99 0.97 0.96

A strong relationship is found in three out of four conditions (c.c. > 0.96). It must be borne in mind that all the regressi(Dns have been calculated with a maximum of six results. One deviating result t^us has a great effect.

.la,"' en eacn n

• -.• -^-.^1 O O l _.er:;L.

•19 'afauiile. iCas ,:'."::•: de !a mise en pace.s.i

,-\L.; C.C. Hontauban H.F.C, van de Ven Detemination aechanical properties of hydraulic asi^altic concrete Slieet Paper

} "

Testing the old asphalt of Vlissingen is found to pose a number of problems. The quality sometimes proves to be so poor that a beam even can rwt be sawn. It is also found that beams are so weak that one load cycle alrea(^ leads to failure. In the beams that could be tested, the quality was fourxi to be so variable (see table in 5.1) that it was pointless to determine the regression lires. The following results were obtained: VLISSINGEN Nf-<Jmax 100000 „ 10000 a

I

1000 lOOd

Maximum Bending stress ömax [MPa]

0*0; 9.6 Hz A 5*0; 9.8 Hz « 5*0; 1.0 Hz x 1 0 t ; 1 . 0 H z

Fig. 14

The calculation of the regression lines provi(Jed the following results:

Temp. Freq. liog(k) a c.c. n

0 5 5 10 9 . 8 9 . 8 1 . 0 1 . 0 9 . 6 - 1 . 4 4 . 2 2 . 9 9 . 9 2 . 2 4 . 7 3 . 5 0.54 0.15 0.92 0.80 6 6 6 6

It is found from the correlation coefficients (and the negative value of Log(k)) that there is virtually no connection here. Each observation will in itself form part of

:-c.-^,. .-CM'::

.''aper C.C. Hontauban H.F.C, van de Ven Deteraination aechanical properties of hydraulic asphaltic concrete

}

16

a possible regression line; however, there will be several regression lines which d^aend (for exanple) on the density. Instead of statistically testing the coefficients of the old and new asphzdt, the testing of the "old" data against the mean level of the "new" data is a possibility. This can be done by setting vp a confidence interval of the "new" data for the regression line and examine vAiat "old" data lie within this interval. Another advantage of this method is that observations at N = 0 (i.e. no beam to be tested) and N = l

(i.e. fracture after one load repetition) can be included. In Figure 15 the results of Vlissingen (Boulevard De Ruyter) are cogoipared with the confidence interval of the regression line of the Delta Flume study. Out of 27 cbservations, six are found to lie outside the interval, indicating that the old asphalt is significxurtly poorer.

Boulavard Da Rgyter •*r _-tr Boulavard 0* Ruyttr • < i — • — n — t t — n — » r t * — t -Boulioard 0» Ruytar •<i 1 t i — t r t i — r r -Bouloard Oa Ruyttr \ • * ! — t — r t — 1 > — r i — n — t

-Fig. 15 Testing of old asphalt against regression interval of new asphalt (S = maximtmn bending stress)

:ar:< ..;

de'a;aL,:,3. .Cas oc de 'a m s e e n oaaes.

Paper Au;"crisi oi.rrarr.\: C.C. Hontauban H.F.C, van de Ven Deteraination aechanical properties of hydraulic as[^altic concrete ^h."-^*

I

17

6. C A D a U A n C N OF SISUCniRAL VALUE.

A conparison of old and new materials alone does not lead to a statement on the residual value of tiie construction. It is necessary to examine vAiat load an old revetroait can withstand before failure occurs.

A possible procedure is indicated below in diagrammaidc form (Figure 16): 0. Wave impact

t^Jlk-I

C. WBhlercurves H E. Design fomnila a=P */3* E V * ' (l-y*)ei»» 6. Miners role

I

Fig. 16: Diagram for calculation of structural value.

Results have been obtained from testing the drilled cores on layer thickness (Fig. 16A), the E modulus (Fig. 16B) and the fatigue bdiaviour (Fig. 16C).

From the fatigue data (N-a), kncwn as the Wöhler curves, and the initial E moduli, F curves are derived (Fig. 16F).

With wave force distributicais (Fig. 16D), design fonmiLa (Fig. 16E) from the Guidelines and layer thickness/E modulus data, the total scale of stress alternations can then be calculated. Wi-üi this scale of stresses and E moduli as input in F curves, it is possible to determine the distributions of the permissible ixunbers of load r^)etitions (Nj). These numbers

'as-r "'."".as .viii i'c: •ic:a\..e!- ^ur : : r a c i - i :3 ^a reuiiie. (Ces ;nc J3 a rmseen pages. ^ . : : i i . -.f — TXtrfiS'- • ^ u j -acoorr. • irs

Paper

-. . ..^.,,.^^_^, Deteraination aechanical 31,^5;

C.C. Hontauban " properties of hydraulic ^ " \ 18 H.F.C, van de Ven - ,- ^ asphaltic concrete c,-...^ -..^ ƒ

c a n statistically b e t e s t e d a g a i n s t t h e w a v e f o r c e s p e c t n m a (rii) u s i n g M i n e r ' s r u l e .

B y a p p l y i n g M i n e r ' s r u l e , t h e p a r t i a l f a t i g u e d a m a g e o f e a c h p a r t o f t h e s t r e s s s p e c t r u m c a n b e calculated. Suranation o f the damage then leads to the Miner's sum, which can be tested against a standard to be specified.

7 . CCËUCUJSICHS A N D R B O C M M E M D K E I C H S

It seems to be possible to determine the quality of an asphaltic concrete revetment by drilling large (250 mm dia.) I cylinders, sawing short beams and subjecting the beams to dynamic bending. Drilling cores have the great advantage that :

i samples can be taken simply and nm-selectively from the

i

revebnent. { The short beams can be c^namically tested in a simple

three-^»int bending set-n;p, force-controlled repeated loading being found to be suitable for causing the fatigue failure of [

the testspecimens without creep hacving a great effect on fatigue. This conclusion ^:plies to a nmbec of load

repetitions < 10*. i Using this method it is possible to determine both

stiffness modulx:s, phase angle and the relationship between stress and number of load repetitions as a function of : temperature and frequency.

As a result of the absence of adequate measuring

i

instruments and software, however, E and 0 have to be measured separately from fat:igue. Soon it will be possible to measure the variation in E/0 during fatigue.

Both with E/0 and with the fatigue data (N-a), it is possible to test old material technologically, a l t h o u ^ a statisticed. model needs to be developed with which the "old" : data can be tested against a confidence interval of the "new" data. For better interpretation it will be necessary to make

'nciCj-r ?i:r •:naGL..;-i;.-. :aa >xfa':;\;- :u :•;;: --^- ••;„;: '!" ...i; auteurs, vtra ..u •-.:;:.:rt. T

ae !a^?^;• a. .CcS ,ra:':a:cra •:ui re ièr.;,-i :.Z:-i •-.z'lZ-^::^-:. ;•; ;c;j-ODiet e'5v:-3r -a ,r--urs crs de !a mise -^n pages.;

Paper Autnor.'si Surname ^ AuraL^r': \ C.C. Hontaiiian j H.F.C, van de V T;H . - u .• -n^r Deteraination aechanical i properties of hydraulic „-.,_ j asjÈaltic concrete T;-'i

. : } " :

a closer study of the qualil^ of the binder during lifetime. This applies both to the recovery method and tiie chemical/^iiysical change in the binder.

It is in principle possible from the total results of drilled cores (layer thickness, E/0, N-<T) to calculate a stnrtural value. For t M s purpose the load spectrum should be known and Miner's rule should be c^iplied. It should then be calculated whether a revetment can still withstand a siper-storm witiiout failing.

Finally it is reocrapaaided that the test results should be linked to visual inspections, to be akae to examine to what extent damage can be explained on the basis of materials testing.

8. REFERBNCES

[1]. Technische Adviescommissie voor de Waterkeringen. Leidraad voor de to^assing van asfalt in de waterbouw. 1984.

[2]. Derks H. en Klein Brebeler M. Gedrag van asfaltbekleding onder golf aanval. Rapport A4-92-19, Watarloc^ikundig Laboratorium, mei 1992.

[3]. Paulmann, G. en Grëitz B. Verformungs- und Zeitfestig^

keitsverhalten von Tragschiditmdschgut mit und ohne Zusatz von Asphaltgranulat bei Biegeschwellbeanspruchung in DïN-System VA. Bitumen No. 3, 1987.

[4]. Larsen T.J. en Nussbavm P.J. Fatigue of soil-cement. Journal of tlie PCA Research and Developaent Laboratory/ mei 1967.

[5]. IBBCHINO. Literatuuronderzoek naar een tarantitatieve relatie tussen buigtrek- en druksteiicte; in het bijzcxider met betreüddng tot de bruikbaarheid voor betonnen wegverhardingen in Nederland. Rapport B-81-449, 1981.

[6]. Hontauban, C.C. Onderzoek naar de dynamische sterkte van waterbouwasfaltbeton. Notitie MAD-N-89040, Dienst Vfeg- en

.Vlars: '.r i a c h oage. ouisiüs ' " = ' These rc-^s '.vi:' net be onv.ea.' Indiquer 3ur cnaqua feLiiiie a • -..•: de ia feu::;e. ,Ces -ndicatons CL, . de !a mise en pages.)

';:"re. :,\ir.'r:oo: :;L.t-cr •-• '4'V are ' c avoid a^-rzrs ••! •:

r''"^>u.'' clu Caere Br. na^^i! a >:ercnt a a s .-ecroau.ta.s.

T •;!:;;. :itie OT aacar, aneet "umoer. v.:5--.eit;rg.

.~ :'3S auieurs, titra da raocort. n-: r - oour cDiet e'éviter las .-^irröurs lors

*>

Paper

yW^-Deteraination aedianical

Author's) Sumam.a ^ . . „ ^ ^ Title of the pacèr ^ w!Ma.«iuou.uii « oiuwu. 3l-,gg^ 1 C.C. Hratauban \ poperties of hydraulic 1 20

Auteurisi Mcrr ƒ H.F.C, van de Ven T i t r e d u - a D o t : - ƒ aspHaltic concrete c ^ . n e ƒ

Waterbouiidainde, november 1989.

[7]. Molenaar J.M.M., Dommelen A.E. van en Pronk, A.C. Theorie van de vierpuntsbuigproef. Rapport MAD-^i-87064, Diaïst Weg-, en W^terbcu»dainde, 1987.

[8]. Kunst P.A.J.C. en Pluim A.J. Stijfheden "Biegeschwell-Versuch" (aan waterbouwasfaltbeton). Notitie 89029-2, Netherlands Pavement Consultants bv, november 1989.

[9]. Wattimena, J.S. Ralibratieonderzoek & software-ont-wikkeling t.b.v. het beproeven van waterbouwasfaltbeton in de driepuntsbuigopstelling. Rapport 90053-2, Netherlands Pavement Consultants bvr, mei 1991.

[10]. Wattimena, J.S. Handleiding van de dri^juntsbuigproef op waterbouwasf Itbetcn. Rapport 90024-a, Netherlands Pavement Consultants bv, september 1990.

[11]. Montaxiban C.C. Onderzoeksplan duurzaamheid waterbouwasfaltbeton. Notitie MftD-N-91069, Dienst Weg- en WaterbouMkunde, novesÉser 1991.

[12]. Versluis, A. Driepuntsbuigproeven aan balken van boarkemen viit de Deltagoot. Rapport 91523-2, Itetiïerlands Pavement Consultants bvr, noventer 1991.

[13]. Hontauban C.C. Asfaltbetohbekleding van de proefvaldten West-Kapelle. Rapport MAQ-Rr88038, Dienst Weg- en Waterbouw-kunde, f^sruari 1989.

[14]. Versluis, A. en Bhairo, J.S. Dri^juntsbuigproeven aan balken van boorkemen lokatie Boulevard de RL^ter Vlissingen. Rapporten 91614-2/92016-2, Netherlands Pavement Consultants, februari 1992.

MarK zr ^acn cac?. JL;? da :ra f'ama ai ;ne yzz: .„zzz- ••^.rr.'i,>j. "it,e :^ aaer. :;• >a; r,.mber. These actes 'A'!]! ~ot be cnrtec. 'hey are to avoia -•••a'' .- caae-seitinc.

inaiauar lur onaaua 'eai-e z:' ïxteneur cu cadre 'zn rzov. ncm aes auteurs, 'iire ai; raaport, n-" de ia feuiHe. .Ces inc caions a'_; r^e sercnt oas 'ec'cca.ïes. ont pour objet e -vite^ 'as .srreurs iors delamiseen pages.)