Category Stone Matrix Asphalt. Theory and Practice

Stage 1: Selecting an Aggregate

Requirements for aggregates to be used in SMA are catalogued in Chapter 5. When differences among densities of aggregates used for composing an SMA aggregate

* 2.36 mm in the United States, 2.0 mm in Europe.

[30] NMAS stands for nominal maximum aggregate size—a sieve one size larger than the first sieve retain­ing more than 10% of the aggregate.

[31] Fill the container with aggregate up to one third of its height, level the surface of the poured aggregate using your fingers, and then tamp the layer down with 25 strokes of the tamping rod, taking care to evenly distribute the strokes over the surface and avoiding hitting the bottom of the container (Figure 7.7).

[32] Having completed the tamping of the first layer of aggregate, fill the con­tainer with a second layer of aggregate—this time up to two thirds of its height—and repeat the tamping procedure.

[33] Calculate the void content in a compacted aggregate according to the formula

[34] In earlier research, other numbers of rotations have been proposed based on Los Angeles (LA) abra­sion loss of coarse aggregate—that is, for LA less than 30%, 100 SHRP gyratory compactor (SGC) rotations can be used, and for LA greater than 30%, 70 rotations (Brown and Cooley, 1999).

* We start with number 3 of the first mixture, because the next mixtures will have a different composi­tion based on HDK content (mixes 1 and 2 with decreased HDK content, mixes 4 and 5 with increased HDK content); so the number 3 is virtually in the middle of the series.

[36] It is being adopted based on the experience of a process engineer (e. g., a typical amount for a given gradation). The term optimum is a little bit exaggerated because the truly optimal content will be described later in the text.

[37] The telescope principle is a term used in a paper by Lees (1969) as the layout principle for grains in a mixture in which the finer particles fill in the remaining voids among coarser ones.

[38] For instance, using a small amount of oil with the density at the room temperature corresponding to the density of binder at the temperature of mixing with an aggregate. So an oil with a density of ca. 0.2 Pa. s at 25°C should be used.

[39] In the Netherlands a gyratory compactor is used for such testing because the Marshall hammer is regarded as being excessively damaging to particles making the skeleton, and for that reason not fit to be used for compacting the aggregate with such a strong skeleton.

* 5.0% of air voids is used only for heavy-duty pavements.

f So-called Van der Baan’s number (i. e., the bituminous number after the EN 13179-2).

[41] Binder 70/100 with Pen@25°C from 70 to 100 dmm. Former road bitumen 80/100 after the Dutch standard NEN

§ Obviously it is about quantities of fixed binder and free binder, whose definitions and significance are discussed in Chapter 3.

[42] We start with 200 rotations of a gyratory compactor. Then we determine such a number of rotations that changes the sample height by 0.2% during the last 10 rotations. Density of the mix reached at the selected number of rotations of the gyratory compactor are a reference (maximum) density for calcu­lating the content of air voids in the SMA. (Voskuilen, J. L.M. et al., 2004.)

[43] The bitumen content for SMA 0/8 amounts to 7.4% (m/m) for the heavy-duty traffic and 7.0% (m/m) for low-volume traffic. Both values have been calculated with 100% of the aggregate mass and the road bitumen 70/100 as a standard binder.

[44] The content of air voids in laboratory compacted samples amounts to 5% (v/v).

• The content of air voids in the compacted coarse aggregate fraction amounts to 36.7% (v/v).

• The content of the coarse aggregate fraction in the aggregate mix amounts to 78% (v/v).

• The FRs ratio equals 0.

[45] With a wetting agent.

[46] Calculate the expected SMA void content based on volume properties.

• Gradation of the aggregate

• Voids in a compacted coarse aggregate skeleton

• Binder content

[47] In this book that energy is usually referred to in 2 x A conventional notation, where A denotes the number of impacts on one side of a sample (e. g., 2 x 50 or 2 x 75).

[48] The first experiments with this type of apparatus were performed from 1939 to 1946 by the Texas Highway Department in the United States. More about the history behind the gyrator compactor can be found in Harman et al. (2001).

[49] Laboratoire Central des Ponts et Chaussees, or Central Laboratory for Roads and Bridges, France

[50] The fixed temperature of draindown testing adopted in the original Schellenberg’s method, 170°C, represents an average SMA production tem­perature for a typical mix with binder having a Pen@25°C = 50-70 dmm. However, it is necessary to note that higher temperatures have been used

for the production of SMA with polymer modified binder. Moreover, in cool or cold months asphalt mixes are usually produced at temperatures that are slightly higher than 170°C.

• It should be emphasized that the AASHTO procedure does not provide for a fixed binder draindown test temperature but makes it conditional on the expected SMA production temperature at an asphalt plant. Such an approach has its advantages. Above all, it ensures that draindown will not occur at the real SMA production temperature, thus the risk is easily estimated. Additionally, draindown testing at a temperature higher than the produc­tion temperature (by 15°C) makes that certainty even stronger since sudden temperature fluctuations happen sometimes, particularly at the beginning of production. Similarly, the European standard EN 12697-18 has defined the draindown testing temperature as 15-25°C higher than the planned mix production temperature.

granulated stabilizer in an oven just prior to mixing it with aggregate (see Section 8.1.4).

• Two more issues regarding draindown testing with the use of a glass beaker are worth raising.

• Before putting hot SMA in a glass beaker, the beaker should be heated in an oven to the test temperature, otherwise the hot mastic of SMA will easily stick to the cool walls of the beaker, falsifying the test results. Many people claim that testing with a beaker is a matter-of-fact mea­surement of mastic-to-glass adhesion.

[53] Let us imagine pouring a mix from the glass beaker onto a tray after warming the mix in an oven for 1 hour. We have weighed all the mate­rial remaining in the beaker (i. e., the mass of the beaker with the SMA residue). The question is, is all the remaining material in the beaker the real ‘potential’ SMA draindown? After all, the rest of the mastic, sand, and grit grains stuck to the walls remain in the beaker. Some material from any asphalt mix would cling to the beaker even if there were no draindown. So, the residue in the beaker should not be considered com­pletely equivalent to drained off mastic. Logic demands, however, that we recognize the material remaining on the bottom of the beaker as draindown. In light of these concerns, the method of measuring mate­rial outside a container— namely, the AASHTO method with a wire basket—deserves consideration.

[54] EN 196-6. Methods of testing cement—Determination of fineness.

[55] The procedure was developed by the Koch Materials Company.

[56] Variant 1—the types of binder and the recommended production tempera­ture range are provided in one specification. This is the simplest way to

[57] Dry mixing

• The dry mixing time should be limited to a minimum since mixing with­out binder substantially increases the rate of wear on the pugmill and pad­dles and promotes the breaking of the weaker grains of an aggregate.

• Increasing the dry mixing time lowers the output of an asphalt plant.

• An excessive extension of the dry mixing time with a stabilizer in a loose form may cause its destruction (pulverization) or grinding down to a filler shape.

• The order of aggregate batching to the pugmill and the moment of the filler delivery has a significant influence on the mixture durability (see Section

• Wet mixing

• An excessive extension of the wet mixing time causes a higher aging rate of the binder.

• Despite a proper mixing time, a granulated stabilizer may not be well – dispersed in the mixture; this can be caused by a poor quality stabilizer, so, it is worthwhile to occasionally check its quality.

[58] The number 2+ is a designation of the level of attestation of conformity (AoC) system according to Annex III of the Construction Products Directive (Council Directive 89/106/EEC). The system describe the tasks of the construction product manufacturer (here, the asphalt mix producer) and the tasks for the notified body (control organization). In system 2+, which is the most often used for construction products, the manufacturer should carry out the initial type-testing of the product (e. g., test for conformity of the mixture with the specification), establish FPC system and, possibly, testing samples taken at the asphalt plant in accordance with a prescribed test plan.

[59] These terms and range of activities have also been described in the European standard EN 13108-21.

[60] When there is excessive binder and mastic

• When the temperature of the mixture is too high

• When there is insufficient stabilizer or it is of poor quality

[61] Polymer modified emulsion, 60% of binder, rapid type. C60BP1-S is an emulsion designation accord­ing to the rules of EN 13808.

[62] Appropriate selection of the mixture gradation relative to the layer thickness

• Efficient operation of the paver (when spreading mechanically)

• Suitable manual spreading in places inaccessible to the paver

• Proper use of rollers

[63] The type of rollers should be compatible with the thickness of layers and the ambient conditions.

• The number of rollers should be compatible with the expected area to be paved and the compacting ability of the rollers. At least two or three rollers for one paver should be expected.

[64] Average per 1,000 m section: not less than 1.3 mm; average for a set of 10 measurements: not less than 1.0 mm (HA MCHW, 2008).

[65] To illustrate some problems, infrared images were used (with temperatures given in degrees Fahrenheit). However, not all of them concern SMA layers; the goal is to show the temperature differences during laydown.

[66] It should be pointed out that an excessive dry mixing time of loose stabilizers causes abrasion of fibers, reducing them to filler (dust) and substantially lowering the stabilizer’s binder absorption effectiveness.

[67] An error in designing the mixture (e. g., high coarse aggregate content, insufficient volume of mastic)

• An error while mixing at the asphalt mixing plant (e. g., incorrect composi­tion of the mixture [i. e., not in conformity with the design])

• An error of placement (e. g., paver setups, manual spreading of the mixture, insufficient temperature of the mixture during laying, undercompaction)

[68] That aspect of problems in design is discussed in Chapter 8.

[69] This remark concerning mixture production and look applies to classic hot mix technology, not the so-called warm mix technology.

[70] Method I—a series of specimens are compacted with different efforts, and their bulk densities are determined.

• Method II—based on a height measurement, a change (i. e., an increase) in the bulk density is determined after completing each stage of compaction.

[71] Limit suggested when evaluating the in-place superpave mix pavement permeability.

[72] Using modified binders is preferred; excessively hard binders are not recommended.

• Proper bonding between the thin course and its sublayer is one of the most important factors determining the pavement’s durability; special tack coats of asphalt emulsions (with enough hard binder) are preferred.

• Usually fewer roller passes are needed for the suitable compaction of a thin course; however, the high speed of laying requires a proper number of roll­ers to keep up with the paver speed.

[73] Instructions on how a recipient can obtain detailed information about the compliance of a mixture with the requirements of EN 13108-5

* Set “+1”: 1.0, 2.0, 4.0, 5.6, 8.0, 11.2, 16.0, 22.4, 31.5, 45.0, 63.0 mm.

t Set “+2”: 1.0, 2.0, 4.0, 6.3, 8.0, 10.0, 12.5, 14.0, 16.0, 20.0, 31.5, 40.0, 63.0 mm.

[75] The sieve systems: basic, +1, and +2 are established in EN 13043.

[76] Full range: Bm

Bmin7.2, Bmin7.4, Bmin7.6.

[77] Full rarrne: V15V 2- V 25V 3V 35 V 4 V 4 5 V 5V 55 V 6V NR

а чи » minA’^’ ’ min^’ ’ min^’^’ ’ min^’ ’ min^’^’ ’ minо ’min^’^’ ’ min^’ ’ min^’^’ ’ min^’ ’ min1-

[78] Full range: VFBmta71, VFBmin74, VFBmin77, VFBmin80, VFBmin83, VFBmin86, VFBminNR. t Full range: VFBm„77, VFBm„80, VFBm„83, VFBm„86, VFBm„89, VFBm„92, VFBmaxNR.

14.5.9 Resistance to Permanent Deformation

Resistance to permanent deformation is one of the most significant properties. Testing is carried out according to the standard EN 12697-22. The equipment used for SMA testing includes a large size device and a small device. selection of device and Test parameters according to EN 13108-20

For SMA tests according to EN 13108-20, methods listed in Clause D.6 are shown in Table 14.1. The appropriate method adopted in an NAD with appropriate test param­eters should be selected from these methods.

The selection of small and large devices is based on Table B.5 of the standard EN 13108-20, which states the following:

• The small device is for testing SMA mixtures designed for axle loads less than 13 tons.


A few years ago, at the beginning of the work on this book about SMA, I did not suppose it would take on such imposing proportions. Meanwhile, over the course of work on the publication, it turned out that the quantity of accessible materials on SMA was really spectacular, and the range of SMA-related subjects was enormous. While carrying out a survey on relevant publications, it became noticeable that SMA was still a fascinating asphalt mixture to many process engineers the world over.

All in all, it would be appropriate to finish briefly, in contrast to the content of the book, which might seem to be a bit verbose here and there. Nevertheless, I do hope that it will help its readers comprehend SMA and clear up any problems that might arise.

It is a matter of course that the examples quoted in the book cannot fully corre­spond with personal experiences of each individual reader. Should anybody like to exchange views about SMA mixture, please get in touch with me by e-mail at sma@ road. pl.

* Road binders (bitumens) B65 and B80 were based on Pen@25°C range according to German DIN standard: B65 was 50-70 dmm, B80 was 70-100 dmm.

[2] Germany, Patent No. 1926808 (1969); United States, Patent No. 3797951 (1971); Sweden, Patent No. 7110151-3 (1972); France, Patent No. 71.28874 (1971); Luxembourg, Patent No. 63688 (1971).

[3] This subject will be discussed in Chapter 7.

[4] When arranging spheres and circles as in Figure 2.6a and b, the passive grains cannot be bigger than 0.41 R. So, in an SMA 0/12-mm mixture. Assuming that the active grains are 8/12 mm, the next smaller size fraction should be just 2/5 mm to prevent shoving the active grains apart. The desired gradation discontinuity is developed by the absence of the 5/8 mm fraction.

[5] For more information, see the discussion of Kjellbase in Chapter 13.

* It is a requirement regarding an SMA surface course after its incorporation; for more information, see Chapter 10.

[7] AASHTO is the standards setting organization; founded in 1914, it was known as the American Association of State Highway Officials (AASHO) before 1973.

[8] See Chapter 11.

[9] In this book definitions of the mastic and mortar will be used as presented here. However, in numerous countries these definitions differ from those adopted in this book; for example, in the United States mastic is called total mortar and mortar is called fine mortar (and contains drainage inhibitor, which is a stabilizing additive). f BBR – Bending Beam Rheometer.

[10] The flow rate method, similar to the European concept, is not standardized in the United States Comparison tests have been conducted and are published in (Tayebali et al, 1996).

[11] Filler grains smaller than the bitumen film on aggregates can behave like a carrier (binder extender); very fine filler makes the mix behave as if there is even more binder present, which may result in such problems as the loss of surface course stability, rutting, binder bleeding, and fat spots.

• Filler grains bigger than the binder film on aggregates behaves like a filling aggregate, forming mastic, and taking part in filling up the voids among chippings.

• An excess of filler leads to mastic stiffening and the increase of cracking susceptibility.

• The affinity between filler and binder influences the durability of the mix (i. e., its sensitivity to water).

• The appropriate ratios of binder and filler, combined with their properties, have an influence on an SMA mixture’s workability and, continuing from that, influence the SMA compaction (or final field density).

[12] Blaine’s test is not very accurate for it does not take into account fine voids (area textures); more pre­cise measurements can be performed with laser devices (Grabowski and Wilamowicz, 2001).

[13] In this example, the content of voids in dry-compacted filler is being tested after Rigden’s method or Rigden’s method modified by Anderson (see Chapter 8).

[14] That state might be called a colloidal system of grains (solid bodies) suspended in binder (fluid body). A blend of binder and filler may be regarded as something in between a gel and an alloy at the working temperature of a road pavement (Harris and Stuart, 1995).

[15] DSR – Dynamic Shear Rheometer. t According to the EN 13043 standard.

[16] According to the EN 12597 standard Bitumens and binder products. Terminology, “hot” applied bitu­mens may be divided into road bitumens (soft and hard), modified (including polymer modified) and special.

[17] Additives that absorb part of the binder (the surplus that is likely to draindown)

• Additives (polymers) that increase binder viscosity at high temperatures, which in turn reduce the risk of its draindown

[18] Cellulose—the most popular (Shown in Figure 4.1)

• Pseudocellulose—made of milled or fragmented waste paper

• Mineral fiber—developed through melting rocks and subsequently process­ing the melted rock to form threads (like rock wool)

• Cellulose-mineral—a blend of cellulose and mineral fibers occurring in various compositions (proportions)

• Cellulose-polymer—a blend of cellulose fibers and different types of poly­mers occurring in various compositions (proportions)

• Cellulose-wax—a blend of cellulose fibers and synthetic waxes, which not only stabilizes but changes the binder viscosity-temperature relationship as well

• Textile—threads of processed and fragmented textile waste products

• Plastics—for example, polypropylene (Shown in Figure 4.2)

[19] This method, which is described in European Standard EN 1097-1, could also be performed without water. In the United States the Micro-Deval method is based on AASHTO T327 (ASTM D6928).

[20] The notation Declared exclusively denotes the necessity of giving a test result without determining any threshold limit for that requirement.

• The notation Category refers to the following:

• An absolute numerical value for a given category in the system presented in EN 13043: the numerical limit is shown in the table following a letter symbol for a given category (see also Table 5.1 for explanations)—for example, a category MBF10 means the requirement is MBF is less than or equal to 10 g/kg according to test method EN 933-9 or PSV50 means requirement PSV is great than or equal to 50 according to test method EN 1097-8).

• A declared value, which means the result of the test is outside the bounds of the last category with a specified numerical value—for example, cat­egory MBFDeclared is used when test result is larger than 25 g/kg. For two specified properties: resistance to polishing (PSV) and resistance to abrasion from studded tires (AN), declared values mean both—result of the test outside the limit or any intermediate value.

[21] Council Directive 89/106/EEC of 21 December 1988 on the approximation of laws, regulations, and administrative provisions of the Member States relating to construction products.

[22] Fiber length: maximum 6 mm

• Thickness or diameter of mineral fibers: maximum 0.005-mm mean test value

• Gradation of cellulose fibers

• Passing a 0.15-mm Alpine sieve (method A): 70% ± 10%,

• Passing 0.85-, 0.425-, and 0.106-mm mesh screen sieves (method B): 85% ± 10%, 65% ± 10% and 30% ± 10%, respectively

[23] SMA position (in a wearing course or in an intermediate course)

• Design thickness of the course after compaction

• Traffic load and the location of the road section (e. g., rural or urban)

• Additional requirements for the SMA course

[24] All four variants have preserved a fixed content of chippings at a level of 75%.

[25] A finer graded coarse fraction is characterized by a greater amount of smaller pores more evenly distributed through the mix, which brings about better interparticle contact and, in contrast, increases the risk of shoving grains aside by larger particles of the sand fraction.

• A coarser graded coarse fraction is characterized by a smaller amount of larger-sized pores unevenly distributed over a mix.

[26] The contents of air voids in the aggregate mix (i. e., VMA), causing a sub­stantial rise in the optimum quantity of binder—the very high volume of air voids in the coarse aggregate skeleton must be filled in with binder

• The SMA resistance to permanent deformation, which is an advantageous effect

[27] These values are of TL-Asphalt 07 and come from the new methods of density measurements (accord­ing to EN); in the old ZTV StB 01 corresponding values were 3-4% (v/v).

Designations used in German guidelines for SMA: N = low and medium traffic, S = heavy traffic (e. g., SMA 11S).

[28] Selecting an aggregate

• Designing a gradation curve that secures the desired interparticle contact (stone-to-stone contact)

• Selecting the gradation corresponding with the criterion of a minimum of air voids in an aggregate mix (minimum VMA)

• Selecting an amount of binder for a target content of air voids in compacted specimens of the asphalt mixture

• Checking for draindown and water susceptibility

Next we will follow the U. S. cycle of design through its successive stages.

Additional Tests of Asphalt Mixture Characteristics

The previous deliberations on FPC use a basic level applied to the standard EN 13108-21 regarding tests of the mixture composition. Meanwhile methods of test­ing other properties of an asphalt mixture are discussed in Annex D (and are very informative). Collected results should be declared and used to support the procedure of extended validation of type testing.

The SMA mixture characteristics that are tested include the following:

• Content of air voids in compacted samples (% v/v) according to EN 12697-8

• Properties of the binder of RAP (only in case of recycled asphalt in an SMA paving mixture) according to EN 12697-4 and EN 12697-3

• Penetration at 25°C according to EN 1426

• Ring and ball (R&B) softening point according to EN 1427

Sampling and testing additional physical properties should be carried out at a frequency selected from the three levels A, B, and C:

Level A—testing every 10,000 tonnes Level B—testing every 5000 tonnes Level C—testing every 3000 tonnes

Selection criteria of the levels have not been determined in the standard. Test results should be stored in FPC files.

Some final remarks follow:

• Sampling of a mixture for additional tests should be conducted in accor­dance with EN 12697-27.

Подпись: European Standards Concerning SMA 275

TABLE 14.3

The Comparison of Specification on SMA 10 or SMA 11 according to the rules of EN 13108-5
















Bitumenbundna lager. 2008:113

Nawierzchnie Asfaltowe 2008





SMA 11 S

SMA 11

ABS 11

SMA 11

SMA 11


Thickness of layer, mm







Method of preparing samples





Impact (Marshall)

Impact (Marshall)





Energy of compaction (strokes/side)


50 50 Gradation (% Mass of Passing by Sieve)



















































The Comparison of Specification on SMA 10 or SMA 11 according to the Rules of EN 13108-5
















Minimum binder content, Bmin




for 50/70 and 70/100




Minimum void content, Vmin







Maximum void content, Vmax







Minimum void filled with binder,








Maximum void filled with binder,







Permitted binder drain-off







Resistance to abrasion by studded tires







Water sensitivity ITSR, %








Maximum proportional rut depth P







Maximum proportional rut depth, testing in air (PRDair)







Maximum wheel tracking slope, testing in air (WTSair)







Resistance to Permanent Deformation

Подпись: 276 Stone Matrix Asphalt: Theory and PracticeSource: Data from EN 13108-5. Bituminous Mixtures—Material specifications—Part 5: Stone Mastic Asphalt. Note: NR = no requirement; SMA = stone matrix asphalt.

• Samples should be prepared with the same method as applied for qualifica­tion tests of a mixture according to a formula (EN 13108-20); careful atten­tion should be paid to using the same method of compacting samples.


For those interested in seeing how the CEN-member states have specified their national requirements for SMA, such a comparison is presented in Table 14.3.

Control of Production Variability

The standard EN 13108-21 contains a second element of FPC—namely, the control of production variability. This control is applied through the determination of the running mean deviation from the target and is described in Item A.5 of the standard. It involves monitoring systematic trends at the production stages of asphalt mixtures to prevent permanent one-sided deviations. The following parameters of asphalt mixtures are under such supervision:

• Mass percentage of material passing through the sieve D

• Mass percentage of material passing through the sieve D/2 or the charac­teristic coarse sieve

• Mass percentage of material passing through the 2 mm sieve

• Mass percentage of material passing through the 0.063 mm sieve

• Soluble binder content

Calculations of the mean value of deviations from the target should be conducted for each of these properties. The mean deviations calculated on an ongoing basis should be compared with admissible values given in Table A.5 of the standard (Table 14.2.).

Running mean analyses are undertaken for the following groups of mixes (clause A.2 of standard):

• Fine grained (D less than 16 mm)

• Coarse-graded (D greater than or equal to 16 mm)

• Mastic asphalt (gussasphalt) and hot rolled asphalt

The mean of the latest 32 analyses should be calculated for each of these groups. By and large, mean deviations exceeding the appropriate values in Table 14.2 indicate a group of nonconforming products (Item 7.4 of the standard applies to them). In these circumstances, suitable corrective measures should be taken; the OCL should be reduced by one level as long as the mean deviation remains outside the permissible range.

Determination of Compliance of a Tested Sample with a Recipe

After extraction, the result of each sample is compared with admissible deviations on test sieves and soluble binder content. A mix may be regarded as compatible with a recipe when all its components are within admissible tolerances.

Table 14.2 depicts an excerpt of Table A.1 of Annex A of the standard demon­strating admissible deviations from a target appropriate for producing SMA mixes with D less than 16mm (e. g., the group of small aggregate mixtures using the single result method). Using admissible deviations, the accuracy of measuring methods has already been taken into account. reaching the ocL Assessment

After assessing the mixture extraction results (the set of 32 results), one can deter­mine the OCL level. According to the standard, there are three OCL level—A, B, and C. They can be roughly interpreted as follows:

• OCL A proves a high rate of compliance with the target for manufactured mixtures.

• OCL B is an intermediate assessment and is still good.

• OCL C is a warning assessment for asphalt plant personnel (and their customers) that the composition of produced mixtures fluctuates considerably.

The OCL assessment is established after 32 consecutive analyses, but it should be kept in mind that they do not have to be samples of only one mixture type; the standard stipulates taking into consideration 32 consecutive results of any produced mixture. The selection of an appropriate OCL then follows, considering the total number of samples that do not conform with requirements

Number of



2 or less

Level A


Level B

7 or more

Level C

In the event that more than 8 of the latest 32 results are nonconforming, the plant’s equipment and procedures should be subject to an immediate and thorough review.

The OCL assessment is a dynamic process because the latest 32 results are always used for calculations. After startup at the beginning of the construction sea­son, the test frequency at the starting level should correspond to the lowest OCL-C of a designated level of requirements (X, Y, or Z). Sampling frequency is subject to change weekly based on the lowest OCL achieved during the previous calendar week.

Minimum frequencies of testing samples in relation to the achieved OCL from the previous week are valid for the next calendar week. The following are OCL sample testing frequencies for the single result method:

OCL-A Level X: 600 tons Level Y: 1000 tons Level Z: 2000 tons

OCL-B Level X: 300 tons Level Y: 500 tons Level Z: 1000 tons

OCL-C Level X: 150 tons Level Y: 250 tons Level Z: 500 tons

Interestingly, the better the OCL, the lower the test frequency. These data also show that for each category (X, Y, or Z), when the OCL advances by one level, the test frequency decreases by half (or twice as much material can be produced between tests).

The startup of a new plant means an operational startup with the OCL-C assessment. A shutdown of longer than 3 months during the season or a major repair of the asphalt plant results in the OCL being reduced by one level from where it was before the stoppage. The OCL level cannot be changed until 32 results have been obtained and the first new OCL assessment is available. In the case of an asphalt plant with a low rate of production, the standard stipulates testing the composition of a mix at least every 5 operating days. Mobile coating plants, after relocation, are treated like plants shut down for 3 months or restarted after a major repair.

Analysis of Mixture Composition

Each sample is subjected to extraction testing, resulting in a gradation of the aggre­gate mix and a content of soluble binder. The sieving operation should be conducted using a set of sieves as in Table 14.2. The standard stipulates testing the aggregate gradation through five sieves (plus possibly a sieve for oversize particles 1.4D). The small number of test sieves speeds up the control analysis but at the same time increases the responsibility of the mix producer to achieve other final SMA proper­ties determined by a recipe. It appears that the proper selection of optional sieves for the coarse aggregates that control the skeleton is of great significance. selection of a method for Estimation of Extraction Results

Prior to the comparison of extraction results with admissible deviations, the method for assessment of extraction results should be selected. The choice should be made between the following two methods:

• Single result method—test results of individual samples are treated inde­pendently and assessed in comparison with admissible deviations from a target in accordance with Table A.1 of the standard; the sets of the previous 32 results form the basis for the OCL assessment (classified as conforming or nonconforming).

• Mean of four results method—test results of individual samples are treated in groups of four analyses of the same mixture, and the mean result of each group is compared with admissible deviations from a target (classified as conforming or nonconforming); after a comparison of results of the eight mean results (e. g., eight groups, each with four results for a total of 32), they are classified as conforming or nonconforming and form the basis for the OCL assessment.

Thus in the single result method, we assess 32 individual results; in the mean of four results method, we assess eight results (means) calculated for groups of four single results.

TABLE 14.2

Comments on Requirements for Permitted Deviations from the Target for Mixtures Produced Using EN 13108-21, for Small Aggregate Mixtures with D <16mm—the Single Result Method

Small Aggregate Mixes (D <16 mm)

Percentage Passing

(Single Result Method for

(Control of Production

(% m/m)

OCL Calculation)



1.4 D

The sieve 1.4D—additional sieve on which requirements for passing 98-100% have been established




The sieve determining maximum particle size in a mix (without oversize particles), (e. g., the record SMA 11 means D = 11.2 mm [at the same time the boundary sieve for oversize particles])

D/2 or characteristic coarse sieve



The sieve indicated in an NAD for the standard EN 13108-5 for each SMA mixture

2.0 mm



Characteristic fine sieve



The characteristic sieve has not been determined in EN 13108-21; it should be established in an NAD for EN 13108-5 for each SMA mixture; the choice has been limited to one of the sieves: 0.125, 0.25, 0.5, or 1.0 mm

0.063 mm



Sieve of filler fraction

Soluble binder content



Soluble binder content with extraction after EN 12697-1

Individual Samples

Tolerance about Target Permitted Mean Deviation Composition from Target

Подпись: European Standards Concerning SMASource: Modified from EN 13108-21. Bituminous Mixtures—Material specifications—Part 21: Factory Production Control. Note: NAD = National application document; SMA = stone matrix asphalt.

These two methods may not be applied in parallel. In other words, only one method can be used at the same time in the same asphalt plant.

Analyses of deviations using only the single result method will be discussed later.

Control of Compliance with an SMA Mix Recipe determination of the Operating compliance Level

The principle of fixing frequencies of control testing depends on the number of deviations found and has been adopted in the standard EN 13108-21. With an established level of requirements (X, Y, or Z), the frequency of sampling is variable, depending on the production accuracy of the asphalt plant (i. e., the number of devia­tions from a recipe). Start in Control

The initial frequency of sampling depends on an established category of conformity (X, Y, or Z), so at first the frequency of sampling is as follows:

• Every 150 tons of a manufactured mixture (for level X)

• Every 250 tons of a manufactured mixture (for level Y)

• Every 500 tons of a manufactured mixture (for level Z)

Samples should be representative of the entire production; the relevant clauses of EN 12697-27 and EN 12697-28 should be taken into account.

G eneral Guidelines for a Control System of Production Homogeneity

The guidelines for controlling production homogeneity are made up of the following two parts:

• Control of compliance with a recipe, accomplished by establishing an oper­ation compliance level (OCL), which is a variable index that estimates the production accuracy and indicates the appropriate mixture test frequencies

• Control of production variability through determining a running mean of the deviation from target (i. e., from a recipe).


In the European system of standardization of asphalt mixtures according to the series of standards EN 13108, the issues of control and quality assurance have been deliberated in EN 13108-21:2006 Bituminous Mixtures—Material Specifications— Part 21: Factory Production Control (with Amendment AC/September 2008). This standard contains an outline of requirements regarding the system of quality assur­ance during the production process of a mix. Generally speaking, the FPC is in conformity with principles from ISO 9001, so that part of the standard will not be discussed here. Requirements for the production control can be found in the nor­mative Annex A, entitled “Tolerances and Test Frequencies for Finished Asphalt,” which stipulates admissible deviations and frequencies of controlling manufactured and delivered mixes.

14.8.1 Levels of Requirements

EN 13108-21 provides three levels of requirements related to the expected accuracy of production—level X, level Y, and level Z.

Level Z is a basic one and its application secures the fulfillment of the requirements connected with the evaluation of conformity. Levels X and Y require assurance of a higher frequency of control testing and can be adopted when such increased control is required. The determination of different levels of requirements for various types of mixes or specified contracts is also possible (e. g., a higher level for bridge deck surfacing). In these circumstances, the selection of an appropriate level (X, Y, or Z) indicates the significance of the minimum risk of nonconformity. Consequently, the frequency of testing production samples depends on the accepted level of require­ments (i. e., the desired level of conformity).


Asphalt mixtures are construction products and the standards EN 13108-1 to 13108-7 are harmonized with the Construction Products Directive 89/106/EEC. The system “2+” has been adopted to evaluate conformity of asphalt mixtures. It consists of the following:

• Initial type testing of each mix produced by its manufacturer

• Certification of a Factory Production Control (FPC) with reference to EN 13108-21 conducted by a notified body, separate for each production site (asphalt plant)

The SMA design process (recipe) should be followed by confirmation that the mixture meets all requirements shown in the relevant categories listed in the NAD; if it does meet the requirements, it cannot enter the market. This set of tests, called initial type testing, is based on requirements of the standard EN 13108-20 (Table B.5), which include the following:

• Binder content EN 12697-1 and 12697-39

• Grading EN 12697-2

• Void content including VFB EN 12697-8

• Void content of gyratory compacted samples EN 12697-31

• Binder drainage EN 12697-18

• Water sensitivity EN 12697-12

• Resistance to abrasion by studded tires EN 12697-16

• Resistance to permanent deformation EN 12697-22

• Resistance to fuel (airfields) EN 12697-43

• Resistance to deicing fluids (airfields) EN 12697-41

Routine (daily) testing of a manufactured mix is conducted using a system of FPC according to the standard EN 13108-21.