Pavement Engineering

Moisture in pavements
DCP analysis and design chart
Simple pavement design chart

Test methods
Bitumen - Australian grades
 - key temperatures

(of cement stabilised layers)

Gravel quality analysis for 
unsealed pavements
Material Test methods 
Asphalt stripping/moisture

Maintenance rolling
of seals 

Reseal and resurfacing selection chart
Seal design: comparison of Australian,
 NZ, South African and UK methods

Surfacing failure - hints for an investigation

Troubleshooting charts
for seals, asphalt, slurries

Austroads guides

Colas bitumen and asphalt 
products and services manual

Metric-Imperial conversions 
Pavement Management School
Shell Bitumen Test Data Chart
World map of Thornthwaite's Climatic Index

World map of Koppen's Climatic Index




The pavement moisture regime reaches an equilibrium under the centre of a sealed road (or runway) over a deep water table. This can be predicted, and my first PhD was on that topic. It is often important to be able to judge how "wet" the layers of a pavement are, and these prediction methods can be used to do that. I have summarised the data and knowledge into one page on moisture. 


I only use this as a very rudimentary rule-of-thumb check, and it is not substitute for proper pavement design. However, from AUSTROADS (2001), the pavement thickness design chart for thin bituminous surfacings is here WARNING - this does not analyse for specific basecourse or subbase performance and the road can also fail in those layers.


DCP chart   (and here is the DCP test method)

Dynamic Cone Penetrometer. DCP.
Great engineer's tool.  A J Scala (1962) got it up and running for subgrades in Australia. Then Burrows and Ed Kleyn (1983) extended its use to all pavement layers in the old Transvaal Provincial Administration.  Dr Frank Netterberg turned it into a serious scientific and research tool. I was involved in redesigning it for improved reliability in 1983.  An unnamed US Corps Engineer took the design drawings back to Vicksburg in 1984 (I gave my set to him personally). And now, in the 21st century, even America has discovered it (complete with the South African drawings) and are promoting it an excellent tool. They even left in the dumb bits of my redesign.

The DCP results can be used to divide a road or runway into uniform sections, get design subgrade strengths and thus feed into FAA design charts, or estimate elastic moduli and feed into the various (APSDS, LED) computer elastic layer design programmes for runway thickness. It feeds into the various road design catalogues (TRH 4, AustRoads). In the field, it can be an important tool to investigate failures and help find their cause. IT IS VIRTUALLY THE ONLY TOOL THAT FINDS THIN WEAK INTERLAYERS such as partially soaked layers or clay lenses - in basecourse, subbase, or subgrade (unless you count Prof Emile Horak's layer analysis of FWD results). It is also very useful for finding problems inside the basecourse, especially at the top where the base may be crushing or too weak to resist seal aggregate penetration and you have seal bleeding. The DCP can even be used to help understand FWD test results.

Above all, the DCP is marvellous for sorting out the confusion that arises when all the various test results are put together. Usually when a pavement has been evaluated, there is so much data that it doesn't make sense. One test shows a section is about to fail, while another test shows it is the strongest section. The DCP is the "honest" tool to help you sort out what is what. Warning - only one group have got a successful way of analysing the results in software - Fritz, Fenella and the team at Modsys (the Rubicon Toolbox software) since most other algorithms get confused by non-standard conditions (which of course is what you're using the DCP to look out for). If you haven't got Rubicon, an hour or two of pencil and ruler and charts plus sweat labour gets most projects analysed. 

I've written up a guide for competent geotechs on how it can used - click here for that Word file, and here for the DCP chart to plot field results and get CBRs (and with the DCP-CBR correlation table). And here to see what a completed chart looks like. WARNING this chart is for the DCP model (fall, weight, cone) as noted on the chart - check yours is the same; and remember that field DCP and CBR are affected (sometimes greatly) by moisture; and that the heterogeneity of natural materials means that there is an imprecision about all this. (here is the DCP test method).

I've got the Rubicon Toolbox software (and if you think I'm pretty pleased with it, well yes, I am), so here are a couple of examples of the output and what they show:

This is an old sealed runway in Australia, gravel basecourse, designed in the DC-3 days. The DCP shows the pavement is weak - make that extremely weak - and its structural capacity is almost nothing. The red bars on the chart show where the pavement has inadequate strength, and the green bars show adequate strength. This is a good example of a challenge to upgrade it. If you simply try and overlay, even with thick asphalt, you'll still have all the rubbish below. I've seen someone do this, adding 160mm asphalt, for a Dash-8 design aircraft. Rather a lot of money for not much of a result. The better option is to dig the whole lot out, and start over. And of course the DCP has now given us an idea of what the design subgrade CBR might be. Click here for that DCP result 1653  (hint: some browsers load this file as 'fit to screen', and it is a bit bigger than that, so click the picture or whatever to show it full size).
Since the better option is to dig the whole lot out, that's what we did. Dug out 0-300mm to spoil, then mixed-in-place 1% cement in the 300-600mm band and compacted. Then brought back 300mm of a good blend of crushed rock/gravel as a basecourse and mixed-in-place 1% cement (for rapid drying) and sealed over it. The DCP test on the rebuilt pavement was started at a depth of 150mm because the top 150mm was just too strong to penetrate (which is sort of CBR > 180, UCS > 1500 MPa), so we drilled out 150mm and started with the DCP.  That's a pretty good pavement, and suited for a Boeing 737-800. Click here for the rebuilt runway DCP result 1460

Here is a different runway, with an old gravel runway in South Africa upgraded with a new gravel basecourse and seal, designed for 100-passenger jets. The seal was bleeding badly. The pavement is not that weak, the basecourse density testing was fine, the base gravel grading was fine and the plasticity was PI < 6. The unsoaked CBR test in the lab was CBR 85. But the DCP and the red bars show the story - the top of the basecourse has inadequate strength, and the stones punched into the base and thus the seal bled. Click here for that DCP result 2300  

And finally here is a real gem in Transvaal - cement stabilised basecourse on a new highway. Sealed surface. The highway goes through hilly terrain in a moderately wet area that is known to give problems with sub-surface water. By unknown means, the constructor was able to omit the sub-surface drainage to save money. Within a year, very shallow pothole failures were occurring in the wet season. The asphalt repair patches also failed in the following wet season, with crocodile cracking in the wheeltracks (along with more seal failures). FWD testing showed the road to be strong. "Water" is the obvious thought. But the failures were occurring in both cuts and fills, and some were occurring in places where it was impossible for sub-surface water to reach. The basecourse density and materials testing at construction showed it was fine, and the plasticity was PI < 4. The basecourse had a UCS of 1500-2500 MPa. So why were the potholes occurring? Now the DCP is not really suited to penetrating cement stabilised materials, and it broke a few times as we tried (as expected), but a welder worked wonders to fix it each time. The DCP and the red bars show the story - the very top of the basecourse has inadequate strength, as it is crushing under traffic (ref. the research by Morris de Beer on crushing of lightly cemented layers). This weak layer is only the top few millimetres, and the DCP just got a glimpse of it, but that was enough for the engineers to go and carefully open up some shallow test-holes personally, and the weak loose material was plain to see and easy to scoop into a pile with one's fingertips.  Click here for that DCP result 1415  and then here for the photo 151 


Asphalt stripping/moisture
It's an old old problem, and it still  is around - stripping and damage due to moisture. The issue is that pavement surfacings  have failed because an underlying asphalt layer has stripped and turned to mush. Technically it is called the separation and removal of asphalt binder from aggregate surface due primarily to the action of
moisture and/or moisture vapour. 

In the identification of the cause of stripping, practitioners have, sometimes, tended to focus their attention on the sensitivity of the aggregate and asphalt system in the presence of moisture. Other practitioners (Dolf) try to blame the bitumen or the contractor or anyone but them. Well, now there is more science to throw at it. There is a good paper PREMATURE FAILURE OF ASPHALT OVERLAYS FROM STRIPPING: CASE HISTORIES by Kandhal (NCAT expert) and Ian Rickards (recently with PRS, now a consultant, and of HiPAVE and APSDS4 fame). The paper was presented at Association of Asphalt Paving Technologists, Florida in 2001. Go read it to learn more about the topic.

My quick checklist to detect the possibility of stripping

Symptomatically, if top layer = seal

or if top layer = asphalt

The case histories in the K&R paper document the effect of pavement saturation. The authors suggest that under saturated conditions all asphalt mixes may fail as a consequence of cyclical hydraulic stress physically scouring the asphalt binder from the aggregate. The authors classify this stripping as a mechanical failure of the asphalt pavement system, and the classical moisture sensitivity tests are irrelevant. While under saturated conditions a less moisture sensitive asphalt system may survive longer, it is probable that failure is deferred and not avoided. 

In one investigation I did [urban road, busy with much truck traffic, adjacent lawn watering so water was going onto the road surface even in the dry season, construction was asphalt on crushed rock base on sand subgrade], we found the presence of large voids in the pavement between the base and the asphalt due to a poor primer seal in-between. This focussed our attention to the capillary suction mechanism, diffusion of moisture through the asphalt and opened up another way of assessing this sort of failure (or is it another manifestation of the problem?). It needs (a) water, (b) permeability, AND (c) voids large enough to defeat the suction within the asphalt due to the moisture differential (wet bottom and dry top), and of course (d) traffic. 

Maintenance rolling
of bituminous seals (chip seals) on airport pavements

A newly constructed bituminous seal or reseal needs to be rolled to embed the stone into the bitumen, and to ‘work’ the bitumen around the stone. On the highway, 20% of the necessary rolling is done at the time of construction (or less if you listen to the contractor's foreman), and the remaining 80% is provided by traffic. On runways and taxiways, the traffic is much less than on highways, and much more rolling should be done at construction (although often it is not). The remainder of the rolling should be provided by maintenance rolling during the first few years of life of the seal, because sure-as-eggs the aircraft traffic isn't going do it. Click here for the Word file which discusses this.

Seal design: comparison of Australian, New Zealand, South African and UK methods

Certain countries lead the world in the use of sprayed seals (chip and spray). At the ARRB First International 2008 Sprayed Sealing Conference in Adelaide, there was a session on national practice with highly respected presenters from each of four countries. For research purposes only, their papers and presentations are here:

Mr Walter Holtrop, AAPA, Australia    paper  presentation
Mr Barry Gundersen, Gundersen Consulting Ltd, New Zealand    paper  presentation
Dr Cliff Nicholls, TRL, UK    paper  presentation
Mr Trevor Distin, Sabita, South Africa     paper  presentation  

The 2nd ARRB Sprayed Sealing conference was held in 2010, and hopefully there will be a 3rd.

Reseal and resurfacing selection chart  
This is a comprehensive chart that myself, Tom van Rijckevorsel [technical and practical aspects of asphalts and slurries], Gerrie van Zyl aka Crokodil Dundee [seals, choice of surfacings, low volume roads design] and Christo Olwagen [development and production] put together for Colas Southern Africa. The idea of such a chart came from those charts in the old Corps of Engineers publications. We've got all the newer surfacing types (such as Novachip from Jean-Claude Roffe at Screg/Colas), and then, for the first time ever in the world, we were able to combine seals, asphalts and slurries in one chart. Click here for that chart.



GRAVEL QUALITY ANALYSIS FOR UNSEALED ROADS From the work by Phil Paige-Green of CSIR and published in many places (including TRH20), the gravel quality for unsealed roads can be assessed by simple lab tests. These are used to calculate the gravel material's shrinkage product and grading co-efficient. These are plotted on a simple matrix, and the suitability of the material found. It then indicates the type of distress mechanism, such as erodes, ravels, corrugates, dusty, or slippery. A neat Excel spreadsheet from Darren Shepherd's paper to calculate and plot this is found here. Phil has been awarded the J D Roberts Award for outstanding research & innovation for his work in this field.

The FWD and the various software programmes (many exist such as Rubicon, Elmod, Elsym5, CIRCLY, and Prof Joe Mahony's free Washington State elastic layer analysis software) will analyse the whole pavement and enable you to perform a design. But for the "back-of-the-cigarette-packet" calculations, this is pretty handy: from AUSTROADS pavement design guide (2001), the transfer function for deflection (D0) to design traffic is shown here.  The Surface curvature index has largely dropped away for thinner surfacings with the 2004 edition of the Austroads (thank goodness).

Test methods  
Some of the South African test methods are online (well done Barry). 
The ones online are in bold - click for them here  TMH methods on-line  

TMH1: Standard methods of testing road construction materials 1986
TMH2: National standard for the spraying performance of binder distributors 1979
TMH3: Traffic axle load surveys for pavement design 1988
TMH5: Sampling methods for road construction materials 1981
TMH6: Special methods for testing roads
(here is the DCP test method)
TMH7: Code of practice for the design of highway bridges and culverts in SA, 3 Parts 1989
TMH9: Pavement management systems: standard visual assessment manual for flexible pavements 1992
TMH10: Manual for the completion of as-build materials data sheets 1993
TMH12: Pavement Management Systems: Standard Visual Assessment Manual for Unsealed Roads 2000

Bitumen - Australian grade- temperatures  
Key temperatures including softening point, rolling temperatures, mixing temperatures. For all Australian grades, multigrades, and PMBs. Click here for that extract from the very useful "AAPA Code of Practice: Manufacture, Storage and Handling of Polymer Modified Binders, 1st edition, June 2004. Melbourne, Victoria."

And when you wanted to know what was in each PMB grade:


Troubleshooting charts
These are a series of troubleshooting charts which were developed when I was at Colas Southern Africa for all types of bitumen and bituminous surfacings. 

Fogspray (also called dilute emulsion)
Polymer modified bitumen
Seals (also called chip and spray)

Colas Products and Services CD and manual
The CD (which is the electronic version of the manual) is intended to provide a quick source of reference to the roads practitioner. The manual was written by myself, Tom, Gerrie, and Christo, with inputs from the old masters Adrian Bergh and Les Davidson, and colleagues at Colas including (in alphabetical order) Philip Cronje, Trevor Distin, Rob Franks, Errol Kotze, Jeff Newell, Johan O’Connell, John Onraët, and Toby Reeder.  The content is based on published materials, internal Colas/Coltech materials, and my own courses at the Universities of Stellenbosch and Witwatersrand, SARF and Sabita. Electronic copies are available from Colas in South Africa. Their website is

It is written for the South African industry, and so uses their jargon and bitumen grades, but 99% is applicable internationally. A conversion for bitumen grades is given in Table 3 below. Some are penetration based grading systems and some are viscosity based. Because the viscosity test is fundamental and the penetration test is empirical, there is no direct relationship between them, and the relationship is approximate.

Australian and South African bitumen specifications

The South Africa bitumen specification is SABS 307 penetration grade, from which an extract is shown in Table 1.

Table 1 : Extract from SABS 307 penetration grade bitumen specification


Test method





Penetration @25oC, 0.1mm






Softening point oC






Viscosity @60 oC, Pa.s.

ASTM D4402





Viscosity @135 oC, Pa.s.

ASTM D4402





The Australia bitumen specification is AS 2008, from which an extract is shown in Table 2 (and see Bit_temp for more properties of these bitumens). Note that the 2013 update to the specification has dropped Class 50, and added Class 240 and 450 (and the multigrades M500 and M1000).

Table 2 : Extract from Australian AS 2008 bitumen specification


Test method

Class 170

Class 240

Class 320

Class 450

Class 600

Viscosity @60ºC, Pa.s

AS 2341.2, .3, .4






Viscosity @135ºC, Pa.s

AS 2341.2, .3, .4




max 0.70


Penetration @25ºC, 5s (0.1 mm or pu)

AS 2341.12

62 min

53 min

40 min


20 min

There is an approximate functional equivalence between South African and Australian bitumen grades, which is included here to aid interpretation of results (Table 3), although this equivalence does not hold at the specification level. And the temperature susceptibility (PI) of the pen graded bitumens is usually different to that of viscosity graded bitumens.

Table 3 : Approximate functional equivalence of bitumen grades


South Africa (pen grade)

South Africa (old viscosity grades)


Hot climate, heavy duty AC hotmix



Class 600, 1000/320 
or use a PMB

Heavy duty AC hotmix



Class 320, 450

AC hotmix or Seal



Class 170 or 320




Class 170 with cutter




Class 170 with cutter

Of course, if you want a stiffer grade of bitumen, here is the backyard chart to blend in some gilsonite.

Pavement Management School
The (world's first and longest running) Pavement Management School, offered annually in South Africa.  I started the School years back with Gerrie van Zyl, Andre van der Gryp, and Prof Hugo. It still runs under Kim Jenkins and Gerrie, and it covers
the latest developments in PMS world wide. Contact Prof Kim Jenkins at Stellenbosch University. They use a hands-on learning experience whereby the candidates learn it in class in the morning, go out and do it on the road in the afternoon, then come back and run it in the PMS system at night. Hard work, great learning, and the wind-up party is tops. 

The School  includes the following:
- classify a surfaced road network,
- make visual assessments in the field,
- determine appropriate condition indices and determine remaining life,
- extend this to unsealed road networks
- do rehabilitation analysis at the network level,
- develop maintenance and rehabilitation alternatives,
- gain experience regarding the latest methodologies (Expert Systems, Neural Network),
- use HDM4 and other locally developed systems to prioritise and to optimise maintenance programmes
   for varying budget levels,
- gain insight into the use of GPS and GIS, as tools of PMS and infrastructure management,
- make presentations at management and politician level

If cement or lime-stabilised materials are exposed to air, the hydration products may react with carbon dioxide thereby reducing the strength of the material by an average of 40 per cent of the unconfined compressive strength (Paige-Green et al, 1990). That report is one of the definitive reports on carbonation and is here and it is substantially better than the later (and now thought to be inaccurate) Botha report. Phil Paige-Green presented a later paper on "The Durability of Stabilised Materials", which is easy to read and gives an excellent picture of stabilisation (including durability and carbonation and crushing; this will become a seminal paper in pavement engineering). Carbonation is associated with a decrease in the pH of the material from more than 12 to about 8.5. The form of this distress presents itself after several weeks, usually in the form of surface disintegration of the primed or unprimed layer. 

The literature notes several risk factors for carbonation: 

  1. Not sealing the road as soon as possible after stabilising (a prime is considered highly permeable and not a seal).
  2. Stabilisation at low cement contents.
  3. Wet/dry cycles in the curing process.
  4. Higher permeability material.
  5. Over-compaction leading to micro-cracking and increased permeability.
  6. A fine grained low plasticity raw material (which in an uncemented state would be considered unsuitable as basecourse).

The testing protocol to find carbonation has already been established by Dr Frank Netterberg  Netterberg's carbonation test


setstats 1