The thin black line

A precis of a presentation by Clare Dring, national product manager at Fulton Hogan, on asphalt thickness and are we paving layers too thinly?

At the introduction of her presentation, Clare Dring said she had recently read an article by Rebecca McDaniel that followed on from research done in the United States back in 2004, detailing what should be our minimum thickness size for asphalt paving. In addition, recent conversations with industry raised the same question.

“This research concluded fine grained mixes should have a minimum layer thickness of three times the nominal size. That would be equivalent to our DG (Dense Graded) mixes.

“And then for coarse grain mixes four times, which is equivalent to our AC mixes.”

Before going into more detail, Clare touched on ‘compaction’, and why the industry agrees it is the single most important factor when it comes to asphalt performance.

“There are a number of reasons why voids, surface texture, densification, permeability and strength characteristics all ultimately lead to well-performing asphalt.

“So, why is it going wrong?

“I think it is to do with design methodology changes from Marshall to gyratory compaction in the lab. Marshall mix design gave binder-rich, finer mixes that were prone to rutting so the US, NZ and other countries changed to gyratory compaction which gives us more rut-resistant mixes. But less binder and less fines can result in slightly higher interconnected voids if the asphalt layer is placed too thinly.

“This is why it is important that we focus on those layer thicknesses and make sure they’re at the depths we need them.”

Clare says confusion comes from the change in the naming convention of asphalt mixes.  In the past maximum stone sizes were used to name mixes but now nominal stone sizes are used.

This is important because the size of the mix is used to determine the minimum thickness required, however the NZTA M/10 specification has always stipulated the minimum thickness.

“If we go back and have a look at where the New Zealand industry has come from, we can see all the way back in 2005 when we were designing mixes using Marshall design methodologies a factor of 2.5 was applied to that mix size to get the minimum layer thickness.

“When we started moving to gyratory mix design, the factor changed to 2.5 for the nominal stone size for the DG mixes and 3 times for the AC mixes.

“Currently where we sit with M/10 2014 we have 3 for DG mixes and -3.5 for AC mixes. But M10 is currently under a review and layer thickness is intended to increase again. It is expected to increase to 3.5-4 times the nominal stone size, which aligns very similarly to the research that was done in the States in 2004.

“We may be a couple of years behind but we’re starting to recognise the need for it here in New Zealand.”

When asphalt is paved too thinly then, generally, the layer cools down too quickly and adequate compaction isn’t achieved, says Clare.

“There is the risk of having stone on stone interaction that inhibits further densification through rolling.

“And you’re left with a poor surface texture, lower densities and high air voids that allows for an increase of water and results in ravelling and premature aging.

We all have skin in the game, there will be a cost, whether economic or reputational.

“When pavements are laid too thickly the asphalt holds its temperature a lot longer and segregation of the mix can occur. In addition, the bottom of the asphalt layer struggles to meet the compaction requirements.

“This can result in non-uniform compaction throughout the mix and typically, after opening to traffic, rutting and shoving occurs.”

Clare provided two examples. One where an AC14 was specified to go down at 35mm. The industry standard at the time was 50mm, representing a 30 percent decrease in layer thickness than what was accepted as best practice.

“This resulted in cracking and water ingress that would reduce the life of that pavement by three to five years.

“And we’re starting to see a lot of this out there where specifications require us to pave the asphalt thinner than best practice.”

Clare illustrated another example where the project has been paved too thickly.

“This is an SMA (stone mastic asphalt) that went down at 65mms and should have been 45mms. Again, that represented a 30 percent increase in what should have been paved.

“Soon after being exposed to traffic there was severe rutting and up to 30 percent deformation.”

Fulton Hogan did a high-level analysis (analysing data from 2014 to now) to try and understand what is actually out there on our network in terms of what was compliant and non-compliant based on the M10 specification.

When analysing the data is was apparent that up to 30 percent of the asphalt currently on the network is paved thinner than the minimum set out in M/10 2014 and was at high risk of premature failure.

When using the minimum layer thickness as proposed in the M/10 2019 as much as 74 percent of the network’s pavement layers have been paved too thinly, indicating a high level of risk and potential for premature failure.

“We haven’t gone into the detail to look at exactly who is driving that risk, whether it’s contractor or client specification, but we are seeing a lot more specifications requiring less than industry standard thickness.”

International industry best practice specifies, or at least suggests, the minimum layer thickness should be three to four times the nominal aggregate size of the asphalt mix says Clare.

“Currently, in New Zealand, it is three to three and a half depending on the mix and there’s the anticipation that we could move to 3.5-4 times the layer thickness.

“But when we look at our network based on those numbers, and what the pavement depths are out there, there is the potential of 30-74 percent of our pavements do experience premature aging because the asphalt layers are too thin.

“So, I guess the challenge to us as the industry? We all have skin in the game, there will be a cost, whether economic or reputational.”