Why asphalt cores fail density, and what to do about it

Most failed cores are a temperature problem wearing a different costume.

The causes, roughly in order of frequency

Cold rolling. Breakdown rolling started late, or rolling continued below the cessation temperature. This is the biggest single cause by some distance.
Not enough compactive effort. Too few passes, or a roller train missing the heavy pneumatic-tyred roller.
Lift thickness wrong for the mix. Thinner than about three times the nominal maximum aggregate size and the stone locks up instead of moving; too thick for the roller weight and the bottom half never gets compacted.
Segregation, mechanical from paver operation and wing dumping, or thermal from uninsulated trucks.
Paver settings and speed. Running too fast, or tamper bars and screed vibration set low, so the mat comes out of the screed at low initial density.
A soft layer underneath. A yielding base or an uncured prime lets the whole mat deflect instead of resisting the roller.
Mix variation from the plant. Low binder content or off-JMF filler and sand ratios.

Reading the core

The core itself will usually tell you which of those it was.

Open and honeycombed at the bottom, tight at the top: the base chilled the underside, or the compaction energy never reached full lift depth.
Porous all the way through: overall cold rolling, under-rolling, or low binder content in the mix.
Coarse aggregate clustered with no mortar around it: segregation — paver operation or end-of-load dumping.
Crushed aggregate faces or micro-cracking at the top: rolling a mat that was already too cold, or over-vibrating a thin rigid lift.
Sheared, oily or stripped bottom face: paved over an uncured tack coat or a wet base. The mat slid instead of compacting.

Measure the core height with calipers at four points while you are at it. If the height is under three times the NMAS — under about 38 mm for a 12.5 mm wearing course — the aggregate had nowhere to go and no amount of rolling was going to reach target.

Temperature is the whole game

Temperature controls binder viscosity, and binder viscosity is what lets the aggregate reorient and expel air. For standard 60/70 or 50/70 penetration bitumen:

Delivery to the paver hopper: 150 to 165 °C
Breakdown rolling: 135 to 150 °C, immediately behind the screed
Intermediate rolling: 110 to 135 °C, pneumatic or heavy vibratory
Finish rolling: 85 to 110 °C, static steel
Cessation: 80 to 85 °C — below this, stop

Below the cessation temperature the binder has stopped behaving as a lubricant. Further rolling adds no density at all; it bridges high spots, fractures aggregate bonds and puts hairline checking into the surface. Rolling a cold mat harder is the most common wrong answer to a density problem.

Roller pattern, weight and passes

Breakdown: 10 to 14 tonne vibratory double drum. High amplitude and low frequency on thick lifts; low amplitude and high frequency on thin ones.
Intermediate: 18 to 24 tonne pneumatic-tyred. The kneading action of rubber tyres is what seals surface voids and works fines into the matrix without crushing stone. If your densities are marginal and you are not running a PTR, that is where to look first.
Finish: 8 to 10 tonne static steel, purely to remove marks.

Establish the pattern on a trial strip — something like two breakdown passes, four to six intermediate, two static finish — and then hold it. Passes beyond the density peak give nothing back, and on cooling mix they shatter aggregate and lose density. Overlap passes by 150 to 200 mm, and reverse smoothly at a slight angle; abrupt reversals on hot mix leave shoving and low-density depressions.

Segregation, moisture and soft support

Segregation shows as isolated pockets or linear bands of coarse rock with no fine mortar. Those cores come back with very low Gmb and air voids over 8% to 10% no matter how many passes the roller made. Thermal segregation is the same problem in a different form: cold streaks in the mat go below compaction temperature before the breakdown roller reaches them.

Moisture — trapped in aggregate at the plant, or rain on the base — flashes to steam under hot mix. The mat bubbles and pushes under the drum instead of consolidating, and cores show stripped binder and blistering at the bottom.

A soft underlying layer gives the mat a trampoline effect. The compaction energy goes into deflecting the subgrade rather than compressing the asphalt. Cores show hairline cracks on the bottom face and poor density regardless of pass count.

What happens contractually

Most highway specifications follow the same three steps.

Referee cores. On a failed core, three more are drilled in the same sub-lot in the engineer's presence, usually equally spaced either side of the original footprint. Their average becomes the sub-lot value.

Pay factor. A sliding scale of payment reduction. As an example of the shape of it: full payment at 97.0% and above; about 95% payment for 96.0% to 96.9%; 80% to 85% for 95.0% to 95.9%; below 95.0% is outside the scale.

Removal and replacement. Below the absolute floor — commonly under 95% of Marshall or under 90% of Rice — no financial adjustment is offered. The lot is milled out to the lift depth, re-tacked and re-paved at the contractor's cost.

Your own specification will have its own numbers. The principle is constant: there is a band where you lose money, and a floor below which you lose the pavement.

What to change tomorrow morning

Move the breakdown roller onto the screed. Kill the 30 m lag so initial compaction happens above 140 °C.
Ballast the pneumatic roller up to 20 to 24 tonnes and add passes over the mat while it is 110 to 130 °C.
Raise screed tamper and vibration frequency to lift pre-compaction out of the screed from around 80% to 85–88% before a roller touches it.
Stop the paver stopping. Match plant delivery to paver speed; every stop is a cold transverse band.
Tarp every truck and ban dumping the paver wings between loads.

If you only change one thing

Move the heavy breakdown roller directly onto the paver screed, and get 80% of your density before the mat drops below 120 °C.

Compaction on a hot workable mat takes fewer passes and less effort, which means less aggregate damage and a more consistent void structure. Every strategy that relies on catching up later is a strategy that fails on the cold days, the long haul days, and the days the plant runs behind.

This guide describes general practice and typical figures. Your project specification, job mix formula and laboratory results govern the actual values on your job. Nothing here is a design or a substitute for the engineer of record.