Ignition Oven and Solvent Extraction Disagree After an Aggregate Change
A plant has changed its coarse aggregate from crushed limestone to a slag blend. The QC technician runs the ignition oven (AASHTO T 308) on a split sample and reports 5.82% asphalt content. The referee solvent extraction (AASHTO T 164) on the same split reports 5.24%. The ignition oven correction factor in the QC plan is 0.38%, established for the old limestone. The ignition oven temperature is 538°C (1000°F), and the sample mass is within tolerance. The aggregate gradation from the ignition oven shows slightly more material passing the 0.075 mm sieve than the solvent extraction, but the difference is small. The technician wants to report 5.82% because the ignition oven is the plant's primary QC method and the result is within the production tolerance. The mix is a 9.5 mm surface mix. The contractor is placing the mix tonight.
1. Which asphalt content is more likely to be correct, and why?
Read the worked answer
The solvent extraction result of 5.24% is more likely to be correct for this sample. The ignition oven measures total mass loss and subtracts an aggregate correction factor to obtain asphalt content. That correction factor is aggregate-specific. The 0.38% factor was established for the old crushed limestone and was not re-established after the switch to a slag blend. If the new slag blend loses more mass at 538°C than the old limestone, the old factor is too low. The ignition oven then subtracts too little aggregate loss and reports an artificially high asphalt content. The 0.58 percentage-point difference is consistent with an understated correction factor. The solvent extraction is not immune to error, especially fines loss, but here the known change in aggregate source is a direct reason to distrust the ignition oven correction.
2. What is the root cause of the discrepancy, and how would you verify it?
Read the worked answer
The root cause is an invalid ignition oven correction factor for the new aggregate blend. Verify it by running representative aggregate samples of the new blend through the ignition oven at the same temperature and procedure used for the mix. Determine the average aggregate mass loss, establish the new correction factor, and compare it with the old 0.38%. If the new factor is about 0.96%, for example, then the reported 5.82% would be reduced by the difference between the old and new correction, bringing it close to the solvent extraction result. Also check the ignition oven calibration, sample basket condition, and whether the sample was split representatively. Verify the solvent extraction for fines loss and complete asphalt removal. Do not simply average the two results; identify which method has the valid correction.
3. What should the plant do for tonight's production before a new correction factor is approved?
Read the worked answer
Do not report the uncorrected 5.82% as the asphalt content. For tonight, use the solvent extraction result or a provisional correction factor based on the new aggregate until the formal correction factor is established and approved. Notify the owner and QC manager of the aggregate change and the invalid correction factor. Increase verification sampling at the plant and at the paver, and watch for signs of high or low asphalt content such as flushing, tender mix, or dry, stiff mix. If the plant must use the ignition oven for rapid control, run it in parallel with solvent extraction on split samples and apply the provisional correction only after it is documented. Update the QC plan and correction factor before the next production lot.