Proofroll Failure: Pumping Silty Clay Under a Warehouse Slab
Subgrade for a 60,000 ft² warehouse slab is a low-plasticity silty clay at 1.5 to 3 ft above the seasonal high water table. The specification requires proofrolling with a loaded tandem-axle dump truck and acceptance with no rutting deeper than 1/2 in. or visible deflection. On a hot, dry day the first pass produces continuous 3- to 4-in. ruts and visible pumping along about 40% of the area, with soil extruding between the tire marks. Field moisture is 4 percentage points above optimum. The contractor proposes placing 12 in. of No. 57 stone over the entire area, re-proofrolling, calling it a stabilization layer, and asks you to sign the subgrade acceptance form. Undercutting to a stable layer would take about 4 ft of excavation, and the schedule is already behind.
1. Does the subgrade pass the proofroll? What mechanism explains what you are seeing?
Read the worked answer
No, it fails. Rutting and pumping here are a strength failure, not a surface-regularity issue. The silty clay is roughly 4 points wet of optimum, which in a low-plasticity fine-grained soil means low effective stress and high saturation. A loaded tandem axle then generates excess pore pressure that cannot dissipate quickly in a low-permeability soil, so the soil loses shear strength and extrudes from beneath the tire.
Against a 1/2-in. criterion, 3-4 in. of rutting with visible pumping over 40% of the area is unambiguous. Two points reinforce it: a hot, dry day is the best case for this subgrade, so conditions will not improve on their own, and re-rolling will remold and weaken the clay further rather than densify it. Stop and escalate to the engineer of record.
2. Evaluate the 12-in. No. 57 stone stabilization proposal.
Read the worked answer
As proposed, it is a guess rather than a design. A granular layer works as load spreading only where the soft zone is shallow and the stone bridges over a firm stratum, and only with a separation geotextile so fines cannot pump into the stone voids; without separation the layer contaminates and loses effective thickness while the subgrade keeps moving. It also adds roughly 90-100 psf of dead load.
The 4-ft undercut depth says the soft zone is deep, and 40% of the area says it is not a local soft spot. Twelve inches of stone will not cure a deep bearing or shear deficiency. Legitimate options for the engineer of record: undercut and replace with compacted granular fill, aerate and dry the clay then recompact, lime or cement stabilization of the upper 12-18 in., or a reinforced stone platform with designed thickness and a drainage path. You cannot sign an acceptance form that converts a failing subgrade into a design change.
3. What acceptance evidence and documentation would you require after any repair?
Read the worked answer
Re-proofroll the full area with the same loaded truck and the specified pass pattern, and require 1/2-in. ruts or less with no deflection or pumping. On a granular platform, test density and gradation of the placed layer and verify thickness and elevation against plan.
Document the date, weather, truck make and gross weight, observer, routes and number of passes, and station-by-station results with photographs. Record the location and depth of any undercut. Put in writing that subgrade acceptance was rejected on the original date and remains pending re-proofroll plus the engineer of record's written approval of the repair method.
Flag the seasonal high water table at only 1.5-3 ft below subgrade. If the repaired platform cannot drain, the same failure returns after the first wet season, so a drainage detail belongs in the repair scope.