Case Study 1 — Forty-One Feet of Rock

The Cottonwood Creek Pier 2 Drilled Shafts, Day 19 Through Day 61

Cottonwood Creek Bridge Replacement, Kestrel Construction Group, and everyone in this case are Tier-3 illustrative composites. The numbers are internally consistent and realistic. No real agency's specification is quoted or paraphrased.

Setup

The project. Cottonwood Creek Bridge Replacement. $18.7 million, state department of transportation as owner, unit-price contract, 210 working days. Kestrel's superintendent is Del Ferraro; the project engineer is Ingrid Sørensen, in her fourth year and on her second heavy civil job.

The item. Pier 2 is founded on eleven 36-inch drilled shafts. The working platform is at elevation 1,262. The plans tip the shafts at elevation 1,206 — fifty-six feet apiece, bearing on competent rock with a nominal four-foot seat. The bid schedule carries the item as 616 LF of 36-inch drilled shaft at $412/LF = $253,792. There is no separate pay item for shaft excavation in rock, which matters enormously and which Ingrid did not notice until day nineteen.

What the documents indicated. Four borings across the pier footprint, bound into the contract, each showing river alluvium — sand, gravel, cobble — over competent rock. Top of rock: elevation 1,202 to 1,210. A general disclaimer in the front of the geotechnical report stated the information was furnished for information only, represented conditions at the boring locations on the dates drilled, and did not relieve the bidder of responsibility for its own interpretation.

How the rate was built. Kestrel priced the item on augering alluvium at about 8 LF per hour with a rig-and-crew cost of roughly $1,680 per hour — $210/LF of variable cost — plus $52,000 of item fixed cost (rig mobilization, crane, casing and slurry setup, integrity-testing setup) spread across 616 LF at $84.42/LF, marked up about 40%. Planned margin on the item: $72,432.

What Happened

Day 19 — the drill stops

Shafts 2-1 and 2-2 drilled clean, six hours each, right on the plan. On 2-3 the rate of penetration collapsed at elevation 1,247 — forty-one feet above the top of rock shown on the logs. The crew fought it for eleven hours and got nowhere. Same on 2-4, and 2-5.

By 6:40 a.m. on day nineteen, eight of eleven shaft locations had refused at roughly the same elevation, and Del had consumed the rock tooling he had budgeted for the entire job.

Ingrid did four things before 7:15.

One — she stopped the operation. Not because the rig could not keep grinding, but because grinding through it would destroy the only thing that could pay for it. Del did not agree, and said so.

Del: "I've got a rig and a five-man crew standing still at sixteen-eight a day."

Ingrid: "And a claim worth ten times that if the resident engineer can put his own tape on this rock before we chew it up. Give me the morning."

Two — she documented. For each of the eight refusals she recorded, in a bound field book with numbered pages: the elevation at which the rate of penetration changed, the clock time, the tooling on the rig at that moment, and a description of the cuttings. She photographed the spoil beside a folding rule with a placard identifying shaft and elevation range, and set aside a labeled five-gallon sample from each. She then asked the agency's shaft inspector to initial each elevation entry. He did.

Three — she sent notice at 6:52 a.m. Subject line: NOTICE — Pier 2 drilled shafts — subsurface conditions differing from contract boring logs — request inspection prior to further excavation. Two hundred and ten words. It identified the location, cited the boring logs and the top-of-rock elevations they showed, stated the elevations actually encountered, stated that work in the affected area had stopped and the conditions were undisturbed, requested inspection before the conditions were disturbed, and reserved rights to additional cost and time. It named no dollar figure and made no argument.

Four — she opened a cost code. Every hour of labor, every equipment hour, and every consumable associated with the event went to a dedicated code beginning that morning, before anyone knew whether there would be a claim.

Days 19–23 — the agency looks

The resident engineer was on site by 9:30 with the agency's geotechnical consultant. They took their own elevations, photographed the cuttings, and split Ingrid's samples. On day 21 the agency issued a written directive: core each of the eight affected shaft locations to verify the material and its continuity. No pay item existed for coring, so the work went on force account.

Del moved the rig to Pier 3 on day 21 and kept two shafts moving there — which mattered later, and not in the way he expected.

The cores came back competent rock, continuous across the eight locations. On day 27 the designer issued revised plans: tip the eight affected shafts at elevation 1,243 — the same nominal four-foot seat, forty-one feet higher.

The measurement nobody expected

Del read the revised plans twice and then called Ingrid from the truck.

Del: "Tell me I'm reading this wrong. We hit rock we didn't buy, we burned three bits, we sat for three days — and they're paying us less."

Ingrid: "You're reading it right. The schedule pays quantity. The quantity went down. Everything else is a different clause."

That sentence is the entire case.

The Analysis: Four Mechanisms, Four Different Payments

Ingrid's analysis fit on one page and identified four live mechanisms.

1. The bid schedule. Pays measured quantity. Operated automatically and against Kestrel: −$121,952. Nothing to argue about; this is the contract working as written.

2. Force account — the eight directed verification cores. A subcontracted driller, four days, every ticket signed by the inspector the day it was worked. Claimed $44,160, allowed $44,160. The cleanest recovery on the job, and the reason was entirely procedural: nothing was reconstructed.

3. The quantity variation clause. A 48.1% underrun on an item representing well over the contract's "major item" threshold. Kestrel's ask was the stranded fixed cost: planned recovery $84.42/LF × 616 LF = $52,000; actual recovery $84.42 × 320 = $27,014; stranded = $24,986. Kestrel did not ask for lost profit on the 296 LF never drilled, because most clauses of this type compensate cost, not anticipated profit. Negotiated settlement: $22,500.

4. Differing site conditions — Type I. This carried the weight. Ingrid built a single one-page exhibit: the four boring logs, the plan profile, and the actual measured refusal elevations, all on the same vertical scale, with the inspector's initials visible on the field-book excerpt. The agency's geotechnical consultant did not contest that the condition differed materially from what the documents indicated. The general disclaimer was raised and did not carry the day — the logs affirmatively indicated a top-of-rock elevation, Kestrel had demonstrably relied on it in building an $8-per-hour-penetration rate, and Kestrel had complied with the clause to the letter.

Time. Kestrel requested 9 working days. The agency granted 6. Two requested days coincided with a high-water event that would have stopped in-stream work regardless — genuine concurrency, and Ingrid conceded it rather than fight it. One day was waiting on a tooling delivery, which is Kestrel's own supply chain. Six working days on a 210-working-day contract, granted in writing, with the shaft activity's float recalculated in the next update.

The honest scoreboard

This is what a well-run differing-site-conditions event looks like when it goes well. Kestrel recovered the overwhelming majority of its cost and less than half of its margin. Nobody gets made whole. You get made close.

What actually produced the recovery

Three things, and none of them are legal skill.

The stop. Three hours of a $16,800-a-day spread — about $5,000 — bought an inspected, undisturbed condition. Everything downstream depended on it.

The elevation record with initials. "We hit rock" is a complaint. "Refusal at elevation 1,247 at 09:40 on shaft 2-3 with a rock auger on the rig, initialed by the agency's inspector" is a claim. The difference cost nothing.

Segregating cost on day one. When the settlement was negotiated on day 61, Kestrel produced a cost code that had existed since day 19. The agency's negotiator trimmed the tooling and standby lines but never questioned whether the costs belonged to the event — because they had never been mixed with anything else.

And one thing that cost Kestrel money: Del's entirely correct decision to move the rig to Pier 3 on day 21 became the agency's evidence that he could have moved it on day 20. Mitigation is required and it is the right thing to do, and it will sometimes be used against you. The answer is not to stop mitigating. The answer is to write down, on the day, why you could not move sooner — which Ingrid did not do, and which cost about $16,800.

Discussion Questions

1. Suppose the contract had carried two pay items — drilled shaft in soil and drilled shaft in rock, at different prices. How much of this event would have been resolved automatically by the bid schedule, and how much would still have required a claim? What does your answer tell you about what to look for in a bid schedule before you price a foundation item?

2. Ingrid conceded two of the nine requested working days on concurrency grounds without much argument. Was that good judgment or a giveaway? Build the strongest case for each position, and identify what you would need to know to decide.

3. The agency disallowed $9,800 of "additional engineering and documentation time" as non-compensable home-office cost. That documentation is the reason the rest of the claim succeeded. Is there a way to structure or characterize that cost so it becomes recoverable, and is it appropriate to try?

4. Del was right that stopping the rig cost real money in real time, and Ingrid was right that continuing would have destroyed the claim. Design a decision rule a superintendent could apply in the moment, without a project engineer present, that gets this right most of the time. What are its failure modes?

5. Kestrel lost $38,796 of expected margin on an event where it did everything correctly. Where, if anywhere, should that risk have been priced — and what would have happened to Kestrel's bid if it had been?

Your Turn

Using only the facts above, write the two-page position paper Ingrid submitted with the claim on day 44. Required elements: a statement of the condition indicated by the contract documents, with specific elevations; a statement of the condition actually encountered, with specific elevations, times, and how each was verified; the Type I analysis in three sentences; an itemized damages table separating the four mechanisms so that no dollar is claimed twice under two clauses; the time request in working days with the concurrency concession stated up front; and a one-paragraph closing that does not accuse anyone of anything.

Then write, on a separate sheet, the three sentences you would delete if this document were going to a mediator instead of a resident engineer, and say why.