How deep did steel piers go across a large Tulsa apartment complex?
Tulsa2025Large apartment complex, 34 multi-unit buildings on about 30 acres, brick veneer with interior stud wallsSlab-on-grade with perimeter stem wall and spread footing; one crawl space building
In short: A large apartment complex in Tulsa, 34 buildings on about 30 acres, was underpinned with several hundred steel push piers plus polyurethane foam and crack stitching in 2025-2026. The engineer's drawings specified galvanized steel piers driven to refusal on bedrock and repeatedly pointed to roof water and drainage; the pier logs show refusal depth varied from 2 to 40 ft across the site 1.
What did the buildings show?
The engineer's drawings marked pier locations and foam-injection areas building by building, but did not include a written narrative, soil borings or a stated cause 1. One building next to a retaining wall was described as appearing to subside on its south side "due to bad site drainage and uncontrolled water from roof" 1.
What did the engineer find was causing it?
No cause was written up. Standard notes on every sheet called for gutters on all eaves and stated that better drainage around each building would improve structural stability. A second engineer, answering a question about the retaining-wall building, specified 16 piers, removal of nearby trees, a French drain behind the retaining walls, and gutters routed to swales away from the walls 1.
What repair was specified?
- 469 galvanized steel push piers at exterior footings, hydraulically driven to refusal at bedrock; plan spacing typically 3 to 5 ft, with some 6 to 7 ft spans. No pier diameter, capacity, safety factor or minimum depth was given 1.
- About 32,400 sq ft of 1.75 lb/cu ft two-component polyurethane foam under interior slabs, after a passing hydrostatic plumbing test 1.
- Concrete and carbon fiber crack stitching set in structural epoxy 1.
- Slab procedures: plumbing test after repair, probe for utilities after drilling, gradual lift checked with survey equipment, no over-jacking 1.
What happened on the job?
Work ran in five phases from September 2025 into 2026. Of 469 planned piers, 396 have both a logged depth and pressure. Buildings that first needed siding or wall demolition for access are the ones with missing logs 1.
Logged results 1:
| Measure | Value |
|---|---|
| Refusal depth, median | 12 ft |
| Middle half of depths | 9 to 14 ft |
| Full range | 2 to 40 ft |
| Driving pressure, median | 9,000 psi |
| Pressure range | 2,500 to 10,000 psi |
| Piers deeper than 20 ft | 14 |
- Phase medians ranged from 10 ft to 20 ft. In one phase every pier went at least 10 ft (median 15 ft); one building's piers ranged from 7 to 40 ft, two of them stopping at 40 ft at 6,000 psi 1.
- In two buildings depths split into two groups: about 4 to 7 ft at 2,500 to 4,500 psi, and about 10 to 13 ft at 7,500 to 9,000 psi 1.
- A few piers refused at 2 to 5 ft at 8,000 to 10,000 psi, flagged in the record for verification 1.
- Where no footing was found and piers were set at the slab, they stopped at 8 to 12 ft and 4,000 to 5,000 psi 1.
- In the last phase, 171 of 192 piers were logged at exactly 10,000 psi; the documents do not state a termination criterion 1.
- Depth and pressure were almost uncorrelated (r = -0.06 across all 396; r = +0.24 excluding piers at 10,000 psi) 1.
- At one building the log notes "foam heaved," and the engineer's table noted a hump in one unit 1.
What does this case show about Oklahoma foundations?
Depth to a pier-stopping layer is not uniform even within one property. Tulsa-area Pennsylvanian rock is a stack of shales, sandstones and limestones whose thickness changes across short distances; one limestone mapped in the area thins from well over 100 ft to about 8 ft southward 2. On this site, piers on the same complex stopped anywhere from 2 to 40 ft down 1.
Water was the cause the engineers could point to. Federal building-science guidance notes that a few inches of rain on a roof produces thousands of gallons of runoff and that water pooled near a building can damage foundations on expansive soils 3. The engineers' standard notes, and their answer for the retaining-wall building, were about controlling that roof water 1.
Related
- Steel Push Piers
- Slabjacking and Polyurethane
- Depth to Rock from Borings
- Bedrock - Tulsa Metro
- Pennsylvanian Shales
- Drainage Solutions
Common questions
Did piers on the same site reach refusal at the same depth?
No. Logged depths ran from 2 ft to 40 ft. Phase medians ranged from 10 to 20 ft, and one building's piers ranged from 7 to 40 ft.
Did deeper piers take more pressure?
Hardly. Across 396 piers the correlation between depth and logged pressure was about r = -0.06; pressure tracked work phase more than depth.
What did the engineer point to as the cause?
No written cause was given, but every drawing called for gutters and better drainage, and one building near a retaining wall was described as subsiding from bad site drainage and uncontrolled roof water.
Sources
- Project record: engineer's pier and foam drawings, field observation tables, RFI response and pier logs, 2025-2026 (private client records) Licensed professional engineers, Oklahoma; repair contractor field logs
- The "Oologah" limestone and associated formations in the Tulsa area (Proceedings of the Oklahoma Academy of Science, 1931) Oklahoma Academy of Science
- Gutters and Downspouts (Building America Solution Center) U.S. Department of Energy / Pacific Northwest National Laboratory
Was this page helpful?
Thanks. Your note goes straight to the person who maintains this library.