Review of Foundations, abutments and footings (Hool and Kinne, Eds., 1923), Section 3: Foundations (Part B)

By Michael Bennett, P.E., M.ASCE (Gannett Fleming – Audubon, PA)

The second half of Section 3 of Foundations, abutments and footings continued the first half’s discussion of various foundation types and their construction.  Walter Cahill, who penned most of the latter portion of Section 3, was a vice president of the Great Lakes Dredge and Dock Company in 1923.  Then as now, Great Lakes was among the USA’s largest contractors for port facilities.  Cahill covered a multitude of facets of timber and concrete pile construction, including their manufacture, driving equipment and procedures, load testing, and common field issues.  Frederick Avery, a bridge engineer of 25 years’ experience, closed out Section 3 with a brief section on sheet piling.  His writings consisted primarily of tabulated cross-sectional properties from the production catalogs of preeminent steel companies of the Roaring Twenties, including US Steel, Jones and Laughlin, and Lackawanna Steel (Avery 1923, Cahill 1923 A, Cahill 1923 B, GLDD 2024, Marquis 1922).

Cahill began his writing by covering the history of pile installation and construction.  He started with the etymology of the word “pile,” which comes from the Anglo-Saxon pil, i.e., sharp stake or arrow; the Latin pilum, i.e., javelin; and the Latin pila, i.e., pillar.  Cahill wrote that timber piles had dominated the construction industry throughout most of human history and that human- or animal-powered drop hammers had mainly been used to drive them.  He added that the technology had been in use for at least 2,000 years, since Julius Caesar’s legions had driven thousands of pila to span rivers in modern France as they conquered it during the Gallic Wars.  Several years later, when Caesar turned his ambitions inward, his pile-driven triumphs in Gaul fueled the Roman Republic’s transformation into an empire (Cahill 1923 A, PBS 2006, Van Houten 1932).

 

 

Cahill continued that pile-driving had remained a game of drop hammers and timber piles until the dawn of the Industrial Revolution.  In 1845, Scotsman James Nasmyth successfully fitted a steam engine to a pile hammer.  Drop hammers did not disappear immediately and remained common in certain sectors of construction even in 1923, but the steam hammer rapidly gained widespread use and kickstarted the modernization of pile foundations.  The pile itself was next up for change.  The reinforced concrete pile debuted in France in 1897 and first appeared in the United States in 1901.  Pioneers of the technology, such as the Raymond Concrete Pile Company, were quick to point out that both precast and cast-in-place concrete piles saved time and money by having higher load capacities than timber piles, which meant installing fewer piles under smaller pile caps.  Steel piles also came into vogue during the early 20th century, with I-beams predominating until Bethlehem Steel manufactured the first H-piles in 1908; their improved buckling strength allowed for deeper driving and better scour resistance at bridges.  During the 1920s, even as Walter Cahill wrote his portion of Section 3, engineers in Weimar Germany pioneered the diesel combustion hammer.  100 years later, diesel hammers make up the bulk of most pile contractors’ equipment (Cahill 1923 A, Van Houten 1932, Warrington 2009).

 

 

 

Cahill offered pile-driving contractors plenty of advice in his write-up, much of which remains sound.  His note that an experienced crew can typically drive 30 to 35 piles in a workday matches the expected output of a seasoned crew a century later.  Cahill’s guidance on the need to treat timber piles to preserve them from wetting-drying cycles, fungi, and marine borers also holds up, as does his suggestion to use creosote for the job. (Compounds such as chromated copper arsenate, aka CCA, have superseded his other recommended preservatives, crude oil and cement.) Modern geo-professionals might be especially interested to read Cahill’s note that creating a bulb or pedestal at the bases of cast-in-place piles allows contractors to take advantage of “the higher bearing capacity generally existing in the lower strata of earth,” an observation since explained by bearing capacity theory (Cahill 1923 A, Cahill 1923 B, Collin et al. 2016).

Others of Cahill’s nuggets of purported wisdom typified the erroneous information that permeated foundation construction before the dawn of modern geotechnical engineering.  He reviewed whether timber piles should be debarked before driving and ultimately sided with a grizzled contractor who stated: “‘If the bark is loose, it comes off in driving, and if it stays on, it is as good as the pile itself and helps develop more friction due to its roughness.’” This take notwithstanding, all modern timber piles are debarked during manufacturing.  Cahill also discussed how contractors could splice timber piles in the field using a “sleeve” of steel pipe, steel fishplates along the piles’ perimeters, or – under extenuating conditions such as swampy environs – iron dowel pins placed at the piles’ centers.  100 years later, the near impossibility of splicing timber piles reliably is so well-established that even the Timber pile design and construction manual, a trade group publication, lists “difficult to splice” as the chief drawback of timber piles (Cahill 1923 A, Cahill 1923 B, Collin et al. 2016).

 

 

Cahill’s write-up also reflected that next to nothing was known about the theory of pile design in 1923.  Contractors working on pile foundations back then had only dynamic formulas and static load tests available for estimating pile capacity.  The first came into vogue after the steam hammer’s debut began standardizing pile-driving.  Subsequent civil engineers had used Newtonian mechanics to generate a plethora of formulas to estimate pile capacity using variables from driving such as the pile hammer’s weight and fall height and the pile’s penetration under each hammer blow.  Most of the dynamic formulas also included a sizable constant referred to as a “factor of safety” to account for the unknowns which then abounded in pile installation.  Cahill mentioned dynamic formulas only in passing since a later section of Foundations, abutments and footings examined them at length.  However, many contractors and engineers were acutely aware even during the Roaring Twenties that, as geotechnical legend Ralph Peck later groused, the formulas’ “variety and number are matched only by their shortcomings.”  Most notably, they considered soil conditions and soil-structure interaction only via the extremely indirect metric of pile penetration.  Vulcan Hammer, a prominent maker of pile hammers during the Jazz Age, went so far as to warn customers about using the formulas in its commercial literature (Cahill 1923 A, Cahill 1923 B, FHWA 2016, Peck et al. 1974).

 

 

 

 

 

 

 

 

 

Static load tests in the 1920s involved installing a pile or pile group at a job site, loading the pile or group (often to several times its intended capacity), and measuring its subsequent settlement at intervals over some length of time.  The tests, which remain in widespread use, show the field behavior of the loaded pile or group and reflect in situ geotechnical conditions to some extent by capturing the soil-structure interaction of the pile/s and the surrounding soil.  They are thus better suited by orders of magnitude to assess pile capacity than dynamic formulas, and the FHWA describes static load tests as “the most accurate method of determining load capacity.”  However, the tests have drawbacks, such as unreliable prediction of pile settlement and multiple approved testing standards being in use.  The latter problem was even worse during the Jazz Age; Cahill reviewed several static load test case histories in his portion of Section 3, and each test seems to have involved a unique procedure.  On this basis, Ralph Peck accurately observed that foundation engineers and contractors of the Roaring Twenties placed “a somewhat misguided reliance on load tests in the field” – although he spared the tests the scorn he directed at dynamic formulas (Cahill 1923 B, FHWA 2016, IBC 2024, Peck 1993).

 

 

 

The challenge of further advancing pile-driving must have appeared Sisyphean to civil engineers of the 1920s.  However, change lay just around the corner.  Walter Cahill unwittingly put a finger on it when he shared an anecdote about a contractor who drove a reinforced concrete pile and found it had cracked at its leading end but not its top.  A few years later, Australian civil engineer David Isaacs began pulling this thread and, ultimately, started unraveling the mysteries of dynamically assessing pile capacity.  The topic was among the first that Isaacs, then in his mid-20s, researched during a distinguished career that included directing Australia’s primary construction research laboratory for 25 years, developing techniques for the structural analysis of welded connections, and designing major bridges across his country.  Isaacs also served in World War II, during which he played a key role in salvaging gold from an Australian liner sunk by German mines (AHP 2021, Cahill 1923 B, McInnes 2023).

 

 

In a 1931 paper on pile dynamics, Isaacs became one of the first civil engineers to point out that dynamic formulas’ factors of safety were inaccurately named.  He pointed out that true factors of safety accounted for aleatory uncertainty, while the “[factors] of ignorance” in the formulas accounted for the epistemic uncertainty then present in pile dynamics.  Isaacs demonstrated the formulas’ limits with an experiment in which he hung several pairs of thin steel rods horizontally.  For each pair, Isaacs pulled the first rod back a given distance, released it and let it collide with the second rod, measured the second rod’s velocity after impact, and compared it to the predicted velocity of the rod after an elastic collision.  He found that the second rod’s measured velocity was consistently lower than Newtonian mechanics predicted, indicating that energy was not conserved during the collision.  Isaacs also observed that the collisions became less and less elastic as he used increasingly longer second rods (Isaacs 1931).

 

 

Isaacs concluded from his experiment with colliding rods that energy transfer during the analogous process of pile driving was best examined using not Newtonian mechanics but wave mechanics.  His conclusion, like so many scientific breakthroughs before and since, was simple yet startlingly profound.  The failure of traditional elastic collision-based approaches to accurately describe the transfer of energy through a long, thin body impacted by a second, similar body explained why the dynamic formulas had never consistently and correctly predicted pile capacity and could never do so.  Civil engineering codes still permit the use of dynamic formulas under certain circumstances, but Isaacs’s discovery made clear exactly why they could not be trusted and why their use should always be discouraged (FHWA 2016, Isaacs 1931).

 

 

Having slain the Goliath of dynamic formulas, David Isaacs then laid out a new method for predicting pile capacity.  First, he derived a series of equations to describe the behavior of impact-induced tension and compression waves through piles and hammers.  Isaacs noted how the waves’ intersections affected their amplitudes to a degree beyond what researchers in 1931 could quantify.  He bypassed this issue by making the simplifying assumption that two intersecting waves’ attenuation would be directly proportional to their amplitude; he added the caveat that further investigation of the phenomenon would be required.  Next, Isaacs continued his wave equation derivations and showed that accounting for pile cushions and helmets made the equations significantly more complicated.  Clearly, adding other pile-driving considerations such as geotechnical conditions and soil-structure interaction into the equations would make them untenably complex to derive, let alone apply.  Instead, he turned his attention to how to expedite solving the equations he had already derived (Isaacs 1931).

Isaacs’s solution for solving his wave equations for piles was just as ingenious as his decision to use the wave equations.  He constructed a mechanical curve-sketching machine to create approximate plots of pile capacity versus penetration for a given pile, hammer, stroke, and cushion combination.  The machine, which foreshadowed the need for more powerful computing tools for such problems, cut the time needed to sketch such a plot from several weeks to only 10 minutes.  Isaacs then considered several hypothetical driving scenarios and compared the pile capacities he predicted using his plots to those predicted with several dynamic formulas.  He deliberately chose values of pile penetration for which the formulas were known to be somewhat less unreliable.  Isaacs found that his plot-predicted capacities for these scenarios agreed well with those predicted using the formulas.  He estimated that the plots predicted capacities with an accuracy of roughly ±30% and recommended a “factor of ignorance” for them of 1.25 – a dramatic reduction from those of the dynamic formulas (Isaacs 1931).

 

 

Like many scientific pioneers, most notably Sir Isaac Newton himself, David Isaacs was well-attuned to his findings’ limitations.  “The new method cannot yet be taken as definitely giving all that is desired,” he concluded, especially “in regard to the relationship between driving resistance and bearing resistance for various classes of ground, and the correlation of load tests with pile formulae.”  Isaacs never revisited pile dynamics in his research, but his discoveries on the subject paved the way for reliable techniques for dynamic pile assessment such as WEAP, PDA, and CAPWAP.  The rise of modern computers made these breakthroughs possible, but all of them stemmed from Isaacs’s painstaking derivations and brilliant curve-sketching machine.  Any of these innovations may well have seemed like science fiction to Walter Cahill, Frederick Avery, and their civil engineering peers in 1923.  The fact that such advances eventually came to fruition is a testament to their and Isaacs’s determination to keep their field moving forward.

 

Acknowledgments

Sebastian Lobo-Guerrero, Ph.D., PE, BC.GE, M.ASCE (A.G.E.S., Inc.: Canonsburg, PA), reviewed the entry’s technical content.  Don Warrington, Ph.D., P.E., M.ASCE (Vulcan Foundation Equipment: Rising Fawn, GA) freely shared his expertise on pile driving’s history and mechanics and suggested exploring Isaacs (1931) in more depth.  Thomas Kennedy (Geopier: Davidson, NC), co-wrote a 2021 version of the entry posted on an independent webpage.

 

References

AHP (Australian Histories Podcast). 2021. “50 – WW2 salvage operation.” Australian histories podcast, Apr. 30. Accessed Jun. 30, 2024. https://www.australianhistoriespodcast.com.au/50-ww2-salvage-operation/ 

Avery, F.H. 1923. “Section 3: Foundations – Sheet piles.” In Foundations, abutments, and footings, G. A. Hool and W. S. Kinne, eds. New York, NY, USA: McGraw-Hill, 198-207.

Cahill, W. 1923. “Section 3: Foundations – Concrete piles.” In Foundations, abutments, and footings, G.A. Hool and W.S. Kinne, eds. New York, NY, USA: McGraw-Hill, 165-198.

Cahill, W. 1923. “Section 3: Foundations – Timber piles.” In Foundations, abutments, and footings, G.A. Hool and W.S. Kinne, eds. New York, NY, USA: McGraw-Hill, 138-165.

Collin, J.G., E.D. Entsminger, R. Shmulsky, and J. McNeel. 2016. Timber pile design and construction manual. Starkville, MS, USA: Southern Pressure Treaters’ Association.

FHWA (Federal Highway Administration). 2016. Design and construction of driven pile foundations, Geotech. Eng. Circular No. 12. Washington, DC, USA: Department of Transportation.

GLDD (Great Lakes Dredge & Dock Company). 2024. “Home page.” Great Lakes Dredge & Dock Company. Accessed Jun. 26, 2024. https://gldd.com/ 

ICC (International Code Council). 2023. “2024 International Building Code, Chapter 18: Soils and Foundations.” ICC Digital Codes. Accessed Jun. 27, 2024. https://codes.iccsafe.org/content/IBC2024P1/chapter-18-soils-and-foundations 

Isaacs, D.V. 1931. “Reinforced concrete pile formulae.” J. Inst. Eng. Aust., 3 (9), 305-323.

Marquis, A.N. 1922. Who’s who in engineering: a biographical dictionary of contemporaries. Chicago, IL, USA: A.N. Marquis and Co.

McInnes, K. 2024. “Isaacs, David Victor (1904-1991).” Encyclopedia of Australian Science and Innovation, May 3. Accessed Jun. 26, 2024. https://www.eoas.info/biogs/P007078b.htm 

PBS (Public Broadcasting Service). 2006. “Julius Caesar.” The Roman Empire in the first century. Accessed Jun. 30, 2024. https://www.pbs.org/empires/romans/empire/julius_caesar.html 

Peck, R.B. 1993. “The coming of age of soil mechanics: 1920-1970.” 1st Spencer J. Buchanan Lecture, Oct. 22. Texas A&M Univ., College Station, TX, USA.

Peck, R.B., W.E. Hanson, and T.H. Thornburn. 1974. Foundation Engineering, 2nd Ed. New York, NY, USA: John Wiley and Sons.

Simplex (Simplex Concrete Piling Company). 1907. “Simplex Concrete Pile: Foundations, wharves, trestles, reservoirs.” Boston, MA, USA: New England Foundation Co.

Van Houten, R.W. 1932. “Piles and pile driving.” B.S. thesis, Newark, NJ, USA: Newark College of Eng.

Warrington, D.C. 2009. “Pile driving introduction.” In Pile driving by Pile Buck, D.C. Warrington, ed. Vero Beach, FL, USA: Pile Buck, 1-21. Accessed Jun. 26, 2024. https://pilebuck.com/pile-driving-by-pile-buck/chapter-1-introduction-3/

Warrington, D.C. 2017. “Isaacs and Glanville: The beginnings of the wave equation for piles.” Vulcan Hammer, Sept. 18. Accessed Jun. 26, 2024. https://vulcanhammer.info/2017/09/18/isaacs-and-glanville-the-beginnings-of-the-wave-equation-for-piles/