Side By Side Axial Load Testing - The International Association of Foundation Drilling

Side By Side Axial Load Testing
Home > Publications > Foundation Drilling Magazine > FOUNDATION DRILLING 2007 Archive > dec/jan 2007

Side By Side Axial Load Testing of Residential Underpinning

by W. Tom Witherspoon, Ph.D., P.E.

President

S&W Foundation Drilling

The following article draws on the doctoral dissertation written by ADSC Board Member Tom Witherspoon, President S&W Foundation Drilling, Richardson, Texas. The dissertation, which was a key part of Witherspoon’s doctoral program, is based on original research performed in Dallas, Texas. Financial support and contributions-in-kind were provided by the ADSC’s Industry Advancement Fund and the members of the ADSC’s South Central Chapter. Witherspoon’s research has garnered a great deal of interest in the geo-engineering and geo-construction community. He has become a much sought after speaker at professional organization conferences. The research which was performed under the guidance of ADSC Technical Affiliate Member, Dr. Anand Puppala, University of Texas at Arlington, is the subject of several technical journal articles. Dr. Witherspoon’s complete research report, including all equations, tables and graphs upon which this article is based, is available though the ADSC’s Technical Library service in DVD format. The catalog number is TL-195 and the price is $20 for non-members, and $10 for members. Witherspoon’s earlier book, “Residential Foundation Performance” is also available through the Technical Library service. To order contact the ADSC office at 214/343-2091. (Editor).

Abstract

For the first time ever, all the common residential underpinning methods were tested on one site in expansive clay for a side by side comparison of axial compressive capacity. Each underpinning element was installed by the contractors who do this work and then tested across the wide soil moisture conditions that are common in North Texas. This article presents the findings of this research with very definite design guidelines for the engineering practitioner. While the detailed engineering analysis is more completely addressed in the Dissertation of W. Tom Witherspoon, the broad overview presented here will none the less provide important information that is shaping an industry.

Introduction

Damage to foundation failure in the US exceeds the combined total of all other natural disaster including tornados, hurricanes and earthquakes. Estimates by Witherspoon place this damage at an approximate $13 billion per year. To address this problem is a large industry that has grown in many cases without ample testing to verify suitability of the underpinning methods that are marketed.

Because the primary problem in clay soil is settlement, piers and piles have been utilized to lift the low section and prevent further settlement. The most common pier piling techniques are addressed at this site across soil moisture conditions that run from dry to wet and wet to dry. This wide range of climatic swings places the North Texas area in what is called a semi-arid climate that runs from drought in the summer to extremely wet in the spring. With expansive clay there is expansion and contraction that many times stresses a foundation beyond original design. If the foundation fails subsequent to soil movement, then underpinning is implemented to mitigate the problem.

The underpinning methods tested at this site include: Drilled Straight Shafts, Belled Shafts, Augercast Piles, Helical Piers, Pressed Steel Piles and Pressed Concrete Piles. Half of the elements were installed in the dry summer and tested in the wet spring and half of the elements were installed in the wet spring and tested in the dry summer.

While underpinning techniques such as drilled shafts have received extensive testing and research to develop standards for design and practice, other methods such as the hydraulically pressed piles have no recognized research available to determine proper design or construction practices. Therefore, this research is ground breaking for the pressed piling industry.

Reaction piers were installed at the site with tie-down rods and reaction beams that provided adequate resistance to axial load capacity testing using the accepted ASTM- 1143 testing method. Each pier and pile was tested to failure with the deflection and ultimate capacity recorded to, and then compared with, empirical methods to determine if the prediction models were accurate.

Site Selection

A site in southeast Irving, Texas was donated for testing where no rock was available for 63 feet. An initial soil boring was taken to confirm suitability and then others were done at time of testing and installation to provide adequate soil characteristics for empirical projections of each pier and pile. Samples from the borings were transported to the soils lab at UTA for testing and evaluation. In addition to the laboratory testing, in-situ methods including Standard Penetration testing and Cone Penetration Testing were engaged to further qualify the soil and provide all data necessary for prediction of performance.

Soil Characteristics

From the soil testing, final soil parameters were established to predict axial compressive capacity. The zone of seasonal moisture change appears to be 10’ at this site for the testing period. UU Triaxial Testing at the UTA soils laboratory confirmed CPT test results and established spring time soil parameters for prediction of capacities.

Underpinning Elements

The six foundation methods selected for this research are the most widely used types of piers and piles in residential foundation repair and new construction. For each method the most common size and installation practices were utilized so that the test results are representative of actual conditions in the field.

Drilled Shafts

Drilled shaft design has been tested in almost every environment with the FHWA design manual by O’Neill and Reese recognized as the most accurate tool for the foundation engineers. For this research the shaft was established at 12 in diameter and 15 ft deep.

Belled Shafts

The belled shaft is normally used to either increase capacity of the straight shaft or because of economic reasons to terminate the shaft above rock that may be deeper than feasible. A common design size is 2.5 diameters and 15 ft deep, which was used for this testing.

Augercast Piles

Many engineers have considered augercast piles to have more skin friction capacity than the straight shafts in clay soil because the grout is pressured through the shaft to push out pier spoil. This test provides a direct comparison in the same soil conditions with end bearing in clay.

Helical Anchors/Piers

A considerable amount of testing of helicals has been done in various soil environments but there has been little testing in expansive clays such as exist in Irving, Texas. As common with residential foundations, this test includes the single 12 in diameter helix and the double helix with an installation torque set at 5,000 ft-lbs.

Pressed Steel Piles

Pressed Steel pilings became popular in foundation repair in soil conditions where subsurface water caused pier hole caving or bearing capacity of the soil was weak. The pilings for this test are 2-7/8 in, which is common for this method on residential foundations. Although the normal drive pressures range from 25,000 pounds to 35,000 pounds, installation pressure for this research was established at 50,000 pounds. This method has never been tested in a research environment in either clay or sand.

Pressed Concrete Piles

This method evolved in Texas in the early 1990s and has become the most popular of the remedial underpinning techniques. The 6 in diameter x 12 in long concrete blocks are stood on end and pushed into the ground using resistance from the house. Here again, the installation pressure was established at 50,000 pounds although this amount of resistance is seldom available.

Comparison of Predicted

to Actual Capacity

Straight Shafts

Cone penetration tests provided a cross section for establishment of soil properties for this site. Four soil borings supplied supplementary information that was complimentary to CPT data and allowed accurate computations of wet season soil properties. These results were used to predict axial capacity of foundations.

It should be mentioned here that terms such as ‘Dry to Wet’ condition refers to underpinnings installed in dry season and tested in wet season. ‘Wet to Dry’ condition refers to underpinnings installed in ‘Wet’ season and tested in ‘Dry’ season.

The observations made at the time of testing and measurement along the shafts revealed that the soil had shrunk away from the shaft to a point of 4 ft below the ground. Therefore, the top 48 in. or 4 ft was only eliminated from the calculations. This is consistent with the approach used to estimate axial loads for drilled shaft, which do not account for the upper 5 ft of side friction. Remedial and construction piers are typically residential foundations and are placed under an approximate 18 in. of soil. Piers in such conditions may not experience soil shrinking away from the perimeter to such depths.

In clay soil, the O’Neill and Reese method for drilled shafts discounts the top 5 ft of the shaft because of possible lack of contact between pier and soil, immobilization of side friction to full magnitudes, and probable active depths and considered this upper layer as a non-contributing zone. Soil surrounding these shafts shrank away from the shaft. This shrinkage was visually observed since the piers were extended slightly above the ground surface and due to a severe drought during the time of summer testing. A thin steel wire (1/16 in. diameter) was pushed adjacent to the pier to a depth between 3 ft. and 4 ft. below the surface. Therefore, the upper 4 ft. of the soil was eliminated from skin friction consideration.

It should be noted here that the soil samples collected from the dry period showed higher undrained shear strengths than the same depth samples collected from the wet spring periods. Increased shear strength was, however, accounted for in the segment below 4ft. Predictions by the drilled shaft models showed a close match with measured load results and small differences between both values are attributed to the use of non-contributing zones for practical reasons. Also, test results from different seasonal installation indicated slightly different values with ‘wet to dry’ condition, which can be regarded as low ultimate loads.

The question for future calculations is how to address the non-contributing upper zone due to skin friction reduction in expansive clay soils such as the ones encountered in this research. If soil sampling and laboratory testing were performed during the wet season, then it would appear that the 5 ft non-contributing zone reduction would result in lower axial compression capacities. If, however, soil testing was performed during the dry season when the zone of seasonal moisture change creates an increase in shear strength, then the 5 ft reduction would be warranted.

Another factor to consider is that in a normal setting the pier is not exposed beyond the surface. In such cases, soil drying may move away soil from piers to a maximum depth of 2 ft. Such depths should be properly established from future research studies. Otherwise, reduction of skin friction of the upper 5 ft can be construed as overly conservative approach when estimating ultimate capacities of drilled shafts.

Belled Shafts

For drilled and belled shafts the normal design allowance is to eliminate friction from the top 5ft. of shaft and the bottom zone of one diameter of shaft and the periphery of the bell. Test results indicate that a better match was made for ultimate axial compressive capacity of piers tested in wet seasons when the shaft friction at the periphery of the bell was only excluded. This implies that there is shaft contact with upper layers during the wet seasonal periods. When testing in the dry season, there is the added deduction of the upper 4 to 5 ft, which resulted in a good match between predicted and measured axial loads.

Augercast Piles

The same design criteria was used with the augercast piles as the drilled and belled shafts in that the top 4’ of skin friction was eliminated with dry season testing but was included when tested in the wet season since soil shear strength was reduced by increased moisture content. This allowance provided a good match between predicted and measured axial loads as was the case for the drilled shafts.

Helical Anchors

Prediction of capacity was made with the individual bearing plate method.

Using this accepted method prediction of test results was very accurate for the single helix. Adjustments had to be made for the double helix since this prediction method shows the trailing helix to have the same capacity considerations as the leading helix, which would have projected an approximate 50% over prediction of axial capacity. Because of the inefficient installation of the helix and disturbance of soil by the leading helix it was necessary to reduce the trailing helix capacity by 80% for this site. When this adjustment was made the prediction models produced an extremely accurate calculation of capacity.

Pressed Steel Piles

Based on soils properties, ultimate load capacity predictions for pressed steel piles would be calculated as follows using equation 2.17 from chapter 2 of the dissertation:

Qu = Rs + Rt

= fs x As + qt x At

= ∑ fs x As + qt x At

Based on the a method, the sleeve friction can be calculated using the following equation.

fs = ca = a x cu a from FHWA Manual

Based on this equation and the cohesion factors from the FHWA manual for driven piles, an example of prediction was made for a 70’ steel pile.

Pressed Concrete Piles

Prediction of pressed concrete pile capacities was made using the same method as for the pressed steel pilings but with more accurate results.

The pressed concrete pilings can be predicted somewhat closer to measured capacities when CPT side friction results are used. This approach performs better than those based on undrained soil strength data but it still requires further calibration and verification. The reasons for better agreement are attributed to the similar quasi-static penetration mechanisms used to push the cone penetrometer and pressed concrete pile.

The important finding from this investigation is that the use of installation capacity is not necessarily resulted in as an ultimate load for all seasons. Rather this method resulted in different depths when installed in dry seasons and produces uniform depths in wet seasons. Due to such high variation in installation depth and their dependency on seasonal installation procedures, as well as lack of calibrated engineering models to predict the axial capacities of the pressed piles, engineers should use their judgment in the selection and use of this method. Further research in this method will help in answering some of these limitations.

Comparisons of Different Underpinning Methods Across All Seasons

Regardless of the season, the drilled shafts and augercast pilings show consistent capacities and higher ultimate axial loads than the rest of the underpinning techniques tested in this research. With the pressed concrete pilings, the depth of penetration is a key element in recording a high axial compression capacity. In the case of pressed steel pilings, there is considerable variation among the lengths of pressed pile systems.

Helical piles did not show any dependency on the type of seasonal installation. Since these foundations derive their capacities from residual shear strength parameters below the helix(s) and not along the pipe stem, seasonal moisture changes do not appear to have a measurable effect on their ultimate capacity.

A comparison of load to deflection was measured for all systems across each season.

Loads vs. Deflection for Different Underpinning Groups and Both Seasons

The deflection point at ultimate capacity was slightly greater for the augercast pile than for the straight or belled shaft. This may be attributable to installation procedures where spoil from the hole is removed by grout pumping and not by visual cleaning of the hole with an auger bit as is the case with the drilled shafts. It should also be noted that this difference may only be measurable because these holes are relatively short where end bearing has a higher contribution to ultimate capacity.

The pressed steel and helical piles showed the greatest amount of deflection in reaching their ultimate load. These results mirror this engineer’s own personal experience in using helical piles for remedial work on foundations of houses. The pressure that was required to “seat” the helical piles prior to applying a load to lift a house appears to be similar to these test results. The amount of deflection recorded for this test, however, was much less than this engineer has witnessed in previous observations. This difference in deflection was probably the result of a connection method used by a different supplier where a looser connection separated slightly as the helix pulled the bar into the ground and created tension in the stem, which was then compressed downward when a load was applied.

It should be noted that the design engineer must consider the amount of deflection recorded to get to ultimate capacity in this underpinning system when deciding to use helical piles for new construction since there is no opportunity to “seat” the pile prior to receiving the building load.

It is also obvious that the pressed steel pile records a considerable amount of deflection prior to reaching its ultimate strength, which is probably the result of a small diameter piling and much less material skin friction than concrete elements. The pressed concrete pilings showed similar deflection behavior to the drilled piers. Therefore, it appears that material skin friction is a key element in side friction of a piling in clay soil.

Summary and Conclusions

Introduction

Foundation distress and subsequent failure will continue to be a problem for homeowners, especially when they are built in expansive clay soils. As a result, underpinning to mitigate deflection problems for slab-on-grade foundations in contact with these expansive soils will continue to be remedial measures adapted in the field. Because of the wide spread use of piers, piling and helical piers, this research was undertaken to address the axial load transfer mechanisms in these foundations. This research will be helpful to provide insights to the practitioner, engineers and homeowners while deciding the proper method of underpinnings for foundation repairs. It should also be mentioned that the research results and conclusions can be further corroborated by conducting additional studies on different expansive soil sites.

The major conclusions and summary information from the present study are summarized in the following section.

Summary and Conclusions

The following conclusions and summary information was obtained from this research conducted on six different underpinning systems installed in an expansive soil zone.

Summary

1. Predictions of axial compression capacity of the drilled shafts, belled piers and augercast piles were close to measured capacities. Hence, the pier or pile capacity procedures using the FHWA-IF-99-025 design manual (FHWA 1999) are considered reliable methods for estimations of axial capacities of these foundations in expansive soil media.

2. Use of a 60° layout plan for field testing of large numbers of piers and piles was proven to be effective and efficient in the field load testing operation.

Conclusions

1. There was a negligible difference between ultimate axial compression capacity in the straight drilled shafts and augercast piles. The skin friction allowance for the drilled shafts should be the same as the one allowed for augercast piles in clay soil. Also, the deflection readings at ultimate capacity were slightly greater for the augercast pile than for the straight or belled shaft. This difference may be attributed to construction procedures in that the augercast pile normally does not produce a clean bottom surface whereas in a drilled shaft, the bottom surface can be inspected before placing concrete by either looking in the hole or running a camera in the case of open holes and by probing when pouring under slurry.

2. There was a negligible difference in ultimate axial compression capacities of the majority of drilled underpinnings between ‘dry to wet’ and ‘wet to dry’ seasonal conditions. While shear strength in upper layers did increase while going from ‘wet to dry’ season, drying induced soil shrinkage from the pier/pile (near surface) might have mitigated the increases in axial capacity.

3. Time of soil sampling has a major bearing on predicted axial compression capacity. If insitu or soil sampling tests is made during the dry season, shear strength parameters might have been increased for upper clayey layers due to desiccation related drying. Conversely, the wet season strength parameters are low and may provide lower, but conservative design parameters.

4. This research also indicates that the non-contributing depth of soil considered for shafts appeared to be important when foundation tests were performed in dry season. The non-contributing lengths from thin wire measurements show that they vary from 3 ft to 4 ft, slightly less than the recommended 5 ft value. If soil sampling and testing is attempted in the wet season, this research indicates that the total shaft depth should be included in the predictions of axial capacity since soil around the upper layers is in contact with shafts.

5. Installation of helical piles shows that the installation process of the helix in clayey medium may not pull into the ground efficiently to prevent augering of the helix and thus producing a trailing section of loose soil. Therefore the helical anchors will many times require seating using pressure from a structure in order to obtain the maximum capacity of the helical piles. With the presence of this void, larger deflection in helical piles installed in new construction jobs must be anticipated in clay soils and the design engineer must allow for this deflection accordingly.

6. In this research, when the installation torque was constant and same for helical piles, there was minor or very little difference in ultimate axial compression capacity between single and double helix piles. The double helix piling did, however, produce a more consistent ultimate capacity.

7. There was no obvious or major difference in the ultimate axial capacities of the helical piles installed in ‘dry to wet’ and ‘wet to dry’ conditions.

8. The Individual Bearing Plate method proved reasonable approach in estimating capacity of the single helix. With the double helix, however, it is necessary to apply a disturbance factor to be included in the axial capacity formulation to simulate disturbed state of soil condition. This disturbance factor was found to be an approximate 80% for this research. Therefore, contributing axial capacity support of the trailing helix was only 20% of the leading helix capacity as measured by area of helix and shear strength of soil.

9. Pressed steel pilings show the greatest amount of deflection prior to reaching their ultimate capacity. Both pressed concrete and steel piling systems yielded consistent at their ultimate capacity when they were installed during the wet period. Deeper penetration depths for these pressed piles were obtained, which indicate that the final capacity of these piles depend on length of the pile, and installation as well as testing seasonal conditions.

10. Pressed concrete pilings appear to perform in an identical fashion as those of drilled concrete piers but with an obvious reduction in axial load capacity due to smaller size of the pressed concrete pile dimensions.

11. When pressed concrete pilings were installed shallower than the zone of seasonal moisture change (between 10 ft and 15 ft for this test), they tend to loose a considerable amount of the installation capacity, i.e. up to 42% of the installation load. When these same pilings are installed below 15 ft in this soil and in these climatic conditions, they gained as much as 37% over installation capacity.

12. Three of the final six pressed concrete pilings were broken either during installation or during testing. It is not known if this failure was due to the movements of reaction beam or due to bending moments caused by lateral soil shrinkage around the piling during the dry period.

13. The pressed concrete pilings used in this research were installed with a #4 reinforcing steel bar passed through the center of the piles with the hole filled with Portland cement grout. Both reinforcement and grouting enhance lateral load resistance as well as flexural capacity of this foundation system. Hence, the present pressed concrete pile results are valid for this type of pressed concrete pile system. The performance of pressed concrete pilings without any reinforcement or bonding may have problems simply due to lesser tensile and flexural resistances.

14. Predicting axial compression capacity of pressed steel pilings does not appear to be accurate when using soil properties from laboratory tests on samples collected from the field. When the continuous CPT profiling with side friction measurements was used to estimate the capacities, they appear to match with the measured ultimate loads. This correlation is especially strong when the pilings were installed sufficiently below the active zone. This indicates that soil around pressed piles are in residual shear strength state, which is well captured by the side friction of CPT. Also, the mechanisms of penetration for pilings and CPT are similar and hence there is a strong correlation between side friction estimation in both methods.

15. Overall comparisons of the six underpinning methods show that the belled shaft has the highest axial load capacity followed by the straight shaft and augercast piles. Though cost comparisons are not included here, it can be qualitatively mentioned that the costs of drilled shafts and augercast piles for underpinning can be more expensive when compared to the rest of the underpinning methods. However, the final selection of the underpinning foundation system should not be based on the cost of installation and construction of them. Such practice may lead to further problems in residential structures in the future.

Recommendations

for Future Studies

The following areas of research are suggested for future studies.

1. Testing of pressed steel and concrete should be attempted using standard installation pressures that would emulate a one story house. These standard pressures appear to be at installation load, between 25,000 lbs and 35,000 lbs.

2. Testing of pressed steel and pressed concrete pilings in expansive clay soil to measure uplift movement and/or pressures.

3. Testing of non-joined pressed concrete pilings to determine axial load capacity and influence of active soil uplift movement against the segmental piling string.

4. Testing of pressed steel and concrete pilings in sandy soil to determine axial compressive capacity over time and water table drawn-down.

5. Further testing of pressed steel and pressed concrete pilings with projections using CPT site data.

6. Comparative testing of augercast piles and drilled straight shafts should be done in sandy soils using both casing and slurry installation for the drilled shafts to compare axial compressive capacity with these installation techniques.

7. Testing of straight drilled shafts should be attempted in expansive clay soils using casings to overcome caving conditions to address if the increase in shaft diameter mitigates perceived skin friction loss along the casing perimeter.

8. Additional testing of drilled shafts across seasons with installation and time of soil testing (in situ or bore hole) to determine/confirm if time of soil testing has an effect on total shaft length consideration.

Contributions in Kind

and Money

This project would not have been possible without the combined contributions of the individuals who donated services, material, equipment, time and money (see sidebar). The magnitude of their contribution was a very humbling event and told me how much they wanted to be a part of this ground breaking research. To have the support of the engineering community across the United States told me this research was important. Support from the foundation underpinning industry to the magnitude reflected below, however, told me that this testing had been needed for a long time and would potentially change the industry. At the very least, this research will create a climate for additional testing that will lead to many more worthwhile discoveries that will not only help an industry but bring greater value to the residential foundation repair market and ultimately increase value to the homeowner and help protect and remediate their greatest value.

Note: At least 5 papers have been written by Witherspoon and Puppala concerning various segments of this research and will be published in the near future.

Donors

1. ADSC, Industry Advancement Fund- Seed money

2. ADSC, South Central Chapter- cash sponsorship

3. Advanced Foundation Repair, Dallas, Texas- 12 pressed concrete piles

4. Allied Drilling company, Ft. Worth, Texas- drilling shafts and belling piers for testing

5. ATS Drilling, Ft. Worth, Texas- Reinforcing steel for reaction piers

6. Cameron Machine Shop- Miscellaneous welding and fabricating

7. Con-Tech Systems, LTD- Donation of 100 ton test ram for use to test in April 2005 and August 2005

8. Custom Crete Concrete- Concrete for reaction piers

9. Dywidag Corporation- 48 reaction bars, machining of bars, engineering and couplings

10. Fox Foundation Repair, Dallas, Texas- pier/pile test plates

11. Fugro Geotechnical Engineers, Dallas, Texas- 4 soil borings

12. Greg In Situ, Houston, Texas – 2 CPT logs and generated analysis

13. Illini Drilled Foundation, Danville, Illinois- augercast piles

14. Lindamood Excavating, Irving, Texas- Hydralift for moving of beams

15. McKinney Drilling Co., Ft. Worth, Texas- Drilling reaction piers

16. Ram Jack Foundation Repair, Dallas, Texas- 12 pressed steel piles, 8 helical anchors and miscellaneous

welding and fabricating

17. N.L. Schutte, Dallas, Texas- Reaction Beams 29 days crane service

18. S & W Foundation Contractors, Richardson, Texas- augercast and drilled pier labor and trucking

19. Texas Shafts, Ft. Worth, Texas- Tying of steel and crane service to set reinforcing steel for reaction piers

20. Mike Trotter General Contractor and the Trotter Companies, Doraville, Georgia.

21. Farrell, Ed – donation of ranch land for test for 16 months and not requiring clean-up of site.

22. Clayton and Johnie Stephens- time, labor and coordination.

Home | Back to Top