To Fill or Not To Fill

Abstract

Foundation underpinning should not be performed without void filling operations when a structure is lifted back to level.

Introduction

In the world of foundation underpinning, specifically in retrofit applications to remediate differential settlement issues in existing structures, there has always been a debate with regard to void space created when the structure was lifted back to level.?Foundation settlement occurs due to many reasons, foremost being the introduction or reduction of moisture around the property, consolidation of fill soils, and seismic or lateral movement of the soils supporting the structure. There are many ways to remediate differential settlement issues, most often via a process called “underpinning”. Underpinning a structure can be performed with a multitude of techniques and products, using both a physical connection to the structure or treatment of underlying soils (or some combination thereof). Different means of underpinning can include the following:

1.????????Steel Underpinning – Steel components driven into the soil, generally attached to a bracket system that is anchored to the footing.

a.????Push Piers – Hollow steel tubes are advanced into the soil, utilizing the weight of the structure to drive the tubes to competent strata and facilitate lift.

b.????Helical Piers – Specially designed and engineered shafts (lengths, leads blades and sizes, thickness, etc) are “screwed” into the earth to a target depth and torque, then attached to a bracket and used to lift the structure

2.????????Concentric Piers – Pier tubes are advanced into the soil utilizing the weight of a structure, however instead of the L-Shaped Eccentrically-Loaded Brackets, a different bracket system drives and lifts from directly underneath the footing, reducing side loading failures.

3.????????Concrete Underpinning – Concrete elements are used to enlarge, strengthen, reinforce, or transfer significant loads into competent strata.

4.????????Caissons – Excavations are cored into minimum bedrock embedment, a rebar cage is dropped, and concrete is used to fill the core to reinforce the foundation and transfer significant loads to competent strata.

5.????????Grade Beams – Concrete is tied into the existing footing using rebar dowels to enlarge, or otherwise strengthen, the existing footing.

6.????????Foundation Replacement – Sections at a time of the footing is replaced with proper reinforcement to support the structure.

7.????????Micropiles – Caissons, the miniature version.

8.????????Concrete-Driven Piles – Concrete cylindrical piles are driven into the soil to push past weak soil zones and transfer significant loads to competent strata.

Soil Treatment – Soil is specifically treated to address its existing failures

o??Compaction Grouting – Injection of a cement sand slurry, cementitious grout, polyurethane foam, or other approved alternative to strengthen the soil

o??Lime Mixing – One of a few different means of stabilizing highly expansive clay soils

o??Deep Soil Mixing – Treating soil with Cellular Concrete, Cement, or Bentonite/Cement mixtures to strengthen frail soil structures.

o??AGSS – Injection of a chemical compound to reverse the ionization of soil molecules in order to prevent the expansion of clay soils.

?

The recommendations of these options are all influenced by different site characteristics, soil properties, and overall goals of a project (and yes, often budget, too). So when a structure has experienced differential settlement, and must be lifted back to level, often the best possible lifting opportunity is presented by the steel underpinning systems, which are a major focus of this article.

Whether one uses push piers or helical piers to lift a home back to as level as practicable, the overall system is designed to physically attach to a footing, to treat the perimeter footing as a beam, being supported by individual columns, which are the piering systems. By all rights and means, the system is designed to be standalone with that regard, generally with the upper 10ft of soil being referred to as unrestrained fill. Most engineering practices for the push piering system finds its Factor of Safety by calculating axial loading and supports of all loads, while based on the conservative assumption that that voidspace created by the friction collar at the base never collapses back in, so the whole system is not designed to be a “Resistance Pier” but rather a “Floating Pier” that never gains that additional soil support. Most helical piers are engineered to act as a beam with embedded columns, providing a degree of tension resistance, lateral support, and axial support.

When a structure experiences subsidence, and a steel underpinning system is used to “lift” the structure back to as level as practicable, there will be an ensuing void directly underneath the footing that spans between each column point, as well as under the existing slab for Slab On Grade Foundation Systems. It goes to follow that lifting a perimeter footing 3” will result in a 3” void. If the piering systems are designed as beams and columns and function well as standalone installations, why then do you see many contractors, engineers, and even municipal regulations dictating that a void fill is necessary under the footing and/or under the existing slab? What benefit is there and what reason to do so? For the remainder of this article, we are going to defer to common Engineering Practices, and the International Building Code (IBC), 2021 edition.

?

Weighing In

This problem was first presented to me by Steven C. Helfrich, P.E., of Helfrich and Associates Engineering and Consulting, back in February of 2006. It had long since been a standard of practice by that time to perform void filling operating after lifting with some means of underpinning by that time, and it drove at-length discussion of “why” and “how” void filling operations were to be completed. After all, it was the oversight and tutelage of Mr. Helfrich that would provide a solid basis for my career platform to lift off of; I owe much of my knowledge to his oversight as I worked for the foundation repair contractor I was involved in for the following 15 years. Helfrich and Associates was, and remains, a consulting service for foundation subsidence and construction practices in the industry, maintaining a substantial reputation and even providing expert witness services in the industry.

The next heavy influence to explore the idea of void filling operations came from an in-depth series of discussions with the Division of the State Architect (DSA) for the State of California, as well as the Office of Public School Construction (OPSC) for the State of California, when an elementary school in Northern CA experienced subsidence and the contractor I worked with was hired to level it back out and stabilize it. Amid the Engineering Design Phase of the project, navigating the nuances of pad footings and a series of support systems that comprised the complicated and comprehensive foundation system, we wound up having a few key meetings with Engineers on our side, and with OPSC and DSA to address this very concern.

The next several influences come from similar experiences with less tutelage than that received by Helfrich and Associates, with input from Mohammed Saleem, PE from Alpine Engineering (in Northern California), Jared Fischer, PE of Waypoint Engineering (from Portland, Oregon), Paul Hayman, PE and Scott Murphree-Roberts, PE of Hayman Engineering (in Columbia, Mississippi), Pablo Naranjo, PE from Engineering Services and Design of SoCal, Inc. (in Rowland Heights, CA), and Jeff Fitch, PE from SFA Design Group (in Livermore, CA). Each of these engineers have designed many projects with me, and we have had many discussions, some at length and some cursory, with which we explored the idea of void filling, means methods and reasons, and the influence of each of those engineers goes into this discussion.

Finally, while much of my experience comes from California and its very applicable California Building Code, which the geography of the State provides a retinue of geotechnical challenges regarding high seismicity and Seismic Classification Zones, soils profiles all over the map, several cities that could be classified as one big landslide zone, and areas of soil support so poor that 100’+ of soil below structures have just about no bearing value, I will be weighing in using the International Building Code (IBC) 2021 to provide substantial discussion points. What this does mean, is that some of the municipal-specific regulations like through the City of Los Angeles, County of Los Angeles, or City of San Francisco, each of which whom have their own Building Codes, will not be addressed in this article, as the County of Los Angeles generally does not even accept Evaluation Service Reports from ICC-ES or IAPMO, without a secondary review, update, and yes even subscription, through their own Evaluation Service, the Los Angeles Research Department reviewing products and processes (generally already-approved Evaluation Service Reports) to issue a Los Angeles Research Report (LARR) to approve products and processes. This is important, because in Los Angeles (to my most recent knowledge), helical piles are not permitted to be calculated to resist any lateral loads, even though by design they do, so it becomes a perfunctory discussion that just sows chaos into this idea, and it’s very limited on a national level, so I’m just going to ignore it for the sake of this discussion.

?

Investigation

For clarity, I’m going to assert right off the bat it’s probably a lot more important to do a void filling operation on a Slab On Grade Foundation System than a Raised Floor Foundation System. Given that Engineering Design by Structural Engineers like Jared Fischer and Jeff Fitch allow the systems to be standalone, using the footing essentially as a beam with columns that are the Deep Foundations of the Steel Underpinning Systems, the “engineering” need for void filling is negligible. The deep foundation system should be able to support the loading, with the Factor of Safety, and there isn’t really an issue otherwise. When we start diving into the Building Code, things start to get complicated, and then, from a Civil/Geotechnical standpoint, we start to find an impasse.

To begin with, we can all agree that a lack of support is less effective or efficient than more-than-adequate support. So by nature, we can agree off the bat that creating a 3” void by lifting a settled foundation 3” is already creating a situation where the soil-foundation connection is interrupted, which by nature means that the foundation is not as supported as it could be. This creates a series of potential problems regarding loading. A raised floor foundation system should already be designed to carry sufficient loads on the subfloor, and if the perimeter footing is to be treated as a beam, with its columns to be the steel underpinning, then the “beam” is already supported, and the subfloor, already designed to be supporting all the service loads, would in theory be sufficient. This is in contrast to a slab on grade foundation system, which would be designed to have the support of soil directly below it, which no longer happens closest to the footing where the piering systems have lifted the foundation the targeted 3” in this scenario. By definition, lifting a SOG Foundation System 3” leaves a 3” void under the slab. Any service loads applied closer to the edges of the foundation, a Grand Piano installed in the corner of the Living Room, granite countertops in the Kitchen, anything heavy that significantly increases Live Loads/Dead Loads at that EXACT spot becomes subject to failures not previously existing until after you removed the support of the soil. This is sort of a cursory discussion point, since there isn’t anything significant that would be derived from such a small area, given the assumption that the slab itself is sound and can support its own loading, however it’s still a discussion point between “good” and “better”.

Now jumping into the nuances of foundation repair, there are a few different issues one may run across with regard to creating void space under a home, which specifically addresses the seismic and lateral resistance of a structure. The DSA and OPSC during our permitting explained to us that creating a void, by way of lifting using underpinning, is to intentionally disrupt the soil-foundation connection, which will indeed adversely affect the seismic and lateral resistance of the overall foundation system. This is less critical on flat lots in flat subdivisions that are not subject to seismic loading or some peak ground acceleration that would otherwise significantly impact the existing structure, however it becomes a discussion point when trying to discuss whether this soil-foundation connection must be resupported, regardless of whether the foundation beam and column system that has been built is implemented.

The first challenge presented by DSA and OPSC is that the deep foundation system will carry significant loading into the soil, however the seismic resilience of the structure will take place at the connection of the underpinning brackets to the footing, and to the slab on grade therein. What this means, impactfully, is that your standard design of dropping some epoxy-anchored bolts or Titen Kwikset anchors in two of the four attachment brackets AT THE FOOTING is technically very insufficient in terms of eccentric loading and seismic resilience. The minimum-approved discussion point during this crux of the conversation was a minimum of 4 bolts, corrosion resistant by nature (whether hot zinc dipped, galvanized, or whatnot), embedded a minimum of 6” into the existing concrete footing, proper procedures of cleaning and prepping, were a base line minimum that almost still was not accepted. The transfer of load onto the piering system supported axial loading with the External Sleeve to do its thing in resisting the bending moments during installation of a push pier, but the void created after the lift presented significant problems since the soil-foundation connection was severed at the moment of lifting (not to mention there was concern regarding the damages that may be resultant of the hydrostatic pressure anchoring down the footing until it broke loose!).

During design, we’re looking at various sections of the IBC, mostly revolving around Chapter 18 for this discussion. Section 1810.2.1 discusses the Lateral Support of the foundation elements, and this is where much of the derived phrasing “10 feet of unrestrained fill” comes from. It explains that a “column supporting a beam”, or, “where ‘deep foundation elements’ stand unbraced in air, it shall be permitted to consider them laterally supported at 5 feet (1524mm) into stiff soil or 10 feet (3048mm) into soft soil unless otherwise approved by the building official on the basis of a geotechnical investigation by a registered design professional.” (2021 IBC 1810.2.1). This is where we assume that a caisson support system has a minimum embedment into competent strata/bedrock of 5’ (“stiff soil” by definition), which provides the anchor you’re looking for to support that system. It is also the assumption made of 10’ of unrestrained fill (“soft soil” by definition) for the majority of steel underpinning options presented. This is why your repair options during settlement issues should always be, at minimum, 10’ deep.

Interestingly enough, when exploring 1810.2.2, specific instructions are provided to consider “stability” in a pile cap, and even at that, allowances are provided for engineering justification for eccentricity and lateral forces. This becomes a discussion point for structural engineers with regard to the piering-foundation beam connection, which could be used to justify or omit void filling operations, if the math adds up. It immediately runs into 1810.2.4 to discuss lateral loading on shafts, which is where a push pin system would typically find a failure, and the soil-shaft connection is nonexistent in the upper 10’ of unrestrained fill, as well as in the resulting void after a lift. With respect to these parameters, micropiles and caissons are much easier to be defined as a sufficient repair than steel underpinning processes. However, steel underpinning has been around since 1838, and written into the IBC/CBC for over 100 years (as helicals), and even Push Pier systems have been found accepted all over the world for the last 35 years.

Hollow stem-auger piering (1810.4.8), socketed drilled shafts (1810.4.9), micropiles (1810.4.10), all utilize Grade Beams to tie into the existing footings, to strengthen and enlarge those footings in an effort to provide additional reinforcement. This makes the lateral and seismic resilience of those repair processes significantly easier to justify, although the steel underpinning systems, while providing better lifting opportunities, are a regularly approved methodology for mitigation differential settlement.

IBC 1807.3.2 discusses design criteria for embedded posts and piles, which becomes a discussion point for raised floor foundation systems but still provides justification for lateral and seismic loading of foundation elements. 1807.3.2.1 discusses unconstrained fill and 1807.3.2.2 discusses constrained fill, both elements of which suggest that once you’ve created a void space and disrupted the soil-foundation connection, you’ve created an unconstrained fill issue that requires attention. While I’ve seen much of this discussion in relation to interior posts on a raised floor foundation system, I have also seen it used as a structural discussion with respect to pouring a helical/push pile into a foundation with a concrete slab. If a void has not been filled, then the opening below the slab would be present and create a higher chance of cracking at the slab once service loads are applied.

IBC 1803.5.11 and 1803.5.12 begin to explore the wonderful world of California Seismicity. It’s not actually just California (given it’s the IBC not the CBC), but this is where it gets tricky and some assumptions have to be made. The geotechnical requirements for seismic classification zones C, D, E, and F dictate specific geotechnical analysis of Slope Stability, Liquefaction, Total and Differential Settlement, and Surface Displacement due to Faulting or Seismically Induced Lateral Spreading or Flow. 1803.5.12 furthers this in zones D, E, and F, in that a geotechnical exploration should explore significant further investigation into potential failures of slopes and retaining walls over 6 feet, the effects of downdrag and liquefaction, lateral soil loads and potential lateral movement on the foundations, and significant ground improvement methods available. Introducing seismic design into a simple voluntary remedial home leveling operation could put significant time and monetary constraints on a homeowner, for a project never intended to resist these items to begin with. By restoring soil connections to the foundations, you are eliminating the very idea that your project is intended to become a seismic retrofit or hillside stabilization project, which eliminates many of these requirements up front.

?“The basis of Ground Improvement is that the project…when completed…will have soil that is stronger and more stable than it was before you started.” (Naranjo, Engineering Services and Design of SoCal, Inc). Whether the void filling operation is in relation to injection of a cementitious grout, cement sand-slurry, structural polyurethane, or other approved alternative, the fact remains that filling a void beneath structural elements to restore soil connections to the foundation will always have a positive effect on the overall foundation system. During permitting, an argument is often made that “restoring” this connection, after “restoring” proper grade on the foundation, is simply “restoring” the design elements that were already present prior to the settlement failure. By this design, we can argue that “the repair system is not designed to either strengthen nor inhibit the seismic or lateral resistance of the structure.” (Saleem, Alpine Engineering). This discussion point often circumvents most of the correction items that may be issued by a Plan Checker or Building Official, as the process of leveling a foundation and addressing axial compression loads only would not take into account subsidiary failures – only as long as proper void filling operations are completed.

?

Void Filling Means, Methods, Materials

Now that we’ve explored the “why” one would void fill on an underpinning job, let’s discuss “what” and “how”. There is much discussion to be had on means and methods for void filling, and one of the safest explanations is to void fill with “cementitious grout, cement-sand slurry, or other approved alternative” (Steven C. Helfrich, Helfrich and Associates). The operative phrase here being “other approved alternative”, which becomes its own discussion point. Void filling operations, especially under slabs, should be completed after the lifting of a structure to maximum practical recovery (in this instance, we continue to evaluate a 3” lift of the structure’s foundation, resulting in a 3” void under the footing, as well as interior slab).

Void filling operations is the process of closing the open space between the soils and the foundations with some form of structural connection. This is most easily achieved with a flowable fill, such as a cement-sand slurry or a cementitious grout. This process is most easily achieved on a slab-on-grade foundation system following a lift, so that the structural fill can be “injected” from the inside, through the slab, until the material flows freely out of the excavated underpinning holes, essentially guaranteeing the voidspace between the slab and the opening of the underpinning hole has been effectively filled.

The process of void filling can be done with a Lightweight Cellular Concrete, a Cementitious Grout, a Cement-Sand Slurry, a Structural Polyurethane Foam (High Density Polyurethane Foam or HDPF), or some other means as approved by an Engineer.?Lightweight Concrete is generally fairly cost prohibitive, but the lightweight nature of the concrete makes it ideal for not disturbing any of the soils below it while still providing structural value to void filling operations. Cementitious grouts have the tendency to be injected via high pressure, which often results in disturbing underlying soils, and the nature of injected concrete/cementitious grouts makes it difficult to control flow and travel of the material. Careful observation of materials used, and surrounding landscapes and hardscapes should be maintained during injection of cementitious grouts. One advantage of using concrete is being able to predict final strengths of the concrete and to specify minimum compressive strengths and other elements. Disadvantages being it is generally heavy, affecting underlying soils inadvertently, and any moisture in the ground causes longer cure times.

Injection of HDPF allows for a low injection pressure, and the filling of the voids with a material whose strength will increase when it finds confining pressure, however that strength is difficult to categorize up-front. Based on this, it is generally accepted to take the polymer’s physical properties in its free-blown state, which is to say in its weakest state, and use that for engineering calculations. Generally speaking, a good HDPF?in its free-blown form will still greatly exceed the strengths of the soil, making it a viable repair option. The hydro-insensitive nature of HDPF will permit the expulsion of moisture from under the slab, adding to the many benefits of this application, and in an ideal scenario will also provide an insulative waterproof barrier below the slab, assuming coverage and travel functions as is intended (Rex Klentzman, URETEK USA, Inc.).

One major distinctive comparison between concrete and HDPF for void filling purposes is providing proof of increased shear strengths between the geopolymer matrix and the concrete that comprises the footing. Courtesy DSA and OPSC for this element of engineering, the concern is that despite restoring the connection between the soil and concrete footing, the shear strength between polymer and concrete, which is negligible at best, cannot be measured in a way so as to support seismic design parameters. In order to properly reinforce a footing for seismic resilience and lateral resistance, concrete elements would need to be doweled and anchored into the existing footing, as opposed to HDPF injection technology for void filling purposes. Thus, the intent of increasing seismic resilience and mitigation liquefiable soils conditions would then be subject to a different injection design, intended to strengthen weak soils while displacing water within the soil, which is another entirely different function and discussion of HDPF injection.

?

Conclusions

While designing a foundation leveling project with steel underpinning creates “good” engineering with appropriate Factors of Safety, sound engineering would call out void filling operations to restore the soil-foundation connection, which will provide ancillary benefits of subverting seismic and lateral design requirements on project not intended to either strengthen or inhibit those loads.