Expansive Errors

The lessons we’ve covered in this long sequence of articles have typically revolved around single, key errors and have generally called for commonsense (and often simple) remedies. In the world of pool construction, however, there are situations in which the problems are far more complex, often rising from multiple missteps and clusters of intertwined failures.

This is one of those situations, and it has to do with a basic pool/spa combination in a brand-new housing development. Although the pool contractor charged only $35,000 for the installation, the associated legal fees and corrective measures shot up beyond $500,000 – and the story isn’t quite over.

Before I get down to specifics, let me start by observing that projects with budgets at such a low level tend to make contractors look for ways to cut corners to increase their margins. This would explain some of the errors discussed here. But there’s also the possibility that the pool contractor was inexperienced and unaware of the impact that soils and the actions of other contractors can have on how a project performs.

Whichever is actually the case here, the cascade of errors were such that it has been difficult to untangle all of the various causes and effects, particularly amid a legal process in which the attorneys, geologists and engineers for both sides seemed determined to muddle and confuse the facts.

I am happy to say that my only role in all of this was as an expert witness on the side of the homebuilder. Indeed, I take great comfort in knowing that while my own firm might occasionally make mistakes, we’ve never been directly involved in anything as awesomely awful as this.

Scenario: The legal proceedings started when the homebuilder was sued by the homeowners, who contended that he’d left them with “bad soil” in their backyard – and that this bad soil had caused their new pool to sink. The homeowners pursued that line because the pool contractor, whom they’d contacted first with their complaint, had assured them he was blameless and immediately pointed a finger at the homebuilder.

The case had gone back and forth for about two years by the time I was brought in. When I finally saw the backyard, my first observation was not that the pool had sunk, but actually that the decks had heaved. As I dug in and reconstructed what had happened, I was utterly and constantly amazed by the number of errors I was uncovering – and by how they were all fitting together.

I believe it all started with the fact that the homeowners, in attempting to save money, had acted as their own general contractor for the exterior work and had hired separate companies to complete the pool construction, the decking and the landscaping. It is clear that the couple didn’t really know what they were doing – and that none of the contractors they’d hired (both licensed and unlicensed crews) had been willing to step up and help them out.

As might be expected in a project run by an unguided homeowner, a number of mistakes were made:

[ ] Mistake #1: Most buyers in new developments fail to notice language in the transfer documents declaring that if they want to build any sort of additional structures on the property, they must obtain soils reports. (In most cases, homebuilders only certify the ground extending for five feet around a home’s primary structure.)

Armed with such a report, an engineer can work with the pool design to accommodate the specific soil conditions found on site. In this case, however, the homeowners did not obtain a soils report – a serious error on its own, but only the first of many.

[ ] Mistake #2: In new housing developments, it’s also common for the soils to be bone dry – basically because the landscaping has yet to be installed and nobody has begun irrigating to any useful or consistent degree.

Even in well-established yards – but especially in new developments – a pool contractor should always moisture-condition the soil before excavation. Some, however, will skip this process because it can take three or four days to saturate dry ground adequately. (With clay soils, which describe the terrain in which these homeowners wanted to place their pool, it’s also advisable to use a wetting agent or detergent to break the surface tension of the water, thereby speeding its absorption.)

These photographs tell part of the tale: The soil surrounding the excavated pool is clearly bone dry, with no water migration from other lots or any evident groundwater source for saturation of any kind (left). Note as well the ramp for excavator access: It was dug off the shallow end of the pool – below and behind where the skimmer was eventually placed – and the backfilled soil was never properly compacted. The homeowners asserted that the pool had sunk into the ground because of ‘bad soil’ the builder had used. In point of fact, however, the pool had mostly stayed put: It was the decks that had heaved – as seen here, by more than a quarter-inch (right).

The idea behind this moisture-conditioning is simple: The added water makes the soil expand before excavation begins, which is important because having that expansion occur later, after structures are in place, can cause serious damage. Nobody thought to do that here – a second big mistake.

In addition, the photographs I saw clearly indicated that there was no groundwater present – something that may have led to another bad omission. Also, shots of the shallow-end area where the excavation ramp had been located showed that the soil had not been compacted. This is important, because moisture will travel through looser soil more readily than it will through compacted soil. This was not necessarily a mistake on its own, but as we’ll see below, it proved to be another contributing factor to a much larger problem.

[ ] Mistake #3: The engineering details for the pool were based on a standard plan and operated with the assumption of non-expansive soil, which turned out to be the wrong choice in the case of this clay-rich backyard.

In that standard plan, however, the bond-beam detail called for three #4 bars. Based on what can be seen in the construction photos of the steel shell, it was evident that the bond beam had been built with #3 bars instead, following an outdated detail that was inappropriate for these soils. In fact, the whole shell was formed using only #3 rebar, and the spacing of the bars was too far apart to comply with what an expansive-soil detail would require.

Once the decking was removed, we could tell that it had been of generally adequate thickness – but that it hadn’t been installed above an adequate layer of non-expansive sand or gravel (left). So when the soil beneath the slab expanded, there was no buffering material – and the deck had nowhere to move but up. Examination of the construction photos also shows that the steel in the bond beam was inadequate for a shell built in these site conditions (right). The engineering detail had called for three #4 bars; #3 bars had been used instead. The spacing of the bars in the shell is also incorrect for these soil conditions.

With no soils report, some degree of basic, reasonable caution on the part of the pool contractor should have led him to think in terms of building a stronger shell; for whatever reason, however, he chose to go with a weaker option – another big mistake.

[ ] Mistake #4: In our forensic work on this project, we discovered that the skimmer had been installed improperly. In a standard detail, a skimmer is wrapped in steel and then encased in concrete, meaning it is fully integrated into the pool shell and moves with it.

The engineering detail shows how the skimmer should have been installed – that is, wrapped in steel, encased in concrete and isolated from the deck with felt or a foam expansion joint. The pool builder had gotten the first two points, but the patio contractor didn’t isolate the deck, creating a second anchor and a source of stress cracking and a large leak.

In this case, the pool builder did all that, but then failed to install an expansion joint around the skimmer that would allow it to move independent of the surrounding deck – in effect anchoring the skimmer to both the shell and the deck. Another major error.

[ ] Mistake #5: The construction photos revealed that no hydrostatic-relief valve was installed. Such a device will open and allow groundwater to flow harmlessly into the pool, usually at the main drain, in the event groundwater exerts enough pressure beneath the pool to move it (generally when the pool is emptied for service).

While this might never have been an issue given the apparent lack of local groundwater, in this case, given the absence of a soils report verifying surrounding conditions, installing such a device would have been a cheap and sensible insurance policy. And as we shall see, the pool contractor’s failure to install a hydrostatic-relief valve eventually contributed to bigger problems.

[ ] Mistake #6: During our investigation, we removed a portion of the concrete deck beyond the back edge of the pool and found that the slab was about three-and-a-half-inches thick. That’s generally acceptable, but we also found that there was no non-expansive material below the slab: Where we should have found a layer of sand or gravel that would allow the slab to shrink and expand and move around, instead we found direct contact with expansive clay soil that would swell when wet.

As water from the leaking skimmer penetrated into the clay soil, the material expanded and gradually lifted the flatwork. In this case, incidental surface water was also part of the picture, which was more problematic than it might have been otherwise because the skimmer had been placed in poorly compacted soil where the excavation ramp had been (left). This led to a great deal of water collecting in the soil in an active zone at the base of the shell. So in the case at hand, the pool did not sink; in fact, the pattern of cracking of the shell shows that, if anything, it was beginning to float.

In a properly engineered project, the soils report would indicate how much non-expansive material to install below the slab. For construction over a moderately expansive clay soil, a typical requirement would be for two to four inches of sand (or gravel and sand). But here, there was nothing at all – another error.

[ ] Mistake #7: Soils scientists often talk about “active zones” and “inactive zones.” An active zone is one in which, if moisture becomes present, swelling of expansive soils will occur. By contrast, an inactive zone lies beneath that level and is a place where the soil moisture is relatively constant and will remain stable.

One of the reasons we isolate pool structures from decks is that the pools are often anchored in inactive zones, while their decks sit atop active zones. (Obviously, the more expansive the soil, the more critical this distinction becomes!)

In this case, we learned that, in the pool area, the active zone extended down approximately four feet. Unfortunately, the shallow-end excavation reached down just three feet, meaning the deep end was in an inactive zone while the shallow end was in an active zone. This created a differential in the expansive forces influencing the pool structure – a mistake with major implications, as we’ll see below.

***

Let’s regroup here and roll through this laundry list of errors, tracing them through the twists and turns of the legal case as it unfolded.

In a short time, the housing development filled with families – and not just our homeowners but also their neighbors all began irrigating their yards. Within weeks, the soil in the vicinity of the pool we’ve been discussing became fully saturated and, naturally and inevitably, began to expand. Given the fact that the soil had not been saturated before construction began, and given the fact that the decks were poured directly atop that expansive soil, the decks heaved upward.

In the initial lawsuit, the homeowners testified that the concrete decks had heaved upward shortly after installation and that the accommodating pool contractor had come in, removed the coping, built up the top of the bond beam and raised the coping so that it was now level with the heaved deck.

The brick steps had not been isolated from the bond beam – in direct contradiction of the structural detail. As a result, the brick started cracking where it came in direct contact with the shell.

The contractor had described this move as a “good faith” gesture and a means of addressing problems related to the “bad soil.” To my skeptical eye, however, it looked as though he was covering his tracks as best he could, hoping the situation would stabilize and that this would be the last he’d hear from these homeowners.

But this wasn’t the end: Because the skimmer had not been isolated from the deck as it should have been, when the deck heaved it pulled the skimmer upward and cracked the skimrner box, resulting in a severe leak. The builder repaired the leak, but only after water had poured out of the gap for several days and had thoroughly saturated a large area around the shell’s perimeter.

Making matters worse, the skimmer had been positioned within the sweep of the non-compacted excavation ramp. This exacerbated the movement of the water around the shell’s shallow end – and as discussed above, no hydrostatic relief valve had been installed to mitigate the pressure. In effect, the loose soil near the skimmer became the water’s path of least resistance, fully saturating an area around the shallow end – a space founded entirely in an active zone.

All the while, the pool contractor and the homeowner were saying that “bad soil” had caused the pool to sink. This seldom happens, of course. In fact, rather than sink, pools far more commonly float, which is the result when hydrostatic pressure builds up around the shell.

So here we have the shallow end of the pool surrounded by expansive, non-compacted soil being rapidly saturated by a leaking skimmer anchored to a heaving deck. The pool shell began to lift and float in the shallow end. Eventually, this stressed the pool shell, which finally began cracking on the hinge point marked by the skimmer.

Ultimately, the excessive saturation of the soil surrounding the pool – caused by the leaking skimmer in the shallow end and abetted by water penetration of areas beneath the decking – had the effect of stressing the shell and making it float, heave, twist and, finally, crack at several stress points around the shell.

These cracks told the whole story: Had the soil beneath the pool been bad enough to cause it to sink (as the pool contractor and the homeowner claimed), one would expect to see a sinkhole effect on the bottom of the pool – something that can occur, for example, when pools under construction are ineffectively mucked out after heavy rains.

In this case, however, it was clear that the site’s expansive soil had caused the concrete decks to heave – and that the supplemental saturation caused by the leaking skimmer had so thoroughly stressed the shell that it began to float before cracking, heaving and twisting.

In many ways, this project failed because too many cooks were involved and ended up spoiling the broth. It had been built without any consideration for soil types. The deck contractor hadn’t layered any non-expansive material below the decks and had failed to saturate the base before pouring the concrete – which certainly didn’t help. But the pool contractor had also failed to moisture-condition the soil before and after the excavation, had used a steel schedule that ran contrary to the plans, hadn’t included a hydrostatic relief valve and had failed to isolate the skimmer properly from the deck.

Once the soil became saturated, the decks heaved. Once the decks heaved, the skimmer leaked, the soil became supersaturated and the pool shell lifted and cracked as a result of hydrostatic pressure and differential forces of the soil.

Lesson Learned: Just as it is said that someone who represents themselves in court has a fool for a client, the same can very often be said of homeowners who act as their own general contractors. If you as a professional see this situation – and if you’re unwilling or unable to step up and explain what the construction process is all about – it is probably better, particularly as a watershaper, not to get involved at all: Finger-pointing and grief too often follows these projects!

Scott Cohen is a construction defect expert witness and president of The Green Scene, an outdoor design/construction firm in Chatsworth, Calif. Past articles in the Lessons Learned series have been compiled in his book, The Candid Contractor. He also provides consultation for clients nationwide and gives seminars on designing landscapes, swimming pools and outdoor kitchens. For more information, go to www.greenscenelandscape.com.