Instrumentation and Monitoring: How Piling Pros Tracks What's Happening Underground

If the shoring wall next to your building moved a little overnight, how would you find out? Probably you wouldn't, at least not until a door stopped closing right or a crack showed up in the drywall. Instrumentation and monitoring is the fix for that. It's a collection of sensors, gauges, and survey points that somebody reads on a schedule, turning movement into a number you can check while there's still time to act.

Piling Pros puts these programs together and runs them on excavation, foundation and underpinning work around South Florida. Most jobs down here end up needing some version of one, since the ground is soft and the water is never far below it.

Table of Contents

The gear: what gets measured

Settlement is the simplest place to start. You fix a plate or a survey point to whatever you're worried about, shoot its elevation every so often, and subtract. When the number keeps shrinking, that thing is sinking.

Sideways movement takes something cleverer, and that's the inclinometer: a flexible tube grouted into a borehole, with a probe you lower down the inside to find out how far the tube has tilted at each depth. You'd think that would be hard to get wrong, but TRB's circular on inclinometers spends quite a few pages on exactly how it gets gotten wrong. If the casing was grouted badly, or somebody forgot to correct the probe readings, the data will say a wall is moving when it isn't, and an engineer ends up making calls off numbers that are bad. That's worse than having no instrument at all.

For groundwater you use piezometers, standpipe or electronic. Of everything on this list, water is the thing we'd watch hardest in South Florida, because a wall designed around one water level doesn't care that the site is delivering a different one.

With a neighbor close by, vibration gets its own transducer, set on their foundation while the piles go in, recording the shaking as it happens. You hear about a problem that same afternoon instead of reading about it in a letter from their lawyer six months later.

Load cells and strain gauges go on the structure itself: tiebacks, struts, piles. They're how you check that the load your engineer calculated on paper is the load the member is carrying once it's in the ground. Pile load testing is a different route to the same question, for foundations nobody is watching all the time.

And if a spec ever lists an instrument your crew has never touched, Montana DOT's geotechnical manual covers most of them. It's long and you won't read it front to back, but it's useful for looking things up.

Why it matters more in South Florida

Soft soil is the first reason. It squishes and shifts more than the stiff ground you'd find on a lot of sites out west, and it's harder to predict, so if you're digging deep close to somebody's building you have a pretty good reason to measure.

The second reason is water. It sits a few feet below grade across most of the region, so dewatering, or just excavating, can change groundwater conditions fast enough that a piezometer reading from last week is out of date by Monday. Throw in storm season, when one heavy rain can shift pore pressure and loads within hours, and you can see why a monitoring plan here gets read more often than one built for calmer ground.

That costs money, and it shows up in the budget. Better to see it coming than to get surprised by a line item.

Can't the super just look?

Sure, and he or she should keep doing it. A super who walks the hole every morning catches things.

The trouble is that eyes only pick up movement once it's big enough to see, and with a slow foundation problem that may be weeks after it started. Instrument readings get compared with a baseline and with trigger levels that were written down before anybody dug. A walkthrough gets compared with what the wall looked like last Thursday in somebody's head, which isn't a great reference.

A historic building next door, a utility line that can't take much, or tight settlement tolerances in the engineer's plans: any of those, go with the numbers.

Terms Worth Knowing Before a Monitoring Proposal Lands on Your Desk

There's a bit of vocabulary you'll run into the first time a monitoring proposal shows up, so here's the short version.

The baseline is the first set of readings, taken before any work starts, and everything after gets compared against it. Without one there's nothing to compare to, and that gets painful the day a neighbor says the crack was your fault. Many projects also photograph the neighbor's existing cracks before work begins, so the baseline is easier to defend.

Trigger levels are the limits, usually set in two tiers. Cross the first and the engineer takes a look at the data. Cross the second and work stops until somebody says it's safe to keep going. Set them too tight and the whole team spends a month chasing false alarms, and set them too loose and you've paid for a program that never tells you anything.

The reference point is the stable spot, outside the area that's going to move, that your survey readings get measured from. The word doing the work there is stable. Put it on something that's settling too and every number measured from it is off, and you won't know until the data stops adding up. Lock-off load checks on a tieback wall follow a similar idea, a known starting number that later readings either confirm or contradict.

Then there's ADAS, short for automated data acquisition. The sensors send readings straight to a computer so nobody has to walk around with a clipboard. Bigger or riskier jobs use it as a matter of course, and plenty of smaller ones do fine reading things by hand a few times a week.

When you don't need all of this

Not every excavation deserves a full program, and paying for one you don't need is a mistake too. A small, shallow dig with nothing sensitive nearby can get by on regular visual checks. If the settlement tolerances are loose and the neighbors are sturdy, survey shots every so often might do the job a continuous sensor would do. Dense, stable soil with a deep water table cuts the scope way down as well, although you don't see that combination much in this part of the state.

What the reports should look like

Readings that land in a folder nobody opens are worthless, so ask up front what you'll receive and how often.

A decent program shows you readings over time with the trigger lines drawn on the same chart, so you can see a trend heading toward a limit before it gets there. It also says who gets the phone call when a limit is crossed, and how fast. If a proposal doesn't answer those two things, ask.

How Piling Pros sets it up

The instrumentation plan gets built right next to the construction plan, and not bolted on once somebody remembers there's a building next door. What goes in the ground, and how often it's read, comes down to three things: the condition of the neighboring structures, the soil and groundwater data, and how touchy whatever's nearby is.

Underpinning leans on this more than most work does. You're reinforcing a structure one section at a time, and the only way to show the rest of it stayed put is to have been measuring the whole time.

Once readings start coming in, they're checked against the trigger levels set on day one, and anything creeping toward a limit gets flagged before it gets there. On jobs built around protecting adjacent structures during urban construction, that record is often what shows a neighbor's movement, if there was any, stayed inside the range everyone expected.

Where it goes sideways

The big one is skipping the baseline. Without it you can't say whether the movement was already there before you showed up, and that gap has cost contractors disputes they would have won.

Next comes instruments that get installed and never read, leaving you with a cable in the ground and nothing else. After that, trigger levels picked without the engineer of record, so you get false alarms all the time or silence until something cracks.

The last one is the reference point nobody ever re-checks. It shows up late, right when the readings stop making sense and someone has to work out why.

Frequently Asked Questions

How often are readings taken?

It depends on what's going on at the site. The baseline happens once, before work starts. During active excavation or pile driving near a sensitive structure you might read daily or continuously, and quieter phases can drop back to weekly.

Who decides the trigger levels?

The structural or geotechnical engineer of record, since they know how much movement the neighboring structure or the wall design can take before a reading goes from normal to worth stopping for.

Do I need this on every excavation?

No.

Small, low-risk digs get by on visual inspection and the odd survey check. A formal program earns its cost where there are sensitive neighbors, tight tolerances, or soil that makes movement more likely.

What happens if a reading goes over a limit?

Depends on the tier. A first-tier exceedance usually means the engineer reviews the data and readings get more frequent. A second-tier exceedance typically stops the affected work until somebody figures out what's happening and whether it's safe to go on.

Is monitoring only for deep holes?

Not at all. Foundation work and underpinning near existing buildings get monitored too, because the question is how close the neighbors are, and depth is only part of that.

What to Do Next

If something sits close enough to your dig that movement would matter, get monitoring sorted out before the first shovel goes in. Send Piling Pros your project details and we'll work through what your site calls for.