How Is Settlement of Adjacent Buildings Measured?
Settlement of adjacent buildings is measured with three types of settlement point, and each answers a different question. The first is the Surface Settlement Marker, a ground-surface settlement point installed by embedding a metal marker in the road, sidewalk or concrete slab around the construction site. The marker head is then leveled in rounds against a benchmark outside the work's zone of influence. The difference from the baseline is the ground-surface settlement at that location.
The second is the Building Settlement Point, fixed to the actual structure: the columns, footings or load-bearing walls of the adjacent building. The difference from surface markers matters directly for interpretation. The ground surface may settle because of loose fill or deteriorated pavement while the building itself, sitting on piles, has not moved. Reading both sets together therefore shows whether the settlement is in the ground surface or in the structure.
The third is the Multi-level Settlement Point, often called a Deep Settlement Point on site, which installs sensors at several levels in a single borehole. This shows which soil layer's compression is causing the settlement seen at the surface. It is used when you need to understand the settlement mechanism, not just monitor the numbers. See an installation example in Case Study: Deep Settlement Point, Bang Na km 5. A Settlement Plate is a different job: it is placed beneath fill to measure embankment settlement and is not used for adjacent buildings.
Installing these instruments is separate from the pre-construction condition survey. Recording existing conditions, preparing a crack map and taking photographs as evidence are part of a dilapidation survey, which is covered separately in Adjacent Building Survey Before Construction. This article covers instruments left in place to give continuous numerical readings throughout the works. For an overview of all geotechnical instruments, see Geotechnical Instrumentation Guide.
How Precisely Can a Tiltmeter Measure Tilt?
The tiltmeters SPN uses for building monitoring have a resolution of ±0.01 degrees, which converts to about 0.17 mm per 1 m of height (tan 0.01° × 1,000 mm). A 12 m tall building whose top moves about 2 mm out of plumb is therefore within the range a tiltmeter can detect. A movement that small is invisible to the eye, and an ordinary plumb bob cannot distinguish it from its own measurement error.
The sensor is fixed to a base plate anchored to a wall or column of the building and measures tilt from vertical on two perpendicular axes, so it shows both the magnitude and the direction of tilt. Tiltmeters have three advantages for adjacent-building monitoring: installation is done within a few hours without drilling a borehole, they can be connected to a wireless system and datalogger for automatic readings, and they can be set to send an alert immediately when a value exceeds the threshold without waiting for the next site visit.
Tiltmeters have three limitations. First, the sensor measures tilt only at the point where it is installed; it does not tell you how many millimeters the building has settled. Second, it does not tell you how the soil under the building is moving. If you need a lateral movement profile through the full depth of the soil behind a retaining wall, you need an Inclinometer, which is installed in a borehole and gives a displacement profile depth by depth. Third, the sensor is sensitive to temperature changes. Installing it where it alternates between strong sun and shade makes the plot swing on a daily cycle, which must be separated from the real trend during interpretation.
Settlement Point or Tiltmeter: Which Should You Use?
Adjacent-building monitoring should use settlement points and tiltmeters together, because they measure different components of the same movement. A settlement point tells you how much the building has settled, while a tiltmeter tells you how much it has tilted and in which direction. A building that settles uniformly usually suffers little structural damage, whereas a building that settles unevenly until it tilts or distorts is the case that causes cracking and needs serious monitoring. Having both sets of data therefore gives a more accurate interpretation than having only one.
← Scroll to see the full table →
| Topic | Surface Settlement Marker | Building Settlement Point | Tiltmeter |
|---|---|---|---|
| What it measures | Vertical settlement of the ground, road surface or slabs around the building | Vertical settlement of the building structure itself | Tilt of the structure from vertical, two axes |
| Installation location | Embedded in the road surface or slabs around the site | Fixed to columns, footings or load-bearing walls | Base plate fixed to a wall or column of the building |
| Typical resolution | Millimeter level, depending on survey order | Millimeter level, depending on survey order | ±0.01 degrees (≈ 0.17 mm/m) |
| Reading method | Level instrument; site visit required every round | Level instrument; site visit required every round | Automatic readings via wireless datalogger |
| Practical frequency | Scheduled rounds, e.g. daily to weekly | Scheduled rounds, e.g. daily to weekly | Continuous, down to hourly |
| Automatic alerts | No | No | Real-time alert thresholds can be set |
| Strengths | Shows the settlement trough around the excavation across the whole layout | Separates building settlement from ground-surface settlement | Sensitive to change; responds immediately during high-risk activities |
In practice, tiltmeters are usually placed on just a few buildings, the most sensitive ones or those closest to the excavation edge, while Surface Settlement Markers are spread in a grid around the whole project, because their cost per point is much lower and they show the full extent of the work's zone of influence. For an overview of the full instrument set used in deep excavations, see Deep Excavation Monitoring in Bangkok Soft Clay.
How Many Points, and Where?
The number and location of monitoring points are set by the design engineer in each project's monitoring specification, based on excavation depth, type of support system, distance to adjacent buildings, soil conditions from the soil investigation, and the sensitivity of each building. There is no ready-made number that applies to every job, but there are a few shared layout principles.
For Surface Settlement Markers, the principle is to lay them out in sections perpendicular to the retaining wall, with each section having several points at increasing distance from the excavation edge out to where the zone of influence is expected to end. This arrangement lets you actually plot the settlement trough and see how far the work's influence extends, rather than just knowing the value at a single point. There must also be at least one reference point outside the zone of influence to check that the benchmark itself has not moved.
For tiltmeters, the principle is to install them at the building corners facing the excavation, with at least two corners on the same building, because a single point cannot tell whether the building is tilting as a whole or twisting. The mounting level should represent the behavior of the main structure: fix the sensor to a sturdy structural member, not a masonry wall that can move on its own, keep it accessible for maintenance, and avoid spots in direct sun or exposed to impact from construction work.
The baseline is the initial reference value for every monitoring point. It must be read before any activity that could affect the building starts, and should be read at least two to three times to confirm that the initial value is truly stable and not still affected by the fresh installation. It should be done at the same time as the pre-construction condition survey, so that the condition records and the numerical data start from the same point.
The 3D view below combines the layout principles in a single scene. There are three things to look at: first, the building markers and the ground-surface markers laid out in a line from the excavation edge; second, the two tiltmeters at the corners of the wall facing the excavation; third, the line of sight of the level instrument tied to a benchmark outside the zone of influence. The right-hand box summarizes the values calculated from the readings. Settlement and tilt in the drawing are exaggerated for clarity.
What Are the Readings Used to Calculate?
Each round's marker level minus the baseline gives the cumulative settlement at that point. The engineer then calculates three comparative values.
- Differential settlement (δ) is the difference in settlement between two markers on the same building, such as points A and B in the figure.
- Angular distortion can be estimated as δ divided by the distance L between the two points. If the whole building tilts as a rigid body, the overall tilt must be subtracted first. This value is more closely linked to cracking than total settlement.
- Tilt (θ) is the angle by which the structure deviates from vertical. The tiltmeter reads this directly on both axes, and it can be used to cross-check the value calculated from markers on opposite sides of the building.
Acceptable values for all three are set by the design engineer according to the structure type and condition of each building.
How Often Should Readings Be Taken, and When Is a Reading Abnormal?
Reading frequency is tied to construction activity, not the calendar. The principle is to read more often during activities that change the stress state in the ground quickly: each stage of excavation, driving or boring piles near the site boundary, dewatering to lower the groundwater, and strut removal, when the retaining wall takes on new behavior. Frequency is then reduced once readings have been stable over several rounds. Tiltmeters connected to a datalogger can be read frequently without adding to the site workload, so they are usually set to record continuously throughout high-risk activities.
An abnormal reading is one that exceeds an alert level, which is set for each project by the design engineer; there is no universal number that applies to every building. Typically there are three levels: a watch level, a level requiring review of the work method, and a level at which activity must stop for investigation. The threshold values come from the designer's analysis together with the sensitivity of each adjacent building. An old building on shallow foundations will naturally have stricter thresholds than a new building on long piles.
What matters more than any single number is the trend. Engineers pay attention to four signals:
- A rate of change that is accelerating instead of slowing down
- Values that do not return to the previous trend after the causing activity stops
- Several points moving in the same direction at the same time, which indicates real behavior rather than measurement error
- Differential settlement between two points on the same building large enough to cause angular distortion, which is more closely linked to cracking than total settlement
Reading instruments alongside crack monitoring therefore gives a more complete picture. For how to read cracks, see Which Structural Cracks Are Dangerous?
Instrument data with a clear baseline and continuous readings helps the team make engineering decisions based on facts, and provides a traceable record of how adjacent structures behaved over each period. However, any claim or dispute is decided on several pieces of evidence taken together and on the judgment of those involved, not on a single set of monitoring data.
What Standards Does SPN's Service Follow, and What Does It Cost?
For settlement points, an international standard already exists: ASTM D6598-19 (Standard Guide for Installing and Operating Settlement Points for Monitoring Vertical Deformations), which covers assembling, installing, reading and maintaining settlement points, both those read by leveling from an external benchmark and those with an internal reference system extending down to stiff soil below the compressible layer (the same principle as a Multi-level Settlement Point). The standard is written mainly for settlement under fill or embankment loads.
For tiltmeters, there is as yet no ASTM or DPT (Department of Public Works and Town & Country Planning) standard that specifically prescribes the installation method, and Thailand does not yet have a DPT standard specifically for adjacent-building monitoring. Work quality therefore comes from three things:
- Leveling surveys follow precise-leveling practice, tied to a benchmark outside the zone of influence. See the scope of SPN's surveying services
- Sensor calibration follows the manufacturer's requirements, before installation and at regular intervals
- A project-specific monitoring specification defines monitoring point locations, reading frequency, alert levels and report format
When the project also installs other instruments, further standards apply, such as ASTM D6230 for movement measurement with inclinometers and ASTM D7764 for pre-installation acceptance testing of vibrating wire piezometers.
SPN provides an end-to-end service: planning the monitoring layout with the design engineer, installing settlement points and tiltmeters, reading the baseline, taking scheduled readings or connecting an automatic reading system with alerts, processing the data into movement-versus-time plots alongside construction activity, through to a report certified by a civil engineer licensed by the Council of Engineers Thailand (COE), with signature and registration number, suitable for submission to the relevant authorities. See the full scope of services on the Geotechnical Instrumentation & Monitoring page.
Fees depend on the number of installation points, instrument type, monitoring duration, number of reading rounds, and whether readings are manual or automatic, so they are assessed per project. Contact us for a free quotation within 24 hours. Prices exclude travel and equipment mobilization, which are charged additionally according to the distance to site.
This content explains the principles of settlement and tilt monitoring of adjacent buildings for general understanding. The number and location of monitoring points, reading frequency, alert levels and decisions to stop or modify work methods must be set by the design engineer and the responsible engineer according to the specific conditions of each project.