These two terms appear in the foundation specification of almost every project, yet many parties still use them interchangeably. The confusion can be expensive — for example, ordering a working pile to be tested at test-pile load levels so that the pile can no longer be used, or using the result of a single test pile to confirm the quality of every pile on the project without further sampling. This article separates the differences point by point, from test load level, timing and number of piles to what to do when a result fails. For an overview of all pile testing methods, see Pile Testing Services

What Are Test Piles and Working Piles?

A test pile is a pile installed specifically for testing, usually before production piling starts, to confirm that the design assumptions are correct — for example, whether the design length actually gives the calculated load capacity, and whether the bearing stratum for the pile toe is at the expected level. Most test piles are not counted as load-bearing piles of the building and are usually placed outside the foundation layout or where they do not affect the structure

A working pile is a pile that will sit under the building and carry real load throughout its service life. Testing these piles is a quality spot-check of the installation work, verifying whether the piles actually built on site perform close to the design, so they must be tested in a way that does not damage them

The diagram below sums up this difference in one view. The cards on the left compare the roles of the two kinds of pile, and the graph on the right shows that the working pile stops at the proof level, while the test pile continues until settlement increases rapidly

Comparison of testing a test pile up to failure with proof-testing a working pile, which stops before failure On the left are two cards. The upper card is the test pile, loaded with a jack to about 2.5 to 3 times the design load or until failure behavior is seen, done before production piling to verify the design. The lower card is the working pile that will carry the real building, proof-tested to about 2 times and then stopped; it must never be tested to failure. On the right is a conceptual load–settlement graph: the test-pile line continues until settlement increases very rapidly at failure, while the working-pile line stops at the proof level and is then unloaded with partial rebound. The graph shows the concept, not actual test results Test Pile vs Working Pile: how far to load Not to scale Test Pile Load to ~2.5–3× design load or until failure Done before piling to verify design / length Working Pile Proof ~2× (200%) design load, then stop Never test to failure it carries the building Concept: load–settlement 1× design 2× 3× 0 Load → ← Settlement Proof range Test range Stop at proof load Unload, partial rebound Continue until rapid settlement = failure Test pile Working pile Unload
Concept of the two types of test: the working pile stops at the proof level in the specification, while the test pile continues until the behavior in the range where the pile begins to fail is seen (the graph shows the concept, not actual test results)

How Do Test Piles and Working Piles Differ?

They differ in 6 respects: purpose, test load, timing, number of piles, testing to failure, and further use under the building. The table below summarizes values commonly found in practice

← Scroll to see the full table →

Test pile vs working pile in practice
Issue Test Pile Working Pile
Purpose Verify the design assumptions; determine the actual behavior and load capacity of the pile Check the quality of installation — whether the piles actually built meet the design
Common test load About 2.5-3 times the design load About 2 times (200%) the design load
When Before production piling starts, so the design can still be adjusted During or after piling, per the sampling plan
Common number About 1-2 piles per zone with different soil conditions The percentage stated in the project specification
Can it be tested to failure? Yes — it is often intended to reach or approach failure to observe the behavior No, because it must go on carrying the building load
Can it be used as a building pile? Generally not counted, because it has already been loaded beyond the service level Yes, as normal, if the test result passes the criteria
⚠️ The "how many times" figures are common values, not legal requirements

The 2.5-3 times and 2 times figures in this article (overall, static test loads are in the range of 2–3 times) are values commonly found in project specifications and general practice; they are not figures fixed by Thai law. The values actually required follow the specification, drawings and requirements of each project, as set by the design engineer

The 2D drawing below is a section comparing the kentledge test setup and the test load levels for the two pile roles. The 3D view that follows is a project layout showing where test piles are placed, how many are done per soil zone, and how working piles are sampled for testing

2D model (side section) comparing static load testing of a test pile and a working pile The left view shows a test pile installed outside the foundation layout before production piling starts. Its kentledge system carries more counterweight because it is tested up to or near failure; the common test load is about 2.5 to 3 times the design load. The right view shows a working pile in a group beneath the future pile cap and building column, drawn as dashed lines; one pile is randomly selected for a proof test, with a common test load of about 2 times, and it must never be tested to failure because it will carry the real building. A bar beside each view compares the test load with the design load. The figures are common values; actual values follow the project specification Side section (2D) Not to scale 1× 2× 3× Test load (× design load) 1× 2× 3× Test load (× design load) Test Pile Before production piling · outside footings Working Pile Sampled per spec · under actual footing Test ≈ 2.5–3× up to onset of failure Not a building pile Proof ≈ 2× (200%) Never test to failure If passed, carries the building Pile cap (future) Bearing stratum
2D model (side section) — left: the test pile is outside the foundation layout and uses more counterweight so it can be tested up to the onset of failure · right: a working pile in a group beneath the actual pile cap is proof-tested and must never be tested to failure. The side bars compare the test load with the design load; the 2.5–3 times and 2 times figures are common values in project specifications, and actual values follow what the design engineer specifies (Image: SPN Soil Engineering)
3D model of a project site layout showing test pile locations by soil zone and working piles sampled for testing beneath the building's pile caps 3D view of a project site with two soil zones. Zone A on the left has a shallower sand layer than zone B on the right, as shown by the soil layers on the cut front face and by boreholes BH-1 and BH-2. A dashed line on the ground surface marks the building footprint, inside which pile caps are arranged in a grid. Test piles TP-1 and TP-2 are outside the building line, one per zone, each with a kentledge system, and are done before production piling. Working piles sampled for testing in the proportion stated in the specification are marked with orange rings, and one pile cap has a small kentledge for a proof test 3D model Not to scale BH-1 BH-2 T TP-1 T TP-2 A: shallow sand B: deep sand Borehole in each zone Different strata = new zone Building footprint Pile caps on a grid 1–2 test piles/zone Outside building · done first Sampled working piles Orange ring · % per spec Proof test under actual cap Never test to failure Sequence 1 Soil boring → zoning 2 Test TPs before piling 3 Adjust pile length/design 4 Install working piles 5 Sample tests per spec
3D model — the project site is divided into two zones based on the soil investigation results; zone A has a shallower sand layer than zone B. Test piles TP-1 and TP-2 are outside the building line, one per zone, done before production piling, while working piles sampled for testing in the proportion stated in the specification are marked with orange rings (Image: SPN Soil Engineering)

Why Are Test Piles Loaded Harder Than Working Piles?

Because they answer different questions. A test pile must find out how much load this type of pile in this soil can carry at most before settlement increases rapidly, whereas a working pile only has to prove that it can carry the load level set in the specification. The ultimate capacity can only be obtained by loading up to or near the point where the pile begins to fail. If the test stops at the service level, you see only the early, almost linear part of the curve, which does not tell you how much reserve capacity the pile has

Once the actual ultimate capacity is known, the design engineer can check back whether the factor of safety used in the design is appropriate. On some projects, test-pile results allow the pile length to be reduced, and the savings on piles can outweigh the cost of testing. On others, the piles turn out to need more length — and it is far better to know before work starts than after half the project has been driven

The method used for most test piles is the static load test per ASTM D1143/D1143M-20e1 and DPT 1251, because it gives the load–settlement relationship directly. A test load of 2.5-3 times the design load requires a kentledge (counterweight) or reaction pile system that is correspondingly larger than for working-pile tests. For the test procedure and how to read the results, see Static Pile Load Test: Testing Pile Capacity by the Static Method

How Far Can a Working Pile Be Tested, and Why Never to Failure?

A working pile may be tested only up to the proof level set in the specification, commonly about 2 times (200%) the design load. This is called a proof load test: the load is raised to the specified level, held for the specified time, and then the total settlement and the residual settlement after unloading are checked against the acceptance criteria. The result is a "pass" or "fail" at that load level, not the ultimate capacity of the pile

The reason for never testing to failure is straightforward. When a pile is pushed until the soil around it and beneath the toe is destroyed, the pile settles permanently and its load capacity drops, even if the concrete itself has not cracked. That pile can then no longer support the building, and usually new piles must be added and the pile cap at that location revised as specified by the design engineer — a cost nobody planned for

For projects that need to check many working piles without the time and space a static test requires, the popular alternative is the dynamic load test (PDA) per ASTM D4945 and DPT 1252-51, which takes much less time per pile. For a comparison of the advantages and limitations of the two methods, see Dynamic Load Test vs Static Load Test: What's the Difference?

When Should You Test, and How Many Piles?

Test piles should be completed, with results in hand, before production piling starts; the common number is 1-2 piles per soil zone. Working piles are sampled for testing during or after piling in the proportion given in the specification. Doing test piles before work starts means the results can still be used to adjust the design. If they are done after the production piles have been driven, the results serve only as supporting documentation and no longer help with decisions, which defeats the main purpose of the work

A "zone" means an area where the soil investigation results differ significantly — for example, where the sand bearing stratum lies at different depths, as shown in the 3D view above. On a large site where the soil investigation results clearly differ from one corner to another, a single test pile should not be used to cover the whole site

For working piles, the number to be sampled depends on the proportion stated in the specification and the test method chosen, as each method has different sampling criteria. For a summary of sampling criteria by method, see How Many Piles Need Testing?, and for the price range of each method, see Pile Testing Prices 2026

If the Test Result Fails, What Should Be Done Next?

The first step is not to find someone to blame but to isolate where the problem lies, by reviewing three sets of data together: the installation record of that pile, the soil investigation results nearby, and the pile integrity test results. If the driving record is abnormal for that pile only, the problem is usually in the installation. But if several piles give similar results that are below the design, it usually signals that the design assumptions or the soil data are off

Common remedies range from reducing the design load assigned to each pile, adding piles per pile cap, and lengthening the piles in the problem zone, to changing the foundation system. The choice of remedy must be made by the design engineer, and the load assigned to each pile must comply with the relevant requirements, such as the Ministerial Regulation on foundations B.E. 2566 (2023). For the step-by-step investigation when a static load test fails, see Static Load Test Failed: What to Do

⚠️ This article provides general information

This content explains the difference between test piles and working piles for general understanding only. The test load level, the number of piles to be tested and the acceptance criteria for each project must be set by the design engineer and stated directly in that project's specification