# Your Tank Foundation's Bearing Capacity Check Passed and Still Isn't What Differential Settlement Will Do
Author: Nitin Verma
Author URL: https://insights.spans.co.in/author/nitin-verma
Published: 2026-11-16
Meta Title: Bearing Capacity vs Differential Settlement Tank Design | Spans
Meta Description: A tank foundation's bearing capacity and uniform settlement are correctly checked against the geotechnical report. Neither confirms the differential settlement that cyclic filling and emptying, and layered subsoil, actually produce.
URL: https://insights.spans.co.in/tank-foundation-bearing-capacity-vs-differential-settlement-wastewater

A clarifier, aeration tank, or equalisation tank at an [ETP](https://spans.co.in/effluent-treatment-plant-etp/) or [STP](https://spans.co.in/sewage-treatment-plant-stp/) gets its foundation checked against two numbers from the geotechnical report: [bearing capacity](https://en.wikipedia.org/wiki/Bearing_capacity), the soil's resistance to shear failure under load, and settlement, the amount the foundation is expected to sink as the soil compresses. Both checks are done correctly, against the loads the structural drawing specifies and the soil profile the borehole data describes. Passing both is a real result. It does not confirm the tank will settle evenly across its own footprint over a service life spent being filled, emptied, and refilled thousands of times, because uniform settlement and [differential settlement](https://en.wikipedia.org/wiki/Differential_settlement) are different questions, and the second one depends on conditions the original check was never built to capture.

![Cross-section of a two-cell tank: cell A, frequently full, settles further over a softer soil layer than cell B, frequently empty, over firmer soil, cracking the shared wall where the settlement diverges](https://spans.co.in/images/blog/tank-foundation-bearing-capacity-vs-differential-settlement-wastewater-inline.png)

## A correct settlement estimate and an even settlement are different claims

Standard geotechnical practice checks bearing capacity and settlement as two separate criteria, and for a foundation as large as a tank mat or ringwall, settlement almost always governs the design rather than shear failure. That settlement figure is typically calculated for the tank at its design load, full, uniformly distributed, a single, defensible number. What it does not model by default is what happens when a multi-cell tank has one chamber full and the adjacent one empty for maintenance, since the stress each chamber transfers to the soil beneath it is genuinely different in that condition, or what happens when the subsoil itself is layered rather than uniform, with a weaker, more compressible pocket under one corner of the footprint than another. Both conditions push settlement toward being uneven across the tank rather than uniform, and a foundation check run once against a single, full-tank load case has no mechanism to flag either of them.

## Why the single settlement number keeps standing in for the real question

A geotechnical report's settlement estimate is a defensible, documented figure produced once, from borehole data and a defined design load case, and reviewing it is straightforward: check the number against the structural engineer's allowable limit and move on. Differential settlement from cyclic operation is a different kind of problem entirely. Published research on tanks subject to repeated filling and emptying describes measurable additional settlement building up over the life of the structure, with one study on tanks over clay noting the extra settlement depends on the number of load cycles as well as the soil's consistency, and a tank can see well over a thousand fill-empty cycles across its service life. A separate case study on granular soil found that cyclic filling and emptying produced permanent settlement that exceeded the original design estimate, and noted plainly that this residual deformation is commonly left out of routine design altogether. None of that shows up in a single settlement calculation run once against one load case, because that calculation was never asked the cyclic-loading question in the first place.

This is the same shape of mistake we described in [the piece on pump and blower foundation mass design](https://insights.spans.co.in/pump-blower-foundation-mass-design-wastewater): a foundation sized correctly against the load case it was given can still fail the condition that actually governs performance, because the number that got checked and the number that actually matters answer different questions. It is close kin, too, to [the piece on underground tank flotation and uplift](https://insights.spans.co.in/underground-tank-flotation-uplift-wastewater): an uplift check run correctly against a single assumed groundwater level says nothing about what happens once the real water table moves away from that assumption, the same gap between a design-basis number and the actual, independently varying site condition it was meant to stand in for.

Picture a two-cell equalisation tank on a site with a layered subsoil, a firmer stratum under one half of the footprint and a softer, more compressible one under the other, information that was in the borehole logs but didn't change the single settlement figure calculated for the tank at full, uniform load. In operation, the two cells are rarely full together, maintenance on one while the other stays in service, seasonal flow variation, routine desludging, so one side of the tank is repeatedly more heavily loaded than the other over thousands of cycles. The softer stratum consolidates further under that uneven, repeated loading than the uniform-load calculation ever predicted, and the tank develops a genuine tilt and cracking at the shared wall between cells, years after a foundation design that passed every check run against it. Nobody re-opens the geotechnical report at that point and finds an error. The number was correct for the case it was asked to solve, which was never the case that actually happened on site. Repairing a cracked shared wall and underpinning a tilted foundation after the fact costs far more than the extra geotechnical and structural review would have, and that gap between a cheap design-stage question and an expensive retrofit shows up just as clearly once it is set against any realistic [project cost](https://spans.co.in/wastewater-treatment-cost/) estimate for the rework involved.

An industry benchmark from steel tank design, API 650's differential settlement tolerance of roughly thirteen millimetres across ten metres of tank circumference, illustrates the kind of magnitude the industry treats as consequential for a tank shell, even though that specific code governs above-ground steel tanks rather than buried RCC structures and the right tolerance for any given concrete wastewater tank has to come from its own structural design, not from a figure borrowed across tank types.

## What actually checking for differential settlement requires

The fix is not re-running the bearing capacity or settlement calculation more carefully. Both were correct. It is asking a second, separate question at the geotechnical stage: given the actual layering in the borehole logs and the actual operating pattern, multi-cell tanks running unevenly loaded, cells taken out of service for maintenance, repeated full-empty cycling over the design life, where is settlement likely to be uneven rather than uniform, and does the structural design have enough flexibility, movement joints, flexible pipe connections, tolerance in the shared walls, to absorb that difference without cracking. That is a question for the geotechnical and structural engineers to answer together before foundation design is finalised, not something a single settlement number from one load case was ever going to answer on its own, and it is exactly the kind of review worth building into early-stage structural design through [Spans' structural engineering team](https://spans.co.in/contact/) rather than discovering the gap in a cracked shared wall years into operation.


---
This blog is powered by Superblog. Visit https://superblog.ai to know more.
---

