Drive out of Juba in any direction after a week of rain and you will see the same thing: hairline cracks running diagonally from window corners, doors that will not close, plinths that have parted company with the walls above them. Almost none of that is bad blockwork. It is the ground moving.
What the soil is actually doing
Large parts of Juba, Rumbek and Wau sit on expansive clay — black cotton soil. The clay minerals in it draw in water and swell, then give it up in the dry season and shrink. Vertical movement of 40–80mm across a season is ordinary. On a bad site we have measured more.
The damage is rarely uniform. One corner of a building sits under a tree that drinks the moisture out of the ground; another sits under a downpipe that keeps it saturated. The foundation is asked to bridge two soils behaving in opposite ways, and the wall above it does the bridging instead. That is where the diagonal crack comes from.
Finding out before you dig
We open trial pits on every site before a foundation drawing is finalised — usually three to five, at least 3m deep, positioned at the building corners and one in the middle. What we are looking for is straightforward:
- The depth at which the black cotton stops and a competent stratum begins
- Atterberg limits on samples from each pit — a plasticity index above 35 tells us to expect serious movement
- The dry-season water table, which on sloping sites is often much shallower than the client expects
A trial pit costs a fraction of what remedial underpinning costs. We have never regretted digging one.
The four things that actually work
Excavate and replace. Where the black cotton layer is shallow — say under 1.5m — the cleanest answer is to take it out and bring in well-graded murram or hardcore, placed in 200mm layers and compacted properly. Properly means a plate compactor and a moisture check, not a gang with rammers on a Friday afternoon.
Take the footings past the active zone. The active zone is the depth to which seasonal moisture change reaches. In most of the sites we have worked in Juba that is 1.2–1.5m. Pad footings founded below it sit in soil that barely moves. This is why the pad excavations on our sites look deeper than clients expect.
Tie everything together with ground beams. Isolated pads on expansive clay still move relative to each other. Reinforced ground beams spanning between the pads turn a set of independent footings into one stiff frame, and the frame redistributes differential settlement instead of transmitting it into the walls. On our G+1 and G+7 work in Juba this is standard, not an upgrade.
Keep water away from the building. The cheapest structural intervention on an expansive-clay site is a working drainage plan: a paved apron at least 1m wide falling away from the wall, downpipes discharging into a channel rather than onto the ground, and no planting within 3m of the foundation. Half the movement problems we are called to inspect are drainage problems wearing a structural disguise.
What we do not do
We do not put a strip foundation straight onto black cotton and hope. We do not accept a 150mm blinding layer where the specification calls for 300mm of compacted hardcore. And we do not let a contractor backfill a pad excavation with the same expansive spoil that came out of it — which happens more often than the industry likes to admit, because the spoil is already on site and hauling murram costs money.
What it means for your budget
Designing properly for expansive soil adds somewhere between 8% and 15% to a substructure budget, and substructure is typically a fifth of the shell cost. Call it 2–3% of the building. Underpinning a cracked two-storey house four years after handover costs many times that, and you cannot occupy it while the work is going on.
If you are planning to build in Juba, Wau or Rumbek and nobody has mentioned trial pits to you, ask why.
OK Estate Construction and Supply Limited carries out foundation design and construction across South Sudan and Uganda. Talk to our team about a site investigation.