Polyurethane crack injection

What the resin is actually doing inside the crack

Water in the crack is a reactant, not an ingredient. A hydrophobic polyurethane consumes the moisture it meets in order to cure and carries none of it into the finished polymer, which is why the cured foam stays dimensionally stable through the wet and dry cycling a Queensland basement wall goes through every single year. That is the whole engineering case for it, and it is a good one.

Here is the part that rarely makes it onto a brochure. Bond to a wet substrate is genuinely poor. The film of water on the crack face acts as a bond breaker, so what you end up with is not an adhesive joint. It is a mechanical plug - foam keyed into the roughness and the taper of the crack and held there by its own expansion against the walls of it.

Everything else follows from that one fact. It is why the geometry is not negotiable: a 13 mm hole at 45 degrees, offset from the crack line by roughly half the element thickness so it meets the crack at mid-depth, drilled to two-thirds or three-quarters of the way through and never out the far face, is what gives the plug something to key into across the full thickness of the wall. It is also why surface ports glued to the face of a wet wall are close to useless. They put foam on the concrete rather than through it, the surface seal will not adhere to wet concrete, and there is no adhesion holding any of it in place.

The second consequence is movement. Polyurethane foam is semi-rigid to flexible - it is not an elastomer. It absorbs the small seasonal opening and closing of a shrinkage crack comfortably. It does not survive continued cyclic movement, which tears the plug open, and no choice of resin fixes an element that is still moving. If movement is uncertain, tell-tales read over 4 to 12 weeks cost a fraction of injecting the same crack twice.

Hydrophobic or hydrophilic - the decision that sets up the whole job

Both are MDI-based polyurethanes and both cure on contact with water. The difference is what happens to that water afterwards, and it changes where each one belongs.

Free-rise is a laboratory number. Confined expansion is 3 to 8 times.

Technical data sheets quote free-rise expansion, commonly 8 to 30 times for hydrophobic grades and up to 1000 per cent for some soil-stabilisation products. Free-rise means the foam was allowed to expand in an open container with nothing holding it back. It is a fair measurement of the chemistry. It is simply not the number that applies inside a wall.

In a confined crack under back-pressure, field practice assumes roughly 3 to 8 times. That is a factor of four against the data sheet, in the direction that hurts. A contractor who estimated resin volume from a free-rise figure runs out of material mid-afternoon, and then either drives back to the yard or quietly stops short of the top of the crack and hopes.

So never estimate resin volume from a free-rise figure on a TDS. Volume is estimated from crack width, element thickness and length, with a budget set per packer before injection starts and a hard stop when a packer takes more than it should. That stop matters: a packer swallowing litres on a 200 mm wall is not filling a crack, it is filling a void or a drainage layer behind the element, and no quantity of resin will seal the face from there. A sensible hard limit on a suspect packer is one to two litres, then move on and have a different conversation about what is behind the wall.

Gel time is set on site, not in the factory

Polyurethane arrives as a single component and its speed is dialled in on site with an accelerator, dosed at roughly 1 to 10 per cent by volume. That range buys anything from about 20 seconds to several minutes before gel, and the choice is a judgement about flow rather than a preference.

A gushing crack under head needs a very fast gel, in the 20 to 30 second region, or the resin is simply washed out of the crack before it can react. On a genuine gusher the flow gets knocked down first with oakum packing or a hydraulic cement plug so the resin has a fighting chance, then the section is injected properly.

A fine, damp crack needs the opposite. Minutes of working time, so the liquid travels the full length of the crack and the full thickness of the wall before it stiffens. Dose that one fast and you seal 200 mm either side of the packer and nothing else - which looks like a completed job on the day and is a leak in March.

Temperature moves the number as well. A dose that behaves at eight in the morning is noticeably quicker by two in the afternoon in an unventilated car park in February, and the crew adjusts through the day rather than mixing to a recipe.

Isocyanates, in plain English

Uncured polyurethane resin contains isocyanates. In plain language: the chemical that makes the resin set is a respiratory sensitiser. It can cause an asthma-like reaction in people who breathe the vapour or mist, and once a person is sensitised they are sensitised for life. There is no growing out of it, and afterwards they react at levels that are perfectly safe for everyone else.

Australia's workplace exposure standard for isocyanates is 0.02 mg per cubic metre averaged over eight hours, with a short-term limit of 0.07 mg per cubic metre over fifteen minutes. Those are limits for the people doing the work, and what they mean on a site is specific: air-supplied or correctly fit-tested cartridge respiratory protection, chemical-resistant gloves and eye protection, forced ventilation in an enclosed basement or lift pit, a current safety data sheet on site for every product, spill and decontamination provisions, and health monitoring for technicians with ongoing exposure.

Drilling for the packers releases respirable crystalline silica as well - exposure standard 0.05 mg per cubic metre over eight hours - which is why the drills run with on-tool extraction or wet, particularly in an enclosed car park or a lift pit where dust has nowhere to disperse to.

For everyone else in the building, the position is much simpler. Once the resin has reacted, which happens within minutes, it is an inert plastic and it is not hazardous. During the work there is a chemical smell that clears with ventilation. We ask that people, pets and food are kept out of the immediate work area while injecting and until the space has been aired, which is normally the same day. Safety data sheets are available for every product used, and a building manager or committee should be asking for them as a matter of course rather than as a favour.

Where polyurethane is the wrong resin

PU is the default for leak sealing, not the answer to everything. These are the situations where reaching for it is a mistake, and what belongs there instead.

• The crack is dry, dormant and the intent is to restore strength. That is epoxy - 60 to 90 MPa cured compressive strength, with bond exceeding the tensile strength of the parent concrete. Foam cannot transfer load and never will.
• The path is a designed movement joint. Filling one with foam defeats its purpose and pushes the movement into the concrete beside it. Backing rod and a movement-capable sealant.
• The element is still moving. Foam absorbs small seasonal movement and is torn open by continued cyclic movement. Monitor first, then decide.
• The defect is a void network rather than a line - honeycombing, an unfilled block core, a large cavity. PU is expensive by volume, contributes nothing structurally, and can disappear into it without ever sealing the face. Microfine cement grout, specified by d95 in microns rather than by the word microfine, is the right fill.
• The crack is finer than PU will enter. An acrylate gel at 10 to 50 mPa.s - about the viscosity of water - permeates where a foam bridges over.
• There is no defined line at all. Broad-area seepage through the body of a wall or slab gives the foam nothing to grip. That is a tanking, drained cavity, curtain injection or excavation conversation.
• The element stays permanently wet and hydrophobic was quoted anyway, or it dries out permanently and hydrophilic was quoted. Both are the wrong grade of the right resin, and both fail slowly enough that nobody connects it back to the decision.
• The cracking is corrosion-driven, following the reinforcement with rust staining or drummy areas. Sealing over expanding steel hides a problem that keeps growing underneath it.
• Anything in contact with potable water without a current AS/NZS 4020 certificate for that specific product. Certificates run five years. Get the certificate, not the rep's assurance.

What a polyurethane job looks like on a quote

Bands below are indicative South East Queensland figures excluding GST, for reading a quote against rather than as a quote. The most useful line is the first one: a short single crack prices as a minimum job, not at a linear metre rate, because attendance covers travel, set-up, the pump and plant and a minimum block of labour whether the crack is one metre long or four.

The loadings are listed separately on purpose. Confined space, traffic management, after-hours penalty and making good are the four line items most commonly buried in a lump sum, and burying them is what makes competing quotes impossible to compare.

What it costs

Indicative only, and excluding GST. Build a line-by-line estimate for your own job with the price book, or read the working behind the rates in the 2026 cost guide.

Questions we actually get asked

Will the foam expand hard enough to crack my wall?

No, and the reason is the number most people have wrong. Free-rise expansion of 8 to 30 times is what happens in an open bucket; inside a crack under back-pressure the real figure is roughly 3 to 8 times, and that expansion is what keys the plug into the crack rather than something that levers the concrete apart. Where damage genuinely can occur it comes from injection pressure, not expansion - a thin precast panel of 150 to 200 mm, a core-filled block wall whose mortar beds blow out before the block does, or a thin slab under hydrostatic uplift. Those get low pressure, and on slender or prestressed elements an engineer nominates the maximum in writing before the pump starts.

The crack is dry at the moment. Can you still use polyurethane?

Yes, and we flush clean water through the packers first to make it work. Polyurethane cures by reacting with water, so a bone-dry crack has to be pre-wet or the resin has nothing to react with. The flush does two other useful things at the same time: it proves the crack is hydraulically connected from one packer to the next, and it clears out drilling dust, laitance and the white salt deposits that would otherwise block penetration. The one time flushing is wrong is ahead of an epoxy injection, because epoxy needs the crack dry and still.

One contractor quoted hydrophilic and another quoted hydrophobic for the same crack. Who is right?

Possibly both, which is unsatisfying but true. There is a real, unresolved disagreement in this industry: one camp argues hydrophilic should always be used because it bonds to damp concrete and chases the leak into fine paths, the other argues hydrophobic should always be used because it does not shrink when the wall dries out. Both camps are commercially motivated. The question that actually decides it on your crack is whether that element will ever dry out permanently. A wall that leaks through the wet season and is bone dry by August wants hydrophobic, because a hydrophilic foam in there will desiccate, shrink and let the water back through. A planter box or a tank wall that never dries wants hydrophilic or an acrylate gel. Ask each contractor which of those two your element is, and see whether the answer is about your wall or about their drum.

How much resin should my job take, and is there any way to sanity-check it?

On a 200 mm element with a crack of ordinary width, packers take modest quantities and those quantities should be broadly consistent along the run. A packer that took four litres did not fill a crack - it filled a void, an unfilled block core or a drainage layer behind the element, and that is a different repair with a different price. Good practice sets a budget per packer before injection starts and a hard stop around one to two litres on a suspect one, then moves on rather than pumping a drum into the ground. If a quote was priced on resin volume, ask what expansion figure was assumed: 3 to 8 times is the real confined number, 8 to 30 times is the data sheet.

Can we paint, render or tile over the repair afterwards?

Yes, once the patching has cured - typically 24 to 48 hours, and longer for a cement-based patch that needs to dry out before painting. One thing to know in advance: nothing bonds to cured polyurethane, so if any resin has bled out onto the face during injection it has to be ground off before any coating goes on. Tell us the intended final finish before we start, because it changes how the packer holes are cut back and filled. And treat the render and paint reinstatement as a separate priced item rather than assuming it is folded in.

How long do we need to stay out of the space?

During injection and until the area has been ventilated, which is normally the same day. The reason is the uncured resin, not the finished one: isocyanates are a genuine respiratory hazard for the people handling them, which is why the technicians wear proper respiratory protection and we run forced ventilation in an enclosed basement or lift pit. Once the resin has reacted, which takes minutes, it is an inert plastic and it is not hazardous. There is a chemical smell during the work that clears with ventilation. We can provide the safety data sheet for every product on site, and in a strata or commercial building the manager should be asking for them as routine.