Does it actually work, and for how long?

The short answer, and the caveat that carries all the weight

Crack injection works. Resin injected properly fills the crack through the full thickness of the wall and stays there. It does not degrade, it is not consumed, and it does not need renewing. On the specific path that was sealed, the repair is permanent for practical purposes.

The caveat is that injection seals a path. It does not stop water existing, remove the water pressure behind the wall, or fix whatever is delivering the water to the wall in the first place. Every honest assessment of whether injection will work is therefore an assessment of the diagnosis, not of the material.

That is why the industry's failure rate is a diagnosis problem dressed up as a materials problem. When people say injection did not work, the overwhelming majority of the time the resin did exactly what it was supposed to do, in a crack that was not the reason the building was wet.

What permanent actually means here

Cured polyurethane in a crack is a closed-cell plastic. It does not oxidise away, it is not eaten by the alkalinity of the concrete, and short of the crack being physically pulled open again it stays where it was put. Cured epoxy in a dry, dormant crack is stronger than the concrete around it, with bond strength exceeding the tensile strength of the parent material, which is why a correctly epoxied element cracks beside the old crack rather than in it if you keep loading it.

So the material is not the limiting factor. Three things are, in descending order of how often they cause trouble: whether the water path was correctly identified, whether the resin was actually delivered into the crack rather than onto it, and whether the resin selected suits what that crack does across a year.

Those map onto the three failure modes below. The first two are workmanship and judgement. The third is a genuine materials trade-off that both sides of the industry argue about, and it deserves to be said out loud rather than buried.

Failure one: the diagnosis was wrong

This is the big one, and it accounts for more returned calls than the other two combined. Water was coming through the wall, a crack was visible, the crack was injected, and the crack was never the reason the building was wet.

The classic version in South East Queensland is the undercroft or basement wall with a vertical crack halfway up it. The crack gets injected. Three weeks later the floor along the base of the wall is wet again, because the actual path was the kicker joint where the wall meets the slab, and always was. The crack was passing a little of the water and the joint was passing most of it.

The other common versions are all situations where the concrete was never the problem: a silted-up subsoil drain, weep holes rendered over during a paint job, a driveway that falls toward the building, a downpipe discharging at the footing, a failed sealant joint on the deck above, or a leaking service that nobody pressure-tested. Sealing the wall in any of those cases raises the water level against the outside of it and pushes the leak to the next weakest point.

There is a version of this that is worse than wasted money. If the crack follows the line of the reinforcement and shows rust staining, spalling or drummy patches, the cause is steel corroding and expanding. Injecting seals the crack while the steel keeps growing. That is a concrete repair scope, not a waterproofing one.

• Water appearing along the base of the wall rather than at the crack, before or after the repair, points at the kicker joint.
• Water that only appears after several days of rain rather than during a shower points at head building up behind the wall, which points at drainage.
• Broad-area damp with no defined line was never an injectable defect at all: that is a tanking, drained cavity or groundwater management problem.
• Rust staining or spalling along the crack line means corrosion, and injection will hide it rather than fix it.
• Damp that shows no correlation with rain at all, in a still, enclosed, humid space, is very often condensation.

Failure two: surface ports on a wet crack

Surface ports are packers glued to the face of a wall over the crack, with a bead of stiff paste capping the crack in between, injected at low pressure of roughly 0.7 to 3.5 bar. They have a legitimate place: thin sections that cannot safely be drilled, low-pressure structural epoxy work, and cracks too fine or too shallow to drill into.

They are the wrong method on an actively leaking wall, for a reason that has nothing to do with opinion. The surface seal is an adhesive bond to the face of the concrete, and the face of an actively leaking wall is wet. Adhesives do not bond through a film of water. The seal either fails to stick at all or blows off as soon as the pressure comes up, and the resin ends up on the wall instead of in it.

Even when it appears to work, what you have is resin in the mouth of the crack rather than through the wall. That will hold in dry weather and against a slow weep. It will not hold against head. This is why the pattern of a cheap repair that lasts one dry winter and fails at the first serious wet season is so common, and it is entirely predictable from the method.

The comparison is not subtle. Mechanical packers are drilled into the concrete, seal inside the wall body against the sides of the hole, take 35 to 140 bar, and deliver resin into the crack at mid-depth. That is the method that produces a repair through the wall.

Failure three: hydrophilic resin in a crack that dries out

This one is not workmanship. It is a materials trade-off, it is real, and it is the failure mode the industry is least willing to discuss in front of clients.

Hydrophilic polyurethane is built on a polyethylene-oxide-rich backbone, which means water is not just the trigger for the cure, it becomes a permanent component of the finished polymer. That is what makes the material so good at what it is good at: it seeks water out, chases the leak into micro-cracks and capillaries a hydrophobic foam would bridge straight over, and bonds tenaciously to damp concrete rather than merely keying into it.

The price is that if the crack dries out permanently, the entrapped water evaporates and the foam shrinks. That is desiccation shrinkage, and the leak comes back. The theory is that it re-swells when the water returns, commonly quoted at two to four times. The practice is less reassuring: shrunk foam can lose its key in the crack and never fully recover the seal it had, because a plug that has pulled away from the crack faces does not re-seat itself just because it got wet again.

Where does that leave a wall in Brisbane or on the Gold Coast? It depends entirely on whether the crack stays wet. A planter box, a tank wall, a lift pit sitting in groundwater or a basement below the water table never dries out, and hydrophilic is defensible and often the better choice there. A retaining wall or undercroft that streams in February and is bone dry by July is being asked to survive a full drying cycle every single year, which is exactly the condition that produces this failure. Hydrophobic foam contains no water in its structure and is dimensionally stable through that cycle.

If your repair works every wet season and fails again after every dry winter, and the sealed line itself is what is passing water, resin selection is a live suspect.

If yours failed, here is how to tell whose fault it was

It costs nothing, it takes an afternoon, and it settles the question with evidence instead of argument. What you need is a record of what was actually sealed: photographs before and after taken from the same positions, and a written description of what was found and what was done. Any competent injection contractor produces that as a matter of course, because it is their own defence in a dispute.

Work through these in order.

1. Ask for the record of the workPhotographs before and after taken from the same positions, a written description of what was found and what was done, and the resin type and product used. If no such record exists, that on its own is most of your answer: a contractor who cannot demonstrate what they sealed cannot show that the work was done to any standard, and cannot argue that a new leak is a different path.
2. Mark and measure where the water is appearing nowPhotograph it, and measure it off a fixed reference point such as a corner, a column or a door frame. Not "the middle of the wall". A dimension. Do this while it is actually wet, ideally during or straight after rain.
3. Overlay the twoPut the new wet point against the photographs of the line that was sealed. Everything that follows depends on this single question: is the water coming out on the line that was sealed, or somewhere else?
4. On the sealed line, between packers, within one wet seasonThis is a workmanship problem. The most likely causes are packer spacing that was too wide for the crack, or resin that never travelled because the crack was not hydraulically connected and nobody flushed to find out. Ask what the spacing was and whether the crack was flushed before injection.
5. On the sealed line, with glued ports and a paste bead visible on the faceThe method was wrong from the start, not just executed poorly. Surface ports on an actively leaking wall cannot hold their surface seal. The correct comparison is not "they got unlucky" but "this was never going to work on a wet wall", and the repair needs redoing with drilled mechanical packers.
6. Off the sealed line, more than roughly a metre away, after the sealed line heldThis is very likely a new path, not a failure. Sealing from the dry side redirects water rather than removing it, so the head that used to escape at your crack now goes looking for the next weakest point. This is physics, not an excuse, and the honest version of it is stated in writing before the work rather than invented afterwards. Check whether it was.
7. On the sealed line, dry all winter and wet again every summerSuspect desiccation shrinkage of a hydrophilic resin in a crack that dries out annually. This is a resin selection judgement, which sits somewhere between a materials trade-off and a diagnosis error depending on how obvious the seasonal drying was. Ask which resin family was used and why.
8. Everywhere at once, broad damp with no defined lineIt was never a crack problem. Water is coming through the body of the wall or slab, which means general porosity, and there is nothing for injection to grip. The options are negative-side tanking, a drained cavity system, curtain injection behind the wall or groundwater management. Somebody sold the wrong scope.

Five things that look like a failed repair and are not

Not everything that is still wet after an injection job is a failure, and misreading these wastes money and burns a relationship with a contractor who did their job.

• A new leak two or three metres away on a wall sealed from the dry side. Redirected water is the expected behaviour of negative-side sealing, which is why a competent contractor puts that in writing before starting.
• A lift pit that leaks again from a different joint. The pit is the lowest point in the building and collects everyone else's water, so a new leak there is frequently a new source rather than a failed seal.
• Efflorescence continuing to appear for a few weeks after the repair. That is salt already dissolved in the concrete coming out as the wall dries. It proves water has been moving through, not that it still is.
• A wall that stays damp to the touch for weeks. Concrete holds a lot of stored moisture and gives it up slowly. This is exactly why moisture meter readings at the same marked points before and after are worth taking, so there is a number rather than an impression.
• A planter box that leaks again after an injection that was expressly sold as an interim measure with the outlet still blocked. That is not a failure, that is the disclosed limitation arriving on schedule, and the answer is the membrane replacement that was flagged at the time.

How long before you know it worked

Timing is diagnostic in itself. When a leak comes back tells you a great deal about why it came back.

Observation for at least 24 to 48 hours after injection is the minimum, and observing through an actual rain event is far better than any amount of dry-weather confidence. Where the geometry allows it, water test: pump a lift pit out and watch, fill and hold a tank against the leakage criteria in AS 3735, flood a planter box.

What to demand as proof, on any injection job

The reason so many of these disputes are unresolvable is that nobody recorded anything. Demand these and the question of whether a repair worked stops being a matter of opinion.

• Resin appearing at the next packer along during injection, which is the direct on-the-day confirmation that resin travelled through the crack rather than into a void.
• Observation for 24 to 48 hours minimum, and through a rain event where the weather cooperates.
• Moisture meter readings at the same marked points before and after, so there is a number rather than a feeling.
• A water test where the shape of the structure allows one: dewater and watch a lift pit, fill and hold a tank to the AS 3735 leakage criteria, flood a planter box.
• Photographs before and after taken from the same positions, together with a written description of what was found and what was done.
• A re-inspection 24 to 48 hours later and again after the next real rain event, rather than a phone call asking how it went.

If yours has failed, what to do next

1. Go back to the original contractor first, with the record and not an accusationA contractor who kept a proper record usually stands behind it, because the record is what lets them see in minutes whether it is their repair or a new path. Ask what changed rather than opening with a demand. If they cannot produce a record, you have learnt something important and you are better off starting fresh.
2. Work out where the new leak is before you get another quoteWalking into a second quote knowing whether the new wet point is on the sealed line or off it changes the conversation entirely, and it stops you buying a second repair for a problem the first one already told you about.
3. Consider paying for a real diagnosis rather than another free quoteA diagnostic investigation with moisture readings, crack monitors installed and read, water testing, scanning and a written report sits in the range of $450 to $1,500. On a wall that has already had one failed repair, that is cheap relative to a second failure, and it filters out anybody whose plan is to inject the same crack again and hope.
4. Fix the cause before you seal anything againIf the diagnosis comes back as drainage, a failed membrane, corrosion, condensation or ongoing movement, spend the money there. Injection over an uncorrected cause is a repair with a scheduled expiry date, whoever does it and however well.

Questions we actually get asked

The contractor says the new leak is a different crack. Is that real, or an excuse?

It is genuinely real, and it is also the easiest excuse in the trade, which is why a record of what was sealed matters so much. Sealing from the dry side redirects water rather than removing it: the head that used to escape at the point you sealed goes looking for the next weakest point in the wall. A new wet spot well away from the sealed line, appearing after the sealed line held, is consistent with that. A wet spot sitting on the sealed line, between two packers, is not. The way you tell the difference in minutes is by comparing the new leak against the before and after photographs of exactly what was sealed. If the contractor cannot produce that record, they are not in a position to make the argument.

Can a failed injection be re-injected, or does it all have to come out?

It can almost always be re-injected, and the old resin does not have to be removed. If mechanical packers were used and are still in place, the same packers can often be injected again. If the spacing was too wide, fresh holes go in between the existing ones. If surface ports were glued to the face, the whole set-out was wrong for a leaking wall and the job starts again with drilled packers, but the cured resin already in the crack mouth is not a problem, it just is not doing much. Cured resin on the face is a different matter: it has to be ground off before anything is rendered or painted, because nothing bonds to it.

The cheap repair worked for six months. Why did it work at all?

Because low-pressure injection through surface ports does put some resin into the mouth of the crack, and a partly filled crack mouth is enough to stop a slow weep in dry weather. What it cannot do is hold against head. When several days of rain put water pressure back on the outside of the wall, the pressure finds the unfilled part of the crack behind the plug and pushes through. That is why the pattern is so predictable: fine through a dry winter, wet again in the first serious wet season. The repair did not degrade, it was never a full-depth seal to begin with.

Should I go back to the original contractor or start with somebody new?

Go back first, unless the wall is showing structural signals or the original contractor cannot produce any record of the work. Two practical reasons. A contractor who documented the work can identify in minutes whether the new leak is on the sealed line, which is information you will otherwise pay for. And a second contractor quoting on a failed repair has an obvious incentive to condemn all of it. Take the record with you either way. If there is no record, treat it as a fresh job and get a proper diagnosis before anybody quotes.

Is it worth paying for an independent investigation before I get more quotes?

On a wall that has already had one repair fail, usually yes. A quote-only site visit is commonly free and gets you a price for the work the contractor wants to sell. A diagnostic investigation with moisture readings, crack monitors read over 4 to 12 weeks, water testing, scanning and a written report runs somewhere around $450 to $1,500 and gets you an answer. Where structural cracking is in play, an engineer's assessment is separate again, typically $1,200 to $3,500 or more. The report is also useful evidence if there is a builder's defect or insurance question attached.

Does a workmanship warranty on injection cover a leak that appears somewhere else on the same wall?

Honestly, no, and any warranty that claims otherwise is either not being read carefully or is not going to be honoured. A warranty on injection can cover the crack or joint that was sealed. It cannot sensibly cover a new path opening elsewhere, because negative-side sealing redirects water rather than removing it, and it cannot cover water finding a way in because a drain silted up or a membrane failed. That limitation is only defensible if it is disclosed in writing before the work and the contractor keeps a record precise enough to tell the two situations apart afterwards. If both of those are true, the exclusion is physics. If neither is, it is an excuse being invented after the fact.

Where to next

• our warranty in full, including the exclusions
• 6 and 12-month re-inspection results
• hydrophilic or hydrophobic resin
• when injection is the wrong fix
• work out what your leak actually is