Negative side waterproofing — and when it is honestly the wrong answer

The physics, stated before the sales pitch

Water moves from high pressure to low pressure through whatever route offers least resistance. A crack, a construction joint or a form tie hole is a low-resistance route through the wall. Injecting resin into it removes that route. The pressure driving the water is untouched.

So the water does what water does: it takes the next route. On a well-built wall with one defect, there is no next route worth taking and the repair is permanent — this is the ordinary case and it is why injection works at all. On a wall with several marginal paths, sealing the worst one raises the flow through the others until one of them becomes visible. Clients who have been sold three separate injection repairs by three different contractors, each of which stopped one leak and started another, are not the victims of bad workmanship. They are the victims of a wall that needed an area treatment and got a line treatment.

This has three practical consequences, and they are the reason this page exists. First, the diagnosis has to establish whether you have a defined path or a diffuse one before anyone quotes — a distinction covered in full at /water-coming-through-concrete. Second, the scope has to name what is being sealed and what is not, with photographs before and after taken from the same positions and a written description of what we found and what we did, so that a leak appearing later can be identified as a new path or a failed repair within minutes rather than argued about for a fortnight. Third, a warranty on negative side work can honestly cover the crack or joint that was sealed. It cannot cover a path nobody sealed, and any contractor claiming otherwise is either not thinking or not telling you.

Positive side, negative side, and why the choice is usually made for you

Positive side is the wet face. On a below-ground wall that is the outside, buried against the soil. A positive-side membrane is the textbook answer because it keeps water out of the concrete entirely — the structure never gets wet, so the reinforcement inside it never sits in moisture, and the membrane is working in compression with the water pushing it against the substrate rather than trying to peel it off.

Negative side is the dry face — the inside, where you can stand. A negative-side repair works against the pressure rather than with it, which is why the seal has to be keyed into the body of the element rather than stuck to its surface. It is also why the packer type matters so much here. A mechanical packer is drilled into the concrete and locked in place by tightening a nut that compresses a rubber sleeve against the sides of the hole; it seals inside the wall and takes high pressure. A surface port is glued to the face over the crack with a bead of filler capping it. On an actively leaking negative face, a surface seal will not adhere to wet concrete and will simply blow off — the resin ends up on the wall rather than through it. On a wet basement wall, surface ports are a warning sign about the whole quote.

In remedial work the choice is rarely a free one. The building is already there, the wall is buried under a driveway, a lift shaft or a neighbour's boundary, and excavating to the positive side would mean removing landscaping, paving, services and often the access the building needs to keep operating. Negative side is chosen because it is the only side available, and the honest professional question is not whether it is ideal — it is whether, for this specific defect, it will hold.

Where negative side genuinely holds

Negative side sealing is a permanent repair, not a stopgap, wherever three conditions are met: the water is arriving by a defined path, the element has stopped moving, and nothing upstream is supplying water that should have been drained away instead.

In practice that means the following situations are ordinary, warrantable negative-side work.

On any of these, the seal is mechanical rather than adhesive — bond to a wet substrate is genuinely poor, because the water film acts as a bond breaker. The cured resin works as a plug keyed into the crack, which is exactly why the hole geometry matters: 13 mm holes drilled at 45 degrees, offset from the crack line by roughly half the element thickness so the hole intersects the crack at mid-depth, drilled to about two-thirds to three-quarters of the thickness and never through the far face.

• A discrete crack in a basement, undercroft or retaining wall, dormant, injected at 150–400 mm packer spacing.
• A construction or cold joint — the line where one concrete pour met the next — at 200–400 mm centres. Joints are wide and connected, so they take more resin than a crack of the same length.
• The kicker joint where a wall meets the floor slab. This is the real path in most leaking basements even when a crack higher up is the visible symptom, and it is injectable at 8–15 linear metres per crew per day.
• Form tie holes — a straight pipe right through the wall, the most direct water path in the building. Filled through the full wall thickness at $45–$120 per hole, with any remaining steel tie cut back 25–40 mm from each face first.
• A lift pit or tank wall, injected from the dry side so the water pressure presses the seal into the element rather than out of it, usually followed by a crystalline or cementitious system over the injected substrate.

When it lasts, and when it desiccates

The most common way a technically correct negative-side repair fails in South East Queensland is not workmanship. It is a resin chosen for the wrong moisture regime.

Hydrophilic polyurethane is water-loving: it seeks water out, chases the leak into micro-cracks and damp pores that a hydrophobic foam would bridge over, and bonds tenaciously to damp concrete. Water is a permanent component of the cured polymer. That is a genuine advantage on anything that stays wet all year — planter boxes, tank walls, structures below the water table. It is a liability on anything that dries out. If the crack dries permanently, the entrapped water evaporates, the foam shrinks, and the leak returns. It is supposed to re-swell on re-wetting, commonly quoted at two to four times, but a shrunk foam can lose its key in the crack and never fully recover.

Hydrophobic polyurethane uses water to cure but does not incorporate it, so the finished closed-cell foam is dimensionally stable through wet and dry cycling. It is the right call for a wall that runs from November to March and is bone dry by June — which describes a great many SEQ basements and undercrofts. Its trade-offs are real too: low elongation, so it is semi-rigid rather than elastomeric and will re-crack under continued cyclic movement, and it can blow out through the crack face or disappear into a void behind the element without ever sealing.

There is a genuine, unresolved disagreement in the industry here. Some manufacturers argue hydrophilic should always be used because of the bond to wet concrete; others argue hydrophobic should always be used because of dimensional stability. Both positions are commercially motivated and neither is settled science. We treat it as a judgement call on the specific crack, and the decisive question is simple: will this crack ever dry out permanently? If yes, avoid hydrophilic. If it stays wet year-round, hydrophilic or an acrylate gel is the better bet.

Acrylate gel sits outside both camps. It has essentially the viscosity of water, commonly 10 to 50 mPa·s, so it permeates soil, capillaries and hairline cracks that no polyurethane will enter, and cured it is highly elastic with elongation often in the hundreds of percent. It has zero structural strength, it must stay damp or several formulations shrink irreversibly, and its gel time is strongly temperature-dependent — a mix that behaves at 8 am can go off in the hose at 2 pm on a Brisbane summer day.

What the Australian standards actually say — and what they do not

This is where most of the market is either silent or wrong, so it is worth being precise. No Australian standard says negative side waterproofing is compliant. No Australian standard says it is non-compliant either. What exists is a set of standards covering adjacent territory, and a National Construction Code that sets performance requirements rather than nominating methods.

Read the middle column for what each document genuinely governs and the right-hand column for the claim it cannot support. If a report you have been given cites one of these standards for something in the right-hand column, the citation is decorative.

The Queensland warranty position

This is the part of the topic the market avoids, and it is worth understanding because it protects the owner and disciplines the contractor at the same time.

For domestic building work in Queensland, statutory warranties are implied into the contract by Schedule 1B of the Queensland Building and Construction Commission Act. They are not optional terms that a contractor can trade away in the fine print. A contract clause reading 'no warranty is given on negative side waterproofing' does not achieve what its author thinks it achieves — an attempt to contract out of a non-excludable statutory warranty is void, and the warranty applies regardless. If you have been handed a quote with wording like that, the wording is not protecting you and it is not protecting them either.

What does exist is a narrow and genuinely useful defence. Under Schedule 1B section 29(6)(a), a contractor is not liable for a breach where the work was carried out in accordance with the owner's written instruction given contrary to the contractor's prior written advice. The words that matter are prior and written. Advice given after the leak reappears is worth nothing. A verbal conversation on site is worth nothing. The advice has to exist, in writing, before the instruction, and the instruction has to be given knowing what the advice said.

That mechanism is the reason the redirection risk needs to be set out in writing, and agreed, before a negative-side scope proceeds. It is not a disclaimer and it is not an attempt to dodge a warranty we cannot dodge. It is the only structure Queensland law provides for an owner to make an informed decision to proceed with a compromise repair, and for both parties to know afterwards what was actually agreed.

Two related points. QBCC licensing applies to trade work valued over $3,300 including labour, materials and GST, so almost every job on this page requires a licensed contractor. Residential work over that threshold also triggers QBCC Home Warranty Insurance obligations, and the premium is the owner's protection, not an administration fee.

What goes in the written advice, before anything is drilled

Six things. It runs to about two pages, it is issued before the works, and it is signed and dated by whoever is authorising the job — owner, building manager, committee or builder. It is deliberately readable by someone with no trade background.

We hand the same document to a homeowner and to a body corporate committee. If it cannot be understood by the person paying for the work, it does not do the job Schedule 1B expects of it.

1. The diagnosis, and what was ruled outWhich of the leak paths we found, how we established it, and the alternatives we eliminated — drainage, a leaking service, condensation, membrane failure, movement. Naming what was excluded is what turns an opinion into a diagnosis.
2. The redirection statement, in plain wordsThat sealing from the internal face stops water entering at the defects we treat, and does not lower the water table, reduce pressure, or prevent water finding another path. No jargon, no acronyms, no small print.
3. Exactly what is being sealedElement by element, with linear metres, packer spacing, resin family and the pressure band for that element type. This is the scope a competing contractor can quote against, which is the point.
4. Exactly what is not being sealedAdjacent joints, tie hole grids, porous zones, the wall around the corner. Where a residual path is known and out of scope, it is named, and the consequence if water takes it is described.
5. The alternatives and their cost bandsIncluding the one we are not doing and why — drained cavity, crystalline tanking, curtain injection, or excavation and a positive-side membrane. If the compliant answer is more expensive than what is being instructed, the owner sees both numbers and chooses.
6. The instruction to proceed, signed and datedThe owner or their representative instructs the works knowing all of the above. That sequence — advice first, instruction after — is what Schedule 1B section 29(6)(a) requires, and it is the only version of it that has any effect.

The four alternatives, and what each is genuinely for

Where a defined-path negative-side repair is not the right answer, there are four options. They are not competitors so much as different tools, and on larger jobs two or three of them appear in the same scope.

Note what none of them do: none of them fix a drainage problem, a leaking service, or a structure that is still moving. Those get corrected first or every option below fails.

The order we work through it, and when we say no

On every below-ground enquiry we work down this list in order. It is deliberately arranged so the cheapest and most permanent options get considered before the ones we are paid to do.

Fix the drainage first. Clear the weep holes, prove the agricultural drain flows, check the fall of the driveway and where the downpipes discharge. On a meaningful share of enquiries that arrive asking for injection by name, this is the entire fix — and sealing the structure without doing it raises the head against the wall and increases the load on a retaining structure that may never have been designed for saturated backfill.

Redirect the source. If water is discharging onto the area from above and can be diverted with a downpipe extension, a re-graded path, a new fall or a diverter, do that. It is cheaper, more reliable and more permanent than anything we can inject.

Seal the defined path. Cracks, construction and cold joints, kicker joints and tie holes, from the dry side, with mechanical packers. This is the core negative-side scope and where it applies, it is permanent.

Manage what cannot be sealed. Where the path is diffuse rather than defined — general porosity, honeycombing, a whole damp face — move to crystalline tanking, a drained cavity, or curtain injection into the ground behind. Crack injection has nothing to grip in those situations and selling it there is selling a failure.

Excavate. Where the membrane is dead across an area, this is the answer, and no amount of negative-side work substitutes for it.

And there are situations where we decline the work rather than sell a compromise. We will not inject an element that is still moving until tell-tales have been read across 4 to 12 weeks and the cause established — a rigid repair in an active crack simply cracks again alongside. We will not inject a crack that follows the line of reinforcement with rust staining or spalling, because that is corrosion driving the crack and sealing over it hides an expanding problem; that is a concrete repair scope. We will not inject a designed movement joint, which needs a backing rod and a movement-capable sealant, not resin. We will not drill a post-tensioned slab without a GPR scan and written engineer approval. And we will not seal the only outlet from a saturated cavity or a planter box with a blocked drain, because the honest question on every negative-side job is where the water will go instead — and if the answer is nowhere, that is a design problem, not an injection problem.

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

Is negative side waterproofing compliant in Queensland?

There is no Australian standard that declares it compliant or non-compliant, which is why the question gets answered badly so often. AS 3740:2021 covers internal domestic wet areas. AS 4654.1 and .2 cover external above-ground membranes. AS 3600 governs the concrete structure itself, including exposure classifications and serviceability crack widths of roughly 0.10 to 0.40 mm. The National Construction Code sets performance requirements for water management rather than nominating a method. So the compliance question is not 'is negative side allowed' but 'does this repair achieve the performance requirement for this building, and is that documented'. Anyone quoting you a clause number that supposedly bans or blesses the method is citing something that does not say what they claim — and if they are citing AS 3740-1994, they are citing an edition that has been superseded twice.

Can we sign something that releases the contractor if it leaks somewhere else?

Not in the way most people assume. For domestic building work in Queensland, the statutory warranties in Schedule 1B of the QBCC Act are implied into the contract and cannot be contracted out of, so a clause saying 'no warranty on negative side work' is void and the warranty applies anyway. What does exist is section 29(6)(a): a contractor is not liable for a breach where the work was done in accordance with the owner's written instruction given contrary to the contractor's prior written advice. The operative words are prior and written. The practical consequence is that the diagnosis, the redirection risk, what is and is not being sealed, and the alternatives with their costs should all be set out in writing and agreed before anyone instructs the work. Ask for that from whoever you engage. It protects your ability to make an informed choice at least as much as it protects the contractor.

You sealed the crack and now it is leaking two metres along. Is that a failed repair or a new one?

That is exactly the question the scope and the record of the works answer. The scope names which defects were treated and which were not, and we back it with photographs before and after, taken from the same positions, and a written description of what we found and what we did. If the new water is emerging at a point we treated, it is our repair and we come back. If it is emerging somewhere we did not treat, it is a new path — which is the redirection effect described at the top of this page, is stated in the written advice before the works, and is a further scope rather than a warranty claim. Without that record the conversation is two people asserting things at each other. With it, it takes about five minutes.

Which standard actually covers a below-ground basement wall?

Honestly, none of the waterproofing standards covers it squarely, and that gap is why the topic is contentious. AS 3740:2021 is internal wet areas. AS 4654.1 and .2 are external above ground. AS 3735 covers liquid-retaining structures such as tanks and pools. The standard that genuinely applies to your basement wall is AS 3600, and it governs the concrete — exposure classification, cover, durability and crack width — not the waterproofing of it. Practically, that means a below-ground remedial scope is assessed against the NCC's performance requirements for water management, the manufacturer's product data and, where injection products are involved, EN 1504-5 as a product conformity standard. A report that acknowledges the gap is more credible than one that papers over it with a confident-sounding clause reference.

Does leaving water in the wall cause damage over time?

It is a fair question and the honest answer is that it depends on the element. Negative-side sealing stops water crossing the wall, but the concrete outboard of the seal stays wet. On a sound wall with adequate cover to the reinforcement that is a durability question rather than a problem — the concrete was designed for its exposure classification under AS 3600 and has been wet on its outer face since the day it was backfilled. Where cover is low, where chlorides are present, or where cracking already follows the line of the reinforcement with rust staining, it becomes an engineer's question, and in that last case injection is the wrong scope entirely because corrosion is driving the crack and sealing over it hides an expanding problem.

Our engineer specified tanking and the waterproofer quoted injection. Which is right?

They are usually answers to different questions rather than competing answers to one. Injection treats a defined path — a crack, a joint, a tie hole — and cannot treat an area. Tanking treats an area and cannot efficiently stop a running crack, because a coating goes onto a substrate that has water coming through it and will not bond or will simply be pushed off. On a lift pit or a tank the correct sequence is generally both: inject the joints and cracks with hydrophobic polyurethane first to kill the flow, then apply the crystalline or cementitious system over the now-dry substrate. If a quote proposes one where the diagnosis calls for the other, ask which leak path each is addressing — the answer will make the difference obvious.

Do we have to tell the neighbours?

For ordinary crack injection inside your own structure, no. For curtain injection, yes, and it should be in the scope rather than remembered afterwards. Curtain injection puts gel into ground you cannot see, and gel travels: into an adjacent basement, a service trench, a stormwater line or a neighbour's property. That is a real legal and insurance exposure, so neighbour notification and photographic before-and-after condition are both part of doing it properly. It is also why curtain work is priced from a trial panel at a rate per point or per kilogram rather than as a lump sum — the take is genuinely unpredictable until you start.