Wet vs Dry Cutting with Diamond Blades

Every diamond blade on the market will cut concrete dry — once. The question is not whether dry cutting works, it is what it costs. Water changes the temperature the diamond runs at, and temperature is what decides whether a segment wears out slowly or turns into a shiny grey ring that will not touch the material. This page covers what water is actually doing, where dry cutting is genuinely the right call, and how to decide before you have ruined a blade finding out.

What water actually does

Water does four separate jobs in a diamond cut, and it is worth naming them because most operators only think about the first:

• It keeps the diamond below its own limit. Diamond is not stable in air at high temperature — above roughly 700 °C it starts converting to graphite, which is soft. The metal matrix around each crystal is meant to wear back and expose new diamond; if the diamond dies before the matrix does, the segment polishes instead of cutting.
• It carries the swarf out of the kerf. A dry kerf packs with fines, and the blade starts grinding steel against concrete rather than cutting. That is where the heat and the wander come from.
• It lubricates the core against the kerf walls. Side wear is the quiet killer of a blade that is otherwise perfectly good: the core thins, the cut drifts, the blade starts binding in its own slot.
• It suppresses dust. This is the one that now has a regulation attached, and the reason the argument has shifted on commercial sites.

Take water away and none of those four jobs disappear. They simply stop being managed, and the blade pays for it.

What heat does to a segment

A dry cut in hard concrete can push the segment surface past 600 °C within seconds — fast enough that you will not feel it through gloves. The failure then runs in a predictable order:

1. The matrix softens before the diamond dulls. Diamond is pulled out of the bond while it is still sharp, so the segment wears fast without cutting faster. The operator compensates with more pressure, which adds more heat.
2. The exposed diamond graphitises. The segment goes glassy, dark grey and slightly glazed. It will not cut mild steel, let alone concrete. This is the stage most people describe as "the blade is finished" — it is usually not, it is dulled, and it can often be dressed.
3. The joint takes the heat. On a brazed blade the filler creeps and the segment lifts. On a welded one the joint survives, but the steel around it can still be softened. This is the same failure mode described in welded vs brazed, seen from the temperature side.
4. The core loses tension. A tensioned core relaxes somewhere around 300–400 °C. After that the blade will not run true, the cut goes out of round, and a 400 mm blade at 2,000 rpm will eventually crack from the arbor outwards.

The practical lesson: the segment is the cheap part. A glazed segment can be dressed. A detensioned core cannot be fixed on site, and a segment that has left the core at speed is an injury waiting for a report.

What dry cutting costs

In our experience with customer returns and site feedback, running a concrete blade dry instead of wet typically costs 30–50% of its life, and 20–30% of its cutting speed. That is not a marketing number, it is the arithmetic of heat: the same diamond, held at a lower temperature, keeps its edge and keeps its bond.

Set that against what water costs. A pump, a hose and a water butt are cheap and reusable. The real cost of wet cutting is logistics — getting water to the cut, and getting slurry away from it. On a ground-floor slab that is trivial. On the fourth floor of a finished building it is a genuine problem, which is why dry cutting is not going anywhere, and why the answer is usually "shrouded and extracted" rather than "dry and hope".

When dry is the right call

Dry cutting is not a compromise on every material. It is the correct method for:

• Brick, block and mortar. Soft, abrasive, low heat. A segmented blade runs dry all day and lasts.
• Aerated concrete and plasterboard-adjacent work. No water wanted anywhere near the substrate.
• Ceramic and porcelain tile with a continuous rim blade and a dust extractor. Water is better for edge quality, but dry with extraction is standard practice in occupied buildings — see the tile blade range.
• Small diameters on site where a 115 mm blade on a grinder is doing a handful of cuts and dragging a hose out is absurd.
• Freezing conditions. Water on a scaffold at −5 °C is a slip hazard and a burst pump.
• Overhead and vertical work where slurry would run down a wall or an operator's arm, and where a shroud and vacuum do the job better.
• Interiors where slurry cannot be disposed of — heritage work, hospitals, anywhere the run-off is a controlled waste.

What dry cutting is not: a way to cut reinforced concrete quickly, or a way to cut a 400 mm deep slot in one pass, or a licence to skip extraction.

When water is not optional

There are cases where dry cutting is simply the wrong tool, and no operator skill recovers it:

• Reinforced concrete, any diameter. The steel interrupts the cut, the segment spends time rubbing rather than cutting, and heat spikes. Wet, or shrouded with a laser-welded blade and realistic expectations.
• Anything above roughly 300 mm diameter. A large blade has a long arc in the cut and nowhere for heat to go. This is why wall saws and floor saws are water machines by design.
• Deep cuts. Past about 50 mm of blade buried in the material, swarf cannot clear itself.
• Hard, dense aggregate — flint, quartzite, granite — where the diamond does the work and heat builds fastest.
• Core drilling. A core bit is a closed cut with almost no airflow; dry core bits exist, but they are a different tool for a different substrate, covered in how to choose a core bit.
• Green concrete. Abrasive, and the dust is alkaline enough to burn skin and eyes.

Side by side

What changes on the machine

Wet and dry are not the same blade with a hose attached. Three things differ:

The bond. A dry-cutting segment uses a softer, freer matrix, designed to wear back and expose diamond before the heat kills it. Run that blade wet and it wears out early, because nothing is slowing the matrix down. A wet bond is harder and relies on water to keep it from glazing — run it dry and it polishes in the first metre. This is the single most common mismatch we see in customer photos.

The gullets. Dry blades are built to throw dust and let air reach the segment; wet blades move slurry, which is heavier and packs differently. A dry slotted blade run with water often ends up with slots full of paste, binding in the cut.

The supply. For hand-held work, a visible stream at the segment — roughly 2–4 litres a minute — is the minimum. Wall saws and large floor saws want an order of magnitude more. For dry work, a shroud that actually seals against the surface plus a HEPA extractor rated for the saw; a shop vacuum with a paper bag is not extraction, it is decoration.

Personal protection does not change: eye protection, hearing protection, and at least a P2/P3 respirator for any dry cut, regardless of what the extraction is doing.

The dust rules that changed the argument

Respirable crystalline silica is the reason this conversation moved from preference to procedure. Cutting concrete, mortar and stone dry releases fine silica that stays airborne long after the cut and scars lung tissue years later. Regulators have responded with exposure limits and, in some jurisdictions, prescribed methods.

In the United States, OSHA's silica rule for construction lists, for hand-held saws, either integrated water delivery or a shroud with a HEPA vacuum as the specified control — dry and uncontrolled is not an option on a covered job. Canada handles it province by province, with Ontario, BC and Alberta all imposing exposure limits and dust-control obligations. Australia has a national silica code of practice that pushes the same way, and the EU has its own occupational exposure limits. The detail differs, the direction is identical.

What that means commercially: on a commercial site in Toronto, Sydney or Manchester, "dry" now means "shrouded and extracted", with a written method and a respirator. Water is often the cheaper way to satisfy the same requirement. If a supplier is still selling you a dry blade with no conversation about dust control, they are selling you a problem for the main contractor.

Five mistakes that kill blades

1. Running a wet bond dry. It glazes in the first few metres, the operator adds pressure, and the segment is polished before anyone works out why.
2. Running a dry blade with water. The slots pack with paste, the blade binds, and the cut wanders. It also rusts the core in the van afterwards.
3. Mistaking a trickle for water. Water at the guard is not water at the segment. On a 400 mm blade the difference is the whole life of the blade.
4. Skipping the break-in. Two or three metres at reduced feed lets the matrix open up and the segment seat. Jump straight into full depth and the first metre decides the rest of the blade's life.
5. Blaming the blade for the rebar. A segmented blade meeting 16 mm bar at depth will glaze whatever the bond. The answer is a laser welded blade, a shallower pass, or a different line — not a discount on the next container.

Deciding in sixty seconds

Four questions, in this order. If any answer is yes, run wet: does the material contain steel? Is the blade over 300 mm? Is the cut deeper than 50 mm? Is the aggregate hard and dense? If all four are no — brick, block, tile, light concrete, small blade — dry is fine, with extraction and a respirator.

If you are between the two, run wet. The pump costs less than the blade, and the blade costs less than the day you lose when it stops cutting. If water genuinely cannot go to the cut, say so at the quotation stage: we will steer you to a dry bond from the saw blade range that suits the material and tell you honestly how much life you are giving up, rather than sell you a wet blade and let you find out on the wall.