Tungsten Carbide Nozzles and Wear Consumables: Why Service Life Beats Purchase Price

2026-09-02

Nozzle inserts, bushings, sleeves and valve seats in WC-Co: how a consumable wears, what that wear does to your process long before the part fails, and why the cheapest line on the invoice is usually the most expensive one to run.

Every plant has a drawer of parts nobody thinks about. Nozzle tips, guide bushings, plunger inserts, valve seats. They get bought by the box, replaced by the operator and never mentioned at a management meeting, yet they are what decides whether a line holds its process window from one shift to the next.

Hydroforce makes these components in cemented carbide on a complete in-house powder metallurgy chain: powder blending, pressing, vacuum sintering and diamond grinding under one roof. What carbide is and how we make it is covered on our cemented carbides page. This article deals with the part nobody writes down, which is what happens to a consumable once it is in service, and how to specify one so that the process survives along with the part.

The consumable paradox

A wear part is bought in order to be destroyed. That makes the usual purchasing logic of comparing unit prices and taking the lower one fairly useless here, because the unit price is the smallest number in the equation.

Look at what a nozzle change really costs:

The line stops. In a blast booth a mixing tube change takes minutes; on a downhole tool or a high-pressure pump it can run to hours.
Somebody has to do it, and that is skilled time rather than the price on the part.
The process shifts. A new nozzle does not behave like the one that came out, so throughput, coverage and consumption all step back to a different point.
Quality drifts in between changes. That is the expensive item on this list, and it is the one nobody sees.

Set against all of that, the difference in price between a steel insert and a carbide one is a rounding error. A hardened steel nozzle in an abrasive stream may last a shift. A tungsten carbide nozzle on the same abrasive runs for hundreds of hours. The gap is one to two orders of magnitude, and it lands on downtime rather than on the parts budget.

How a nozzle actually fails

Most specifications treat a nozzle as a part that will eventually break. Abrasive nozzles almost never break. They erode from the inside, evenly and invisibly, and they keep working the whole time they are doing it.

As the bore opens up, four things happen at once:

Exit velocity falls, because the same supply pressure now feeds a larger throat. Cleaning rate, cutting rate and impact energy all drop with it.
Consumption rises. A wider throat passes more air, more water and more abrasive for the same amount of work, so the media bill climbs while output goes the other way.
The pattern spreads. On cutting this shows up as edge quality; on surface preparation, as uneven coverage.
Nothing alarms. There is no noise, no leak and no failure code, so the operator compensates by working slower and the loss disappears into normal variation.

By the time somebody notices, the part has been costing money for a long while. That is the reason nozzles are replaced against a measured bore diameter rather than against their appearance or a date in a maintenance plan.

What bore growth actually costs

Dry blasting shows the whole cascade clearly, and the numbers for it are published.

Take a standard No. 4 nozzle with a 1/4 in (6.35 mm) bore. New, it passes roughly 81 CFM at 100 psi. Worn to 5/16 in (7.9 mm), which is 1.6 mm of growth and invisible to the eye, the same nozzle needs about 137 CFM to hold that same 100 psi. Air demand has gone up by close to 70 %, and the compressor was sized against the original figure.

One of two things follows. Either the compressor keeps up, and the energy bill quietly rises by that margin. Or it does not, and pressure falls instead. Published guidance puts the productivity loss from a drop of 100 to 80 psi at roughly half.

None of this is visible from where the operator stands. The nozzle looks fine, the gauge reads a little low, and the job simply takes longer.

Published replacement thresholds run between 0.8 and 1.6 mm of growth depending on whose practice you follow, which is a narrow enough band to work with. Measure the bore, pick your threshold, replace against the number. Carbide does not remove that discipline, it stretches the interval between measurements from hours to weeks.

Waterjet cutting behaves the same way on a different scale. A focusing tube opens up by roughly 2.5 µm per hour of cutting and the stream widens with it, which is why one study of discarded tubes found around 85 % of them taken out of service for wear rather than breakage. The tube does not fail. It drifts out of tolerance.

What carbide buys you, and where it stops

Cemented carbide is a composite of hard tungsten carbide grains held in a metallic binder, usually cobalt. Hardness normally lands between 1,400 and 2,000 HV, well beyond anything a hardened steel will reach, while the binder keeps enough toughness in the part for it to survive real handling and real impact.

For a consumable that is close to an ideal combination. Carbide is not a single material, though, and there are duties where it is the wrong choice.

Condition in service	What matters	Direction to take
Pure abrasive erosion, no shock	Hardness, fine grain, smooth bore	Low-cobalt fine-grained grade
Abrasion plus pressure cycling	Balance of hardness and rupture strength	Medium-cobalt, medium grain
Repeated impact, percussion	Fracture toughness before hardness	Higher-cobalt, coarser grain
Acidic or chloride media	Binder chemistry, not hardness	Nickel-bearing binder instead of cobalt

Two limits deserve stating outright. Carbide is brittle next to steel, so it handles erosion superbly and point impact badly, and a dropped part or a poorly supported mounting will chip where steel would have dented. In acidic media the cobalt binder leaches out before the carbide grains are touched at all, which makes the failure chemical rather than mechanical and puts it beyond anything hardness can fix. Both are dealt with when the part is specified, not once it is in the field.

Harder ceramics do exist. Silicon carbide and boron carbide outlast tungsten carbide in blast nozzle service by a wide margin, and for pure erosion in a protected mounting they are the better answer. Tungsten carbide keeps its place wherever the part also has to be tough, carry pressure, hold a ground surface, or sit brazed or press-fitted inside an assembly, which covers most of what goes into working machinery.

The parts nobody specifies carefully enough

Nozzles get attention because they are visibly consumable. The rest of the wear set usually gets inherited from an old drawing and never looked at again.

Bushings and guide sleeves. Press-fit or slip-fit, they set the alignment for everything downstream of them. A steel bushing that has worn oval shows up as a quality problem somewhere else entirely, and it usually gets chased for weeks in the wrong place.

Valve seats and cores. In high-pressure systems the seat takes erosion and repeated seating impact together. It is the mixed-duty case in its purest form: hardness on its own chips, toughness on its own erodes.

Seal rings and mechanical seal faces. Here the ground surface does the work rather than the bulk material. Diamond-ground carbide holds Ra 0.1–0.4 µm and goes on holding it, which keeps the sealing interface predictable across the whole service interval rather than only at the start of it.

Plunger inserts and sleeves. High-pressure pumps put abrasive fluid and reciprocating contact on the same surface. The standard answer is a carbide insert in a steel body, so that the wear surface is carbide while the structure stays steel.

Choke and flow-restriction inserts. A component whose function is defined by a passage diameter is, by definition, a component that fails when that diameter changes. Anywhere throttling happens in an abrasive or erosive stream, the insert material sets how long the flow characteristic stays where it was designed to be, and how often somebody has to go in and put it back.

Geometry and mounting decide the rest

Material choice accounts for roughly half the result. The other half is design detail that rarely reaches us in an enquiry.

Bore finish. A rough bore gives erosion somewhere to start, since every peak is a point where abrasive can bite. Ground and polished bores wear more slowly, and more evenly, than as-sintered ones.

Inlet geometry. The convergent section governs how the abrasive stream forms. Get it wrong and the wear concentrates into a single band instead of spreading along the throat, which cuts service life without anything being wrong with the material itself.

Concentricity. A bore that is off centre wears asymmetrically, spreads the pattern early and pulls the process off target well before the nominal diameter limit comes up.

The joint. This is where most carbide consumables actually die. The insert survives, but the braze lets go, or the press fit relaxes under thermal cycling, or the steel body erodes away around a perfectly good carbide tip. Interference, braze alloy and body material need specifying alongside the insert rather than after it.

Where the parts come from

Powder blending sets the grade. Pressing and vacuum sintering at 1,350–1,500 °C set density and microstructure, and the part shrinks 15–20 % along the way, which is designed into the tooling instead of being corrected afterwards. Diamond grinding is the only method that will machine sintered carbide at all, and it is what takes a part from ±0.15 mm as sintered down to ±0.01 mm with Ra 0.1–0.4 µm on functional surfaces. Where the duty is critical, hot isostatic pressing closes the residual micro-porosity that would otherwise give erosion its first foothold in service.

With those stages under one roof, changing a grade, a bore finish or a piece of geometry is an internal decision rather than a negotiation across three suppliers with three lead times.

Every batch gets verified before it leaves us: Rockwell hardness on finished parts, metallographic microscopy for grain size and binder distribution, X-ray for internal porosity on critical components, coordinate measurement to under 3 µm. Consumables are ordered again and again, so consistency between batches is what actually matters. A nozzle that behaves differently from the last one is a process problem even when it passes inspection on its own merits.

Parts run from 0.001 kg to 300 kg and from Ø 3 mm to 500 mm, with minimum wall thickness down to 0.1 mm, in quantities from prototypes to repeat series.

What to send us

Quoting a wear part takes slightly different information from quoting a structural one, because what governs the answer is the duty rather than the drawing alone:

Drawing or sample, with bore diameter, geometry, critical surfaces and required finish.
The medium: what flows through the part, abrasive type and particle size, concentration, chemistry.
Operating conditions: pressure, temperature, flow rate, continuous or cyclic duty.
How it mounts: press fit, braze, threaded body, and what the mating part is made of.
What is happening now, meaning the current material and how long it lasts. This is the single most useful line in any enquiry, because it turns an abstract specification into a target we can beat.
What “worn out” means to you, as a bore diameter or a drop in performance at which the part comes off the machine.

Given that, we can propose a grade, a geometry and a finish, and tell you honestly where carbide earns its place and where something else would serve you better.

Need carbide nozzles, bushings or wear inserts for a specific duty? Send your drawings and operating conditions to office@hydroforce.ee and our engineering team will come back with a grade recommendation and a quotation.