Reading hydrocyclone wear: what the apex and the vortex finder each tell you

A hydrocyclone is usually treated as a single wear part with a single replacement interval. It is not. The feed chamber and inlet see the slurry at its highest velocity and least organised flow; the conical section sees a progressively tightening spiral; the apex sees the coarsest fraction concentrated into the smallest opening. Each region wears from a different cause and therefore reports a different fault.

The practical consequence is that a single question — how long does the liner last — cannot be answered usefully. Ask instead which region reached its limit first, and whether that region is supposed to be the sacrificial one. When the answer is no, the wear is telling you about the circuit, not about the ceramic.

The apex opening controls the underflow, so it is designed to be consumed and replaced. What matters is the shape it takes. An opening that grows while staying round has done its job and simply needs a shorter replacement interval or a more resistant insert. An opening that goes oval, or that thins on one side only, is reporting something upstream: an uneven feed distribution, a partially blocked inlet, or a cyclone hanging out of true on its mounting.

Photograph the apex from the same position at every inspection and keep the images with the date. A sequence of comparable photographs establishes whether the opening is growing steadily or accelerating, and acceleration is itself diagnostic: it usually means a change in feed, not a change in the liner.

The vortex finder is often replaced without being examined, which discards the most informative surface in the vessel. Wear on its outer wall belongs to the descending spiral and is ordinary. Wear on the inner wall or at the lower lip means coarse material is entering the overflow — the separation is degrading, and the wear is a consequence of that, not its cause.

This distinction has a downstream cost that dwarfs the part. Coarse material reporting to the overflow loads the next stage, and the loss is usually paid in recovery or in extra grinding rather than in liner spend. Inspect the inner surface, note where the lip has receded, and treat that observation as process information.

In a lined cone the ceramic faces usually survive longer than the lines between them. Slurry finds any step, any recessed joint and any mismatch at a section flange, and once it has a purchase there it works behind the element. A cone that failed at a flange or along a joint line did not run out of ceramic; it lost its continuity.

This is why a like-for-like replacement sometimes fails faster than the original. The second installation inherits a flange face that is no longer flat, or a shell that has distorted slightly, and the new elements sit with a step the first set never had. Check the sealing faces and the shell geometry during the outage, not after the next failure.

Cyclones rarely work alone, and a cluster gives you a controlled experiment you did not have to design: identical units on the same feed, differing only by position on the distributor. When one or two units consistently wear faster, the cause is almost never the liner supplied for those units. It is where they sit — a distributor that favours certain outlets, a shorter or straighter feed run, or a valve that does not close fully.

Record the position of every unit alongside its wear observation. A map of the cluster with dated findings identifies the favoured outlets within a couple of campaigns, and a distributor correction protects every unit at once — which no change of lining material can do.

Once the three regions are read separately, the specification usually changes shape. A vessel wearing evenly at the apex needs a replaceable insert and a defined change interval. One wearing at the joints needs attention to flange geometry and installation method before any material discussion. One wearing unevenly needs the distributor examined. Only the first of these is answered by ordering a different ceramic.

Bring the dated photographs, the cluster map and the observed wear shape to the technical discussion. A supplier who can see where and how the vessel failed can propose a construction that addresses it; a supplier given only a part number can only propose the same part again.