Disassembly procedure

1. The strip is the inspection

Taking a coupling apart is the only opportunity to find out why it wore — and most of that opportunity is destroyed in the first ten minutes by whoever cleans it before anyone looks.

The sequence matters more than any individual technique in it. What follows is ordered to preserve the diagnosis, not to get the parts off fastest.

2. The two things that hurt people

• Stored energy. A train isolated electrically can still turn under the weight of a load, a fan can windmill on a draught, and a coupling being split stops holding the two halves in relation to each other. Lock off, and physically restrain anything that can rotate or drop.
• Heat. Hub removal involves a component at a few hundred degrees that looks identical to one at ambient, in a working area, near grease. Anyone who has heated a hub off has a story about forgetting which one was hot.

3. Sequence

1. Take alignment readings as found — before any foot or flange bolt is slackened. Slackening one bolt on a machine with a soft foot changes the geometry you were about to measure.
2. Photograph the assembly in place.
3. Remove the guard.

3.1 Splitting the sleeve

Where the grease evidence is won or lost.

1. Position the coupling so the lubrication plugs are accessible, and put a container under it. The grease that comes out is the diagnosis and is worth more than the time it takes to catch.
2. Sample it before it contacts anything else.
3. Mark the sleeve halves and hubs for relative position before they separate. Orientation is information, and once the parts are on a bench it cannot be recovered.
4. Slacken the flange bolts progressively in a cross pattern, not one at a time, and keep them with the coupling rather than in the general fastener bin.

With the sleeve off, examine both hubs in place before either is removed. Contact patch position on each, and whether the two agree, is the observation that localises a fault to one end of the train — easiest to make while the parts are still oriented as they ran.

4. Getting the hub off

A clearance-fit hub with a setscrew or retaining plate comes away with a puller and little drama. An interference-fit hub does not.

Heat removes the interference rather than overcoming it. The hub is heated quickly, so it expands before the shaft underneath has time to follow, and pulled off with a puller already loaded and ready. Induction is ideal because it heats the hub and not the shaft.

An oxy-acetylene flame is not: it heats unevenly, will readily exceed the temperature at which the hub material is affected, and a hub that has been overheated should not go back on regardless of whether it came off.

4.1 Why force is the wrong tool

The instinct to reach for a larger puller is the wrong instinct. Force fights the interference pressure across the whole bore surface, and the load required grows faster than the hub tolerates. What gives way first is usually the hub flange, the puller, or the shaft end.

Where a hub will not move with heat and a loaded puller, the joint has usually seized through fretting or corrosion. The decision then is whether to cut the hub off — frequently the correct answer, and always cheaper than a damaged shaft.

5. What to inspect once it is apart

6. What can go back

• Seals — never. Cheapest component in the assembly, most common origin of the next failure. Replaced whenever the coupling is open.
• Flange bolts — a rated part carrying the full transmitted torque. Replace after a fault or a fretted joint, and never substitute from the stores bin.
• Hubs and sleeves — reusable if the wear pattern is consistent with the hours run and no crack indication is found. A hub that has been flame-heated or driven off cold should not go back.

Whatever is reassembled, the alignment is part of the repair, not a separate job. A coupling refitted to the geometry that wore the last one out wears the same way, on the same schedule.