How to Measure Load Chain Wear on a 5 Ton Harrington CB050 Hoist

A chain that still lifts, still runs quiet and still looks straight can already be past the point where it should come off the hoist. That is the expensive half of the problem: the half where the hoist keeps working and nobody has a number to argue with. The other half is the crew that scraps a chain with years left in it and pays for a part nobody needed. Both halves come from the same habit, judging chain wear by eye.

This is written around one machine: the Harrington CB050 5 ton hoist, and the chain wear group it shares with several other CB models. Nothing here is a universal rule for every 5 ton chain hoist on the market. Within that scope, the verdict comes down to two caliper readings compared against the discard row for that exact model code, and how the chain looks is not one of those two readings.

The cost of guessing at chain wear

Suppose the guess goes the expensive way, and the bill is a chain, a downtime window and a maintenance tech's afternoon, because a chain judged by eye gets replaced on suspicion rather than on measurement, which means the shop pays for a part it did not need and loses a bay it could have used. Not a small bill. And it comes back every quarter.

Suppose the guess goes the other way, and the cost is quieter but worse: the hoist keeps running on a chain that measures past the line the manufacturer drew for it. Nothing fails on the day you skip the check. The chain just keeps losing material where the links bear on each other, which is the exact outcome the manual's replace instruction exists to prevent.

What makes this worth writing down is that the check itself is small. Two numbers, one comparison, a line in a log. The expensive part is not the measuring. It is the years of decisions made without it, each one paid for in either a wasted chain or a chain that should already be off the hoist.

What pitch and wire diameter actually tell you

A load chain is a row of links that bear on each other under load, and every lift removes a little material at exactly those contact points. That is the whole mechanism. Wear does not thin a chain evenly the way it wears a belt; it eats the surfaces where steel meets steel, and the chain gets longer as a result.

The reason the manual watches two dimensions instead of one is that the two failures move in opposite directions. Pitch, the P dimension, is a length measured along the chain, and it grows as material disappears from the bearing surfaces. Wire diameter, the d dimension, is the thickness of the link material itself, and it shrinks. One number drifts up while the other drifts down, which means a single question about whether the chain looks stretched cannot cover both. A chain can be lengthening and thinning at the same time.

Two limits rather than one stretch allowance, then. A chain that fails one can still pass the other, so a check built on only one of them misses half the wear it was supposed to catch.

One thing this article does not carry is the diagram that fixes where the pitch span starts and ends. That figure is not in the material in front of me, so no link count and no span length is printed here. Where you place the caliper jaws comes from your own copy of the manual, and it has to, because a pitch reading taken across the wrong number of links is a number with nothing attached to it.

Get the right discard row before you measure

Before any jaw touches the chain, read the product code off the hoist itself, then open the chain wear table in the manual for that model and find the row it belongs to. Skipping that step makes everything after it pointless.

That table groups models rather than listing one row per capacity. In the version in front of me, CB050 sits in the same row as CB025, CB080, CB100, CB150 and CB200, while CB020 has a row of its own. The grouping is the reason you match by code and not by tonnage: two hoists that both feel like a 5 ton machine can land in different rows, and values pulled from the wrong row are still numbers, still printed, still authoritative-looking, and still wrong for that unit.

Say it plainly: the extracted copy of that table carries an unresolved column mapping problem, so the discard values are deliberately not reproduced here. Take them from the printed table or the PDF for your own unit, off the row you just matched. The procedure travels; the values do not.

Taking the two readings so they mean something

Repeatability is not a formality. A reading you cannot reproduce next quarter is a reading you cannot trend, and trending is what turns a quarterly check into a replacement plan with a date on it instead of a surprise.

Clear the chain before the jaws go on, because grit packed between the links holds the span open and inflates the pitch reading. What the caliper ends up measuring includes the dirt. Your manual's lubrication check is where that cleaning expectation comes from, since it calls for every link surface to be coated and free of dirt and grime, and clean steel is also what makes a measurement mean anything.

Measure the same span every time, and write the span down beside the reading. If the first inspection used a four link span and the next used six, the two numbers are not comparable, which means the log stops being a log and becomes a pile of unrelated measurements nobody can act on.

Alongside that dimension check, the manual calls for a load sheave inspection, which means a clean reading on a worn sheave is only half the picture. Note which section of chain you measured, so next quarter's reading lands on the same links instead of a random stretch of chain.

One more: a reading taken on a link sitting in a groove tells you about the groove, not about the link.

Comparing your numbers to the discard values

With two numbers on paper and the matched row in front of you, the comparison is short. Pitch should not be greater than that row's discard value. Wire diameter should not be less than it. Fail either one and the chain comes off; pass both and it goes back into service.

What follows that check in the manual is replace, and it is paired with inspecting the load sheave, plus the idle sheave on multiple fall hoists. That pairing is not decoration: a worn chain and a worn sheave work on each other, which is why the manual puts both in the same inspection line.

Keep the log entry small: model code, span measured, both readings, which links, date, initials. Six fields, and next quarter's comparison takes a minute instead of an argument about what anybody remembers.

Three visual checks that catch what calipers miss

Surface condition comes first, and the list is specific: gouges, nicks, dents, weld spatter, corrosion, deformed links, abrasion, and significant wear on the surfaces where links bear on one another. Any of those, and the action called for is replace. This is where a chain that passes the caliper can still fail, because a nick or a corrosion pit is damage that never shows up as a span or a thickness.

Lubrication is a visual and an auditory check at once. Every link surface should carry lubricant and be free of dirt and grime, and the chain should not crack or pop while it is lifting a load. Noise during a lift is information about the chain, not background. Where this check fails, the manual routes you to clean and lubricate rather than straight to condemning the chain.

Reeving is the third. The chain should be reeved correctly through the load sheave, and on multiple fall hoists it should be installed properly and free of twists. A twist changes how the links sit and bear, which means the numbers you log are describing a chain running in a condition the reeving check exists to rule out.

Who should not run this check themselves

This check is not for anyone who cannot put their hands on the manual for their exact model code. Measuring against the wrong row is worse than not measuring at all, because it produces a number that carries authority and has no basis behind it, and somebody will pull a hoist on it. It also is not much use in a shop where a replace verdict cannot be acted on that week, since the reading would only document a problem nobody can close.

It fits the in-house maintenance team that has the manual, a caliper, a place to log the readings, and the standing to take a hoist out of service when a number says so. For that team, two readings and ten minutes settle a question that otherwise runs on opinion, and next quarter's readings settle it faster.