Crimp charts: why they cannot be trusted blindly — and how to build your own

Beginner operators often look for a single chart that tells them what diameter to crimp any hose to. No such chart exists. Each fitting manufacturer has its own, some hose manufacturers have their own, and the values for one and the same size differ. This is not a compiler's error but a property of the joint itself. Let us look at why this happens and how to work with it.

Where the crimp diameter comes from

A finished joint consists of several layers: the fitting ferrule, hose rubber, braids, rubber again, the nipple. The correct crimp diameter is the one at which all layers are compressed to the designed degree: the braids have settled on the nipple with the required interference, and the rubber has been compacted but not cut. And each layer has its own manufacturer with its own tolerances.

Ferrule wall thickness differs between factories by tenths of a millimetre. So does the thickness of the rubber layers of the hose. Add up these variations and it becomes clear why a fitting manufacturer's chart honestly works only with its ferrules — and, as a rule, with the hoses on which it developed its charts.

Typical problems — and what to do about them

Problem 1: hose from one brand, fitting from another

The most common situation on the market: the hose came from one supplier, the fittings from another, and no ready chart for this pair exists. Using the value from the fitting manufacturer's chart without checking is risky: the ferrule is theirs, but the hose is different, with a different rubber thickness.

The solution is a pair confirmation procedure. It takes about ten minutes and one or two metres of hose:

1. Take the diameter from the fitting manufacturer's chart as the starting point — the ferrule is theirs, after all.
2. Crimp two or three trial samples and inspect them: the ferrule must be compressed evenly, without a "waist", rubber extruded at the end or cuts.
3. Measure the diameter and compare with the starting point.
4. Pressure-test the samples on a bench — at one and a half times the working pressure, with a hold. For critical applications it is reasonable to take the first sample to destruction: You will see the real margin of the pair.
5. Record the confirmed combination — hose (brand, type, size), fitting, diameter, die set. This is the first line of Your own chart.

Problem 2: the charts contradict each other

You open the hose manufacturer's chart and the fitting manufacturer's chart — and the figures differ. This is normal: each measured with its own components. Give priority to the chart of whoever made the ferrule: it is its wall that determines how much metal has to be deformed. Use the second chart as a guide to the range, and let the test make the final decision.

Problem 3: where and how to measure

After crimping the ferrule is not perfectly round — it carries the flats left by the dies. Hence a typical mistake: one reading "as it comes" and confidence that the diameter is within tolerance.

The right way: measure at the centre of the ferrule (at the ends the diameter is always slightly different), put the caliper jaws on the flats, take two readings turned by about 90° and look at both. A noticeable difference between them is a sign of die wear or uneven closing, and a reason to check the machine rather than to average the figures. Manufacturers usually set tolerances in tenths of a millimetre, so the instrument must read to 0.05–0.1 mm, not "by eye".

Problem 4: the diameter drifts over time

It happens that a proven pair begins to give a diameter at the tolerance limit. There are usually three causes: wear of the die working faces, dirt in the die seats of the head, or a change of hose batch. The first two are cured by machine care — we wrote about die condition in a separate article. The third is a rule: a new hose batch, let alone a new supplier, means a quick re-check of the pair even if the size is "the same".

Problem 5: crimping "to the stop" with no chart at all

Sometimes operators crimp "to the stop" — until the ferrule stops turning. This approach is dangerous: an under-crimped ferrule may hold during installation and be pulled out under pressure on the machine, while an over-crimped one may cut the braid, and the joint fails later, in service (more in the article on crimping defects). If there are no data at all, this is a case for the procedure from Problem 1, not for intuition.

Your own chart — the shop's main asset

The result of all the solutions above is a log of confirmed combinations. Over time it covers more than ninety per cent of Your orders and becomes the shop's main technological document: a new operator works to verified figures, not to someone else's charts. A convenient form has four columns: the pair (hose + fitting), diameter, dies, date and test result. On CNC machines the same chart lives directly in the machine's memory as saved programmes: the operator selects the pair from the list and the machine sets the diameter itself.