7 min read

Dimensional control: plug gauges, pin gauges and ring gauges

A plug gauge measures nothing. It answers yes or no, and that is what makes it more reliable than a micrometer on a production run. Provided you know which limit it checks.

Contrôle dimensionnel : jauges tampons, piges et jauges bagues. Les méthodes et tolérances ISO 286
From Ø 0.5775 to Ø 20.465 mm: the same inspection principle, from the watchmaking gauge to general engineering.

A plug gauge measures nothing. It gives you no dimension, no figure, no value to write on a report. It answers one question, yes or no: is your part on the right side of the limit? That is precisely why it is faster and more reliable than a micrometer.

Provided you know which limit it checks, and why two gauges are needed where one instrument would seem to do.

Double-ended ISO plug gauge, GO end and NO GO end on one body
A double-ended ISO plug gauge carries both ends on a single body. The longer one is the GO end, the shorter one the NO GO end: they are told apart by hand, without reading the markings.

Two gauges, two different questions

How a limit gauge works cross-section of a bore being checked GO NO GO at the lower limit diameter full form, over the full length at the upper limit diameter point contact, one dimension only The GO end is deliberately longer: that is how they are told apart by touch.
This is Taylor's principle. The GO end checks the maximum material condition and several defects at once, straightness and roundness among them. The NO GO end checks the minimum material condition, on one dimension at a time.

The practical consequence is simple. If the GO end passes and the NO GO end is stopped, the part is good. If both pass, the bore is oversize. If neither passes, it is undersize. No reading, no interpretation, no possible error.

It also explains why the GO end must have the full form and full length of the bore. A slightly oval or barrelled bore can give the right dimension at two points and still refuse a complete cylinder. A micrometer does not see that.

Reading an ISO 286 dimension

A drawing almost never says "10 mm bore". It says Ø10 H7. That notation carries three separate pieces of information.

Anatomy of a dimension: Ø10 H7 Ø10 H 7 Nominal size, in mm Position of the tolerance zone Width of the zone: IT7 Ø10 H7 = 10.000 to 10.015 mm Upper case for a bore, lower case for a shaft. H puts the lower limit on the nominal size.
The letter places the zone, the figure gives its width. H is the most common case because a standard drill or reamer naturally works above the nominal size.

On a shaft the same grammar is written in lower case. A Ø10 g6 falls between 9.986 and 9.995 mm, entirely below the nominal size, which guarantees clearance with an H7 bore. That is the classic sliding fit.

The same tolerance is not the same dimension

Here is the point many people discover late. An IT7 is not a value, it is a grade of precision relative to the diameter. The larger the part, the wider the interval, because machining large is inherently less precise.

The same IT grade widens with diameter tolerance width in micrometres, per ISO 286-1 IT6 IT7 0 to 3 mm 3 to 6 mm 6 to 10 mm 10 to 18 mm 18 to 30 mm 30 to 50 mm 6 10 8 12 9 15 11 18 13 21 16 25 0 5 10 15 20 25 µm
An IT7 is 10 µm below 3 mm and 25 µm between 30 and 50 mm. An IT7 in watchmaking and an IT7 in general engineering do not call for the same inspection equipment.

The ten per cent rule

A gauge is itself a machined part, and therefore has a tolerance. Established practice gives the gauge ten per cent of the part tolerance, split between the two ends. And that share is taken from inside the zone, never outside it.

The gauge tolerance is taken from inside lower limit upper limit parts accepted GO gauge NO GO gauge No out-of-tolerance part can pass the check. But up to ten per cent of good parts may be rejected. That is a deliberate choice: the gauge protects the customer, not the producer.
On a plug gauge the tolerance is plus on the GO end and minus on the NO GO end. On a ring gauge it is the reverse, since the material is on the outside.

A second allowance applies to the GO end alone: the wear allowance, of the order of a tenth of the gauge tolerance. The GO end rubs at every check and loses material, and therefore diameter. The allowance is added in the direction opposite to that wear, so the gauge stays correct for longer. Below a part tolerance of roughly 0.09 mm it becomes negligible and is no longer applied.

Set of Cary plug gauges presented in a workshop case
A complete set in its case. The step between two consecutive gauges, what the catalogue calls the progression, is the first selection criterion: it sets the smallest tolerance the set can sort.

Choosing from the catalogue

Four decisions, in this order.

DecisionThe optionsWhat settles it
Progression0.001 mm or 0.01 mmA 1 µm step is only justified if your tolerance falls below 10 µm. Otherwise it is a more expensive set for a resolution you cannot use
MaterialSteel or hard metalHard metal for series inspection, where wear on the GO end becomes the limiting factor. Steel for occasional use
HandleWith or withoutThe handle keeps your hand off the gauge and so limits expansion. Without a handle for mounting in a fixture
Pin gauge length15 mm or 32 mm32 mm to reach the bottom of a bore or pass through a thickness. 15 mm where access is clear and rigidity matters

Two families sit side by side at Azurea. The sets with a regular progression, where each gauge is a nominal diameter, are for measuring and fine sorting. The ISO plug gauges are toleranced directly to a class from the drawing, and are therefore ready to check an H7 with no calculation. The Cary ring gauges cover external inspection, in steel or hard metal.

What throws an inspection off

Temperature first. All standardised dimensions apply at 20 °C. Steel expands by about 11 µm per metre per degree, which is 0.11 µm on a 10 mm bore for one degree of deviation. Nothing at all. But a gauge held in a full hand rises by about ten degrees, and those 1.1 µm change status depending on the tolerance being checked.

What ten degrees of hand warmth cost on a Ø10 share of the tolerance consumed by 1.1 µm of expansion IT4 IT5 IT6 IT7 IT8 27% 18% 12% 7% 5% 0 10% 20% 30%
The orange marker sits at 10%. To the right of that line, hand warmth weighs more than the uncertainty of the gauge itself: from IT6 onwards, the gauge is set down and left to return to workshop temperature.

Force next. The GO end must enter under its own weight, without you pushing. Forcing turns an attribute check into a subjective judgement, and lets out-of-tolerance parts through.

Wear last. A GO end thins with use and ends up accepting bores that are too small. There is no visible sign: only periodic inspection of the gauge itself reveals it. That is the reason for the wear allowance, and the reason a set of gauges has a functional expiry date even when it looks new.

In short

A micrometer gives you a dimension and leaves you to judge. A limit gauge leaves you nothing to judge, and that is exactly what is asked of it on a production run. The two are complementary: the micrometer to understand a drift, the gauge to sort without thinking.

Which leaves the rule that sums up all the rest. Choose the gauge from the tolerance on the drawing, never from the nominal size. The tolerance is what decides the progression, the material and whether you need an ISO version.

So, gauges or instruments on your production runs? Tell us in the comments.

Tolerance values per ISO 286-1:2010, system of limits and fits. The GO and NO GO principle follows Taylor's principle, carried over into the ISO 1938 series on limit gauges. Product characteristics per the Azurea and Cary catalogues. All standardised dimensions refer to a temperature of 20 °C.