Fits

What a fit is

A fit is the relationship between two mating parts: a shaft, the enclosed or male member, and a hole, the enclosing or female member. The surface where the two meet is the mating surface.

Both members always carry the same basic size. What differs is their tolerances, and those alone decide whether one part runs freely inside the other, slides with no play, or has to be pressed in. A fit is therefore inseparable from tolerance.

Why tolerances exist: interchangeability

No part is ever produced at exactly its basic size. Every part comes out somewhere inside a tolerance zone, a band of acceptable dimensions.

In volume production the shaft and the hole are made separately, by different people, on different machines, in different places and at different times. For them to assemble correctly with no fitting work at the bench, each one has to land inside limits agreed in advance. That property is called interchangeability, and it is the reason the whole system of limits and fits exists.

Clearance and interference

The governing quantity is clearance, defined as hole size minus shaft size. Its sign is what names the relationship:

Because each member owns a whole tolerance zone rather than a single value, every fit is described by two extremes, obtained by pairing the worst limits against each other:

The diagrams below are labelled in Greek: Χμ is the maximum clearance and Χε the minimum clearance.

The four kinds of fit

Depending on where the two tolerance zones sit relative to one another, and therefore on how much freedom is left between the members, four cases arise.

1. Clearance fit

The hole is larger than the shaft in every possible pairing, so even the minimum clearance stays positive. The parts slide and rotate freely.

2. Sliding fit

The limiting case of the clearance fit, where the two zones just touch at the zero line and the minimum clearance falls to zero. It allows assembly and axial sliding by hand, but it is not meant for continuous rotation under load, since in the worst pairing there is no gap left to carry a lubricant film.

3. Transition fit

The two zones overlap, so the minimum clearance turns negative, meaning interference, while the maximum clearance stays positive. Depending on the actual sizes of each pair, the assembly ends up either slightly loose or slightly tight. It gives accurate centring and remains a joint you can take apart.

4. Interference fit

Even the smallest permissible shaft is larger than the largest permissible hole, so interference is always present. Assembly requires a press or a temperature difference, heating the hole or chilling the shaft, and produces a practically permanent joint that transmits torque with no key or other locking element.

ISO groups these into three classes (clearance, transition, interference), since the clearance and sliding cases together make up the clearance fit.

The ISO 286 system of limits and fits

In practice fits are not invented case by case· they are specified through the standard ISO 286 code. Each tolerance zone is described by two things:

A dimension is then written as Ø32 H7 for the hole or Ø32 g6 for the shaft. Every value in the system applies at the reference temperature of 20 °C· with tight grades and large sizes, how far the inspection room drifts from that temperature genuinely matters.

Basic hole system (H) and basic shaft system (h)

To arrive at the fit you want, you hold one member fixed and vary the other.

Basic hole system (H)

By far the more common of the two. The hole carries a fundamental deviation of zero, so its lower limit coincides with the basic size and rests on the zero line, and the character of the fit is set entirely by the shaft. It is also the cheaper route: the hole is produced with standard drills and reamers and checked with standard plug gauges, while the shaft, which is easy to turn or grind to any size, does all the varying.

Basic shaft system (h)

Here the shaft carries the zero fundamental deviation, so its upper limit coincides with the basic size, and the hole is varied instead. It is the less common arrangement and earns its place when the outside diameter arrives finished and is not machined at all: drawn or ground bar stock, one long shaft carrying several components (pulleys, gears, collars, spacers), or the outer ring of a rolling bearing seated in a machined housing bore.

Common fits and where they are used

Basic hole system, the preferred ISO 286 combinations.

The same fit characters exist in the basic shaft system with the roles reversed: C11/h11, F8/h7, G7/h6, K7/h6, P7/h6 and so on.

Inspecting fits with go/no-go gauges

Since a fit comes out of two parts, you inspect it by checking the hole and the shaft separately against their own limits: if both members are within tolerance, the intended fit follows automatically. In volume production that check is not made by measuring the actual size, but with limit gauges, plug gauges for holes and snap gauges for shafts.

A limit gauge has no scale. It does not tell you what the dimension is, only whether it falls inside the permitted limits. It has two ends of fixed size:

The part is accepted when the GO end passes under its own weight, with no force applied, and the NO GO end does not. That confirms the dimension lies between the two limits without reading a single number. Different gauge types serve the two members.

Gauges come in single or double form, with the two ends separate or combined in one tool. Each gauge is marked with the basic size, the fit class and grade (for example Ø60 H7), the limit values in µm and often the words GO and NOT GO. Gauges carry manufacturing tolerances of their own, and the GO end gets an additional wear allowance, since it is the end that rubs on every single check.

Their design follows Taylor’s principle (1905): the GO end checks the maximum material condition and should check as many features at once as possible, meaning size, roundness and straightness together, which is why it has full form and full length. The NO GO end checks the least material condition and only one feature at a time. In practice the GO end is the decisive one: if it does not pass, the part is rejected without exception, because it will simply not assemble.

Thread tolerances and thread gauges

The same logic carries over to screw threads, except that the fit no longer hangs on a single diameter but mainly on the pitch diameter, together with the pitch itself and the flank angle. A tolerance system parallel to ISO 286 applies here too: for ISO metric threads (ISO 965) the class is written as 6H for the internal thread (nut) and 6g for the external one (bolt), where the number gives the grade and the letter the position of the zone, upper case for internal and lower case for external threads, exactly as with plain features. A typical bolt and nut pairing is therefore written 6H/6g.

Inspection is again done with go/no-go gauges, in a dedicated form: thread plug gauges for nuts and thread ring gauges for bolts. The GO end has the full profile and full length and must run on freely along the whole thread, confirming that the part will assemble. The NO GO end has a truncated profile and checks the pitch diameter alone· under ISO 1502 it may not engage more than two full turns, while American practice (ASME B1.2) allows up to three on certain lengths. Inch threads follow an equivalent class system, 2A and 3A for external threads, 2B and 3B for internal ones.