How to Identify an Unknown Thread with Calipers and a Pitch Gauge

Step 1 - measure the major diameter

Close a caliper on the crests of the thread - the outer edges - and you have the major, or outside, diameter. For an external thread this is the single most useful number you can take, and it is easy to get right if you avoid two mistakes: measuring across a damaged or burred crest, and measuring across a rolled thread's slight upset near the head. Take the reading over clean thread in the middle of the threaded length, and take it twice at ninety degrees.

Expect the reading to come in slightly under the nominal size. Standards specify the nominal as the basic major diameter and allow real fasteners to sit below it, so a healthy M8 typically measures somewhere around 7.8 to 7.9 mm and a 1/4 inch bolt around 0.245 to 0.249 inch. A thread measuring a couple of tenths of a millimetre under nominal is normal; one measuring over nominal is either a different size or a plated fastener.

Internal threads are harder. You cannot get calipers onto a tapped hole's major diameter, and the minor diameter you can measure sits well below the nominal - an M10 × 1.5 hole has an internal minor diameter of about 8.38 mm. The practical method is to find a bolt that runs in cleanly by hand and identify the bolt instead. If you have to work from the hole alone, measure the minor diameter and compare it against the internal minor diameters listed on the size pages.

Step 2 - measure the pitch or count the TPI

Metric threads are specified by pitch: the distance from one crest to the next, in millimetres. Inch threads are specified by threads per inch. They describe the same geometry from opposite ends and convert directly - pitch in mm = 25.4 ÷ TPI.

A thread pitch gauge is the fastest and most reliable tool. Try leaves until one seats along the thread with no light visible under any tooth. A leaf that nearly fits is not the answer; the difference between 1.25 mm and 1.5 mm is obvious once the right leaf is in your hand. Gauges come in metric and inch sets, and having both saves a great deal of arguing with yourself.

Without a gauge, lay a steel rule along the thread with the zero mark in a valley and count the crests over one inch - count gaps rather than crests to avoid the classic off-by-one. Over 25 mm, count crests and divide 25 by the count for an approximate pitch. Both methods are workable on coarse threads and get unreliable above about 24 TPI, where a gauge is genuinely necessary.

One caution: metric threads are never specified in TPI. ThreadRef converts and labels the value as a conversion for exactly this reason - an M6 × 1.0 works out at 25.4 TPI, which is not a number any tap is sold by.

• Pitch gauge: the leaf must seat with no light under any tooth.
• Steel rule: count the gaps over one inch, not the crests.
• Convert with pitch (mm) = 25.4 ÷ TPI.
• Above roughly 24 TPI, use a gauge - counting is not accurate enough.

The pairs that catch people out

Two failure modes account for most misidentifications, and they are mirror images of each other. In the first, two threads share almost exactly the same diameter and are separated only by pitch. In the second, two threads share almost exactly the same pitch and are separated only by diameter. Knowing which situation you are in tells you which measurement has to be precise.

M8 against 5/16 inch is the classic diameter collision: 8.0 mm against 7.938 mm, a difference of 0.062 mm that no ordinary caliper reading through workshop conditions will resolve. The pitches are not close at all, though - 1.25 mm for M8 coarse against 1.411 mm for a 5/16-18 UNC - so the gauge settles it instantly.

M5 against #10 is the mirror case. The pitches are effectively identical: M5 × 0.8 has a pitch of 0.8 mm and a #10-32 UNF works out at 0.794 mm, a difference of six microns that no pitch gauge will ever show. The diameters, however, are 5.0 mm against 4.826 mm - 0.174 mm apart, which a caliper reads comfortably. Here the diameter is the discriminator and the pitch is useless.

The lesson is to take both measurements every time, rather than deciding from whichever one you happened to take first.

Step 3 - check the form, the taper and the hand

Diameter and pitch place a thread in a size. The form places it in a family, and getting that wrong is what damages sealing surfaces. ISO metric and Unified threads both use a 60° included flank angle, which is why they look so similar and why their sizes cross-contaminate so easily. Whitworth-derived threads - including BSPP and BSPT - use 55° with rounded crests and roots, and a good pitch gauge set will include 55° leaves that show the difference.

Taper is the other structural check, and it matters most on pipe threads. NPT and BSPT are tapered: the diameter grows along the thread at 1 in 16, which is 3/4 inch per foot, so measuring across the first full thread and again three or four threads along gives you a visibly different number. BSPP is parallel and measures the same all the way along. A tapered male thread seals by wedging into the mating flanks; a parallel one seals on a gasket or an O-ring, and swapping them produces a joint that either leaks or cracks the fitting.

Pipe threads also break the naming convention people expect. The trade size is a legacy pipe bore, not a thread dimension. A 1/2 inch NPT has a major diameter of 21.34 mm - about 0.84 inch - so if you measure 21 mm and go looking for a 21 mm thread you will find nothing. Measure it, then match it against the pipe tables.

Finally, check the hand. Almost everything is right-hand, but left-hand threads turn up on rotating shafts, gas equipment, turnbuckles and some bicycle and machine tool parts. Look at the thread from the side: a right-hand thread's crests run up to the right. A left-hand thread will refuse to start in a right-hand hole, which is easily misread as the wrong size.

• 60° flank angle: ISO metric, UNC/UNF/UNEF/UN, NPT.
• 55° flank angle with rounded crests: BSPP (G) and BSPT (R).
• Tapered: NPT and BSPT, at 1 in 16. Parallel: BSPP and all machine-screw threads.
• Pipe trade size is not the thread diameter - 1/2 in NPT measures 21.34 mm.

Matching the measurements to a size

Once you have a diameter and a pitch, the match is a lookup. The thread identifier takes both numbers, accepts either pitch in millimetres or TPI, and ranks every metric, Unified and pipe thread in the database by how well it fits. It deliberately weights pitch more heavily than diameter, because pitch is what separates a coarse thread from a fine one at the same size and is usually the more precisely measurable of the two.

Read the ranking with your measurement uncertainty in mind. If the top two candidates are 0.06 mm apart in diameter, your caliper has not chosen between them and the pitch has. If they are separated only by pitch and your gauge fitted cleanly, trust the gauge. When two candidates remain genuinely tied, the deciding test is physical: a correct thread runs in freely by hand for its full engagement.

That last point is the one worth ending on. Force applied to a thread that will not start is how good parts become scrap. Resistance in the first turn is not something to overcome - it is the measurement telling you the identification is wrong.