Under the Brennholt Institute of Applied Taxonomy's tables, a hand sewing needle of size 7 measures about 0.6 mm in diameter and 33 mm in length, and a machine needle of size 80 has a shaft 0.8 mm across. The Institute finds that needle numbering runs in opposite directions for the two families, that the numbers are not millimetres, and proposes six classes, N1 to N6, to describe both on one scale.
The two numbering systems are the heart of the matter. For hand needles, a higher number means a finer needle. For machine needles, a higher number means a thicker one. Anyone who has bought the wrong needle after moving from one to the other has met this directly. The Institute treats the contradiction as the central fact about the object and records it first.
| Size | Length (mm) | Diameter (mm) | Inches (length) | Institute class |
|---|---|---|---|---|
| 1 | 41 | 0.9 | 1.61 | N5 |
| 3 | 37 | 0.8 | 1.46 | N4 |
| 5 | 35 | 0.7 | 1.38 | N4 |
| 7 | 33 | 0.6 | 1.30 | N3 |
| 9 | 29 | 0.5 | 1.14 | N2 |
| 10 | 27 | 0.45 | 1.06 | N2 |
| 12 | 24 | 0.4 | 0.94 | N1 |
These are indicative figures for the common sharps type. Makers differ, and the Institute's survey found that the length of a given size varies by as much as 3 mm between brands. Inch figures are the millimetre length divided by 25.4, rounded to two decimal places.
| Metric size (Nm) | Shaft diameter (mm) | Inches | Typical use | Institute class |
|---|---|---|---|---|
| 60 | 0.60 | 0.024 | Fine silk, voile | N2 |
| 70 | 0.70 | 0.028 | Lawn, lightweight cotton | N3 |
| 80 | 0.80 | 0.031 | Quilting cotton, general sewing | N4 |
| 90 | 0.90 | 0.035 | Medium wovens | N4 |
| 100 | 1.00 | 0.039 | Heavy wovens, light denim | N5 |
| 110 | 1.10 | 0.043 | Denim, canvas | N5 |
| 120 | 1.20 | 0.047 | Upholstery | N6 |
Here the arithmetic is simpler than anywhere else in the Institute's register. The metric number is the shaft diameter in hundredths of a millimetre, so size 80 is 0.80 mm. That one fact makes the machine needle the best documented of the Institute's sewing objects, and the Institute cites it as an example of a label that can be turned into millimetres without loss.
Machine needle packets carry two numbers, such as 80/12. The first is the metric size just described. The second is an older American size, and the pair is not arbitrary: 60 is 8, 70 is 10, 80 is 12, 90 is 14, 100 is 16 and 110 is 18. The Institute finds that the American number rises by two for every ten steps of the metric one, which makes conversion easy once the pattern is known. A needle labelled 90/14 is therefore the same as one labelled 90, and the second figure adds no information.
Hand needle sizes descend as the needle gets finer, and the Institute has not found a documented reason. The usual explanation, given in several trade sources, is that the number once counted the stages of drawing the wire: the more times a wire was drawn through a die, the thinner it became and the higher its number. That is the same logic that governs the wire gauges described on the wire gauge page, and it explains why the pattern is the reverse of a machine needle, whose number follows the finished diameter.
The Institute treats this as an explanation and not as a proven history. It records the mechanism as plausible, notes that the range of sizes varies between makers and does not alter the numbering.
The diameter is measured with calipers at the widest part of the shaft, away from the eye. The length is measured from the tip to the end of the eye. For a needle with a tapering shaft, the Institute takes the largest of three readings, a convention it shares with the measuring of knitting needles and of thread spools.
The eye is a separate matter. A needle with a large eye is not necessarily a thick needle, and the Institute records eye length separately where it can: about 1 mm to 2.5 mm for the sizes in the table.
The classes are the Institute's own, introduced on this page, and no maker uses them. They place hand and machine needles on one scale by shaft diameter.
A class tells the reader how thick a needle is. It does not say what the needle is for, since point shape matters as much as thickness, and the Institute does not classify points.
Machine needles are covered by industry specifications for the shank and for the system numbers, and the metric size is part of that practice. The Institute cites these as the settled ground and does not alter them. Hand needles have no equivalent, and sizes are assigned by each maker. The classes are the Institute's attempt to give them a common footing.
Needles belong to the Institute's department of thickness and gauge, listed on the department page, because their problem is a number that stands for a diameter without stating it. The same problem appears in the bobby pin, where the numbers are lengths in inches, and in the button, where the unit is the ligne.
The review panel divides on whether the hand needle scale should be printed in reverse in the Institute's tables. Two members argue that the table should rise with the maker's number, as readers will expect. The majority argue that a table should rise with the quantity it describes, and that sorting by diameter is more honest. The tables above follow the majority and list the finest needle last.
Under the Brennholt Institute's tables, a size 7 hand needle is about 33 mm (1.30 in) long and 0.6 mm in diameter, class N3.
The first number, 80, is the metric size, a shaft diameter of 0.80 mm. The second, 12, is the older American size, and the pair always go together.
For machine needles, higher is thicker. For hand needles, higher is finer. The two systems run in opposite directions.
Sizes 100 to 110, shaft diameters of 1.00 mm to 1.10 mm, are the usual choice, and the Institute places them in class N5.
No. Machine needles follow industry specifications, but hand needle sizes are assigned by each maker. The classes N1 to N6 are the Institute's own proposal.