Inch Fraction Converter (Decimal and mm to 1/16, 1/32, 1/64)

Inch Fraction Converter (Decimal and mm to 1/16, 1/32, 1/64)

Decimal inches or millimetres to the nearest fraction on a tape measure, with the residual printed in thousandths of an inch and in millimetres so you can decide whether it matters. Whole inches, numerator and denominator come out as separate numbers, reduced to lowest terms — and feet and inches go in as well as out.

Inches, fractions and the residual

Decimal or mm → nearest 1/2 to 1/64
Only used in the feet-and-inches mode. A US framing dimension is written this way: 5 ft 7 3/8 in.
A tape measure is marked in sixteenths; a machinist’s rule goes to sixty-fourths. The finer the division, the smaller the residual — and the residual is what this page is for.
Not a circuit: one inch of tape measure, to scale, with the tick hierarchy a real tape uses — tallest at the halves, then the quarters, then the eighths, then the sixteenths. Every division on an inch rule is reached by bisection, which is why every denominator is a power of two and why reducing a fraction on a rule is only ever halving. The long line is your own measurement, placed to the nearest sixty-fourth of an inch, so you can see directly how far it sits from the nearest marked division — which is the residual printed below it.
3.93701inExample

100 mm, rounded to the nearest sixteenth

Advertisement

Multiply, round, then halve until it stops

numerator = round(fraction × denominator)  ·  reduce: while the numerator is even, halve both  ·  residual = numerator ÷ denominator − the measurement  ·  inches = millimetres ÷ 25.4, exactly
round
to nearest, ties away from zero. So the residual is never larger than half a division: ±0.03125 in at sixteenths, ±0.0078125 at sixty-fourths
reduce
every denominator here is a power of two, so reduction is nothing but halving the pair while the numerator stays even. Six halvings cover everything up to sixty-fourths, which is why this needs no loop
25.4
millimetres in an inch, exactly, since the 1959 international agreement. Nothing on this page is an approximation except the rounding you asked for
thou
a thousandth of an inch, 25.4 µm. Half a sixteenth is 31.25 thou and half a sixty-fourth is 7.8125 — those are the two numbers worth memorising

Worked example

100 mm, rounded to the nearest sixteenth
100 ÷ 25.4 = 3.93701 in, exactly, because the inch is exactly 25.4 mm
The fractional part is 0.93701, and 0.93701 × 16 = 14.992, which rounds to 15 — so the fraction is 15/16
15 is odd, so 15/16 is already in lowest terms: the answer is 3 whole inches and 15/16
15/16 is 0.9375, so the fraction reads 3.9375 in — which is 0.492 thousandths of an inch over, or 0.0125 mm
Half a sixteenth is 31.25 thou, so this is one of the luckiest roundings on the rule: the residual is one sixty-third of the worst case

Every sixty-fourth, reduced, in decimal inches and millimetres

Sixty-fourthsReduced fractionDecimal inchesMillimetresThousandthsReduced denominator
11/640.0156250.396915.664
21/320.0312500.793731.332
33/640.0468751.190646.964
41/160.0625001.587562.516
55/640.0781251.984478.164
63/320.0937502.381293.832
77/640.1093752.7781109.464
81/80.1250003.1750125.08
99/640.1406253.5719140.664
105/320.1562503.9688156.332
1111/640.1718754.3656171.964
123/160.1875004.7625187.516
1313/640.2031255.1594203.164
147/320.2187505.5562218.832
1515/640.2343755.9531234.464
161/40.2500006.3500250.04
1717/640.2656256.7469265.664
189/320.2812507.1437281.332
1919/640.2968757.5406296.964
205/160.3125007.9375312.516
2121/640.3281258.3344328.164
2211/320.3437508.7312343.832
2323/640.3593759.1281359.464
243/80.3750009.5250375.08
2525/640.3906259.9219390.664
2613/320.40625010.3187406.332
2727/640.42187510.7156421.964
287/160.43750011.1125437.516
2929/640.45312511.5094453.164
3015/320.46875011.9063468.832
3131/640.48437512.3031484.464
321/20.50000012.7000500.02
3333/640.51562513.0969515.664
3417/320.53125013.4937531.332
3535/640.54687513.8906546.964
369/160.56250014.2875562.516
3737/640.57812514.6844578.164
3819/320.59375015.0812593.832
3939/640.60937515.4781609.464
405/80.62500015.8750625.08
4141/640.64062516.2719640.664
4221/320.65625016.6687656.332
4343/640.67187517.0656671.964
4411/160.68750017.4625687.516
4545/640.70312517.8594703.164
4623/320.71875018.2562718.832
4747/640.73437518.6531734.464
483/40.75000019.0500750.04
4949/640.76562519.4469765.664
5025/320.78125019.8438781.332
5151/640.79687520.2406796.964
5213/160.81250020.6375812.516
5353/640.82812521.0344828.164
5427/320.84375021.4312843.832
5555/640.85937521.8281859.464
567/80.87500022.2250875.08
5757/640.89062522.6219890.664
5829/320.90625023.0187906.332
5959/640.92187523.4156921.964
6015/160.93750023.8125937.516
6161/640.95312524.2094953.164
6231/320.96875024.6062968.832
6363/640.98437525.0031984.464
641/11.00000025.40001,000.01
The whole of a machinist’s rule in one table, computed here. The last column shows what the reduction does: thirty-two of the sixty-four numerators are odd and cannot reduce at all, sixteen reduce to thirty-seconds, eight to sixteenths, and so on down to 32/64 = 1/2. That pattern is why reduction by repeated halving terminates in six steps and needs no arithmetic beyond a test for evenness.

Round metric sizes and the nearest fraction at three denominations

mmInchesNearest 1/16Error (thou)Nearest 1/32Error (thou)Nearest 1/64Error (thou)
10.039371/16 in23.131/32 in-8.123/64 in7.50
20.078741/16 in-16.243/32 in15.015/64 in-0.62
30.118111/8 in6.891/8 in6.891/8 in6.89
50.196853/16 in-9.353/16 in-9.3513/64 in6.27
60.236221/4 in13.781/4 in13.7815/64 in-1.85
80.314965/16 in-2.465/16 in-2.465/16 in-2.46
100.393703/8 in-18.7013/32 in12.5525/64 in-3.08
120.472441/2 in27.5615/32 in-3.6915/32 in-3.69
150.590559/16 in-28.0519/32 in3.2019/32 in3.20
200.7874013/16 in25.1025/32 in-6.1525/32 in-6.15
250.984251/1 in15.7531/32 in-15.5063/64 in0.12
301.1811019/16 in6.4019/16 in6.4019/16 in6.40
401.5748025/16 in-12.3025/16 in-12.30101/64 in3.32
501.9685031/16 in-31.0063/32 in0.2563/32 in0.25
752.9527647/16 in-15.2647/16 in-15.26189/64 in0.37
1003.9370163/16 in0.4963/16 in0.4963/16 in0.49
1505.9055147/8 in-30.51189/32 in0.74189/32 in0.74
2007.8740263/8 in0.9863/8 in0.9863/8 in0.98
2509.84252157/16 in-30.02315/32 in1.23315/32 in1.23
30011.81102189/16 in1.48189/16 in1.48189/16 in1.48
A positive error means the fraction is larger than the metric size. Note how the sixty-fourths column is not always four times better than the sixteenths one: rounding is luck, and 100 mm happens to fall almost exactly on 3 15/16 while 15 mm is 28 thou off its nearest sixteenth. The only rows with no error at all are the ones where the metric figure is itself derived from an inch — 25.4, 12.7, 6.35.

The divisions themselves, for reference

DivisionDecimal inchesmmThousandthsSixty-fourths
1/64 in0.0156250.396915.6251.0000
1/32 in0.0312500.793731.2502.0000
1/16 in0.0625001.587562.5004.0000
1/8 in0.1250003.1750125.0008.0000
1/4 in0.2500006.3500250.00016.0000
1/2 in0.50000012.7000500.00032.0000
1 in1.00000025.40001,000.00064.0000
1 thou0.0010000.02541.0000.0640
1 mm0.0393701.000039.3702.5197
1 m39.3700791,000.000039,370.0792,519.6850
The last two rows are the conversion the other way: a millimetre is 0.03937 in, which is 2.52 sixty-fourths — so a millimetre is a bit more than a fortieth of an inch and there is no fraction on any rule that matches it. That single fact is the reason mixed-unit work goes wrong.

A fraction is a nearest value, and the residual is the answer

Converting a decimal to a fraction is not a conversion, it is a rounding. 3.937 in is not 3 15/16 in; it is 3 15/16 in minus half a thousandth. Most of the time that does not matter and occasionally it matters enormously, and the only way to know is to be shown the difference. So this page prints the residual three ways — in thousandths of an inch, in millimetres, and as a share of the measurement — and puts the worst possible residual for your chosen denominator next to it. Half a sixteenth is 31.25 thou, which is 0.79 mm; half a sixty-fourth is 7.81 thou, which is 0.198 mm. Those are the bounds, and nothing you round can ever be further out than that.

The arithmetic is short. Multiply the fractional part by the denominator, round to the nearest whole number, and that is your numerator. Then reduce: because every denominator on a rule is a power of two, reducing is nothing more than halving both numbers while the numerator is even. 14/16 halves to 7/8 and stops because 7 is odd. 48/64 halves to 24/32, then 12/16, then 6/8, then 3/4, and stops. Six halvings covers every case up to sixty-fourths, which is why this needs no loop and no greatest-common-divisor algorithm.

Why the numerator and denominator are separate numbers here. This page’s calculator computes numbers; it has no way to assemble the text “3 15/16”. Rather than print something that looks like a fraction and is not, it gives the whole-inch part, the reduced numerator and the reduced denominator as three rows, plus the unreduced numerator over the denominator you asked for. That is more honest and, for anybody actually working, more useful: the unreduced numerator is the number of tape divisions to count, and the reduced pair is what you say out loud.

Metric to fractions is where the trouble is. Almost nothing round in millimetres is round in inches, because a millimetre is 0.03937 in — about two and a half sixty-fourths, and no rule is marked in those. The exceptions are the sizes that were inches to begin with: 25.4 mm is exactly 1 in, 12.7 mm exactly 1/2, 6.35 mm exactly 1/4, 3.175 mm exactly 1/8. Everything else carries a residual, and the residual does not shrink smoothly as the numbers get bigger — it is a sawtooth, which the chart above shows over a hundred millimetres. 100 mm lands within half a thousandth of 3 15/16 in, which is luck; 15 mm is twenty-eight thousandths off its nearest sixteenth, which is luck in the other direction.

Feet and inches. A US framing dimension is written as feet, whole inches and a fraction — 5 ft 7 3/8 in — and that is three numbers, not one. This page takes them as three inputs and gives them back as three outputs, with the decimal inch (67.375) and the millimetre figure (1711.325, exactly) alongside. For the trade scales that live on the same fractions, see the drill bit size converter, where a 1/4 in drill, an E drill and a 6.35 mm drill are one bit with three names, and the wire gauge converter for the gauge numbers that sit between the fractions. The figures behind this conversion come from a published table, not a definition, so a different source may give slightly different numbers. The reference below is the one used here.

Advertisement

Frequently asked questions

What is 100 mm in inches as a fraction?

3 15/16 in, and it is 0.49 thousandths of an inch — 0.0125 mm — under the fraction. Exactly, 100 mm is 3.937008 in. Half a sixteenth is 31.25 thousandths, so this is one of the closest roundings on the whole rule; most metric sizes are nowhere near this lucky.

How do I convert a decimal to the nearest 1/16?

Multiply the decimal by 16 and round to the nearest whole number; that is your numerator in sixteenths. 0.6 × 16 = 9.6, so 10/16, which reduces to 5/8 — and 5/8 is 0.625, so you are 25 thousandths of an inch over. The reduction is just halving while the numerator stays even.

How accurate is a fraction on a tape measure?

The rounding is bounded by half a division: ±31.25 thousandths of an inch at sixteenths, ±15.6 at thirty-seconds, ±7.8 at sixty-fourths. In millimetres those are ±0.79, ±0.40 and ±0.20. Whether that is accurate enough is a question about your job, not about the arithmetic — and for most carpentry the tape’s own hook play is the larger error.

Why do inch rules use sixteenths and sixty-fourths and not tenths?

Because halving is the one operation you can do with a pair of dividers and no arithmetic: mark the middle, mark the middle of that, and keep going. Every division on an inch rule is reached by bisection, which is also why every denominator is a power of two and why reducing a fraction on a rule is only ever halving.

What is 5 ft 7 3/8 in in millimetres?

1,711.325 mm, exactly. Five feet is 60 in, plus 7 makes 67, plus 3/8 makes 67.375 in, and 67.375 × 25.4 = 1711.325. The inch has been exactly 25.4 mm since 1959, so this is not a rounded figure.

What does “thou” mean?

A thousandth of an inch, 0.001 in, which is 25.4 µm. Machinists also say “mil”, which means the same thing in the United States and something completely different in continental Europe, where a mil can be a millimetre. This page says thou and gives the millimetre figure next to it.

Is 1/64 the finest division I will see on a rule?

On a general steel rule, usually yes, and often only over the first inch or two. Below that you leave the rule and go to a caliper or micrometer reading in thousandths or hundredths of a millimetre, because a sixty-fourth of an inch is 0.4 mm and you cannot reliably split that by eye.

Related calculators

References

  1. National Institute of Standards and Technology, Refinement of Values for the Yard and the Pound, Federal Register notice of 1 July 1959, which fixed the international inch at exactly 25.4 mm and the avoirdupois pound at exactly 0.45359237 kg. Both exact values are used throughout this batch; every inch-to-millimetre figure here is therefore exact, not approximate.
  2. ISO 3:1973, Preferred numbers — Series of preferred numbers, which defines the R5, R10, R20 and R40 series. Copyrighted; cited by number. The R10 series used here — 1, 1.25, 1.6, 2, 2.5, 3.15, 4, 5, 6.3, 8 — is the rounded decimal geometric series with ten steps per decade and is recomputed on this page from 10^(n/10) rather than copied.