Inches to Fraction Calculator

Convert decimal inches to the closest binary fraction.

Convert GCD reduced Error shown
Rate this calculator · 4.5 (4)

Decimal in → Fraction

GCD-reduced output · error in in & mm

Instructions — Inches to Fraction Calculator

1

Pick a precision

1/16″ matches a standard tape measure. 1/32″ suits hardwood joinery and hardware. 1/64″ is the working tolerance for machinist drill charts and CNC bit selection.

2

Enter decimal inches

Type any positive decimal, for example 0.625, 3.375 or 1.4375. The calculator multiplies by the chosen denominator, rounds, and reduces the fraction with the Euclidean algorithm.

3

Read the error

The output shows the rounded fraction, its decimal value, and the rounding error in inches and millimetres. For 0.333″ at 1/16, the error is 0.0205″ (0.52 mm).

Mental math: multiply the decimal part by 16. 0.4375 × 16 = 7, so 0.4375″ = 7/16″.
Max error at 1/16: 1/32″ = 0.0156″ = 0.40 mm.

Formulas

Decimal-to-fraction conversion is a single rounding step followed by a GCD reduction. ASME Y14.5 (geometric dimensioning) and machinist references treat 1/64″ as the smallest practical fractional tolerance for general manufacturing.

Numerator from decimal
$$ n = \text{round}(d_\text{frac} \times D) $$
Take the decimal part (0.375 from 3.375), multiply by the denominator D (16), and round. 0.375 × 16 = 6, so the fraction is 6/16.
GCD reduction
$$ \frac{n}{D} = \frac{n / \gcd(n,D)}{D / \gcd(n,D)} $$
Reduce 6/16: gcd(6,16) = 2, so 6/16 = 3/8. The Euclidean algorithm computes gcd by repeated remainder: gcd(a,b) = gcd(b, a mod b).
Rounding error
$$ \epsilon = \left| d - \left(w + \frac{n}{D}\right) \right| $$
The absolute difference between the original decimal d and the rounded value. For 0.333″ at 1/16, ε = 0.0205″ = 0.52 mm.
Maximum error at denominator D
$$ \epsilon_\text{max} = \frac{1}{2D} $$
Worst-case rounding error is half the smallest tick. At 1/16: 0.03125″ / 2 = 0.0156″ (0.40 mm). At 1/64: 0.0078″ (0.20 mm).
Fraction back to decimal
$$ d = w + \frac{n}{D} $$
To go the other direction, divide numerator by denominator and add the whole-inch part. 3 3/8″ = 3 + 3/8 = 3.375″.
Defined inch (1959 treaty)
$$ 1\text{ in} = 25.4\text{ mm (exact)} $$
The International Yard and Pound Agreement of 1959 fixed the inch at exactly 25.4 mm. Every fractional inch is therefore an exact metric value: 1/16″ = 1.5875 mm.

Reference

Common Fractions — Decimal & Millimetres
FractionDecimal (in)Millimetres
1/640.0156250.3969
1/320.031250.7938
1/160.06251.5875
1/80.1253.1750
3/160.18754.7625
1/40.256.3500
5/160.31257.9375
3/80.3759.5250
7/160.437511.1125
1/20.512.7000
9/160.562514.2875
5/80.62515.8750
11/160.687517.4625
3/40.7519.0500
13/160.812520.6375
7/80.87522.2250
15/160.937523.8125
11.00025.4000

Where each precision applies

Choose the smallest fraction that meets your tolerance. Tighter precision means a smaller rounding error but a less readable fraction.

Trade vs. precision
FieldTypical
Rough framing1/8″
Woodworking1/16″
Joinery / hardware1/32″
Machining (drill bits)1/64″
Plumbing thread1/8″
Max error per precision
NearestMax error
1/2″0.250″ / 6.35 mm
1/4″0.125″ / 3.18 mm
1/8″0.0625″ / 1.59 mm
1/16″0.0313″ / 0.79 mm
1/32″0.0156″ / 0.40 mm
1/64″0.0078″ / 0.20 mm

Source: NIST Handbook 44 and ASME machinist tolerance guidance. Standard CNC tolerances are typically ±0.005″ (0.13 mm), tighter than any fractional precision.

Article — Inches to Fraction Calculator

Inches to Fraction Calculator: From Decimals to Tape-Measure Marks

An inches-to-fraction calculator rounds a decimal value (like 0.625) to the closest binary fraction (5/8″), with the denominator restricted to a power of two: 2, 4, 8, 16, 32, or 64. Maximum rounding error is half the smallest tick, so 1/16 work tops out at 0.79 mm error and 1/64 caps at 0.20 mm.

The binary system survives because most physical tools, from steel rules to drill indexes, are graduated in halves of halves. The inch was fixed at exactly 25.4 millimetres by the 1959 International Yard and Pound Agreement, which makes every fractional inch an exact metric value (1/16″ = 1.5875 mm), useful when crossing between imperial and metric drawings.

What inches to fraction conversion does

The calculator takes a positive decimal inch value and returns the closest fraction whose denominator is a power of two. For 0.625″ at 1/16 precision: 0.625 × 16 = 10, so the fraction is 10/16, which the calculator reduces to 5/8″ via the Euclidean algorithm. The conversion has two steps: rounding to the chosen tick, then simplifying with the greatest common divisor.

Power-of-two denominators are not arbitrary. Tape measures, rulers, drill sets, and wrench rolls are all manufactured around halves: 1/2″, 1/4″, 1/8″, 1/16″, 1/32″, 1/64″. A 1/3″ or 1/5″ mark does not exist on any standard imperial tool because no useful tool divides into thirds or fifths.

Did you know

A standard US machinist drill index contains 29 fractional bits from 1/16″ through 1/2″ in 1/64″ steps. The full set has 64 sizes from 1/64″ to 1″. Decimal-equivalent charts published by university engineering departments accompany every set so machinists can match a fractional bit to a decimal-spec hole.

The inches-to-fraction formula

The math is two operations: round, then reduce. Round the decimal part times the denominator to the nearest integer, that becomes the numerator. Then divide both numerator and denominator by their greatest common divisor to get the reduced form.

Inches-to-fraction shorthand (at 1/16)
0.125″ = 1/8″
0.25″ = 1/4″
0.375″ = 3/8″
0.5″ = 1/2″
0.625″ = 5/8″
0.75″ = 3/4″
0.875″ = 7/8″

The Euclidean algorithm reduces a fraction by computing gcd(a, b) = gcd(b, a mod b) until one of the values is zero. For 10/16: gcd(10, 16) = gcd(16, 10) = gcd(10, 6) = gcd(6, 4) = gcd(4, 2) = gcd(2, 0) = 2. Divide both by 2 and you have 5/8. The algorithm dates back to Euclid’s Elements (c. 300 BCE) and is still the most efficient method for finding GCD.

Choosing the right inches-to-fraction precision

Precision should match the tool you can read. Standard tape measures show 1/16″. Combination squares and precision rules go to 1/32″. Machinist scales reach 1/64″. Going finer than the tool you are using to read the result wastes precision and creates false confidence.

  • 1/2″ (max error 6.35 mm): mental estimates, lumber sizing
  • 1/4″ (max error 3.18 mm): rough framing, drywall
  • 1/8″ (max error 1.59 mm): rough framing, plumbing threads, garage shelving
  • 1/16″ (max error 0.79 mm): finish carpentry, woodworking, hardware
  • 1/32″ (max error 0.40 mm): joinery, cabinet making, fine metalwork
  • 1/64″ (max error 0.20 mm): machinist drill charts, hand-tooled metal
  • Below 1/64″: use decimal inches and the standard CNC tolerance of ±0.005″ (0.13 mm)
Tolerance is not the same as precision

A 1/16″ reading is precise to 0.0625″, but the machinist tolerance for a part can be much tighter. ASME Y14.5 (geometric dimensioning and tolerancing) treats fractional inch dimensions as having an implied tolerance of ±1/64″ unless a tighter callout is specified. For functional fits (bearings, gears, sliding parts), always switch to decimal inches with explicit numerical tolerances.

Inches to fraction in different trades

Different trades pick different default precisions because their tools and materials demand it. Carpentry and rough construction tolerate 1/8″ or 1/16″ because wood moves with humidity (typically 1% per 4% moisture change) and joints are designed with play. Machining and hardware fitting demand 1/32″ or finer because metal parts must mate without forgiveness.

Sewing patterns frequently use 1/8″ seam allowances and 1/4″ hems. Quilting tools have 1/8″ ticks specifically so half-square triangles cut precisely on the bias. Plumbing thread sizes (1/2″ NPT, 3/4″ NPT) describe the nominal pipe ID, not any actual diameter, but the tap and die sets that cut these threads work to 1/32″ pitch tolerance.

Tip

Lumber dimensions are quoted nominal, not actual. A “2×4” measures 1 1/2″ by 3 1/2″ after planing. A “1×6” is 3/4″ by 5 1/2″. American Softwood Lumber Standard PS 20 codifies these differences. If your decimal-inch design assumes nominal dimensions, you will be off by 1/2″ on every board.

Rounding error in inches-to-fraction work

Maximum rounding error is always half the smallest tick. At 1/16″ precision the worst-case error is 1/32″ = 0.0313″ = 0.79 mm. For a single dimension that error is usually invisible. Stack four cuts of 1/16″ rounding and the worst-case drift grows to 0.0625″ (1.59 mm), enough to misalign mortise-and-tenon joinery.

Some decimal inputs have no exact binary fraction. The classic example is 0.333″ (one-third inch). At 1/16: 5/16″ with 0.0205″ error. At 1/32: 11/32″ with 0.0108″ error. At 1/64: 21/64″ with 0.0049″ error. The binary system never closes on one-third, which is one of several reasons machinists abandon fractions for decimal inches whenever a part needs to mate with metric parts or non-binary lengths.

1/16″ max error
0.79 mm
Tape measure tick
1/64″ max error
0.20 mm
Drill index tick
CNC tolerance
0.13 mm
±0.005″

Reading a tape measure

Standard US tape measures use four tick lengths between whole inches. The longest tick (after the inch line) marks 1/2″. The next-longest marks 1/4″ and 3/4″. Shorter ticks mark 1/8″ positions, and the shortest marks 1/16″. Reading a fractional inch is a matter of counting ticks from the nearest whole inch, then reducing mentally: 4/16″ reads as 1/4″, 12/16″ reads as 3/4″.

Some metric-friendly tapes include a 10ths-of-an-inch row alongside the fractional row, which is convenient when copying from a CAD drawing in decimal inches. NIST Handbook 44 specifies tolerance classes for commercial measuring tapes: a Class I tape (the strictest) must be accurate to ±1/32″ over 36 inches.

Common inches-to-fraction mistakes

  • Forgetting to reduce: 4/16″ should be 1/4″. The calculator reduces automatically, but hand-written cut lists often skip the step.
  • Wrong precision for the tool: writing 7/32″ on a cut list when your tape reads only 1/16″ means the carpenter has to guess.
  • Stacking rounding errors: ten cuts at 1/16″ precision can drift 5/16″ cumulatively. Lay out from a single reference edge when possible.
  • Mixing fractions and decimals: a drawing labelled “0.5″” should mean 1/2″ exactly, but in joinery it may imply a tolerance class. Confirm with the engineer.
  • Treating 1/3″ as binary: there is no exact binary fraction for 1/3″. Use decimal inches (0.333″) and accept the rounding, or redesign around binary multiples.
  • Confusing nominal lumber: a 2×4 is not 2″ × 4″. It is 1 1/2″ × 3 1/2″ after planing.

FAQ

Multiply 0.375 by your denominator. At 1/16: 0.375 × 16 = 6, giving 6/16. Reduced by gcd(6,16)=2, the answer is 3/8″. The calculator does the reduction automatically using the Euclidean algorithm.
1/16 inch. Most US tape measures and rulers show ticks down to 1/16″, with the 1/2″ tick longest, then 1/4″, then 1/8″, then 1/16″. For rough framing, 1/8″ is enough. For hardwood joinery, drop to 1/32″.
Find the greatest common divisor (GCD) of 12 and 32. gcd(12, 32) = 4. Divide both: 12 ÷ 4 = 3, 32 ÷ 4 = 8. The reduced fraction is 3/8″. The Euclidean algorithm finds the GCD by repeatedly taking remainders.
7 ÷ 16 = 0.4375 inches, which equals 11.1125 mm exactly (since the inch is defined as 25.4 mm). 7/16″ is a common bolt-hole drill size and a frequent shim thickness.
Historical reason. Binary subdivision (halving a length, then halving again) predates decimal notation by centuries. It is also physically intuitive: splitting a board in half, then half again, naturally produces 1/2, 1/4, 1/8, 1/16. Carpenters used this system long before standard rulers existed.
1/16 inch on a standard tape measure (about 1.6 mm). Precision machinist rules go to 1/32″ or 1/64″. Engineering scales are usually decimal (1/10, 1/50, 1/100), not fractional.
Maximum error is 1/32″ = 0.03125″ = 0.79 mm. For tighter tolerance, choose 1/32″ (max error 0.40 mm) or 1/64″ (max error 0.20 mm). For CNC work, drop to decimal inches and use ±0.005″ as the standard machining tolerance.
There is no exact binary fraction for 0.333″. At 1/16: 5/16″ (0.3125, error 0.0205″). At 1/32: 11/32″ (0.34375, error 0.0108″). At 1/64: 21/64″ (0.328125, error 0.0049″). One-third inch is genuinely awkward in the binary system, which is one reason machinists prefer decimal inches for non-binary lengths.