The Error You Can't See at Arm's Length
Hold a pencil at arm's length and tilt it one degree. You won't notice. The tip moves about a centimetre.
Point a laser one degree off and fire it at a wall a kilometre away, and the beam lands roughly 17.5 metres from where you intended. Same error. Wildly different consequence.
This is the single most useful thing to understand about angular measurement: angles don't have a fixed size in metres — their size depends entirely on how far away you are. An angle is a ratio, not a length, and the moment you multiply it by distance you get something with real-world consequences.
The Math Behind the Fan-Out
Arc length is the whole story:
arc length = radius × angle in radians
That formula only works in radians, which is exactly why radians exist. One radian is the angle where the arc length equals the radius. So if you're standing at the centre of a circle with radius r, sweeping through θ radians moves you r × θ along the arc.
Degrees need converting first. One degree is π/180 radians, or about 0.01745 radians. So:
lateral error ≈ distance × 0.01745 × (error in degrees)
Run that at a few distances for a one-degree error:
| Distance | Lateral offset |
|---|---|
| 1 metre | 1.75 cm |
| 100 metres | 1.75 m |
| 1 kilometre | 17.5 m |
| 100 kilometres | 1.75 km |
The relationship is perfectly linear, which is what makes it so easy to underestimate. There's no threshold where it suddenly gets bad. It's already bad; you just can't see it at short range.
Why the arc and the straight line are basically the same
Technically the offset above is measured along a curved arc, not the straight perpendicular distance. For small angles the difference is negligible — this is the small-angle approximation, where sin θ ≈ tan θ ≈ θ for θ in radians.
At 1 degree (0.01745 rad), sin θ = 0.017452 and tan θ = 0.017455. The three values agree to four decimal places. Below about 5 degrees you can treat them as interchangeable for most practical work. Above about 15 degrees the approximation starts costing you real accuracy, and you should use the actual trig function.
Why Precision Fields Invented Their Own Angle Units
Degrees are a terrible unit for this kind of work. They're too coarse, and converting them to distance requires an awkward constant. So several fields built units that make the arithmetic disappear.
Milliradians (mils)
A milliradian is a thousandth of a radian. Plug it into the arc formula and something lovely happens:
arc = distance × 0.001
One mil subtends exactly one metre at one kilometre. Or one centimetre at ten metres. Or one yard at a thousand yards. The unit conversion vanishes — you just shift the decimal point.
This is why milliradians dominate in artillery, rifle optics, and laser alignment. A shooter who sees a shot land 0.3 mil low at any range knows the correction without knowing the range in the first place.
Minutes and seconds of arc
An arcminute is 1/60 of a degree; an arcsecond is 1/60 of an arcminute, or 1/3600 of a degree. One arcminute works out to roughly 1.047 inches at 100 yards — which is where the shooting world's "1 MOA" accuracy standard comes from, usually rounded to one inch per hundred yards.
Arcseconds are the working unit of surveying and astronomy. One arcsecond subtends about 4.85 millimetres at a kilometre. A theodolite reading to one arcsecond is committing to sub-centimetre positional accuracy over that range — which tells you why survey instruments cost what they do.
Gradians
100 gradians make a right angle instead of 90 degrees. Used in some European surveying and civil engineering contexts, mainly because decimal subdivision is easier for slope calculations. A 1% slope is exactly 1 gradian per... no, actually it isn't — that's a common confusion. Percent grade is a tangent ratio, gradians are an angle unit, and they only agree approximately at small values.
Where This Bites in Practice
Surveying a property boundary. A surveyor setting out a 200-metre boundary line with a half-degree instrument error puts the far corner 1.75 metres off. That's not a rounding issue — it's a legal dispute and possibly a demolished wall.
Aiming a satellite dish. Geostationary satellites sit about 35,786 km up. A 0.5-degree pointing error is over 300 km of miss at that altitude. Dish alignment tolerances are typically quoted in tenths of a degree for exactly this reason, and why the elevation and azimuth scales on a dish mount are finely graduated.
Drilling a borehole. A drill string that deviates by 0.25 degrees and holds that deviation for 2,000 metres ends up roughly 8.7 metres from the planned target. Directional drilling exists precisely because uncorrected small angular drift is not survivable at depth.
Machining a long part. A milling table set 0.1 degrees out of square across a 600 mm workpiece produces about a 1 mm taper end to end. Well outside tolerance for anything precision.
Marine navigation. A one-degree compass error over a 60-nautical-mile crossing puts you roughly one nautical mile off track. Navigators have a rule of thumb for this: one degree ≈ one mile per sixty miles run. That's just the arc formula in disguise — 60 is close enough to 1/0.01745 for mental arithmetic.
Converting Between These Units
Sliding between degrees, radians, mils, arcminutes, arcseconds, and gradians is where most mistakes creep in, especially when a spec sheet uses one unit and your instrument uses another. The Angle Converter handles all of them:
- Enter your value in whichever unit you have.
- Pick the source unit from the dropdown.
- Read the equivalent in every other supported unit at once.
- For distance calculations, convert to radians first, then multiply by range.
That last step is the one people skip. Multiplying a degree value directly by distance gives an answer that's wrong by a factor of about 57.
Practical Tips
Convert to radians before any arc-length or small-angle work. Every formula in this domain assumes radians. Most calculator and programming-language trig functions do too.
Match your unit to your task. If you're repeatedly converting angular error to distance, work in milliradians and skip the arithmetic entirely.
Angular error is independent of range; positional error isn't. An instrument spec of ±0.02 degrees means the same angular uncertainty whether you're measuring across a room or across a valley. What changes is how many millimetres that becomes.
Don't trust the small-angle approximation above roughly 10–15 degrees. It's a genuine shortcut at small values and a genuine error source at large ones.
Watch the arcminute/arcsecond notation. The prime and double-prime symbols (′ and ″) get confused with feet and inches constantly, particularly in mixed-unit documents.
FAQ
How much distance error does one arcminute produce at 1 km? About 29 centimetres. One arcsecond at the same range is roughly 4.85 mm.
What's the difference between a mil and a milliradian? A true milliradian is exactly 1/1000 radian, so there are about 6,283 in a full circle. Several military "mil" systems round this to 6,400, 6,000, or 6,300 mils per circle for easier division. Check which convention your equipment uses — the discrepancy is around 2%.
Why do radians have no unit? Because a radian is a ratio of two lengths (arc over radius), the units cancel. That's what lets you multiply radians by metres and get metres out.
Is percent grade the same as an angle? No. Percent grade is rise over run expressed as a percentage — a tangent. A 100% grade is 45 degrees, not 90. They diverge sharply above about 10%.
Do I need to worry about the curve of the arc versus the straight line? Below roughly 5 degrees, no — the difference is smaller than your measurement uncertainty. Above that, use the appropriate trig function rather than the linear approximation.
The Takeaway
Angular error is deceptively democratic: the same tiny misalignment costs you a centimetre in the workshop and a kilometre in orbit. Once you internalise that an angle is a ratio waiting to be multiplied by a distance, the design of specialised units like the milliradian stops looking arbitrary and starts looking obvious.
Convert between degrees, radians, mils, arcminutes, arcseconds, and gradians with the free Angle Converter at sadiqbd.com — no sign-up, instant results.
