Sundial Maths: Solar Angles & Timekeeping

A sundial turns a shadow into a clock, but the shadow is not labelled with hours by nature. Geometry must connect the Sun’s apparent motion, Earth’s rotation, latitude, longitude and the orientation of a gnomon. Time-zone rules and the equation of time then connect local apparent solar time to a modern clock.

That chain makes sundials a delightful answer to “Why is mathematics important?” Angles become time. Trigonometry converts a rotating celestial picture into hour lines on a flat surface. Periodic functions explain why sundial noon and clock noon drift apart through the year.

This article is a learning guide, not a navigation or precision-time standard. Never look directly at the Sun, even through cameras, binoculars or improvised filters. A safe sundial investigation observes shadows on the ground.

One word, time, three different ideas

Everyday conversation treats time as one quantity, but sundials expose several definitions.

Apparent solar time

Apparent solar time follows the actual Sun. Apparent solar noon occurs when the Sun crosses the local meridian and is highest for that day. A correctly aligned sundial is fundamentally an apparent-solar-time instrument.

Mean solar time

The apparent Sun does not move uniformly against our clock through the year. Mean solar time imagines a fictitious mean Sun moving uniformly. The difference between apparent and mean solar time is called the equation of time.

Civil time

Civil time uses time zones, legal rules and atomic timekeeping infrastructure. Everyone within a time zone normally shares the same clock time even though their longitudes differ.

The U.S. Naval Observatory’s explanation of the equation of time describes it as the variable part of the difference between time kept by an ordinary clock and time kept by the Sun. NIST likewise notes that the difference between apparent and mean solar time can reach roughly 16 minutes in magnitude.

A sundial therefore should not be dismissed as “wrong” when it differs from a watch. It may be answering a different time question.

Earth’s rotation converts time to angle

Earth rotates approximately 360° relative to the mean Sun in 24 hours.

360° ÷ 24 h = 15° per hour.

This creates hour angle, H:

H = 15° × (local solar time − 12 h).

At apparent solar noon, H = 0°.

At 10:00 apparent solar time:

H = 15° × (10 − 12) = −30°.

At 15:00 apparent solar time:

H = 15° × (15 − 12) = +45°.

Sign conventions vary, so a writer must state whether morning angles are negative or positive. Geometry can be correct under either convention if it is applied consistently.

Why 15° is only the beginning

It might seem that hour lines should be spaced 15° apart on every dial. That is true for some equatorial dial arrangements, but not for a horizontal dial. Projection onto a surface changes the spacing.

This is the same mathematical idea seen in map projections: equal angles or equal intervals in one coordinate system do not necessarily remain equally spaced after projection onto another surface.

Did You Know?

Modern official time does not use the Sun as its direct oscillator. NIST’s explainer What Determines the Length of the Day? contrasts intuitive solar time with modern time definitions based on atomic standards. A sundial links astronomy, while an atomic clock links frequency to a defined atomic transition.

The gnomon points toward the celestial pole

The gnomon is the shadow-casting element. On a common horizontal sundial, the straight edge that marks time is called the style. To make the hour-angle geometry work, the style is aligned parallel to Earth’s rotation axis.

In the Northern Hemisphere it points toward the north celestial pole; in the Southern Hemisphere, toward the south celestial pole.

For a horizontal dial at latitude φ, the style’s angle above the horizontal plane is approximately |φ|.

At latitude 52° N, the style rises about 52° above the horizontal toward north.

At latitude 1.35° N, near Singapore, the style is only about 1.35° above the horizontal toward north. That is extremely shallow. A small alignment or construction error can be large relative to the intended angle.

Why latitude appears

Latitude measures the angle between the equatorial plane and the line from Earth’s centre to the observer, or equivalently the angular height of the celestial pole above an ideal horizon.

Because the style must be parallel to Earth’s axis, its altitude equals the observer’s latitude in the idealised geometry.

Near-equatorial complications

Near the equator:

the polar style is nearly horizontal;
the Sun can pass north or south of the observer at different seasons;
shadows may switch sides around noon;
the Sun may be nearly overhead on certain dates; and
a conventional small horizontal dial can be hard to read accurately.

This does not make a sundial impossible. It makes design choice and alignment more important. An equatorial ring, vertical arrangement or carefully designed two-sided dial may be more educational for some settings.

The horizontal sundial hour-line formula

For an ideal horizontal sundial with a polar-pointing style, the angle θ of an hour line from the noon line satisfies:

tan θ = sin φ × tan H.

Here:

φ is latitude;
H is hour angle; and
θ is the angle of the hour line from the local-noon line on the dial plate.

Use a consistent degree or radian mode on the calculator.

Worked example at 40° latitude

Find the 10:00 solar-time line.

At 10:00:

H = −30°.

Then:

tan θ = sin 40° × tan(−30°).

sin 40° ≈ 0.6428.

tan(−30°) ≈ −0.5774.

tan θ ≈ −0.3711.

θ ≈ arctan(−0.3711) ≈ −20.4°.

The 10:00 line is about 20.4° from the noon line on the morning side.

For 14:00, H = +30°, giving θ ≈ +20.4°. The ideal geometry is symmetric about noon.

Hour-line table at 40° N

Solar time	Hour angle H	Hour-line angle θ
09:00	−45°	about −32.7°
10:00	−30°	about −20.4°
11:00	−15°	about −9.8°
12:00	0°	0°
13:00	+15°	about +9.8°
14:00	+30°	about +20.4°
15:00	+45°	about +32.7°

The differences are not all 15°. Projection compresses lines near noon at this latitude.

Near Singapore

Using φ = 1.35°:

sin 1.35° ≈ 0.02356.

At H = 45°:

tan θ = 0.02356 × 1 = 0.02356.

θ ≈ 1.35°.

Even three hours from noon, the hour-line angle is only about 1.35° from the noon line under this ideal formula. Many lines crowd together. That is a mathematical reason why a conventional horizontal polar-style sundial near the equator is challenging to read.

Solar altitude explains shadow length

A vertical stick can be used as a simple shadow instrument, even though equally spaced clock-hour marks are not produced automatically.

If a vertical stick has height g and the Sun’s altitude above the horizon is α, shadow length L on level ground is:

L = g ÷ tan α.

If g = 1.0 m and α = 30°:

L = 1 ÷ tan 30° ≈ 1.73 m.

If α = 60°:

L = 1 ÷ tan 60° ≈ 0.577 m.

As the Sun rises higher, the shadow becomes shorter.

Solar altitude from latitude, declination and hour angle

An ideal astronomical relationship is:

sin α = sin φ sin δ + cos φ cos δ cos H.

Here δ is solar declination, the Sun’s angular position north or south of the celestial equator.

At local solar noon, H = 0°, so cos H = 1. The Sun’s noon altitude is approximately:

α_noon = 90° − |φ − δ|,

when interpreted for the ordinary range.

Equinox example

At the equinox, δ is approximately 0°.

At latitude 40° N:

α_noon ≈ 90° − 40° = 50°.

A 1 m vertical stick has a noon shadow:

L = 1 ÷ tan 50° ≈ 0.839 m.

Near latitude 1.35° N:

α_noon ≈ 88.65°.

L ≈ 1 ÷ tan 88.65° ≈ 0.0236 m, or about 2.36 cm.

Near an overhead Sun, a vertical-stick shadow becomes very short and its direction becomes sensitive to tiny levelling and measurement errors.

Solar declination changes through the year

Earth’s axis is tilted relative to its orbital plane. The Sun’s apparent declination changes between roughly +23.4° and −23.4° over a year.

A simple classroom approximation is:

δ ≈ 23.44° × sin(360° × (284 + n) ÷ 365),

where n is day number. Different approximations use different phase constants and leap-year treatment.

This formula is useful for plotting, not precision ephemerides.

Declination changes:

sunrise and sunset directions;
daylight duration;
solar altitude;
shadow length; and
which side of an east–west line the noon Sun occupies near the tropics.

The analemma

Photograph the Sun from the same place at the same civil clock time over a year—with safe professional methods, never by looking through an optical device—and its positions trace a figure-eight-like curve called an analemma.

One dimension comes mainly from changing declination. The other reflects the equation of time. The figure joins two periodic effects into one geometric pattern.

For students, a safer alternative is to calculate and plot declination against equation-of-time values from an authoritative table.

Longitude and the equation of time

A sundial reads local apparent solar time. A watch reads civil time. Converting between them usually needs at least:

a longitude correction relative to the time-zone meridian;
the equation of time for the date; and
any daylight-saving or legal offset.

Longitude correction

Earth rotates 15° per hour, so:

1° of longitude corresponds to 4 minutes of mean solar time.

If a location is 5° west of its time zone’s standard meridian, local mean solar noon occurs about:

5 × 4 = 20 minutes

later than 12:00 standard time, before the equation-of-time correction.

Locations east of the standard meridian reach mean solar noon earlier.

Equation of time

Apparent solar time differs from mean solar time because:

Earth’s orbit is elliptical, so orbital speed varies; and
Earth’s axis is tilted, so motion along the ecliptic projects unevenly onto the celestial equator.

The correction varies through the year and changes sign. It is not a constant “sundial error.”

The U.S. Naval Observatory gives an example: near the end of July, the equation of time is about −7 minutes under its sign convention, so a sundial is about seven minutes behind local mean solar time. Always check the stated convention before adding or subtracting.

NIST’s Time and Frequency A to Z entry notes that the apparent Sun can precede or follow the mean Sun by as much as about 16 minutes.

A safe conversion template

Rather than memorising signs, write meanings:

calculate what the longitude implies for local mean solar noon;
read the equation-of-time definition used by the source;
test the correction on a known example; and
then apply any legal clock offset.

Sign errors are more likely when formulas are copied without definitions.

Worked example: why solar noon is not 12:00 in Singapore

Singapore uses UTC+8 civil time. The corresponding central meridian for a simple 15°-wide zone is:

8 × 15° = 120° E.

Singapore is near 103.8° E. The longitude difference from 120° E is:

120.0° − 103.8° = 16.2°.

Singapore lies west of the nominal standard meridian, so local mean solar noon is later than 12:00 civil time by:

16.2 × 4 minutes = 64.8 minutes.

Ignoring the equation of time, mean solar noon is therefore around 13:04:48.

The equation of time can then shift apparent solar noon earlier or later by several additional minutes depending on date and sign convention.

This explains a familiar tropical observation: the shortest shadow may occur closer to 1 pm Singapore time than noon.

What the estimate assumes

The calculation assumes:

longitude 103.8° E as a representative location;
UTC+8 without daylight saving;
4 minutes per degree;
no equation-of-time correction yet; and
an unobstructed local horizon is not needed for meridian crossing.

Different parts of Singapore differ slightly in longitude, creating small time differences. A precise prediction should use exact coordinates and authoritative solar data.

Clock noon, mean noon and apparent noon

For a given date:

civil noon is fixed at 12:00 by the clock;
local mean noon reflects longitude within the zone; and
apparent noon adds the date-dependent equation of time.

Separating these layers prevents the common statement that “the Sun is late.” The Sun follows apparent solar geometry; the legal clock follows a regional convention.

Alignment is a measurement problem

A correct hour-line drawing can still give poor readings if the dial is not installed correctly.

Key sources of error include:

the dial plate is not level;
the style angle does not match latitude;
the noon line is aligned to magnetic north instead of true north;
the gnomon is twisted;
the printed scale is distorted;
the time is read from the wrong edge of a thick shadow;
the observer applies the wrong longitude sign; or
the equation-of-time value uses another convention.

True north and magnetic north

A compass points approximately toward magnetic north, not geographic true north. The angular difference, magnetic declination, depends on place and changes over time. Nearby metal and electronics can disturb a compass.

For a classroom dial, true-north alignment can be found by observing the local meridian direction from the shortest-shadow method over a suitable interval, using safe observation and careful timing. The minimum shadow method itself needs repeated data because one measurement may miss the exact minimum.

Uncertainty near the equator

Suppose a Singapore polar style should rise 1.35° but construction uncertainty is ±0.5°. The relative angular uncertainty is large:

0.5 ÷ 1.35 × 100% ≈ 37%.

That does not translate directly into a 37% time error, but it shows why precise physical construction is difficult. Choosing a dial geometry suited to low latitude may be wiser than demanding unrealistic accuracy from cardboard.

Shadow width

The Sun is not a mathematical point. Its apparent disk has angular width, so a shadow edge can be soft. A thick gnomon creates two edges. Decide which edge or centre line represents time and use it consistently.

Measurement resolution should match the apparatus. Reporting time to the nearest second from a broad pencil shadow is false precision.

How students can build and test a paper sundial

Choose a geometry suited to the learning goal. A simple vertical-stick shadow plot teaches solar motion. A horizontal polar-style paper dial teaches projection but becomes difficult near the equator.

Step 1: State the location

Record latitude and longitude from an authoritative map. Do not publish a child’s precise home location; a school or approximate public-area coordinate is enough.

Step 2: Choose the dial model

For a horizontal dial, calculate hour lines using:

tan θ = sin φ tan H.

For a shadow-tracking activity, use a vertical stick and measure the tip position rather than pre-drawing equal clock hours.

Step 3: Draw a coordinate baseline

Mark the noon line and a perpendicular east–west line. Check the printer scale with a ruler.

Step 4: Construct the style

Set its angle to |φ| for the ideal horizontal model. Near-equatorial classes may use a larger demonstration globe model first so the geometry is visible.

Step 5: Align safely

Place the dial on a stable, level surface. Align the noon line to the true local meridian using a teacher-approved method. Do not stare at the Sun.

Step 6: Record observations

At intervals, record:

civil time;
shadow reading;
date;
weather;
whether the shadow edge was clear; and
any movement of the apparatus.

Step 7: Apply corrections

Calculate the longitude correction and obtain an equation-of-time value from an authoritative source. State the source’s sign convention.

Step 8: Analyse residuals

Residual = observed civil time − predicted civil time.

Plot residual against time of day. A constant offset suggests one kind of error; a changing pattern may suggest alignment, levelling or hour-line geometry issues.

Step 9: Report honestly

Give uncertainty and limitations. A paper dial that reads within several minutes can be a successful model if its construction and shadow width cannot support finer precision.

A data-analysis example

Suppose a class predicts civil time from a corrected sundial and records:

Trial	Clock time	Sundial-based prediction	Residual
1	10:00	09:54	+6 min
2	11:00	10:56	+4 min
3	12:00	11:58	+2 min
4	13:00	13:00	0 min
5	14:00	14:02	−2 min

The residual changes systematically rather than randomly. A simple mean residual is:

(6 + 4 + 2 + 0 − 2) ÷ 5 = 2 minutes.

Subtracting two minutes from every prediction would remove the mean bias but not the trend. The trend may indicate that the hour-line scale or alignment is rotated.

Fit a line:

residual ≈ intercept + slope × hours from noon.

A non-zero slope points toward a scale or orientation effect. Scattered residuals on a cloudy day may instead reflect reading uncertainty.

This is calibration thinking: not merely asking whether an instrument is wrong, but examining the structure of its error.

Repeated observations and confidence

Suppose the class reads the same predicted mark on five clear days after applying date corrections. Residuals are +3, +5, +4, +2 and +6 minutes.

The mean residual is:

(3 + 5 + 4 + 2 + 6) ÷ 5 = 4 minutes.

The deviations from the mean are −1, +1, 0, −2 and +2 minutes. Their squared values sum to 10. The sample standard deviation is:

√(10 ÷ 4) ≈ 1.58 minutes.

The four-minute mean suggests a systematic offset, while the roughly 1.6-minute scatter describes repeatability in this small sample. Correcting the mean offset may improve agreement, but only after identifying a defensible cause such as orientation.

If every observation was made by the same person from the same viewing angle, the sample does not capture all uncertainty. A second observer, another time of day and a re-levelled setup can reveal additional effects.

Students can compare:

repeatability: same method under similar conditions;
reproducibility: changed observer or setup;
accuracy: agreement with an appropriate reference; and
resolution: the smallest shadow change that can be read.

A dial can be repeatable but biased. It can show the same four-minute error every day. That distinction appears throughout science, from laboratory sensors to satellite instruments.

What the model does not include

The ideal formulas assume:

a defined flat dial plane;
a correctly polar-aligned style;
known geographic coordinates;
an ideal horizon for some solar observations;
standard astronomical approximations; and
negligible movement of the dial.

High-precision solar work also considers atmospheric refraction, exact ephemerides, Earth-orientation parameters, elevation and coordinate definitions.

A sundial cannot normally show one universal civil time without location- and date-specific correction. Time zones are human conventions; daylight-saving rules are legal conventions. The shadow does not know them.

The equation of time is not caused by one effect

It combines the elliptical orbit and axial tilt. Explaining it only as “Earth speeds up and slows down” is incomplete. Explaining it only as tilt is also incomplete.

Solar noon is not always halfway between sunrise and sunset on a clock display

Under ideal definitions, apparent solar noon relates to the Sun crossing the meridian, but observed sunrise and sunset times are affected by refraction, solar disk definition, horizon altitude and terrain. Civil clock corrections must also be consistent.

A sundial is not a longitude finder by itself

Comparing local solar time with a reference time can help infer longitude only when the reference clock, date corrections and observations are accurate. Historically, obtaining reliable time at sea was a major challenge.

Equal hours are a mathematical convention

Modern clocks divide a day into 24 equal hours. That feels natural now, but a shadow invites another possibility: divide daylight into twelve parts from sunrise to sunset.

Those are unequal, or seasonal, hours. A daylight hour is longer in a season with longer days and shorter in a season with shorter days.

Suppose sunrise is at 06:30 and sunset at 19:00. Daylight lasts:

12 h 30 min = 750 minutes.

One seasonal daylight hour would be:

750 ÷ 12 = 62.5 minutes.

On another date, daylight may last 11 hours:

660 ÷ 12 = 55 minutes.

A dial marked for seasonal hours answers a different question from a dial marked for equal hours. Neither division is forced by the shadow itself; each reflects a timekeeping convention.

This is an important mathematical lesson. Before comparing readings, ask whether the scales define the same unit.

Local time and railway time

When communities used local solar time, towns at different longitudes had different noons. Travel and communications made that inconvenient. Standard time zones let a broad region coordinate one civil clock.

If two locations differ by 3° of longitude, their local mean solar times differ by about:

3 × 4 = 12 minutes.

Under one time zone, their clocks can still agree exactly. The standard improves coordination by deliberately ignoring some local solar difference.

Mathematics supports both systems:

geometry and astronomy define local solar time;
longitude converts between local meridians;
agreed offsets define zones; and
atomic standards realise uniform seconds.

The choice of civil convention is social and legal, not an error in astronomy.

Reading historical statements carefully

A historical record saying an event happened at “six o’clock” may not map directly to 06:00 or 18:00 on a modern clock. Researchers must ask:

Were hours equal or seasonal?
Did the day begin at midnight, sunset or another boundary?
Was the time local or standard?
Which calendar was in use?
How accurate was the instrument?

This is source criticism joined to quantitative reasoning. Numbers can look exact while their definitions have changed.

NIST historical context

NIST’s Timekeeping and clocks FAQs discusses early divisions of daylight and night and links to broader time-and-frequency publications. Its historical material helps students see that measurement systems develop in response to human needs as well as physical regularities.

Designing a dial as an inverse problem

The usual exercise calculates where hour lines should go. A more challenging activity starts with measured shadow lines and asks what can be inferred.

Possible unknowns include:

dial rotation away from true north;
gnomon angle;
effective latitude used in the construction;
constant clock offset; and
reading bias caused by a thick gnomon.

Students can choose parameters that minimise squared residuals:

sum of (observed line angle − predicted line angle)².

That is a least-squares fitting problem.

If every measured line is rotated by roughly the same amount, the dial may have an orientation offset. If error grows farther from noon, the assumed latitude or scale may be wrong. If morning and afternoon errors are asymmetric, the gnomon may be twisted or the plate uneven.

Fitting cannot automatically identify one true cause. Different error combinations may produce similar residuals. The experiment needs physical inspection and independent measurements.

This is how many scientific inverse problems work: observations are used to estimate hidden parameters, but identifiability and uncertainty must be examined.

Guidance for parents, students and educators

For students

Draw a labelled diagram before using trigonometry. Identify the dial plane, style, noon line, hour angle and latitude.

Keep solar, mean and civil time in separate columns. Most confusion comes from mixing them.

For parents

A shadow investigation is an inexpensive way to connect maths with the sky. Help with a stable setup and safe observation. Celebrate a careful error analysis more than an apparently perfect reading.

For educators

Use a globe and a skewer through its axis. Rotate the globe while keeping the light direction approximately fixed. This makes the polar-style alignment intuitive.

At low latitude, do not force a tiny cardboard gnomon to deliver unrealistic precision. Compare dial geometries or model a higher-latitude site mathematically.

For interdisciplinary learning

Sundials connect:

mathematics through trigonometry and periodic functions;
physics through light and rotation;
geography through latitude, longitude and time zones;
history through changing time conventions;
design through legibility and fabrication; and
computing through solar-position algorithms.

The project becomes stronger when every discipline retains its own evidence and vocabulary.

Common misconceptions

“Sundial hour lines are always 15° apart”

Hour angle changes 15° per solar hour, but projection onto a horizontal dial produces non-uniform hour-line spacing.

“The gnomon should stand vertically”

A vertical stick can track shadows, but a conventional horizontal sundial’s time-marking style is aligned parallel to Earth’s axis.

“Solar noon is 12:00”

Only under particular longitude, date and civil-time conditions. Time-zone longitude and the equation of time usually create a difference.

“The equation of time is a mistake in the sundial”

It is the real seasonal difference between apparent and mean solar time under a stated convention.

“A compass gives true north”

It gives magnetic north approximately and can be disturbed. True-north alignment requires correction or another method.

“A longer shadow means it is later”

Shadow length usually decreases toward solar noon and increases afterwards. The same length can occur in morning and afternoon.

“More precise calculator output makes a precise sundial”

Construction, alignment and shadow width may dominate the error.

Frequently asked questions

What mathematics is used in a sundial?

Angles, ratios, trigonometric functions, coordinate geometry, periodic functions, unit conversion, interpolation, calibration and uncertainty analysis.

Why is hour angle 15° per hour?

Earth rotates about 360° relative to the mean Sun in 24 mean solar hours, giving 360 ÷ 24 = 15° per hour.

Why does the gnomon angle equal latitude?

A polar style is parallel to Earth’s rotation axis. The celestial pole’s altitude above the ideal horizon equals the observer’s latitude.

What is the equation of time?

It is the date-dependent difference between apparent solar time and mean solar time, arising mainly from Earth’s orbital eccentricity and axial tilt.

Why is solar noon around 1 pm in Singapore?

Singapore’s longitude is about 16° west of the nominal 120° E meridian for UTC+8, giving roughly 65 minutes of longitude offset, plus a seasonal equation-of-time adjustment.

Can a sundial work on a wall?

Yes, but vertical declining or reclining dial geometry differs from the horizontal formula and must account for the wall’s orientation and tilt.

What happens on a cloudy day?

A direct shadow may be unreadable. A sundial has no independent oscillator to continue displaying time.

Is the Sun directly overhead every day at the equator?

No. Near the equator, overhead Sun conditions occur only when solar declination matches latitude, on particular dates.

How accurate can a school sundial be?

Accuracy depends on geometry, construction, alignment, correction, shadow definition and reading. A well-analysed several-minute uncertainty may be more honest than an unsupported one-minute claim.

Useful next reading

U.S. Naval Observatory: The Equation of Time for an authoritative definition and dated example.
NIST: Solar time and the equation of time for the connection to timekeeping.
NIST: What Determines the Length of the Day? for solar, stellar and atomic perspectives.
Why Mathematics? | Eclipses, Orbital Geometry and Shadow Paths for another application of shadows and celestial geometry.
Why Mathematics? | Astronomy, Parallax and Measuring Cosmic Distances for angles used as a measurement tool.
Why Mathematics? | School Commutes, Maps and Route Planning for coordinate and location reasoning closer to everyday life.

A final thought

A sundial is a small mathematical model of a rotating planet. Its line and shadow connect a child standing on one patch of ground to latitude, longitude, orbital motion and the global history of timekeeping.

Its greatest lesson is not that clocks should be replaced. It is that measurements have definitions. Once apparent solar time, mean solar time and civil time are separated, the “wrong” shadow becomes a precise invitation to think.

Why Mathematics? | Sundials, Solar Angles and Timekeeping