Theodolite, total station and GPS

In short

A theodolite measures horizontal and vertical angles precisely; a total station adds an electronic distance meter (EDM) and a processor so that angles, distances and three-dimensional coordinates come out of one instrument; GNSS (GPS, NavIC and others) fixes position from satellite ranges without any line of sight between stations. Together they form the backbone of modern control surveys, traversing, setting out of buildings, roads, canals, pipelines, metro alignments and cadastral work. In Indian universities this topic carries the angle-measurement theory, instrument adjustments, tacheometry and GNSS fundamentals; in GATE it appears as numericals on angle reduction, trigonometric levelling, stadia formulae, EDM corrections and DOP. Employers who test it include L&T, AFCONS, Tata Projects, NHAI/NHIDCL consultants, RITES, Survey of India, state PWDs and survey firms running CORS-based RTK for SVAMITVA and metro projects.

Explained

The theodolite: parts, axes and adjustments

• Definition: an instrument that measures horizontal angles and vertical angles, and is also used to prolong a straight line, measure magnetic bearings, and do rough levelling and tacheometry.
• Transit vs non-transit: a transit theodolite's telescope can be revolved through 180° in the vertical plane (plunged/transited) about the horizontal axis; a non-transit one cannot. Almost all modern instruments are transit type.
• Reading systems: vernier theodolite (open graduated circle + verniers A and B, least count 20″ or 10″), micro-optic/optical-scale theodolite (glass circles, micrometer, 1″–10″), electronic digital theodolite (encoder circles, LCD display, 1″–7″).
• Main parts: levelling head with tribrach and three foot screws; lower plate (graduated circle) with lower clamp and tangent screw; upper plate (carrying verniers) with upper clamp and tangent screw; plate bubbles; standards (A-frame); telescope on the trunnion (horizontal) axis; vertical circle with index frame (T-frame, verniers C and D) and altitude bubble; shifting head; optical or laser plummet.
• Four fundamental axes: vertical axis, horizontal (trunnion) axis, line of collimation, axis of the plate level.
• Required relations (basis of permanent adjustments): plate-level axis ⟂ vertical axis; line of collimation ⟂ trunnion axis; trunnion axis ⟂ vertical axis; altitude-bubble axis ∥ line of collimation so that the vertical circle reads zero on a horizontal sight.
• Temporary adjustments (every setup): (1) setting up — tripod roughly level, instrument over the station; (2) centring — plumb bob/optical plummet/laser over the station mark, using the shifting head; (3) levelling — bubble parallel to two foot screws, centre it, turn 90°, use the third screw, repeat; (4) focusing — eyepiece first to remove parallax (sharp cross-hairs), then objective to focus the image.
• Permanent adjustments (workshop/field tests): plate-level adjustment; collimation adjustment by double sighting (reversal); trunnion-axis adjustment by the spire test on a high well-defined point; altitude bubble and vertical-circle index adjustment.
• Least count = (smallest main-scale division) ÷ (number of vernier divisions). Example: 20′ divisions, 60-division vernier → 20″.

Measuring angles with a theodolite, and its errors

• Ordinary (single) method: set the vernier to 0° and sight the back station using the lower clamp; then release the upper clamp, bisect the forward station, read; repeat on the other face and mean.
• Repetition method: the same angle is measured several times, each time cumulatively added to the circle without resetting to zero; the final accumulated reading ÷ number of repetitions gives the angle. It reduces graduation errors, reading errors and eccentricity, and is used for precise single angles.
• Reiteration (direction) method: with several stations to observe, all are sighted in order and the horizon is closed back on the first; the misclosure is distributed. Faster where many directions radiate from one station.
• Changing face (face left + face right) eliminates: collimation error, trunnion-axis error and vertical-circle index error. It does NOT eliminate a non-vertical vertical axis (imperfect levelling) — that must be corrected by levelling properly.
• Reading both verniers A and B (180° apart) and meaning eliminates eccentricity of verniers/centres.
• Instrumental errors: imperfect graduations, eccentricity, imperfect adjustment of plate levels, collimation and trunnion-axis errors, vertical index error, eccentric telescope.
• Personal errors: inaccurate centring and bisection, imperfect levelling, parallax, wrong vernier read, slip of clamps, wrong booking.
• Natural errors: wind vibration, heat shimmer and lateral refraction, unequal expansion, settlement of the tripod.
• Theodolite traversing: included-angle method (check: Σ interior = (2n − 4) × 90°, Σ exterior = (2n + 4) × 90°), deflection-angle method (roads, railways, canals), or direct bearings. Latitudes and departures are computed and the closing error is adjusted by Bowditch's rule (proportional to length) or the transit rule.

Tacheometry and trigonometric levelling with the theodolite

• Tacheometry obtains horizontal distance and difference in elevation by instrument readings alone, without chaining — valuable in broken, steep or waterlogged ground.
• Stadia system: fixed stadia hairs, staff intercept s; multiplying constant K = f/i (commonly 100) and additive constant C = f + d (≈ 0 in modern anallatic/internal-focusing telescopes).
• Tangential system uses two vertical angles to two targets (no stadia hairs); subtense bar method uses a horizontal bar of known length and the measured horizontal angle.
• Field constants are determined by observing known distances on level ground and solving D = Ks + C for two sets.
• Trigonometric levelling: heights from a measured vertical angle and distance; for sights beyond about 300 m the combined curvature and refraction correction (always additive to the computed height difference) must be applied.
• Reciprocal observations (angles measured both ways between two stations) cancel curvature, refraction and index error almost completely — the standard method across rivers and valleys.

Total station: EDM principle, functions and corrections

• What it is: electronic theodolite + EDM + onboard microprocessor + memory/data card, giving slope distance, horizontal and vertical angles, and derived horizontal distance, height difference and coordinates on one display.
• EDM principle 1 — phase comparison (most survey EDMs): a modulated infrared/laser beam travels to the reflector and back; the instrument measures the phase difference of the returned wave. Distance 2D = nλ + Δλ, where n is an unknown whole number of wavelengths resolved by using several modulation frequencies (coarse and fine). The wave speed used is c = c₀/n_air, so air refractive index (hence temperature, pressure, humidity) directly scales the result.
• EDM principle 2 — pulse/time of flight: a short laser pulse is timed, D = c·t/2. Used for long reflectorless ranges and in scanners.
• Classification of EDM: by carrier — microwave (tellurometer, works in fog, needs an operator at both ends), infrared (short range, prism), laser; by range — short (< 3 km), medium (3–25 km), long (> 25 km).
• Reflectors: corner-cube prism with a prism constant (e.g. −30 mm, −17 mm, 0 mm) that must be entered correctly; reflective tape; reflectorless measurement to walls/rock, typically tens to a few hundred metres.
• Typical specifications: angle 1″–7″ standard deviation; distance ± (2 mm + 2 ppm × D) to a prism, poorer reflectorless. Treat these as model-dependent.
• Onboard programs: traverse and coordinate computation; free station / resection (occupying an unmarked point and fixing it from two or more known points); stake-out (setting out) by coordinates or angle-distance; REM (remote elevation of inaccessible points); MLM (missing line measurement between two observed points); offsets; area; road alignment and COGO.
• Robotic/motorised total station with ATR (automatic target recognition) locks onto and follows the prism, allowing one-person survey and machine guidance.
• Corrections applied: instrument zero/index error and scale error (calibration on a baseline), prism constant, atmospheric ppm from temperature and pressure, slope-to-horizontal reduction, earth curvature and refraction, and the grid scale factor when reducing to a map projection.
• Advantages: speed, direct 3-D coordinates, no booking/arithmetic blunders, digital transfer to CAD/GIS, works in low light, one instrument for angle + distance + levelling.
• Limitations: needs clear line of sight and a steady tripod, battery dependent, costly, calibration drift goes unnoticed, and because data is captured electronically a wrong instrument height or prism constant corrupts everything silently — field sketches and independent checks are still required.

GNSS: how satellite positioning works

• Segments: space segment (satellites), control segment (master and monitor stations that track orbits and upload corrections), user segment (receivers and antennas).
• GPS constellation: nominally 24+ satellites in 6 orbital planes, about 20,200 km altitude, 55° inclination, orbital period ≈ 11 h 58 min, so at least four satellites are visible anywhere; carriers L1 1575.42 MHz, L2 1227.60 MHz, L5 1176.45 MHz with C/A (civil) and P(Y) (military) codes.
• Pseudorange positioning: the receiver correlates the satellite code with its own replica to get travel time. Because the receiver clock is cheap, the measured range contains a clock bias — hence four unknowns (X, Y, Z and clock error) and a minimum of four satellites for a 3-D fix (three suffice if height is fixed).
• Carrier-phase measurement is about 100 times more precise than code ranging but carries an unknown whole number of cycles — the integer ambiguity — resolved by differencing between receivers and satellites; a lost lock gives a cycle slip.
• Error sources: satellite clock and ephemeris errors; ionospheric delay (largest; removed by dual-frequency combination or modelled by Klobuchar/SBAS); tropospheric delay (modelled, wet component troublesome); multipath from walls, water and vehicles (choke-ring antenna, good site, elevation mask of about 15°); receiver noise; antenna phase-centre offset. Selective Availability was switched off in May 2000.
• Geometry — DOP: position accuracy ≈ DOP × range error (UERE). GDOP, PDOP, HDOP, VDOP, TDOP. PDOP under about 4 is good; satellites bunched together give a weak, elongated solution.
• Datum and height: GNSS gives coordinates on the WGS-84 ellipsoid (a = 6,378,137 m, 1/f = 298.257223563) and ellipsoidal height h, not the levelled height. Orthometric (MSL) height H = h − N, where N is the geoid undulation from a geoid model; over India N is large and negative (roughly −30 m to −100 m), so this reduction can never be skipped. India's older framework used the Everest spheroid/Indian datum; Survey of India has been migrating to a WGS-84/ITRF-based national datum — check current SoI documentation for the datum name and epoch in force.

GNSS survey methods, Indian systems and the policy framework

• Absolute (single-point) positioning: one receiver, 5–10 m — navigation, reconnaissance, small-scale mapping.
• DGPS: corrections from a known base, metre to sub-metre — GIS asset mapping, hydrography, forestry.
• Static: two or more receivers logging for 30 minutes to several hours; post-processed baselines at a few mm + 0.1–1 ppm — geodetic and primary control.
• Rapid static, stop-and-go, kinematic and PPK fill the range between static and real-time work.
• RTK: base transmits corrections by radio/GSM to a rover; centimetre fixes in seconds, typically ± (8–10 mm + 1 ppm) horizontally — setting out, topographic survey, machine control.
• Network RTK / CORS / VRS: a network of permanent reference stations streams corrections over the internet (NTRIP), removing the need for a local base. Survey of India has been building a national CORS network (a programme of the order of a thousand stations, used in the SVAMITVA rural property-mapping scheme); coverage figures change, so verify current status.
• PPP (precise point positioning) uses precise orbit and clock products with a single receiver for decimetre to centimetre results after convergence.
• Indian systems: NavIC (IRNSS) — a regional constellation of 7 satellites (3 geostationary + 4 geosynchronous) broadcasting in L5 and S band, covering India and about 1,500 km around it, with a Standard Positioning Service around 20 m and a restricted service; the newer NVS series (NVS-01, 2023 onwards) adds L1 for mass-market receivers. GAGAN is the ISRO–AAI satellite-based augmentation system (SBAS) certified for aviation approaches. AIS-140 vehicle tracking devices for commercial vehicles are required to be NavIC-capable.
• Other constellations: GLONASS (Russia), Galileo (EU), BeiDou (China), QZSS (Japan) — multi-constellation receivers are now standard and greatly improve DOP in urban canyons.
• Indian standards, codes and policy: Survey of India (under DST) is the national survey agency and publishes topographic specifications and geodetic manuals; BIS has instrument specifications (for example IS 2976 for vernier theodolites — confirm the current edition before citing). The National Map Policy 2005, the Guidelines for acquiring and producing geospatial data, February 2021 (which deregulated geospatial data for Indian entities, with threshold accuracy conditions around 1 m horizontal and 3 m vertical for finer data and ground-truthing restricted to Indian entities) and the National Geospatial Policy 2022 (high-accuracy national topographic and urban mapping targets by 2030) govern practice. Land-record survey work follows state revenue/land-records rules; project control is usually specified in NHAI/metro/railway tender documents.

Key points

• Theodolite - measures horizontal and vertical angles; transit type can plunge the telescope through 180°.
• Least count - main-scale division ÷ number of vernier divisions; 20′/60 = 20″.
• Four axes - vertical, trunnion, line of collimation, plate-level axis; their perpendicularity relations define the permanent adjustments.
• Temporary adjustments - setting up, centring, levelling, focusing (eyepiece for parallax, objective for image).
• Face left + face right mean - cancels collimation, trunnion-axis and vertical index errors, but not a tilted vertical axis.
• Repetition method - accumulates the angle on the circle to beat graduation and reading errors; reiteration closes the horizon for many directions.
• Verniers A and B - meaning the two eliminates eccentricity of centres.
• Tacheometry - D = Ks cos²θ + C cos θ; K = f/i ≈ 100, C = f + d ≈ 0 for anallatic telescopes.
• Curvature and refraction - combined correction 0.0673 D² m (D in km), always added to the computed height difference; cancelled by reciprocal observations.
• EDM - phase-comparison instruments measure phase difference of a modulated wave; pulse instruments measure time of flight.
• ppm correction - roughly 1 ppm per 1 °C and per 3.5 hPa of departure from the instrument's standard atmosphere (use the maker's formula for precise work).
• Prism constant - a wrong entry shifts every distance by a fixed amount (e.g. 30 mm) with no other symptom.
• Total station accuracy - typically ±(2 mm + 2 ppm) for distance, 1″–7″ for angles.
• Free station/resection - fixes the instrument itself from known points, so no setup over a marked station is needed.
• Four satellites minimum - because the receiver clock bias is a fourth unknown alongside X, Y, Z.
• DOP - position error ≈ DOP × range error; good geometry means satellites well spread in the sky.
• GNSS heights - ellipsoidal; H = h − N using a geoid model, with N strongly negative over India.
• NavIC - 7-satellite Indian regional system in L5/S band, ~20 m SPS; GAGAN is the Indian SBAS; RTK/CORS gives centimetre work in real time.

Formulas and examples

Example 1 — angle by repetition (vernier theodolite). Main-scale division 20′, vernier of 60 divisions. Three repetitions of angle AOB give a final accumulated reading of 182°15′45″, starting from 0°00′00″.

• LC = 20′/60 = 20/60 min = 20″.
• 182°15′45″ = 655,200 + 900 + 45 = 656,145″; ÷ 3 = 218,715″.
• 218,715″ = 60.754167° = 60° + 0.754167 × 60′ = 60°45.25′ = 60°45′15″.
• Note the result is quoted to 5″ although the LC is 20″ — repetition legitimately improves the mean, but only if centring and levelling are good.

Example 2 — total station trigonometric levelling with curvature and refraction. Slope distance S = 1,250.000 m, zenith angle Z = 86°30′00″, instrument height 1.520 m, target height 1.700 m, RL of station A = 245.600 m.

• Horizontal distance H = 1,250 × sin 86°30′ = 1,250 × 0.998135 = 1,247.669 m.
• Vertical component V = 1,250 × cos 86°30′ = 1,250 × 0.061049 = 76.311 m.
• Curvature + refraction: D = 1.247669 km, D² = 1.556678, C = 0.0673 × 1.556678 = 0.105 m.
• Height difference = V + hi − ht + C = 76.311 + 1.520 − 1.700 + 0.105 = 76.236 m.
• RL of B = 245.600 + 76.236 = 321.836 m. (Ignoring the 0.105 m correction would have given 321.731 m — an error of over 10 cm at 1.25 km.)

Example 3 — EDM atmospheric and specification errors. Same 1,250.000 m line, instrument standardised at 15 °C and 1,013 hPa, measured at 38 °C and 1,002 hPa. Instrument specification ± (2 mm + 2 ppm).

• Using the rules of thumb (≈1 ppm per °C, ≈1 ppm per 3.5 hPa): temperature +23 °C → +23 ppm; pressure −11 hPa → +3 ppm; total ≈ +26 ppm.
• ΔD = 1,250 × 26 × 10⁻⁶ = +0.0325 m (32.5 mm) — i.e. 1 part in 38,462, unacceptable for primary control, negligible for topographic detail.
• Random error from the specification: 2 mm + 2 × 1.25 = ±4.5 mm. So the systematic ppm error is seven times the instrument's own noise — always enter temperature and pressure.
• For an RTK rover quoted at ±(8 mm + 1 ppm) on a 12 km baseline: 8 + 12 = ±20 mm horizontally.

Example 4 — stadia tacheometry. Anallatic telescope, K = 100, C = 0. Staff readings 1.150, 1.655, 2.160 m; vertical angle +6°24′; RL of instrument axis = 100.000 m.

• Intercept s = 2.160 − 1.150 = 1.010 m (check: mid reading 1.655 = mean of 1.150 and 2.160 ✓).
• D = 100 × 1.010 × cos²6.4° = 101 × 0.987575 = 99.745 m.
• V = 100 × 1.010 × sin 12.8°/2 = 101 × 0.221549 / 2 = 11.188 m.
• RL of staff station = 100.000 + 11.188 − 1.655 = 109.533 m.

Example 5 — GNSS accuracy and height reduction. A handheld receiver reports HDOP 1.2, VDOP 1.9, with a range error (UERE) of 3.5 m; ellipsoidal height h = 215.480 m and the geoid model gives N = −52.400 m.

• Horizontal error σ_H ≈ 1.2 × 3.5 = ±4.2 m (1σ); vertical σ_V ≈ 1.9 × 3.5 = ±6.7 m — vertical is always the weaker component because all satellites are above the horizon.
• Orthometric height H = h − N = 215.480 − (−52.400) = 267.880 m above MSL. Reporting 215.480 m as the "level" would be wrong by 52 m.

Where it is used

• Infrastructure construction: L&T, AFCONS, Tata Projects, Dilip Buildcon set out piers, piles, metro tunnels and formation levels with total stations, and carry alignment control by RTK against CORS corrections.
• Highways and railways: NHAI/NHIDCL and RVNL consultants run GNSS control traverses and total-station topographic and cross-section surveys for DPRs, and check as-built geometry.
• Land records and cadastre: state revenue departments and Survey of India use GNSS/CORS and ETS for cadastral resurvey, boundary demarcation and the SVAMITVA property-card programme.
• Mining, quarries and dams: monthly volume surveys, slope and dam deformation monitoring with robotic total stations and permanent GNSS monitors; NHPC, Coal India and NMDC sites.
• Utilities, telecom and power: transmission-line tower spotting and sag profiles, pipeline and canal alignment, DGPS-based underground utility mapping for smart-city GIS.
• Job roles: survey engineer, chainman-to-surveyor career path, ETS/GNSS operator, geodetic engineer, QS/setting-out engineer, GIS-survey coordinator; Survey of India and state PWD recruitment (and GATE Civil) test exactly this material.

Interview questions

• Q: What does changing face eliminate, and what does it not? A: It cancels collimation error, trunnion-axis error and vertical index error; it does not cancel an out-of-plumb vertical axis, which only correct levelling can fix.
• Q: Why use the repetition method instead of measuring once? A: Accumulating the angle on the circle averages out graduation, eccentricity and reading errors, so the mean is far more precise than the instrument's least count.
• Q: How many satellites are needed for a 3-D GPS fix and why? A: Four — three for X, Y, Z and one more to solve the receiver clock bias, which is an unknown because receiver clocks are not atomic.
• Q: A total station reads every distance 30 mm short. What is the likely cause? A: A wrong prism constant (or an uncalibrated instrument index error) — a constant offset independent of distance.
• Q: Why must temperature and pressure be entered in a total station? A: The EDM computes distance from the speed of light in air; air refractive index changes with temperature and pressure, giving a scale (ppm) error that grows with distance.
• Q: Why is GNSS height not the same as level height? A: GNSS gives height above the WGS-84 ellipsoid; levelling refers to the geoid (MSL). H = h − N, and N is of the order of tens of metres and negative over India.
• Q: What is HDOP and what value is acceptable? A: A measure of how favourably satellites are distributed; error ≈ DOP × range error. Values near 1 are excellent and work under about 4 is normally acceptable.
• Q: When would you prefer a total station over RTK GNSS? A: Under tree canopy, indoors, in tunnels, in urban canyons, and whenever relative accuracy of a few millimetres between nearby points is needed.
• Q: What is a free station (resection) and why is it useful? A: The instrument is set up anywhere convenient and its own coordinates are computed by observing two or more known points — useful on congested sites where control marks are inaccessible.
• Q: What is NavIC and how does it differ from GPS? A: NavIC (IRNSS) is India's regional system of seven satellites in geostationary/geosynchronous orbits broadcasting in L5 and S band, covering India and about 1,500 km around, with roughly 20 m standard accuracy — regional, not global like GPS.
• Q: Why are reciprocal vertical angles observed? A: Observing both ways between two stations cancels curvature, refraction and index error, giving a far better height difference over long sights.
• Q: What is the integer ambiguity in carrier-phase GNSS? A: The unknown whole number of carrier cycles in the satellite-to-receiver range; resolving it ("fixed" solution) is what turns decimetre results into centimetre results, and a cycle slip destroys it.

Common mistakes

• Using N instead of N/2 style slips in angle work: forgetting to divide the accumulated repetition reading by the number of repetitions, or meaning face-left and face-right readings that differ by 180° without reducing them first.
• Writing the stadia formula as D = Ks cos θ instead of D = Ks cos²θ, or using sin θ in place of (sin 2θ)/2 for V.
• Omitting the curvature-and-refraction correction on long trigonometric-levelling sights, or applying it with the wrong sign (it is always additive to the computed rise).
• Reporting GNSS ellipsoidal heights as reduced levels, or mixing WGS-84 coordinates with old Everest-datum maps without a datum transformation.
• Leaving the prism constant, instrument height or target height from the previous job in the total station — every subsequent coordinate is wrong and nothing looks wrong on screen.
• Confusing precision with accuracy: a 1″ instrument badly centred over the station, or an RTK rover with a "float" solution, gives tight repeatability on a wrong position.
• Trusting electronic data with no field sketch, no closing check on a known point and no closure of the traverse — total-station and RTK blunders are invisible without an independent check.

Related topics

Surveying: chain, compass and levelling · Curves and setting out · Photogrammetry and drone surveys · GIS and remote sensing · Theory of errors and survey adjustment · Geodesy, map projections and datums.