The two instruments measure different things
The "GNSS or total station" argument comes up on every second site, and it is usually held by people who own only one of them. The difference is simple. A total station measures angles and distances to a target it can see through the telescope, and it works relative to points you have given it. A GNSS receiver measures its position relative to satellites and needs open sky, not a line of sight to a ground point. Nearly every practical difference follows from that.
A total station does not care whether you are in a pit, under a bridge or on a ground floor without windows. It cares whether there is a straight line between itself and the prism. A GNSS receiver does not care about a line to anything on the ground, but if there is a walnut tree crown above it, work stops.
Where GNSS wins outright
Open terrain, long distances, many points. Surveying a parcel for a project, staking a road axis across a field, checking an embankment, walking the crest of a dam. One person with a receiver on a pole covers in a day what would take two people and several setups with a total station. No orientation, no moving the instrument, no hunting for line of sight.
IMU tilt makes it faster still, because you stop centring on every point. For the FOIF A90 and the A66 MAX the manufacturer states tilt survey up to 60°, which in practice means a kerb edge gets measured without kneeling. If corrections come from a network, you do not even carry a base. This is the cheapest centimetre in surveying, provided the sky is clear.
Where the total station is the only option
A construction pit ten metres below street level, between sheet pile walls. Setting out columns in a hall where the roof is already on. A facade where you have to mark the levels of the openings. A tunnel, an underground car park, a room. In all of these places GNSS gives you either nothing or something you should not trust.
The total station is at home there. You orient it from two or three known points and from then on every point is stable to the millimetre relative to them. Reflectorless measurement adds something GNSS cannot offer: you measure the edge of a slab, the corner of a building or the bottom of an excavation without sending a person there. The RTS362N lists reflectorless range up to 1000 metres, and the RTS332 comes in 600 and 1000 metre versions. On a real site you rarely shoot that far without a prism, but it does mean that closer targets with poor reflecting surfaces are picked up reliably.
There are in-between cases too. A street in an old neighbourhood with trees on both sides: the receiver will fix at the junctions and lose the solution between them. There, colleagues usually measure the junctions with GNSS and then use them as control for the station along the street. It works well, as long as the control points are measured twice, at different times.
Height accuracy
Many arguments get settled here. With GNSS the vertical accuracy is worse than the horizontal, and you can see that in the specifications of every receiver, ours included. For a topographic survey, cadastre or axis stakeout it does not matter. For a foundation level, a sewer gradient of three per mille or a rail installation it does. There a level, or a total station with good height orientation, is mandatory, and the GNSS result is only a check against gross error.
The reverse is also true. A total station oriented from bad points will give you beautiful millimetres relative to something that is itself displaced. The station's accuracy is relative to the control. That is why on a large site the control is often measured with GNSS in static mode, and then everything inside is done with the station.
One site, two instruments
This is what it looks like on a road job. In the morning you walk the alignment with the receiver and stake the axis points and the expropriation boundaries in the open section. In the afternoon you reach the culvert under the embankment. There you set up the station, orient it from two points you measured with GNSS in the morning, and measure the invert, the wing walls and the levels of the exposed pipes. Same person, same processing software, no waiting for another crew. Record which points were measured with which instrument. Six months later, when someone asks why two levels differ by three centimetres, that is the first thing you will want to know.
On a residential build it is the other way round. GNSS is useful at the start, for the excavation and the layout around the plot, and once the first slab is out almost everything is done with the station. Axis stakeout, formwork checks, the level of every slab. The receiver stays in the case for checking the external connections and for the as-built survey at the end.
If you have the budget for only one instrument, choose by where you spend more days. Open terrain and several sites a day: a receiver. A construction site where you will be for months, with pits and buildings: a total station. And when the time comes for the second one, have a look at the total stations category or give us a call so we can fit it to what you already own.
