Total Station Setup and Stakeout

A total station will happily lay out a whole building in the wrong place, and every point will agree with every other point. The instrument does not know you picked the wrong control point, typed the wrong prism setting or set up on a bumped nail. That is why the checks matter more than the buttons.

You have seen what a total station does. This guide is about using one with less supervision: setting up, proving the setup, staking points and catching mistakes before a crew builds on them. Menus differ by brand, so learn your instrument from its manual and with hands-on time. Only a licensed land surveyor can establish or certify property lines. You work from control set by the surveyor. Specific check values come from the Caltrans Surveys Manual as a working example, and your job may set others.

What You Will Be Able to Do

• Set up and orient on control, or by resection.
• Prove a setup with independent checks before staking anything.
• Enter prism and weather settings and explain what happens if they are wrong.
• Work out distance and direction from coordinates by hand.
• Tell grid from ground, and ask the right question before you start.

Words You Will Use

Setting Up on a Point

1. Check the control. Is the point undisturbed? Has anything hit it? Do you have the current coordinate list?
2. Set the tripod over the point with a level head, legs pushed in firmly.
3. Center and level. Use the optical or laser plummet and the leveling screws, and repeat until both stay correct.
4. Measure the instrument height. Enter it. Enter the target height.
5. Choose your job file. Check the coordinate system and units, including US survey feet or international feet.
6. Enter weather and prism settings.
7. Orient. Aim at the backsight, set the direction, and measure it.

Prove It Before You Stake

Orientation alone proves nothing. You have only told the instrument which way to look.

1. Measure the distance to the backsight and compare it with the distance from the coordinates.
2. Shoot a third known point. Compare its measured position with the control list. This is the check that catches a wrong backsight or a bumped control point.
3. Check elevation the same way if you will stake grades.
4. At the end of the setup, reobserve the backsight. Also check your plummet and heights. If the instrument moved during the day, everything since your last check is in doubt.
5. Overlap setups. Shoot a few points from two different setups where you can.

If a check does not agree, stop. Do not start staking and hope.

Why a Short Backsight Hurts

Direction errors grow with distance.

Worked example. Your backsight is only 40 feet away, and the point is 0.02 foot off center. That is a direction error of 0.02 / 40 = 0.0005. Now stake a point 300 feet away: 300 x 0.0005 = 0.15 foot off, nearly 2 inches. So pick a backsight at least as long as your longest stakeout shot where you can.

A useful number: 1 minute of angle is about 0.03 foot at 100 feet. An angle error you cannot see at the instrument can be plain to see on the far side of a building.

Settings That Quietly Ruin Distances

Prism constant. Prisms are not all the same. Each has a constant that the instrument adds to or removes from each distance, usually given in millimeters. If you change prisms and not the setting, every distance is wrong by the same amount. A 30 mm mistake is about 0.1 foot on every shot. A backsight distance check catches this, if you actually do it.

Weather (ppm). Light travels at a slightly different speed depending on air temperature and pressure. The correction is entered in parts per million (ppm), and many instruments compute it from the values you type in. One ppm is 0.001 foot in 1,000 feet. Caltrans notes that about 1.8°F of change makes 1 ppm of error, so enter it at the start of each day and again if the day warms up.

Reflectorless mode. A beam with no prism can bounce off the wrong thing, such as a leaf, a piece of rebar or the near edge of a wall. Use it for shots you can check, not for a building corner the crew will build from. Check which mode you are in.

Working From Coordinates

You should be able to check the instrument with a calculator. If you can do this math, you can spot a bad point on the spot.

Worked example. You are on point A, at N 5,000.00 and E 5,000.00. You want point B, at N 5,120.00 and E 5,090.00.

1. Change in north: 5,120.00 - 5,000.00 = +120.00.
2. Change in east: 5,090.00 - 5,000.00 = +90.00.
3. Distance: the square root of (120 x 120 + 90 x 90) = the square root of 22,500 = 150.00 feet.
4. Direction: the angle whose tangent is 90 / 120 = 0.75, which is about 36.87 degrees. That is 36 degrees, 52 minutes, 12 seconds. Both changes are positive, so the azimuth is in the northeast quarter.

If the data collector says 150.00 and 36-52-12, your point is right. If it says 15.00 or 323 degrees, something was typed in wrong.

Angles use degrees, minutes and seconds: 60 minutes to a degree, and 60 seconds to a minute. To convert 36.87 degrees: 0.87 x 60 = 52.2 minutes, and 0.2 x 60 = 12 seconds.

Grid Versus Ground

Large projects are often laid out on a State Plane grid, a map projection of the curved earth. Distances on that grid can differ a little from real distances on the ground. The difference is captured in a combined scale factor.

Worked example. The combined factor on a job is 0.99985. A grid distance of 1,000.00 feet is 1,000.00 / 0.99985 = 1,000.15 feet on the ground. If a designer drew the building in ground distances and you staked from grid coordinates, a long building could be off by 0.15 foot per 1,000 feet.

At the start of any job, ask what system the coordinates are in and whether they are grid or ground. It takes one question. The surveyor of record or the field engineer can tell you.

Staking Points

1. Pull the right data. Confirm the file name, the date and the plan revision.
2. Plan the order. Work from the instrument outward, so you can keep your best geometry.
3. Keep the pole plumb. Hold the bubble in the circle. A 6-foot pole leaned just 1 degree is about 0.1 foot off at the top.
4. Stake and measure. Set the point, then shoot it again to confirm what you set.
5. Mark it the way the crew expects. Include the point number, the offset and what it represents.
6. Record it. Save the staked position, not only the design position.

Staking offsets. Excavation will take out a corner stake, so offset points are common. An offset of 5 feet along a known line is easy to measure back from. Write the offset on the stake, and tell the people who will use it.

Check as you go. Building lines give you free checks. After you set the corners, tape the sides and compare the diagonals. Those should match the plan to within your tolerance. A tape check is an independent measure, and it finds errors the instrument cannot.

Robotic and GPS rovers. A robotic total station follows the prism, so one person can work alone. The same checks apply. GPS rovers need a site calibration and a known check point at the start of each session. Under trees or next to buildings, their elevations are the first thing to go, so use a total station or level there.

Resection

Sometimes the best place for the instrument has no control point under it. A resection works out where the instrument is by shooting two or more known points.

• Use at least three points when you can. The extra point gives you a check.
• Spread the points around you. Points bunched in one direction give poor geometry.
• Read the residuals, which are how badly the points disagree. If one is far off, one of your points has moved, or its coordinates are wrong.
• Then treat it like any other setup, and check it on another known point.

Warning: Stakeout puts you beside moving machines and traffic with your eyes on a screen. Wear high-visibility gear, agree on a plan with the operators, and look up every time you move. Keep long poles and rods away from overhead lines.

Common Errors

Check Your Understanding

1. You orient to the backsight and everything looks good, but the third control point is 0.28 foot off. What do you do?

Stop. One of those points has moved, or the setup is wrong. Check the coordinates and your occupied point, and shoot another known point.

2. From N 1,000.00, E 2,000.00 you are staking N 1,060.00, E 2,080.00. What distance should the instrument show?

Change in north = 60, change in east = 80. The square root of (3,600 + 6,400) is 100.00 feet.

3. You changed prism poles after lunch, and all your distances now run 0.1 foot long. What is the likely cause?

The prism constant setting no longer matches the prism. Check it, then shoot a known distance.

4. The combined factor is 0.99990. A grid distance is 2,000.00 feet. How long is it on the ground?

2,000.00 / 0.99990 = 2,000.20 feet.

5. Your backsight is 25 feet away and your stakeout is 400 feet the other way. What is the concern?

A tiny centering error on the short backsight turns into a big error at 400 feet. Choose a longer backsight.

Key Takeaways

• A setup is not proven until it has checked in on a third known point.
• Reobserve the backsight at the end, and recheck your heights.
• A wrong prism constant shifts every distance by the same amount.
• Enter ppm at the start of the day, and again if the air changes.
• Learn to check coordinates by hand with your calculator.
• Ask if a job is on grid or ground before you stake anything.