We touched on it in an earlier post, but the number of control points you calibrate to matters. Surrounding the site with control is important - but if three site control points encircle the project, is that enough? The answer is NO. You would never have guessed, but our site control discussion will take us to a bar stool for Principle 3.
3. Calibrate to a minimum of five control points.
Have you ever noticed that a three legged stool never rocks, but a four legged stool does? If one leg of a three legged stool is a little short the top might be imperceptibly off level but the stool is still stable, it's not obvious your favorite bar stool has a problem. If you have a short leg on a four legged stool it will rock every time.
Your GPS system's error checking works on much the same concept. Using only three control points might place your site in the world horizontally but vertically there is no way to check for errors. A fourth control point allows the system to compare the points against each other and find errors. A fifth point is important for redundancy and even better error checking. More than five points is best, but with five you can be confident that your site control has the project in the right place vertically and horizontally.
Let's walk through each control point as you shoot them in to understand how each point helps to locate your project in the dirt. Keep in mind our analogy to using pins to place your site on a globe (add link).
When the first point is placed, it gets your site roughly where it needs to be but the system cannot rotate it correctly or place it vertically. Imagine placing that first "pin" in your project (the paper) on the globe. You could rotate the project in any direction.
In the picture below you will see that site control point one (yellow triangle) has been measured in. The site is held at that point but it is not rotated correctly. The system cannot determine what is "north" on the virtual plans vs. the real world.
The second point sets the rotation of the site - now it knows which way is north - but has no way to check if the site is rotated correctly. You will notice that once you have two control points shot in you can stake out to the others close enough to find them.
Below you can see that two points have been shot in and the site is now correctly rotated. Notice that the entrance driveways on the right side of the plans now line up with the existing street.
Point three gives the system the ability to error check site rotation and adjust it vertically. It still can't check control vertically (remember a 3 legged bar stool never rocks). After you have three good points measured in you will see that horizontally staking out to new points is really close, but vertically it will still be off.
Shooting in the 4th control point will allow the system to error check the control points vertically. You will start seeing vertical errors in your control point readout. You now have the project fully "pinned" to the globe.
In this picture, if you drew a line between each control point you will see that they box in the working area on this project, as described in our Principle 2 blog post.
Measuring in the fifth point gives your GPS equipment full redundancy to check all of the points against each other. If you think you have a bad point you can even turn them off one at a time to determine which one it is.
Don't stop measuring in site control after you touch on the 5th hub. Each point you continue to shoot in will help improve the machine control system's accuracy and error checking. You may need more points to box in your site, too. It is best to shoot them all in at the beginning of a project even if you are not working in the outlying areas yet - those shiny new points might be disturbed by a lawnmower, farmer, homeowner (I have never met one that likes lathe in his yard) or even taken out by your own crew.