For those who may wish a more detailed explanation of the use of the total station, the following description may be helpful.
A total station (Fig. 5) is a combination electronic transit and electronic distance measuring device (EDM). With this device, as with a transit and tape, one may determine angles and distances from the instrument to points to be surveyed. With the aid of trigonometry, the angles and distances may be used to calculate the actual positions (x, y, and z or northing, easting and elevation) of surveyed points in absolute terms.
A standard transit is basically a telescope with cross-hairs for sighting a target; the telescope is attached to scales for measuring the angle of rotation of the telescope (normally relative to north as 0 degrees) and the angle of inclination of the telescope (relative to the horizontal as 0 degrees). After rotating the telescope to aim at a target, one may read the angle of rotation and the angle of inclination from a scale. The electronic transit provides a digital read-out of those angles instead of a scale; it is both more accurate and less prone to errors arising from interpolating between marks on the scale or from mis-recording. The readout is also continuous; so angles can be checked at any time.
The other part of a total station, the electronic distance measuring device or EDM, measures the distance from the instrument to its target. The EDM sends out an infrared beam which is reflected back to the unit, and the unit uses timing measurements to calculate the distance traveled by the beam. With few exceptions, the EDM requires that the target be highly reflective, and a reflecting prism is normally used as the target. The reflecting prism (Figs. 5 and 6) is a cylindrical device about the diameter of a soft-drink can and about 10 cm. in height; at one end is a glass covering plate and at the other is a truncated cone with a threaded extension. It is normally screwed into a target/bracket on the top of a pole; the pointed tip of the pole is placed on the points to be surveyed.
The total station also includes a simple calculator to figure the locations of points sighted. The calculator can perform the trigonometric functions needed, staring with the angles and distance, to calculate the location of any point sighted.
Many total stations also include data recorders. The raw data (angles and distances) and/or the coordinates of points sighted are recorded, along with some additional information (usually codes to aid in relating the coordinates to the points surveyed). The data thus recorded can be directly downloaded to a computer at a later time. The use of a data recorder further reduces the potential for error and eliminates the need for a person to record the data in the field.
The determination of angles and distance are essentially separate actions. One aims the telescope with great care first; this is the part of the process with real potential for human error. When the telescope has been aimed, the angles are determined. Only then does one initiate the reading of the distance to the target by the EDM. That takes only a few seconds; the calculations are performed immediately.
The total station is mounted on a tripod and leveled before use. Meanwhile, the prism is mounted on a pole of known height; the mounting bracket includes aids for aiming the instrument. The prism is mounted so that its reflection point is aligned with the center of the pole on which it has been mounted. Although the tip of the pole is placed on the point to be surveyed, the instrument must be aimed at the prism. So it will calculate the position of the prism, not the point to be surveyed. Since the prism is directly above the tip, the height of the pole may be subtracted to determine the location of the point. That may be done automatically. (The pole must be held upright, and a bubble level is attached to give the worker holding the pole a check. It is not as easy as one might expect to hold the pole upright, particularly if there is any wind; as a result, multiple readings may be required. Because of that problem, the sighting method chosen at Pompeii was, if possible, not to begin by sighting on the prism itself but on the tip of the pole where it touched the ground. The angle from north would then be fixed and unaffected by the movement of the pole. Then the aim of the telescope could be raised to the level of the prism, adjusting only the angle of inclination.)
In Pompeii a Topcon total station was used,* and we quickly learned a few features of the equipment. (Mr. Eiteljorg had driven to Charlottesville to learn the idiosyncrasies of the instrument in May, but a malfunctioning battery cut the session short, and a few "simple" or "trivial" processes turned out to be neither simple nor trivial without practice.) For instance, leveling the total station is more difficult than we had realized (and spongy soil is devastating, since the instrument is naturally unstable if its support is), and it depends upon accurate adjustment of the bubble level built into the instrument. We also learned that datum points were more difficult to select than expected, since they had to be repeatable; that is, we had to be able to find them again and again with absolute accuracy - this year and next.
When the instrument is set up and turned on, it sets itself to be pointing to zero degrees (north) when power is first supplied. The user must then re-set the instrument to zero degrees when it is actually pointing north; we learned that there is no secondary battery for back-up. When the battery dies, the instrument must be re-set for zero degrees.
Fortunately, these lessons came in the first day or two, and we had no more surprises. (One problem was on-going, however. There are two adjustment knobs for rotating within the horizontal plane. One rotates the telescope to make a sighting, with the readout of angles displaying changes. The other, however, permits the user to rotate the entire instrument and to keep the current angle unchanged during the process. That effectively re-orients the zero or north setting. That can be very helpful when setting up or re-setting the instrument, but, of course, it can be devastating if one makes that adjustment by mistake and thereby changes the north setting. This particular instrument was designed in such a way that it was too easy to re-set the instrument when one only wanted to make a sighting.)
Since we were dealing with standing architecture, the prism pole was often inadequate for our work. The pole is designed to be placed on the survey point in a vertical position; it cannot be placed on a point on the face of a wall. In fact, a prism pole can rarely be placed against the face of a wall because of the bulk of the prism, the pole, and the target to which the prism is attached. We devised two alternate methods for dealing with points on a wall. One involved the use of reflecting tape instead of the prism. Since we were working at such short range, bicycle reflecting tape would reflect the infrared beam well enough to permit the EDM to make a reading. It was a bit slower than using the prism, but it worked. (Bicycle reflectors worked, but their back surfaces were not in the same plane as their reflecting surfaces; so the measurements they generated were from a point too near the face of the reflector by a few millimeters.)
The other method for dealing with points on a wall involved the use of the prism without its pole and target. We could simply position the prism against the point on the wall to be surveyed and take the shot. However, the prism is designed to work on the pole - to give a reading to the center of the pole rather than the back of the prism. In this case, that meant that the prism gave a reading some 13 mm. behind the backmost point of the prism housing. We fashioned a shim with a 13 mm. thickness to attach to the back of one of the prisms (fortunately, we had two prisms). Then the prism could be placed against the point in question and a reading made. The only problem - and the reason reflecting tape was sometimes preferred - was that the prism could not always be placed in a corner and sometimes could not be placed correctly while continuing to face the transit and EDM for reflecting the infrared beam.
When using reflecting tape or a prism without a pole, the tape or prism hides the point to be surveyed. So we aimed the telescope at the point to be surveyed before interposing either tape or prism and maintained the aim of the instrument while putting the tape or prism in position. That reduced the possible error for angular measurement. In the case of the prism, after it was put into position, the transit operator would direct the person holding the prism so that it was aimed directly back at the instrument. (That would have required a walkie-talkie had we been working in a larger area.)
The survey information was recorded by hand, and the data were then entered into the AutoCAD model. We were able to use the data directly, no matter where the machine had been set up for a given session, thanks to an AutoCAD feature called the user coordinate system. Using that AutoCAD feature, each set of data could be entered accurately, regardless of the transit set-up point. (It is unnecessary to describe that process here, but a complete description is available from CSA.)
This process is not necessary if a data collector with the most modern of capabilities is available. (See The Ustica Excavations - A Total Station, AutoCAD at Work.) The data collector can automatically orient all new points to a pre-existing set of survey coordinates. But the process we developed worked well and easily, and it gave us a check of our own accuracy as we manipulated the model and created the alternate user coordinate systems. We also put it into practice in a way designed to make it obvious to the user if he was not entering the data correctly. More important, we can use equipment with various levels of sophistication.
*The instrument used measures to within 5 seconds for vertical and horizontal angles. The electronic distance measuring device (EDM) measures to within 5 mm. and 3 parts per million; so the error will be no more than than the sum of 5 mm. and 3 parts per million of the measured distance from instrument to prism. Instruments are available which measure to tighter tolerances, but for short-range work such as we were doing at Pompeii -- nothing we measured was more than 100 m. from the instrument and most of the work was within 25 m. - the accuracy of the transit and EDM were more than sufficient. The EDM error at 100 m. would be no more than 5 mm. (3 parts per million at 100 m. adds less than a mm. to the maximum error). At 100 m. an error of 5 seconds in an angular reading would make only a 2 mm. error in position; at 40 m., the angular error drops below 1 mm. For the vast majority of the work, then, the maximum theoretical error was the error of the EDM, 5 mm. Of course, human error may add to machine error.
Next Article: A Database About Databases
Table of Contents for the August, 1994 issue of the CSA Newsletter (Vol. 7, no. 2)
Table of Contents for all CSA Newsletter issues on the Web
Go to CSA Home Page