Page 24 - 4798

Systematic Errors
These are repetitive errors that are caused by imperfections in the surveying
equipment, by the specific method of observation, or by certain environmental
errors or cumulative errors.
Under the same conditions of measurement, systematic errors are constant in
magnitude and direction or sign (either plus or minus). They usually have no
tendency to cancel if corrections are not made.
0
For example, suppose that a 30-m steel tape is the correct length at 20 c and
0
that it is used in a survey when the outdoor air temperature is, say 35 c. Since steel
expands with increase in temperatures, the tape will actually be longer than it was
0
at 20 c. And also transits, theodolites and even EDM are also subjected to
systematic errors. The horizontal axis of rotation of the transit, for instance, may
not be exactly perpendicular to the vertical axis.

Accidental Errors
An accidental or random error is the difference between a true quantity and a
measurement of that quantity that is free from blunders or systematic errors.
Accidental errors always occur in every measurement. They are the relatively
small, unavoidable errors in observation that are generally beyond the control of
the surveyor. These random errors, as the name implies, are not constant in
magnitude or direction.
One example of a source of accidental errors is the slight motion of a plumb
bob string, which occurs when using a tape to measure a distance. The tape is
generally held above the ground, and the plumb bob is used to transfer the
measurement from the ground to the tape.

Most Probable Value
If two or more measurements of the same quantity are made, random errors
usually cause different values to be obtained. As long as each measurement is
equally reliable, the average value of the different measurements is taken to be the
true or the most probable value. The average (the arithmetic mean) is computed
simply by summing all the individual measurements and then dividing the sum by
the number of measurements.

The 90 Percent Errors
Using appropriate statistical formulas, it is possible to test and determine the
probability of different ranges of random errors occurring for a variety of
surveying instruments and procedures. The most probable error is that which has
an equal chance (50 percent) of either being exceeded or not being exceeded in a
particular measurement. It is sometimes designated as E 90.

22