Measurement and Evaluation

of EDM Calibration Baselines

by

Ibrahim Chami
 
 

Supervisor and Web Page:

Assoc. Prof. J. M. Rüeger


 
 



This project involved, amongst other things, the completion of a precise levelling survey and of a alignment survey of the UNSW EDM Research Baseline in Regents Park, Sydney, Australia.

Precise Levelling Survey

The aim of the precise levelling survey was to re-measure the elevations of the pillar tops on the EDM calibration baseline, with a standard deviation of ±0.3 mm for one kilometre double run levelling. In order to achieve such a high accuracy, high precision equipment was used and appropriate measuring procedures were adopted. The instrument used for the precise levelling survey of the Regents Park EDM Calibration Baseline was the Zeiss Ni1 S/N372342 precision automatic level. The Zeiss Ni1 level has a built-in plane parallel plate micrometer. It is mostly used for work of very high precision. The manufacturer quotes a standard deviation of ±0.2 mm for one kilometre double run levelling with the Zeiss Ni1. Two 3 m WILD GPLE3 invar staffs, with struts and with 10 mm graduations, were also used. To accurately measure to the pillar tops, a short WILD GNL92 invar staff of 0.92 m length, also with 10 mm graduations, was used. The BFFB measuring procedure was used. This method requires two staffs. The sighting distances were kept below 25 m at all times.

The collimation error of the level was adjusted to 2.2" on 28 June 2001. The levelling of the baseline was carried out on 2, 3 and 5 July 2001 by Messrs. I. Chami, G. Collyer and M. London. The weather was sunny on the first day and overcast on the others. From the differences of the section height differences in the forward and backward runs, a 1 km double run precision of +/-0.34 mm per square root kilometre was computed. The precision of a single B-F observation at each set-up was +/-0.08 mm. The observations were subjected to a least squares adjustment to obtain the new pillar elevations. In doing so, the individual B-S observations from all set-ups were used as observations and the elevations of all but one pillar and the change points as unknown parameters. The adjustment included 168 height difference observations and 53 unknown parameters and had a degree of freedom of 115. Since the UNSW EDM Research Baseline in Regents Park is 980 m long, the precision quoted for Pillar 1 and the minimum constrain solution gives, with ±0.56 mm per square root kilometre, another indication of the achieved precision. The connections to the pillar tops are included in this value whereas they are not in the perviously calculated value of ±0.34 mm per square root kilometre. When connecting to the KERN centring plates on the concrete pillars, the short staff was set up on the three foot rests, in turn. Since the pillar plates are no longer horizontal, the residuals of these measurements are a measure of the tilt of the KERN centring plates rather than the measuring precision.

Pillar
Elevation

Min.Constraint

Solution

Free Net

Solution

...
...
Precision
95% Conf. Int.
Precision
95% Conf. Int.
...
(m)
(mm)
(mm)
(mm)
(mm)
1
24.3016
±0.56
±1.11
±0.32
±0.63
2
31.2857
±0.52
±1.02
±0.25
±0.50
3
30.1950
±0.49
±0.97
±0.22
±0.44
4
28.5052
±0.46
±0.92
±0.20
±0.39
5
26.3305
±0.42
±0.83
±0.18
±0.35
6
23.9588
±0.34
±0.67
±0.19
±0.37
7
22.2536
±0.20
±0.40
±0.27
±0.54
8 (Datum)
22.0000
±0.00
±0.00
±0.32
±0.64
 

Alignment Survey

The aim of the alignment survey was to establish the pillar coordinates at right angle to the baseline and to determine any lateral movements of the pillars since the last alignment survey. The alignment survey was to be measured to a high accuracy, therefore high precision equipment and effective measuring procedures were needed. A WILD T3000 precision electronic theodolite was used in the alignment survey, together with KERN traversing targets. The WILD T3000 has a standard deviation of ±0.6" or better for a direction measured in two faces. The Alignment survey was measured with a technique called interlocking traverses. For interlocking traverses, horizontal directions and EDM distances are measured, at successive traverse stations, to a number of stations in the backward direction and to a number of stations in the forward direction, starting at one endpoint of the baseline and finishing at the other. A precision of ±0.3 mm (free net solution) was aimed at for the lateral coordinates.

As a preparation, the precision of zenith angle measurements (after DIN18723) was determined on the UNSW Campus on 27 June 2001. A precision of ±1.31" was achieved for a single zenith angle measurement in two faces. This is more than twice the value specified by the manufacturer for the WILD T3000 precision theodolite. The alignment survey on the UNSW EDM Research Baseline in Regents Park was carried out on 4 July 2001 by Messrs. I. Chami, G. Collyer and M. London. On the baseline, the precision of a single direction measurement in two faces varied from ±0.47" on Pillar 1 (best) to ±0.81" on Pillar 2 (worst), with a pooled value of ±0.60". Two arcs of directions were measured on each pillar. The number of targets sighted to at each pillar varied from 4 to 5. All data were recorded electronically on WILD GRM10 REC modules. The UNSW GRECV software was then used to merge the FL and FR measurements and to produce the input file for the least squares analysis using the UNSW FIXIT2 software. The least squares analysis included 70 directions and 28 distances as well as 30 unknown parameters. The distances were measured (in all combinations) on 19 April 2001 by Messrs. I. Chami, G. Collyer and M. London with the WILD TC1610 S/N 365568 electronic tacheometer.

Pillar

Northing

Min.Constraint

Solution

Free Net

Solution

...
...
Precision
95% Conf. Int.
Precision
95% Conf. Int.
...
(m)
(mm)
(mm)
(mm)
(mm)
1
1000.0000
±0.00
±0.00
±0.30
±0.61
2
999.9901
±0.29
±0.59
±0.12
±0.24
3
999.9972
±0.35
±0.71
±0.12
±0.24
4
1000.0018
±0.45
±0.91
±0.21
±0.42
5
999.9972
±0.49
±0.99
±0.27
±0.55
6
999.9914
±0.42
±0.85
±0.24
±0.49
7
1000.0012
±0.21
±0.42
±0.14
±0.28
8
1000.0000
±0.00
±0.00
±0.21
±0.42

The table gives the result of an least squares analysis where the directions were weighted separately for each station according to the precision of the direction measurements on the station in question. The Northings of the Pillars 1 and 8 were kept fixed and, thus, define the axis from which the offsets are measured. Because of the orientation and origin of the local coordinate system (Pillar 1: E = 0, N = 1000); Pillar 8: E = 980, N = 1000) and since Pillar 8 is, geographically, due West of Pillar 1, Northings larger than 1000 m indicate that the Pillar is geographically South of the Line from 1 to 8; Northings smaller than 1000 m indicate that the Pillar is geographically North of the Line from 1 to 8. The expected precision of the Northing coordinates was achieved.

Further Information


For more information contact:

 
A/Prof. J. M. Rüeger (supervisor)
Email: J.Rueger@unsw.edu.au
 
Mail:
School of Surveying and Spatial Information Systems
University of New South Wales
UNSW SYDNEY NSW 2052
Australia
 
Phone: +61-2-9385-4173

Fax: +61-2-9313-7493