Yesterday we ran a complete Sun tracking, logging 3D surfaces of the received signal along the path. The aim of these experiments is to have a better understanding of the offsets. We suspect that they are dependent on the positions. How 2D projected surfaces are produced has been explained in a previous post.
Experimental conditions:
-
PB8 start position : (124.597, 42.677)
-
PB8 last position : (, )
-
Main offset alignment on IS-905 (207,30.4)
- Voltage of the excitation signal (Artus resolvers) : 18V
-
AZ speed value: 0.015 °/s
-
Received noise frequency: 4GHz, bandwith 300KHz, 10Hz filter
-
Spectrum analyser: 20dB / 0-4V full scale
-
AZ-EL frame for a 2D exploration (º): (-0.3,0.3) / (-0.3,0,3)
-
Number of sampling zones in one frame: 10×10
-
Number of measured points in a sampling zone : 5
-
Mean time to build an 2D image: 20 min
Considering EL, we overestimate the position of the target before the apogee and underestimate it after. One explanation could be related to the equilibrium and the weight of the reflector. Another option is to incriminate the push/pull mechanism and the engine.
The AZ case is again (same output was observed during a previous Moon tracking, see that post) much more complex to explain. The only option we have for now is to say that the algorithms behind these images are wrong…
