are used in a control loop, the error in the loop is the different between the current star position and the original or target star position.
The line uses pins 2,3,7 of a 25 way D-type and goes from SP1 on the Autoguider VME rack right hand card to a terminal port on the TCS vax ( usually lpvb). The VME rack has a female connector and the cable a male plug.
Baud rate = 9600 baud, 8 bits , no parity.
The packet structure is... XXXXYYYYSTTTTC
and contains
xxxx 4 decimal ascii chars for x error, 2.2 microns in focal plane = 1 unit
yyyy 4 decimal chars for y error, 2.2 microns in focal plane = 1 unit
s 1 character for time flag ascii 0 for ok ascii - for -ve time which
signals bad guide data.
tttt 4 decimal chars for time estimate till next packet, 100 units = 1 second
c 1 character for end of line = ascii hex 0D carriage return.
If the time estimate is 0000 then the TCS should drop out of autoguide back into manual tracking. If the time estimate is -ve ie xxxxyyyy-1234 then the x,y coordinate is ignored If the time estimate is +ve ie xxxxyyyy01234 then the sample is used to track the telescope. A timeout of the expected time given times 3 is used, after which the TCS drops back into manual tracking ( without complaining ).
The coordinate system is arranged to be in the detector focal plane with Y being up at sky PA of 0. The x and y coordeinates are the coordinate of the centroid of the star image on the detector. One unit in the focal plane = 2.2 microns The TCS knows about the field scale for each focal station. The origin is the bottom left of the detector array, so that the x,y coordinates are always positive.
The TCS will ignore any samples which are significantly different from the current position. The TCS has a set of image area boundaries for each configuration. Samples outside this area will cause an error to be issued to the operator console and the TCS will drop back into manual tracking.