Posted By James Rowden on 1/12/2010 at 3:46 PM
I have been tinkering with this for over 2 weeks now. I think I have it figured out but figured if I'm wrong, someone here could spot the error of my logic.
TxDot uses surface coordinates for most all of their jobs using a combined scale factor for an entire county. I have a list of the scale factor used for each county in Texas. The SPC grid coordinate is simply multiplied by the scale factor to get the surface (or close to surface) coordinates. No truncating of the left digits. It gives a coordinate that is about 1300 feet different in my area.
First I created a box 10 min x 10 min around my work area and converted the lat-lon to surface coordinates. I entered both sets of coordinates into Survey Controller and forced a calibration. This created a Transverse Mercator projection that was not on a parallel plane so I threw it out.
Next I tried creating a Lambert projection using a larger radius and quickly figured out I was not smart enough to do that (but it was fun).
Next I entered the SPC projection from the library in Survey Controller, selected the zone, and for coordinates selected "Ground (keyed in scale factor)". Then I entered the TxDot scale factor for the area, left the false N and E at 0.00. I selected the lat and long of the area I am working in for the project location. This gave me surface coordinates that matched the SPC grid coordinates at the point selected for "Project Location", not the 1300 foot difference I needed. I had done this last year and knew what to expect.
Next I changed the project location to the lat and Long of the origin for my zone. This brought it closer but still not right. I could have calculated a North and East offset to make it work but if I went to a different county with a different calculated scale factor then the offset would change.
Next I calculated the Lat Long of the point that would be N 0.000 E 0.000 of the SPC grid. I tried doing it on the NGS website but got a error. I converted it successfully using the SPC83.exe program downloaded from NGS.
3d26'32.41401" N Lat
105d58'59.53412 W Long
I really didn't expect it to work since it is so far out of the actual zone but it did. I had entered lat long for points covering an area 75 miles wide. Displaying them in N and E gave me coordinates that matched what I had calculated by 0.003 feet or less. The coordinates are in US Survey Feet so the 0.003 could be round off error from converting from meters.
I can set this up as a template file and change the scale factor for the county I am working in. Any county in the Texas Central Zone will have the same "project location".
I know I still need to check and recheck to match the coordinates on a job and may still have to do a site calibration if the control does not match my projection.
Do you see any problems with doing this? Was there an easier way?
Sorry to be so long winded.
edit: the NGS online conversion will not work with 0 0 but is will work with 0.000 0.000
To read the rest of this thread go to www.i-boards.com/bnp/pob/messages.asp?MsgID=1518373&ThreadID=143202&IsResponse=False#1518373.
Surface Coordinates and Lying to Survey Controller
January 13, 2010