Welcome, Guest. Please login or register.

Login with username, password and session length
Forums Search:  


Author Topic: Garmin Waypoints into AutoCAD  (Read 8328 times)

kbellis

  • Jr. Member
  • **
  • Posts: 24
  • Karma: 0
    • View Profile
    • glimpsed
Garmin Waypoints into AutoCAD
« on: January 11, 2010, 11:19:27 AM »
What is the best means of getting waypoints from my Garmin Colorado and Oregon into AutoCAD?

I've got AutoCAD version Land Desktop 2004, MapSource Version 6.15.7 and Terrain Navigator Pro version 7.5.

Thanks for any advice.

leszekp

  • Mapper
  • ***
  • Posts: 103
  • Karma: 8
    • View Profile
Re: Garmin Waypoints into AutoCAD
« Reply #1 on: January 11, 2010, 11:49:41 AM »
Garmin's MapSource program offers the option to save data imported from your GPS as a DXF file. If you don't have MapSource ...

http://freegeographytools.com/2007/garmin-mapsource-for-free

http://www.gpsfiledepot.com/tutorials/how-to-install-mapsource-if-it-didnt-come-with-your-gps/

kbellis

  • Jr. Member
  • **
  • Posts: 24
  • Karma: 0
    • View Profile
    • glimpsed
Re: Garmin Waypoints into AutoCAD
« Reply #2 on: January 11, 2010, 01:33:21 PM »
Thanks!

That worked fairly well.

Do you know if the coordinate for the point of insertion for the symbol in the dxf is the coordinate of the waypoint? When I compare the UTM values of the waypoint in 1) the Garmin unit, 2) MapSource and 3) AutoCAD, I noted the following:

1) E554150 N4912561 (Garmin)
2) 19 T 554152 4912561 (MapSource)
3) E554152 N4912561 (AutoCAD)

Why would there be  a 2 meter difference in the easting?

leszekp

  • Mapper
  • ***
  • Posts: 103
  • Karma: 8
    • View Profile
Re: Garmin Waypoints into AutoCAD
« Reply #3 on: January 11, 2010, 02:18:46 PM »
Should be the waypoint coordinate; don't know why you'd get a 2-meter shift.

Boyd

  • Expert Advisor
  • *****
  • Posts: 3814
  • Karma: 43
    • View Profile
Re: Garmin Waypoints into AutoCAD
« Reply #4 on: January 12, 2010, 08:15:38 AM »
Don't know if this could be it, but my understanding is that Garmin's map format only supports about 5 meter accuracy (32 bit coordinates IIRC), so 2 meters could be rounding error.

kbellis

  • Jr. Member
  • **
  • Posts: 24
  • Karma: 0
    • View Profile
    • glimpsed
Re: Garmin Waypoints into AutoCAD
« Reply #5 on: January 12, 2010, 08:48:44 AM »
Hey Boyd, thanks for the tip - can you point out further reading on this subject?

One of the things I've been playing around with is setting up various user grids wherein specific values for the central meridian are entered for each coordinate system (ddd mm.mmm); e.g., W68d30.000'. Invariably, the next time I'm looking at this field the last place has changed; i.e., 30.001' - as an example. But there are times, when it stays put at 30.000'.

Now mind you, I'm not trying to make a silk purse out of my CO or OR, but to just better understand their limitations. In conversations with another surveyor on the subject of setting up SPC on Garmins, limitations due to Garmin's rounding rules may have evolved somewhat from those rules used 10 years ago or more.

Boyd

  • Expert Advisor
  • *****
  • Posts: 3814
  • Karma: 43
    • View Profile
Re: Garmin Waypoints into AutoCAD
« Reply #6 on: January 12, 2010, 10:32:17 AM »
Looks like it's 24 bits.... See page 23 here: http://cgpsmapper.com/download/GM8DocV2.pdf

Quote
When declaring the location of objects in the world, the highest precision available uses 24 bits of resolution, which resolves to an accuracy of about 2.5 meters. Consumer GPS units have a best-case accuracy of about 8 meters. The header attribute of Level0=24 means that objects declared at level 0 will be stored with 24-bit accuracy, or a location of +- 2.5m.

If you look at Garmin's handheld specs, they only claim accuracy of 10 meters. I have played around a lot "surveying" my own land, and this is probably a reasonable expectation over the long haul. Sometimes you do get better results, but probably not on a consistent basis.

So if you have a stake in the ground at a known location, this means every time you visit it, your position should be reported within a circle that has a 10m (33 foot) radius. However, it would be possible for two of those readings to be 20m (66 feet) apart and still be within spec (one reading could be 10m south of the stake and another 10m north of it).

So (IMO) an error of 2 meters is not going to matter very much if you're using data from consumer handheld GPS'es. That should be less than the "noise" in the data.

 

anything