Arcpad and nmea stream

[roberthaught1]

We have a gps system setup where the gps signal is grabbed through the nmea stream, from a leic abackpack. In arcpad 7.1. it worked fine, however in arpad 10 all the elevation are high, about 5 foot exactly. The antennae height is set to 0 (Zero) on both setups. Any suggestions?

[TimHopper]

Robert,

A few questions -

1. What mobile device are you using?

2. What XY and Z coordinate systems are you using?

3. Are you using 'Height above ellipsoid' or 'Mean Sea Level'?

4. If you're using MSL, did you specify a geoid separation value?  There are online calculators which can give you the value for your location.

GPS Preferences > GPS Height tab > Geoid Separation

GPS Height

5. When you say that the values are high, are you determining this by prior collected data you know to be correct?

6. How long are you waiting on each point before you take it.  Also, are you averaging?  If you wait on a point for several minutes (even give it 10-15 minutes on a single spot as a test) does the accuracy of the elevation improve?

Thanks.

[roberthaught1]

Robert,

A few questions -

1. What mobile device are you using? Trimble nomad 800

2. What XY and Z coordinate systems are you using? HARN  83 Florida West Zone Foot, NAD 83

3. Are you using 'Height above ellipsoid' or 'Mean Sea Level'? Height above ellipsoid

4. If you're using MSL, did you specify a geoid separation value? There are online calculators which can give you the value for your location. We have a leica backpack GPS that we grab the NMEA 0183 stream via a serial cable. We have a VRS sytem for RTK

GPS Preferences > GPS Height tab > Geoid Separation. We have to post process the elevations, to be correct as per the grid (elevational geiod). We use 0 for the antenna height and geiod seperation. Because the anntenae height is already figured in teh NMEA stream

GPS Height

5. When you say that the values are high, are you determining this by prior collected data you know to be correct? When we post process, we get a geoid_ z values that when you add to the geiod grid you get the real elevation. Example: In 7.1 we got 5.3 for a elevation , with arcpad 10 we get 10.3. We have determined it is the actuall arcpad software causing this. We setup with 7.1 and 10 on the same spot, and the geoid_z was off. We are using certified benchmarks to test.

6. How long are you waiting on each point before you take it. Also, are you averaging? If you wait on a point for several minutes (even give it 10-15 minutes on a single spot as a test) does the accuracy of the elevation improve? We are not averaging, we use a VRS system with survey grade leica backpack, we wait until the accuracy is within .10 then take the shot, via the NMEA stream at the time

Thanks.

[roberthaught1]

Any help? anyone else using the NMEA stream

[BrettBlack1]

Have you checked with the company that operates the VRS to see if there have been any other issues with vertical measurements reported? I have seen errors in the vertical from other RTN's, but never that much of an error. Is the NMEA message also calculating your geoid? The same message that provides your antenna height can also send the geoid separation value, which could explain the error.

What is your antenna height? If 5', then that is your culprit and the software or your process is compensating for the antenna height twice. I would review your process to see if this is in fact the case.

If that isn't it, check the geoid model process.

I hope this helps.

[BrettBlack1]

Also, you could easily view the NMEA string to see EXACTLY what is being sent. Use a product like My Mobiler's video screen capture and run NMEA/Rangefinder Debug to view the NMEA string and see what is being sent.