Why does my KMZ export have the wrong location?

Continuing the discussion from Survey points random shapes:

@SSD,

The file you shared, BROMHOF1.skp, was not geo-located.

By default, every model has a Location, even when the model is not Geo-Located.
That default location merely gives the shadow engine necessary data to work with.

Unless you changed things, your KMZ ended up at SketchUp’s default location:
USA
Boulder (CO)
40.018309N
105.242139W

You can geo-locate the model via Window > Model Info > Geo-location > Add Location
OR
Window > Model Info > Geo-location > Set Manual Location
The Add Location method imports a Google Earth Terrain model and a flat, single face Snapshot.

If you don’t need the terrain model, do not bring it in; geo-locate the model manually.
Best practice: Never bring it into the model things you don’t need.

Hi, I also have this problem when creating a SU model onto a large Geolocated image from the SU Geo-Location facility.

I then export as kmz and view in GE. But over the space of 300m there is already a 10% placement error.

I need to set out a 200 acre farm and then be able to export as a kmz and view in GE, but this error over such a short distance does not allow me to do so.

Any suggestions?

Thanks, Andy

To help add a little to what Geo posted above: You have to make sure you are using the same datum to compare locations. GE uses the WGS 84 datum and in that zero meridian does not go through Greenwich; I think England uses the Ordiance Survey. Showing my ignorance , the SU geo- facility I have never seen mentioned before. I do know the US version uses data tables in the program by the International Assoctiation of Oil and Gas. Even the location for Boulder Colorado is off in lat by abt 45 m, long is very good, as compared to the USGS ( US Geological Survey). The data tables also have US state plane and UTM for the conformal mapping corrections
Bottom line before making comparison make sure you are comparing apples to apples. My guess the ordance survey probabaly address plate tentonic issues also in the data tables since you are in England. Even in my location , close to Boulder, survey crews tell me Trimble sends correction data over phone lines so they do not have to go back to office to get datum shift info.

Thanks for this, I use the sketchup tool

File
Geo-Location
Add Location

this then gives a google earth based map from which you can then grab a section relevant to the site. This is geo located. I then can either export my SU file as either a kmz file or preview in google earth, though either way creates the same result.

The result is that the location of an item drawn in SU is not correctly located when viewing in Google Earth. Additionally, if there are several drawings, that have a distance between them of say 300m then once viewed on Google Earth the distance has increased to 304m and the drawings are not in the correct place as indicated from SU.

I need to create a large plan in SU and need it to be correct when viewed from GE. In addition, I now do not know which is correctly showing the geolocated information, SU or GE.

My problem is that SU and GE simply don’t mesh with their information, despite the apprent use of GE information within SU to create the geo location.

If there is a correction to be done, that would be something of astart, but I have no idea where to start withing SU to create that correction in the first place.

Thanks, Andy

[quote=“wowo.sketchup, post:2, topic:15472”]
I then export as kmz and view in GE. But over the space of 300m there is already a 10% placement error.


Additionally, if there are several drawings, that have a distance between them of say 300m then once viewed on Google Earth the distance has increased to 304m and the drawings are not in the correct place as indicated from SU.


My problem is that SU and GE simply don’t mesh with their information, despite the apprent use of GE information within SU to create the geo location.

If there is a correction to be done, that would be something of astart, but I have no idea where to start withing SU to create that correction in the first place.[/quote]

Hi Andy,

I think there’s a simple explanation to the distance differences you perceive.
That is, notice the difference in how GE and SU measure distance.

GE’s Ruler Tool measures distance following earth’s contours “on the ground”.
SU’s Tape Measure Tool measures distance in a perfectly straight line, point-to-point.

Thus, unless the farm is perfectly flat as the proverbial pancake, the distance on the ground in GE will be greater than the straight line distance measured in SU.



Currently, I find the geographic coordinates of models in SU 2016 and GE Pro appear to correlate quite well.

Here’s a handy plugin to find the L,Lo of any point in SU.
When installed it appears here: Tools Menu > Latitude/Longitude of point
Then to compare location in GE, the Add Placemark Tool provides the L,Lo of a point.

latlon.rbz (2.4 KB)

See this link that will help you under stand what is happening 15. Control Points and Datum Shifts | The Nature of Geographic Information although you need to follow the examples since the links have some very important data.
We have talked some general info but actual data needs to be presented if you want to do a real error analysis and then you have the issue what is the truth. I would not consider that either Google Earth nor SU for obvious reasons. The link I noted above will show you how to access the USGS data bases and is what I usually use. If you input geo-location for say Boulder Colrado what SU uses could very well not be the data sited by the USGS plus we have no idea what you are using.
As noted Su uses two differnt conformal transformation ( projection of oblate spheroid Earth) to flat maps. One called UTM which is basically a cylindrical transformation and is used for locations out side CONOUS and State Plane which is more accurate and used for CONOUS. Those are not accurate over the whole Earth so they broken up into sections so you have to make sure you use the correct one to get " accurate data" Ref http://www.uwgb.edu/dutchs/UsefulData/ConvertUTMNoOZ.HTM.
BTM inputting location in just seconds is not accurate enough think USGS uses down to 100th sec.