Grid to Ground

Discuss all Construction & Architecture related issues here.
User avatar
Phill
V.I.P Member
V.I.P Member
Posts: 653
Joined: Wed Feb 06, 2008 7:19 am
16
Full Name: Phillip Nixon
Company Details: SKM
Company Position Title: Surveyor
Country: Australia
Location: Sydney
Has thanked: 7 times
Been thanked: 45 times

Re: Grid to Ground

Post by Phill »

Ok

So I did a few calculations. I used the rough coordinates of Sydney as that is where I am to calc the values.
Based on the highest peaks in the world with an elevation of 7200 meters above sea level a change in elevation of 50m results in a change in scale factor of 0.0000078144 (0.9989616849-0.9989538705). Over 50m this change equates to 0.4mm. At sea level this is almost the same (within 0.002mm). I would consider this change in scale negligible over this distance but if you want to take it into account then there should perhaps be some way.

Note here that I am not asking Cyclone to calculate the scale, it is a value that I will give it for the entire project or on a scan by scan basis. It would be my responsibility to calculate the value and give it to Cyclone, and as noted above the scale factor would only affect the x and y values, not the z value which would remain the same (except for perhaps the tiny adjustments shown here). It would also be a combined scale factor, taking elevation as well as position into account.

Now refraction, I have used a standard value of 0.13 over our test distance of 50m I get a change in height of 0.2mm.

Really the distances should be adjusted to local atmospherics as well. I think most scanners adjust for temperature (looking at the manuals the HDS7000 eludes that it is taken into account, the C10 has no mention of it) but not pressure. The ppm range here from -5degrees to 40degrees Celsius, and from 900hpa to 1050hpa is from -30 to +30, so if the default is the median at the extremes this would be 1.5mm over 50m. You used to be able to put these things in using the cyclone scan controller on the laptop, but I don’t think you can on the newer units (I'll have to check)

The only other thing that I am aware of that might apply here is grid convergence, which is the difference in azimuth from true north to projection north. Unless you are taking a backsight of the stars or the sun or using a gyroscope then you will most likely be taking your backsight in a projected or local system anyway and convergence doesn't need to be applied.

In regards to the truivew (I can’t speak for scene here) if the ptx file is adjusted the truview will give grid positions and distances.

In regards to mismeasurement due to having a projection on grid, if the client has specified that they want the data, either extracted or cloud based, on a projection then the mismeasurement is their problem, and they should be aware of it. All you have done is fulfil their brief. Your example of clearance is a bad one, as there is not much change to height when projecting to grid as shown above, but it is definitely a problem with clients who specify they want their data in a projection without understanding the ramifications.

So we have:
-Scale factor to grid: Up to 20mm over 50m (using 0.9996 which is a mid UTM projection scale factor)
-Change in scale factor due to change in height: 0.4mm over 50m
-Refraction of the vertical angle: 0.2mm over 50m
-Scale factor due to atmospheric change: Up to 1.5mm over 50m

So looking at all those numbers the scale factor is the biggest one by quite a long way, which is why I mentioned perhaps it should be something we are able to do in cyclone.

The other thing that I haven’t mentioned here is those doing traversing or cloud to cloud over long distances with just scanners should be concerned about those small numbers shown above, as they do come into effect. Those who use total stations to pick up the targets and do their traversing should have refraction and atmospherics dialled in anyway.

Cheers

Phill
geowac
V.I.P Member
V.I.P Member
Posts: 353
Joined: Thu Jan 23, 2014 10:07 am
10
Full Name: Stefan Wachsmann
Company Details: scantaxi.de
Company Position Title: supervisor
Country: Germany
Linkedin Profile: No
Location: Germany
Been thanked: 1 time
Contact:

Re: Grid to Ground

Post by geowac »

Hi Phill,

thank you for your calculations, I think, I will put them to my records :D
Phill wrote:In regards to the truivew (I can’t speak for scene here) if the ptx file is adjusted the truview will give grid positions and distances.
Thats worth to know, I fear SCENE didn't allow scaling anyway - I have to take a look.

By the way: Is it possible to scale the scans in Cyclone? Will the exported TruView be right too? If you import the scaled PTX, there is no chance to have an "unscaled" TrueView ;)
Phill wrote:Your example of clearance is a bad one, as there is not much change to height when projecting to grid as shown above, but it is definitely a problem with clients who specify they want their data in a projection without understanding the ramifications.
May be you are right I should had to think a little more about this or explain better :(

But I thought of a clearing for big ships (about 42m) and uniform scaling. The allowed mismatch to the drawing is about 15 mm (in my case - I have to take in account the temperature, wind, the load/stress (?), ...) ... You are right, the Z-Value will not be affected like the XY-Values.


I think, best is to scale each scan in xy (not z), but transform every "local" scan to the grid. with a length of 50m per scan you may be right, that the errors are minimized.

My misinterpretation was to scale the whole project and put in to the grid. With a distance of 50m per scan, you will get minor deviations in the overlapping area far from the station (as in real map projections, too).

Thank you for your calculating work!

So we have an answer for the question of the initial post 8-)
- scaling only in xy (if possible)
- scaling single scans
- transform each scan to grid
- deliver pointcloud

Is this what we researched?

Stefan
User avatar
bwiab
I have made 80-90 posts
I have made 80-90 posts
Posts: 88
Joined: Thu May 08, 2014 6:48 pm
9
Full Name: Emilio Ramirez
Company Details: Metro
Company Position Title: BIM Supervisor
Country: USA
Linkedin Profile: Yes

Re: Grid to Ground

Post by bwiab »

Thanks for all the input. Very informative. We are in Denver so elevation +/- 5200 feet. The scale factors we have been using are

Grid to Ground .99979199
Ground to Grid 1.000208053

Those were calculated by someone who left the company long before I started working here.

After reviewing these posts and discussing with my surveyor, our decision was to not scale the point clouds because they reflect the actual condition of the ground, structures, coordinates etc. That being said, we have scheduled a meeting with a consultant for later this month to continue the debate. Thanks again!

Think I'm ready to join the flat earth society! :lol:
Post Reply

Return to “Construction & Architecture”