RD 16 increment accuracy [Archive] - Nodal Ninja Forum

PDA

View Full Version : RD 16 increment accuracy



waters
04-05-2012, 12:22 PM
I am thinking of using a RD 16 with a Leica Disto to do some horizontal triangulation / mapping. How accurate are the angles between click stops, i.e. 3.75 degrees?

nick fan
04-06-2012, 08:14 AM
I am thinking of using a RD 16 with a Leica Disto to do some horizontal triangulation / mapping. How accurate are the angles between click stops, i.e. 3.75 degrees?

I never do such measurement. each unit varies.
0.1 deg should be easily achievable for single detent (>7.5 deg).
accuracy decrease with double detent (e.g. 3.75deg).

But your should be able to calibrate it with a cylindrical pano, provided you always start with a fix point, eg 0 deg.

Nick

Hugh
04-06-2012, 11:29 AM
Hello Waters,

Interesting application!

You could revert back to surveying techniques used some decades ago, but with a twist.
Before EDM we measured angles with a theodolite and distances with a tape.
Reverse the process.
Take your pano shots as usual, then create a Mirror Ball looking to the Nadir and you will have a "protractor" for your angles and the distances with the DISTO to selected points.
The angles measured from the point directly at the Nadir will be correct.
If you have a straight line with features along it (e.g. wall with windows/doors) then a DISTO shot to each end will fix it and intersecting rays from your Mirror Ball image will fix each feature.

waters
04-06-2012, 04:23 PM
I have a room with a bowed wall and want to recreate the " topography " in my CAD program. If the gradations on a R16 are relatively accurate I will have an angle and two known lengths to map it.

Wim.Koornneef
04-06-2012, 11:02 PM
.....If the gradations on a R16 are relatively accurate I will have an angle and two known lengths to map it.

Hello Waters,

Some time ago I tested the accuracy of my (!) RD16 rotator by shooting a pano on a very heavy and ridged tripod and I found a difference in yaw (calculated by PTGui) of less then 0.1 degree between the clickstop positions.

The accuracy of the repeatability of the rotator is even more precise, for this I stacked 10 images shot with a 5D + Tokina 10-17@ 12 mm in the same position (after doing a full rotation for each shot) in a layered photoshop document. It turns out that the "shift" between the images was max 2 px with an avarage shift for all images of 1 px.

(!) I guess accuracy can differ a bit between production models but I doubt it will be much.

Hope this info is of any help.
Wim

Hugh
04-07-2012, 02:41 AM
I have a room with a bowed wall and want to recreate the " topography " in my CAD program. If the gradations on a R16 are relatively accurate I will have an angle and two known lengths to map it.

Hello Waters,
This may interest you:
http://www.hugha.co.uk/PhotoInt/PhotoInt.htm
http://www.hugha.co.uk/PhotoInt/PhotoInt360PanoramicOptic.htm
Unfortunately the development of the DISTO was some decades away in the early 1970s!
Thanks Wim for your input (as always a great source of useful information), which definitely suggests that the accuracy of the RD16 will meet Waters needs.
The only problem I see is that the direction of the DISTO mounted on the RD16 is governed by the click stops so points can be measured along the curved wall at each click stop, but fixing corners and features that are not at the click stops is a different matter so this is where a Nadir Mirror Ball image can be used.
I use Pano2VR to create the Mirror Ball - maybe you do not have this software or another to create this image.
If this is the case i would be happy to oblige if you send me the Equirectangular.
As a matter of interest, which CAD package are you using and does it have the functionality of using "background images"?