On Fri, Nov 25, 2011 at 10:57 AM, dzach <[email protected]> wrote: > Hi > I would like to use two of the 45deg elbow pieces in a camera design. I > was under the impression that the front ends can change from portrait > orientation to landscape and back. However, after looking at the STEP files > for the 10338 sensor (A and D revisions), I am not sure if this is feasible > since the front ends are not square and the mounting holes not equal in the > two dimensions. > > To make sure I understand what is involved, I have the following questions > regarding orientation of the front ends: > > 1. Is it feasible to change _mechanically_ the front end orientation > from portrait to horizontal with the proposed 45deg setup and any of the > sensors? >
Dimitrios, to "change mechanically" we'll need to make different parts, we do not have them. I do not think it is possible to modify existent parts > > 2. Are there any differences in the electronics and/or programming between > 10338A and 10338D front ends? > 10338D has extra identification EEPROM that is used by some software, but they are compatible otherwise > 3. If there are differences, do they influence the orientation of the > front ends and how? > The 10338A is a square board (we did not change the mechanical size of the board since the very first camera, and sesnors and other components became smaller, less there needed on the sensor board, so most of the board became empty. On the other hand when working on Eyesis camera we needed the sensor board to be as small as possible, at least in one direction, so they are now rectangular, not square. The sensors in this application are in portrait mode, and there are several reasons for that, you may find some illustrations in the blog post - http://blog.elphel.com/2011/02/eyesis-4pi/ - you'll get less problems with ERS in such mode, additionally you will cover larger vertical angle from a single row of cameras. You may need landscape orientation of multiple cameras if you use a pair of them for automatic distance measurements (i.e. in automatic car/robot driving). Even if you'll get more visible ERS distortions, when the two cameras are horizontally spaced and precisely aligned, each line will be scanned at the same time and can be used for distance measurements. But that is true only for such applications, if you do the full 3d reconstruction and/or want to have panoramic images with smaller distortions, the portrait orientation is preferable. Andrey > I would be grateful if you could point me to some documentation regarding > this mater, in case I've missed it. > > Thank you > > Dimitrios > > ______________________________**_________________ > Support-list mailing list > [email protected].**com <[email protected]> > http://support.elphel.com/**mailman/listinfo/support-list_** > support.elphel.com<http://support.elphel.com/mailman/listinfo/support-list_support.elphel.com> >
_______________________________________________ Support-list mailing list [email protected] http://support.elphel.com/mailman/listinfo/support-list_support.elphel.com
