It seems to me that the camera position is not at the radius of the dome and, when rotated, the distance to the dome and angle to the surface will change. It would have to be moved forward with spacers to be in the optimal position. I assume the camera should always be at right angles to the dome surface.
The end disk is almost at the end cap and the servo will limit how far the camera can be moved so spacers under the camera bracket would be the solution.
Does it really matter?
I should add that a wide angle camera looking through a flat glass should have increasing refraction the father away from the center the light is entering. This would only distort the image slightly but white light is made up of all the colors which defract at different angles causing color fringing.
A dome would correct these problems which I assume is why BlueRobotics went to the trouble of making them.
My camera position may not be in the same position as the standard BlueRov2 because I am using the 400 mm Aluminum tube. I ‘think’ I allowed for the extra 100 mm but a different thickness material for the tray and bulkheads may have resulted in a different position.
The lens of my camera sits about 1/4" into the dome as seen from the side. Is this about where the BlueRov2 camera sits?