Title: Lateral x-ray: 4 marks vs 2 marks
1Lateral x-ray 4 marks vs 2 marks
- Test perfomed using the plate 57 of one
brickwith 4 x-ray frontal marks and 4 x-ray
lateral marks - The distances between maps in micron are shown in
the following slides (as depend on plate
position). - Legend- MAPFXR4 map evaluated using 4 x-rays
frontal marks- MAPLXR4 map evaluated using 4
x-rays lateral marks- MAPLXR2 min/max X side
map evaluated using 2 out of 4 x-rays lateral
marks located on the plate side where x is
minimum/maximum. () - () A map is a transformation between stage
coordinates and nominal (or detector or
what.ever.you.want) coordinates - () The same mark positions of the case with 4
marks are used to not introduce effects due to
the mark position accuracy. The map is obtained
by simply setting the other marks as NOT FOUND.
2MAPFXR4 MAPLXR4
Distance between maps (micron)
The rough positions of the marks used are shown
in black (1 map) and in white (2 map)
3MAPFXR4 MAPLXR2 minX side
Distance between maps (micron)
The rough positions of the marks used are shown
in black (1 map) and in white (2 map)
4MAPFXR4 MAPLXR2 maxX side
Distance between maps (micron)
The rough positions of the marks used are shown
in black (1 map) and in white (2 map)
5Comments
- The distance between MAPFXR4 MAPLXR4 is below
10 micron over the whole emulsion plate - The map evaluated with only 2 lateral marks get
worse and the distance can increase up to 50
micron on the opposite side of lateral marks . - Let compare the 4-marks frontal x-ray map
with a 2-marks frontal x-ray map obtained by
simply setting as NOT FOUND the 2 marks with
higher x positions - And what happen if we evaluate the distance
between this map and the 2 marks lateral x-rays
map?
6MAPFXR4 MAPFXR2 minX side
Distance between maps (micron)
The rough positions of the marks used are shown
in black (1 map) and in white (2 map)
7MAPFXR2 minX side MAPLXR2 minX side
Distance between maps (micron)
The rough positions of the marks used are shown
in black (1 map) and in white (2 map)
8Final comments
- Changing from a 4 marks map to a 2 marks map
introduces errors in the position estimation.
This is also true if you have 4 marks map but you
didnt find 2 marks. - This is not expected if you dont change the map.
For examples if you use only a 2 marks map for
the entire brick (i.e. in the scan-back). - This is due the different number of free
parameters available for the transformation (6 in
the case of 3 and more marks, 4 in the case of 2
marks). The scaling of Y coordinates cannot be
well estimated by measuting only 2 marks (with
quite the same Y). - 4-marks lateral marks are necessary if you dont
want to manage these large errors (at the
beginning of scanback and at the end of the
scanforth or every time you change the map).