U.S. patent application number 09/812476 was filed with the patent office on 2001-09-20 for method of setting a position of an object of measurement in layer thickness measurement by x-ray fluorescence.
Invention is credited to Kaiser, Karl-Heinz, Robiger, Volker.
Application Number | 20010022829 09/812476 |
Document ID | / |
Family ID | 7635117 |
Filed Date | 2001-09-20 |
United States Patent
Application |
20010022829 |
Kind Code |
A1 |
Kaiser, Karl-Heinz ; et
al. |
September 20, 2001 |
Method of setting a position of an object of measurement in layer
thickness measurement by x-ray fluorescence
Abstract
The invention relates to a method of setting a position of an
object of measurement in layer thickness measurement by X-ray
fluorescence in which a beam of an optical recording device is
projected into the beam of the X-radiation and in which the surface
of the object of measurement is recorded and output as an image
comprising a number of image points, with the distance between the
surface and the collimator being changed by an absolute amount of a
path of movement, with changes in brightness of the image points
being recorded in at least one measuring plane during the at least
one change of the distance between the surface and the collimator,
with the maximum of the difference in brightness of the image
points of an image being ascertained after the at least one change
of the absolute amount of the distance, and with the distance
between the collimator and the object of measurement being set to
the position of the ascertained maximum of the difference in
brightness.
Inventors: |
Kaiser, Karl-Heinz;
(Boblingen, DE) ; Robiger, Volker; (Sindelfingen,
DE) |
Correspondence
Address: |
M. Robert Kestenbaum
11011 Bermuda Dunes NE
Albuquerque
NM
87111
US
|
Family ID: |
7635117 |
Appl. No.: |
09/812476 |
Filed: |
March 19, 2001 |
Current U.S.
Class: |
378/45 ;
378/205 |
Current CPC
Class: |
G01N 2223/076 20130101;
G01N 23/223 20130101 |
Class at
Publication: |
378/45 ;
378/205 |
International
Class: |
G01N 023/223 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 17, 2000 |
DE |
100 13 048.8 |
Claims
1. Method of setting a position of an object of measurement in
layer thickness measurement by X-ray fluorescence in which a beam
of an optical recording device is projected into the beam of the
X-radiation and in which the surface of the object of measurement
is recorded and output as an image comprising a number of image
points, characterized in that: the distance (26) between the
surface (14) and the collimator (17) is changed by an absolute
amount of a path of movement, changes in brightness of the image
points are recorded in at least one measuring plane (32) during the
at least one change of the distance (26) between the surface (14)
and the collimator (17), the maximum of the difference in
brightness of the image points of an image is ascertained after the
at least one change of the absolute amount of the distance (26) and
the distance (26) between the collimator (17) and the surface (14)
of the object of measurement (16) is set to the position of the
ascertained maximum of the difference in brightness.
2. Method of setting a position of an object of measurement in
layer thickness measurement by X-ray fluorescence in which a beam
of an optical recording device is projected into the beam of the
X-radiation and in which the surface of the object of measurement
is recorded and output as an image comprising a number of image
points characterized in that: the distance (26) between the surface
(14) and the collimator (17) is changed by an absolute amount of a
path of movement, changes in brightness of the image points are
recorded in at least one measuring plane (32) during the at least
one change of the distance (26) between the surface (14) and the
collimator (17), the maximum of the difference in brightness of the
image points of a image is ascertained after the at least one
change of the absolute amount of the distance (26) and a direction
opposite to the recording of the change in brightness is chosen for
the changing of the distance (26) and the changing of the distance
(26) is stopped immediately after the maximum of the difference in
brightness is reached.
3. Method according to claim 1, characterized in that individual
measurements for ascertaining changes in brightness of the image
points of a image are carried out during a changing of the distance
(26) between the collimator (17) and the surface (14).
4. Method according to claim 1, characterized in that, for
ascertaining the maximum value of the difference in brightness in a
measuring plane, the image points are determined in a differential
method according to the function F=.SIGMA.(y.sub.1-y.sub.right
neighbor).sup.2+.SIGMA.(y.sub.1-y.- sub.upper neighbor).sup.2,
where y.sub.1 is the brightest value of the image points.
5. Method according to claim 1, characterized in that the changing
of the distance (26) between the collimator (17) and the surface
(24) of the object of measurement (16) corresponds to an absolute
amount of the measuring path in which at least the focal point (31)
lying at the distance (26) from the collimator (17) is passed
through.
6. Method according to claim 1, characterized in that the
individual measurement is ascertained from a number of individual
images at a preferably freely selectable time interval and in that
an average value is formed from the values of the individual
images.
7. Method according to claim 1, characterized in that during the
individual measurement the speed for the changing of the distance
between the collimator (17) and the surface (14) of the object of
measurement (16) is kept constant.
8. Method according to claim 1, characterized in that the changing
of the distance (26) in an approximate search for a first
ascertainment of a maximum of the difference in brightness is
carried out at an increased speed.
9. Method according to claim 1, characterized in that the distance
between the collimator (17) and the surface (16) is reset after a
first intermediate position (42) at a resetting speed to a second
intermediate position, which is preferably close to the starting
position.
10. Method according to claim 1, characterized in that the return
speed is set equal to or greater than that of the approximate
search.
11. Method according to claim 1, characterized in that the
precision search is carried out a speed reduced in comparison with
the approximate search.
12. Method according to claim 1, characterized in that, after
passing through the focal point (31) between the collimator (17)
and the surface (14) of the object of measurement (16), the
precision search is stopped and the surface (14) of the object of
measurement (16) is set to the Z coordinate assigned to the
maximum.
13. Method according to claim 1, characterized in that, before a
commencement of the approximate search, a preset distance between
the collimator (17) and the surface (14) of the object of
measurement (16) is increased.
14. Method according to claim 1, characterized in that the changing
of the distance (26) between the collimator (17) and the surface
(14) takes place along a Z coordinate and the individual
measurement of an image is assigned to a point in time of a
corresponding Z coordinate.
15. Method according to claim 2, characterized in that, during the
precision search, a comparison with the ascertained maximum value
of the approximate search is carried out, and in that, when there
is a minimum in the difference of the maximum values, the changing
of the distance (26) between the collimator (17) and the surface
(14) of the object of measurement (16) is stopped.
16. Method according to claim 1, characterized in that the image
points recorded in an image are recorded separately for
ascertaining the difference in brightness individual fields (51,
52, 53, 54, 55).
17. Method according to claim 16, characterized in that a value of
the maximum change in brightness is recorded for each field (51,
52, 53, 54, 55).
18. Method according to claim 16, characterized in that at least a
division into four fields (51, 52, 53, 54) is chosen and, for the
characterization of a tilting, the coefficient
K.sub.1=(z51'+z53')/(z52'+- z54')-1 and the coefficient
K.sub.2=(z51'+z52')/(z53'+z54')-1 of individual fields are
formed.
19. Method according to claim 16, characterized in that the
coefficients K.sub.1 and K.sub.2 are evaluated in accordance with
the formula (K1).sup.2+(K.sup.2).sup.2<C.
20. Method according to claim 19, characterized in that the
constant is ascertained by measurements of a reference surface.
21. Method according to claim 16, characterized in that the
difference in brightness within each field (51, 52, 53, 54, 55) is
recorded and compared with the neighboring field (51, 52, 53, 54,
55) and in that the angular deviation is converted into coordinates
by which a tilting table receiving the object of measurement (16)
is activated.
22. Method according to claim 16, characterized in that, for
ascertaining the orientation of convexly or concavely curved
surfaces of objects of measurement (16), a central field (55) and
four fields (51, 52, 53, 54) assigned to the side edges or the
corners are provided, the central field being evaluated for setting
the distance (26) and the at least four adjoining fields (51, 52,
53, 54) being evaluated for ascertaining the tilting.
Description
[0001] The invention relates to a method of setting a position of
an object of measurement in layer thickness measurement by X-ray
fluorescence according to the precharacterizing clause of claim
1.
[0002] Setting the correct position of the object of measurement
with respect to the primary X-radiation and with respect to the
detector is crucial for the correctness of the measurement when
measuring thin layers or multiple layers. For such layer analysis,
an X-ray fluorescence radiation of individual elements of a
specimen is detected and converted into layer thickness(es) and
composition(s). Apparatuses which have an X-ray tube in a housing
which is substantially opaque to X-rays are used, with emergence of
an X-ray beam being provided through an opening. The extent of the
X-radiation is restricted to a specific surface area of the
specimen by a collimator. In this surface area, an object of
measurement is positioned at a defined distance from the collimator
on a table which is movable with respect to the measuring head,
comprises an X-ray tube, collimator and the other components
required here. The measuring head has, furthermore, a proportional
counting tube or a detector, which serves for recording the
fluorescence radiation of the irradiated area of the surface.
[0003] The distance between the collimator and the surface of the
object of measurement has to be set to a specific distance for
exact measurement, in order that the fluorescence radiation can be
recorded with sufficient intensity.
[0004] DE 40 03 757 discloses an adjustment in which the collimator
itself is used directly as part of the adjustment. In this case, it
is provided that the tip of the collimator is moved against the
specimen, with the collimator yielding correspondingly on account
of a resilient suspension. Subsequently, further relative movement
between the specimen and apparatus is stopped and the device draws
the collimator away from the specimen again. The distance between
the collimator and the specimen can be set by the amount of spring
deflection of the collimator. This apparatus has the disadvantage
that damage to the surface can occur. In addition, there are
inaccuracies in the setting of the distance on account of
production inaccuracies and the paths to be moved along, with a
cumulative effect of the errors occurring.
[0005] A brochure from the company Veeco Instruments Inc., 1997
edition, likewise discloses an apparatus for measuring thin layers
by X-ray fluorescence analysis. In this case it is provided that a
beam of an optical recording device is projected into the beam of
the X-radiation, in order to be able to view the object of
measurement. In the case of this method, a laser beam is used for
setting the critical distance for the reproducibility of the
measurements. This laser beam falls obliquely on the surface of the
object of measurement. During upward and downward movement of the
object of measurement, the point of impingement of the laser beam
shifts for example from right to left on the surface. Cross wires
are superimposed in the recording device and are adjusted to the
X-ray beam. As soon as the laser of the laser beam projected onto
the surface of the object of measurement coincides with the cross
wires, the exact working distance is set. This upward and downward
movement of the object of measurement in relation to the measuring
head can be performed manually by an operator, with considerable
deviations in the said setting being obtained in these measurement
results.
[0006] Furthermore, this brochure discloses an automatic laser
focusing which is intended to increase the reproducibility of the
exact setting. This automatic laser focusing method of setting the
measuring distance with respect to a surface of an object of
measurement has the disadvantage that the surface impinged is only
indistinctly visible in the case of highly reflective surfaces,
which leads to an inaccurate height setting. The finite size and
unsharpness of the specimen surface impinged by the laser leads to
setting errors. Furthermore, an additional laser and corresponding
shielding are required.
[0007] It is also disadvantageous in the case of both methods
mentioned that tilting of the specimen surface cannot be
recorded.
[0008] The invention is therefore based on the object of providing
a method of setting a point of impingement of an X-radiation on an
object of measurement defined by a distance of a collimator from
the surface of the object of measurement which, on its own, makes
an exact setting of this distance possible.
[0009] This object is achieved according to the invention by the
features of claim 1. The steps provided according to the invention
for carrying out the method allow an automatic setting of the
surface of an object of measurement at a defined distance from the
collimator to take place, with a high degree of accuracy of
reproduction being obtained for the position of the surface of the
object of measurement with respect to the collimator. In addition,
additional sources of error can be eliminated by recording the
changes in brightness of the image points of an image, as is the
case for example with laser focusing with regard to the point of
impingement. Furthermore, the accuracy of reproduction in
comparison with manual focusing can also be significantly improved.
By evaluating the changes in brightness of the image points during
the changing of the distance between the surface of the object of
measurement and the collimator, automatic setting can take place
without additional apparatus. For this purpose, the electronic
recording device, which has a beam projected into the beam of the
X-radiation, is used in order that the exact setting of the
distance between the collimator and the object of measurement is
carried out. By ascertaining the maximum difference in brightness
of the image points of the images recorded, a fixed, defined
distance of the surface of the object of measurement from the
collimator can be set. The beam of the electronic recording device
is advantageously adjusted in such a way that the focal point lies
in a measuring plane which is at the exact distance from the
collimator. When a maximum difference in brightness is ascertained,
it can be ensured that a sharp image is recorded by the recording
device, and that, as a result, the defined distance has been set.
The advantageous assignment of the change in brightness of the
image points in a measuring plane to a Z coordinate makes it
possible that, after ascertaining a maximum difference in
brightness of the image points of an image while moving over the
path, an exact setting of the distance can be carried out by
positioning the surface of the object of measurement and the
collimator with respect to each other.
[0010] The object of the invention is similarly achieved by an
alternative method according to the features of claim 2. The
recording of the differences in brightness of the image points of
at least one measuring plane and the ascertainment of the maximum
take place in analogy with the method according to claim 1. As a
difference from the latter, an assignment of the image in a
measuring plane to a Z coordinate is not envisaged. The maximum
difference in brightness of the image points of an image is
advantageously ascertained and the distance between the surface and
the collimator is changed once more, a change in direction being
envisaged here. During the changing of this distance, the
difference in brightness of the image points of an image in
respective measuring planes in turn approaches the maximum. As soon
as a comparison establishes that the current maximum coincides with
the maximum ascertained when the distance was changed the first
time, the changing of the distance is interrupted, whereby a
focusing of the image and consequently a defined distance of a
collimator from the surface of the object of measurement is
set.
[0011] According to an advantageous refinement of the invention, it
is provided that individual measurements for ascertaining changes
in brightness of the image points of an image are carried out
during a changing of the distance between the collimator and the
surface and the individual measurements are carried out at freely
preselectable intervals in time or virtually continuously. As a
result, the amount of information to be processed, on the one hand,
and the speed of the changes in the absolute amount of a preferably
freely settable preselectable path, on the other hand, can be
determined.
[0012] According to a further advantageous refinement of the
invention, it is provided that, for ascertaining the maximum value
of the difference in brightness in a measuring plane within a path,
the image points y.sub.1 to y.sub.N are determined in a
differential method according to the function
F=.SIGMA.(y.sub.i-y.sub.right neighbour).sup.2+.SIGMA.(y.sub.i-y-
.sub.upper neighbour).sup.2, where y.sub.1 is the brightness value
of the image points used. As a result, the difference in brightness
between a right neighbour and an upper neighbour can be
ascertained, so that the entire information of the image points is
recorded when the difference in brightness is formed. This
ascertained function value is evaluated for the comparison with
other function values ascertained by individual measurements.
[0013] According to an advantageous refinement of the invention, it
is provided that the changing of the distance between the
collimator and the surface of the object of measurement corresponds
to a path in which at least the exact distance between the
collimator and the surface of the object of measurement is passed
through. On account of the advantageous setting of the beam of the
electronic recording device, the focal point of which lies in the
surface of the object of measurement which corresponds to the exact
distance of the collimator from the surface of the object of
measurement, it is made possible that a first unsharpness, for
example above, and a further unsharpness of the focal point, for
example below the surface of the object of measurement, is
obtained, whereby the maximum of the difference in brightness lying
at the focal point can be ascertained with certainty.
[0014] According to a further advantageous refinement of the
invention, it is provided that the individual measurement is
ascertained from a number of individual images at a time interval,
and that an average value is formed from the values of the
individual images. As a result, possible disturbing influences such
as noise on account of divergent values can be minimized.
[0015] According to a further advantageous refinement of the
invention, it is provided that the changing of the distance is
retained during the individual measurement. As a result, changing
takes place without any jerks or jolts, whereby the quality can at
the same time be increased for the recording of the changes in
brightness of the image points. Furthermore, depending on the time
intervals, real-time recording can take place for the individual
measurement.
[0016] According to a further advantageous refinement of the
invention, it is provided that the changing of the distance in the
approximate search for a maximum of the differences in brightness
is carried out at an increased speed. As a result, an approximate
position of the exact distance to be set between the collimator and
the surface of the object of measurement can be ascertained in
first approximation.
[0017] According to a further advantageous refinement of the
method, it is provided that, for a precision search, the distance
between the collimator and the surface of the object of measurement
is reset after carrying out the approximate search to a second
starting point at a resetting speed. This resetting speed is
advantageously greater than the speed of the approximate search, so
that rapid carrying out of the setting is made possible.
[0018] According to a further advantageous refinement of the
method, it is provided that the precision search is carried out at
a reduced speed in comparison with the approximate search. This can
make it possible for the individual measurements for ascertaining
the function value F to be carried out in much closer steps. After
carrying out the precision search, the maximum is ascertained by
calculating the zero crossing of the first derivative as an
approximation by interpolation. Due to possibly image-typical
uncertainties, such as for example noise, several maximums may
formally occur, but are prevented by smoothing.
[0019] According to a further advantageous refinement of the method
according to claim 1, it is provided that, after the precision
search, the maximum of the approximate search and of the precision
search are compared with each other and a path of movement by which
the distance between the collimator and the surface is changed by
during the precision search after passing through the maximum is
calculated. As a result, after the precision search, a direct
setting of the correct distance can be obtained.
[0020] According to a further advantageous refinement of the method
according to claim 1, it is provided that, before the commencement
of the approximate search, a preset distance between the collimator
and the surface of the object of measurement is increased by an
absolute amount. In this way it can be ensured that, in the
subsequent approximate search, a maximum is passed through with a
high degree of certainty, it being observed during the increase in
the distance whether the difference in brightness decreases. As a
result, it can be established at the same time that the starting
point for carrying out a measurement lies below the focal point of
the exact distance in order to permit a reliable setting
thereafter. If the differences in brightness were to increase, the
process would be stopped and an indication given to the user that
another position is being preselected in order to carry out the
setting.
[0021] According to a further advantageous refinement of the method
according to claims 1 and 2, it is provided that the image points
ascertained for recording the difference in brightness in an image
are recorded separately in individual fields. This makes it
possible for the orientation of the surface of the object of
measurement to be ascertained by comparison of the individual
fields with one another. The positionally correct orientation is of
significance in particular in multiple layer measurements and in
the measurement of very thick layers. Recording the orientation of
the specimen surface makes it possible to compensate for
inaccuracies from an ideal orthogonal orientation of the measuring
plane with respect to the X-ray beam.
[0022] For this, it is advantageously provided that the value of
the maximum change in brightness is recorded in every field. As a
result, a comparison between the individual fields can be made
possible. If, for example, two fields neighbouring each other have
the same change in brightness, it can be concluded from this that
this area has no difference in height. If a number of fields have
an approximately equal value of a change in brightness, it is
ascertained that the planar surface of the object of measurement
has a positionally correct orientation, in other words is
positioned orthogonally with respect to the X-ray beam.
[0023] According to a further advantageous refinement of the
invention, it is provided that at least a division into four fields
is chosen and, for the characterization of a tilting, the
coefficient from a right-hand pair of individual fields and a
left-hand pair of individual fields and the coefficient from an
upper pair and a lower pair of individual fields are formed. This
characterizes the tilting or orientation of the surface of the
object of measurement. It is advantageously provided that the sum
of the squared coefficients is compared with a constant which is a
measure of the orthogonality of the surface with respect to the
X-ray beam. Depending on the constant, the tolerance can be
pre-formed such that it is greater or smaller.
[0024] According to a further advantageous refinement of the
invention, it is provided that the differences in brightness within
each field are recorded and compared with the neighbouring fields
and the orientation is ascertained, a table with an inclination
correction being activated in an XY plane with respect to the
collimator. This can make possible an adjustment of the orientation
of the surface deviating from the ideal plane with respect to the
X-ray beam.
[0025] Further advantageous embodiments are specified in the
further claims.
[0026] Particularly preferred embodiments of the method are
described in more detail with reference to the following drawings,
in which:
[0027] FIG. 1 shows a schematic view of an apparatus for measuring
thin layers by X-radiation,
[0028] FIG. 2 shows a schematic view of a beam of an electronic
display device, the focal point of which lies in a surface of an
object of measurement,
[0029] FIG. 3 shows a schematic representation according to FIG. 2,
in which the focal point lies above the surface,
[0030] FIG. 4 shows a schematic representation according to FIG. 2,
in which the focal point lies below the surface,
[0031] FIG. 5 shows a schematic representation of successive method
steps for setting the distance between the collimator and the
surface of the object of measurement,
[0032] FIG. 6 shows a schematic representation of an alternative
sequence of method steps according to FIG. 5
[0033] FIG. 7 shows a schematic representation of the sequence of a
further alternative method of setting a point of impingement of an
X-radiation on an object of measurement,
[0034] FIGS. 8 and 9 show a schematic representation of an image of
an individual measurement with a measuring field divided into, for
example, four individual fields, for ascertaining the orientation,
and
[0035] FIGS. 10 and 11 show a schematic representation of an
alternative arrangement of individual fields within an image for
ascertaining the orientation in the case of convex or concave
surfaces.
[0036] Represented in FIG. 1 is an apparatus 11 for measuring thin
layers or for layer thickness analysis by X-radiation, in
particular by X-ray fluorescence radiation. Such an apparatus 11
has an X-ray tube 12 for generating X-rays in a housing 13. An
X-ray beam leaves via an opening in the housing 13 and impinges on
a surface 14 of an object of measurement 16. A specific surface
area of the X-radiation is restricted on the surface 14 of the
object of measurement 16 by a collimator 17 arranged at a defined
distance 26 from the surface 14. The fluorescence radiation emitted
by the irradiated specimen is recorded and evaluated by a
proportional counting tube 18 or some other detector.
[0037] The apparatus 11 has an electronic display device 21, the
beam 22 of which is projected into the beam of the X-radiation via
a semi-transparent mirror 23 and is directed onto a surface 14 of
the object of measurement 16. This display device 21 allows the
image of the surface 14 to be reproduced on a monitor (not
represented in any more detail).
[0038] For the accuracy of the layer thickness measurement it is
necessary that an exact distance 26 between the surface 14 of the
object of measurement 16 and the collimator 17 is set, determining
the point of impingement of an X-radiation on an object. This
distance 26 is fixed on one occasion on an apparatus. It is
subsequently necessary for this distance 26 to be repeatedly set
exactly. One reason for this is that a specific position of the
proportional counting tube 18 for recording the secondary radiation
being emitted is necessary in order to record a minimum intensity
of the radiation. The components, such as for example the tube 12,
housing 13, collimator 17, proportional counting tube 18, form a
measuring head 27. The object of measurement 16 is arranged with
respect to the said measuring head on a table 28 which is movable
in three dimensions. The following embodiments are described on the
basis of a fixed measuring head 27 and a movable table 28, in
particular in the Z coordinate, in other words reducing or
increasing the distance 26. It goes without saying that the table
28 may similarly be fixedly arranged and the measuring head 27
variable with respect to it, or that both the measuring head 27 and
the table 28 may be moved with respect to each other, or that part
of the movement of the table 28, part of the movement of the
measuring head 27 or some other variable pattern of movements may
be envisaged.
[0039] The beam 22 from the electrical display device 21 to the
surface 14 is represented in FIG. 2. In this setting, a focal point
31 lies in a measuring plane 32, which in this position corresponds
to the surface 14. The display device 21 ascertains a sharp image
of the surface 14. In this position of the surface 14, an exact
position of the distance 26 between the collimator 17 and the
surface 14 is also obtained. Consequently, the exact working
distance 26 with the greatest possible focusing is obtained.
[0040] The image recorded by the display device 21 is read out in
individual pixels. This may take place for example by means of a
CCD camera chip, it being possible for signals digitized via a
Frame-Crapper card to be transferred to a graphics card, which can
take place without processor support.
[0041] Depending on the position of the surface 14, the measuring
plane 32 formed by the focal point 31 may lie above the surface 14,
as is represented for example in FIG. 3, or below the surface 14,
as is represented in FIG. 4. The further the measuring plane 32 is
away from the surface 14, the greater the unsharpness becomes and
the smaller the differences in brightness between the individual
image points y.sub.N become, N being the number of image points
which are read out for the evaluation, still to be explained below,
in a measuring field 36. The further the measuring plane 32 is away
from the surface 14, the unsharper the recorded image becomes and
the smaller the differences in brightness between the respectively
neighbouring image points become. Seen from the converse viewpoint,
this means that, when the measuring plane 32 is arranged in the
surface 14, the differences in brightness are at the greatest and
this maximum stands on the one hand for the focusing of the image
and on the other hand, by the correlation with the position of the
surface 14, for an exact distance 26.
[0042] Consequently, all the image points are read out and a
function value F is determined in accordance with the following
equation: F=.SIGMA.(y.sub.1-y.sub.right
neighbour).sup.2+.SIGMA.(y.sub.1-y.sub.uppe- r neighbour).sup.2,
where y.sub.i is the brightness value of an image point. This is
compared for example with a right neighbour on the one hand and an
upper neighbour on the other hand. It may similarly be provided
that, instead of the right neighbour, the left neighbour is chosen
and, instead of the upper neighbour, the lower neighbour is chosen.
Consequently, the sum of the differences in brightness can be
ascertained by the function value F.
[0043] In a first exemplary embodiment of the method, it is
provided that a position of the surface 14 with respect to the Z
axis is recorded. It follows from this that the function value F is
changing, from which it follows that F=f(Z). This results in a
first embodiment for carrying out the method of setting a point of
impingement of an X-radiation on an object of measurement. In the
case of this embodiment, the Z coordinate along which the X-ray
beam runs is taken into consideration as a further characteristic
variable. The method of recording the point of impingement of an
X-radiation on an object of measurement with a defined distance 26
between the collimator 17 and the surface 14 can take place in the
following way:
[0044] The table 28 with the object of measurement 16 is
transferred into a position in which the measuring plane 32 lies
above the surface 14 of the object of measurement 16. This starting
position 41 is represented in FIG. 5. Then the table 28 is moved
along the Z coordinate towards the collimator 17, until a first
intermediate position 42 is reached, which corresponds for example
to a position according to FIG. 4. The absolute amount of the path
is freely selectable. However, it has a minimum path of movement,
in order that the exact distance 26 between the collimator 17 and
the surface 14 is passed through with certainty. The speed of
movement may take place relatively quickly in a first method step,
which takes the form of an approximate search. Between the starting
position 41 and the first intermediate position 42, individual
measurements are advantageously carried out continuously, the
individual measurement being ascertained from, for example, two or
more individual values within a defined interval, so that the
individual measurement comprises an average value of a number of
individual values. These individual measurements of the images are
evaluated in a way corresponding to the function F. After running
through the approximate search, a maximum is ascertained by
calculating the zero crossing of the first derivative as an
approximation by interpolation. This first maximum is stored.
Subsequently, the table 28 is transferred into a second
intermediate position 43, from where a precision search takes place
up to the third intermediate position 44. The travelling speed of
the table 28 for the precision search is much slower than in the
case of the approximate search. Individual measurements of the
images are in turn carried out and evaluated in a way according to
the method of the approximate search. Furthermore, the maximum is
again ascertained. On the basis of the recording of the Z
coordinate, the position 44 of the table is then known. Similarly,
the Z coordinate of the maximum of the precision search is known
and is advantageously compared with the approximate search.
Subsequently, the table 28 is transferred from the second position
44 directly into a position 45, whereby the exact distance 26
between the surface 14 of the object of measurement 16 and the
collimator 17 has been set.
[0045] The advantage of this procedure, in which the approximate
search and the precision search have an identical direction of
movement of the table 28, is a higher degree of accuracy of the
ascertainment of the maximum and consequently of the exact distance
26. Similarly, the approximation to the maximum can likewise have
the same direction of movement as that of the approximate search
and precision search.
[0046] A further alternative way of carrying out the method is
represented in FIG. 6. Before the commencement of the measurement,
a first path of movement between a starting point 40 and the
starting position 41 is covered. By reducing the difference in
brightness of the image points between the points 40 and 41 on
account of the unsharpness becoming greater, it is ensured that the
table 28 is moved away from the collimator 17. In this way, it can
at the same time be ensured that the table 28 does not run against
the collimator 17, as long as the absolute amount of the
unsharpness becomes greater. Following this, the method steps with
respect to FIG. 5 can be carried out.
[0047] A further alternative refinement of the method of setting a
point of impingement of an X-radiation to a specific distance 26
between the collimator 17 and the surface 14 of the object of
measurement 16 is described in more detail below and explained by
way of example on the basis of FIG. 7:
[0048] The approximate search and precision search, as described in
FIGS. 5 and 6, and also a reverse movement from a first
intermediate position 42 into a second intermediate position 43
take place by analogy with the embodiment of the method according
to FIGS. 5 and 6. The present method takes place without the
assignment of the function value F to a Z coordinate. After the
approximate search between the starting position 41 and the first
intermediate position 42, a rapid resetting takes place into the
second intermediate position 43. Subsequently, a precision search
takes place; during this, the function value F with the maximum of
the function value F is compared with that of the approximate
search. Once the maximum of the function value F is reached during
the precision search, which corresponds substantially to the
approximate search, the precision search is then stopped. As a
result, the surface 14 has again been positioned with respect to
the collimator 17 at the distance 26. Stopping of the table 28 may
also take place if the maximum is slightly exceeded, in order to
ensure that no unsharpnesses are interpreted as a maximum.
[0049] A further alternative of FIGS. 5 and 6 may consist in that
the approximate search begins at the first intermediate position 42
and is carried out up to the second position 43. Subsequently, the
precision search takes place up to the intermediate position 44 and
the positioning to point 45.
[0050] All the abovementioned embodiments of the method according
to FIGS. 5 to 7 share the common feature that the approximate
search and/or precision search can be repeated one or more times,
it also being possible for the speed of movement and number of
measurements during an approximate search and a precision search to
be varied. The more often a search for the maximum is carried out,
the more exactly the distance 26 from the collimator 17 can be set.
Depending on the accuracy requirements, consequently one or more
method steps can be strung together in order to increase the
accuracy still further, with an increase in the time taken before
the surface 14 of the object of measurement 16 is positioned in a
final position with the exact distance 26 from the collimator
17.
[0051] For the evaluation of the image points of an image, an area
36 which is at least the size of the point of impingement of an
X-radiation is advantageously chosen. The size of the image may
optionally be set. Similarly, this evaluation of the optional image
points may take place if the surface has been zoomed or made
visible in an enlarged projection on a monitor.
[0052] The more differentiated the image areas are chosen to be for
the evaluation of the difference in brightness of the image points,
the more certainly setting can be made to the correct distance
26.
[0053] Furthermore, it may advantageously be provided that the
abovementioned method of ascertaining a maximum with respect to the
difference in brightness of the image points in an area is
developed in such a way that an orientation of the surface 14 is
recorded. For this purpose, it is advantageously provided that the
area 36 is subdivided into four individual fields 51, 52, 53, 54,
for example according to FIG. 8, the maximum of the difference in
brightness for each field being ascertained separately.
[0054] An optimum alignment of the surface 14 is obtained when the
measuring plane 32 of the surface 14 is aligned orthogonally with
respect to the Z coordinate. A Z value is assigned for each field
by a comparison, of for example fields 51, 52, 53, 54 according to
FIG. 8, or their maximum, by the first derivative. If these values
are the same within a certain error tolerance, it is concluded that
the orientation is virtually ideal, i.e. perpendicular with respect
to the optical axis of the viewing optics of the display device 21
or with respect to the X-ray beam. This can be monitored for
example by standard deviations of the Z values.
[0055] The characterization of the tilting or ascertainment of the
orientation with regard to the deviation from the ideal measuring
plane may take place for example by a coefficient K.sub.1 being
formed between the right and left fields 51, 53, 52, 54 in
accordance with the equation K.sub.1=(z51'+z53')/(z52'+z54')-1, and
a coefficient K.sub.2 between the upper and lower fields 51, 52,
53, 54, whereby the equation K.sub.2=(z5l'+z52')/(z53'+z54')-1 is
formed. The tilting of the surface of the object of measurement can
be ascertained by the test in accordance with
(K1).sup.2+(K2).sup.2<C. A constant C is provided here,
ascertained empirically in advance for an ideal plane and an
aligned specimen in accordance with the formula
X=(K1).sup.2+(K2).sup.2. This may comprise series of measurements,
for example 5, 10, 15 or 20 measurements. The constant C is then
intended to be, for example, three times the average value of X.
Once the conditions are satisfied, the focusing to an average value
(z51'+z52'+z53'+z54')/4 can take place.
[0056] Alternatively, it may be provided that, by reading out the
individual image values within the fields 51, 52, 53, 54, the
degree of tilting is ascertained in order thereafter to adjust a
tilting table which has, for example, two degrees of freedom of
movement, so that the surface can be positioned orthogonally with
respect to the Z coordinate.
[0057] An alternative embodiment to FIG. 8 is represented in FIG.
9. Similarly, a number of fields 51, 52, 53, 54 may be provided in
rows and columns, in order to record larger planar measuring fields
36 with respect to their tilting by an individual measurement.
[0058] Further alternative arrangements of fields 51, 52, 53, 54,
55 for the reading out of an image of an individual measurement are
represented in FIGS. 10 and 11. These arrangements are
advantageously provided for the recording of convexly or concavely
curved surfaces. A focusing of the image advantageously takes place
on the basis of the derivative of the z55 value of the central
field 55. The tilting can be monitored by analogy with the fields
51, 52, 53, 54 described in FIGS. 8 and 9. Further combinations and
arrangements of the fields for ascertaining the orientation and the
shape of surfaces can similarly be provided.
* * * * *