Note: Descriptions are shown in the official language in which they were submitted.
383
-1- RCA 73, 558
AUTOMATIC STRIPE WIDTH READER
Back~round of the Invention
This invention relates to measurement of the
S widt~ of stripes and spacin~ between stripes in a regular
periodic pattern, and particularly ~o a reader for
performing such measurements.
Although the present invention may be used
to read the widths of many different types of regular
periodic patterns, it hereinafter will be described
with respect to reading the widthsof opaque, light
absorbing lines on a color picture tube faceplate panel
of the matrix type before application of phosphor
elements of a viewing screen.
-Color picture tubes of the line screen matrix
type have been commercially available for several years.
The screens of such tubes comprise alternating lines
of red, green and blue light-emitting phosphors, each
separated from the other by light absorbing stripes
called the matrix. In forming the tube screen, the
matrix is applied first to the inner surface of a
tube faceplate panel and then the phosphor lines are
applied. The matrix and the phosphor lines are formed
in a photographic process which uses the shadow mask
of the tube as a photomaster. Each color-emitting set
- of phosphor eIements requires a different light source
.~ .
- ~ location to ensure placement of the elements at
locations that will be struck by electrons from an
associated electron gun. Since the matrix is applied
before the phosphor elements are applied, formatLon
of the matrix requires~three separate exposures to
~;~ ensure that the holes in the matrix for the phosphors
are in the proper locations O Because of this multiexposure
method, the width of the matrix stripes and the spacing
between stripes may vary from the ideal width and
spacing desired. Since light output and color purity
are at least partially dependent on the matrix stripe
i width and spacing, it is advantageous to determine
this stripe width and spacing prior to completion of
a screen.
,. .
; ', -
~73~3
1 -2- RCA 73,558
Summary of the Invention
Apparatus is provided for determining the
widthsand spacings of substan~ially parallel opaque
stripes on a substrate, wherein the stripes are
separated by openings. The apparatus comprises a
light source for illuminating the substrate and
photodetecting means positioned to receive light
from the i huminated substrate. Means are included
for scanning the photodetecting means transverse to
the stripes and for sweeping the photodetecting means
in a direction substantially parallel to the stripes.
~he system further includes means ~or converting the
outpu~ of the photodetecting means into a quantized
signal, means for dividing the quantized signal into
separate signals representing stripe widths and
opening widths between stripes, and means for converting
the stripe width and opening width signals into a
signal representing center-to-center spacing between
openings~
.~ .
Brief Description of the Drawings
FIGURE 1 (Sheet 1) is a side view of the aeneral
physical structure of a stripe width reader.
FIGURE 2 (Sheet 1) is a Partial ~lan view ~f a t~e
faceplate having an ideal matrix pattern thereon.
.
~ 30
~ - '
. ~ .,
.
:
~7383
-1 -3- RCA 73,558
FIGURE 3 (Sheet 1) is a partial plan view of a tube
faceplate having an actual matri~ pattern thereon.
FIGURE 4 (Sheet 2) is a circuit diagram for calcu-
5 lating spacing distances Sn from inputs related to stripewidth An and the spacings bet~een stripes Bn.
FIGURE 5 (Sheet 2) is a small portion of a tube
~aceplate panel having a stripe pattern thereon with an over-
lay showing the area of the panel to be measured.
FIGURES 6 and 7 (Sheet 3) are a front view and a
cutaway side view, respectively, of a camera unit.
FIGURE 8 (Sheet 4) is a circuit diagram showing
the interconnection of circuits in the present stripe width
reader.
FIGURE 9 ~Sheet 2) shows waveforms of quantized
signals relat~d to stripe widths and stripe spacings.
FIGURE 10 (Sheet 5) is a diagram of a data process-
ing circuit.
Detailed Description
FIGURE 1 illustrates the general physical arrange-
ment of a stripe width reader 10. Various reader components
located on a chair-llke stand 12 having a table section 14 and
a vertical extension 16. A faceplate panel 18 is shown posi-
tioned on the table section 14 on a sliding tray 20. The
25 panel 18 is held on the tray 20 by movable jaws, not shown.
The tray 20 may be clamped in positions which determine the
area of the panel I8 to be measured. A camera unit 22 is
suspended directly abo~e the panel 18 by a cable 24 which
passes through pulleys on an arm 26 which is attached to the
30 vertical extension 16. Vertical position of the camera unit
22 is controlled by a hydraulic mechanism 28 acting on the
cable 24. For a stripe width reading, the hydraulic mechan-
ism 28 lowers the camera unit 22 until it touches the sur~ace
of the panel 18. Camera control electronics 30 are located at
35 the top of the vertical extension 16 and are connected to the
camera unit 22 by electrical leads 32.
The panel 18 is illuminat-ed from below by an
incandescent light source 34. Light from the source 34 passes
through a Fresnel lens 36 which acts as a condenser. This
40 lens 36 is large enough to accommodabe the camera field of
-view taking into account the variable angle at which the camera
- may rest on the panel. A filter 38 also is included on the
~73~3
1 -4- RCA 73,558
lens 36 to cut out any infrared component which would cause a
deterioration of image resolution.
The sensitive element in the camera unit 22 is a 172
5element line-scan photodiode array with elements spaced 15~m a-
part.The matrix image is magnified about 7 times so the array
scans a line about 3.556 mm long. The image is
deflected across the array, in a direction perpendicular to the
array direction, by a rotatable mirror. The effective distance
ept on the matrix during measurement is about 2.54
mm . Width measurements made on the video signal during the
sweep time are agyregated to form an average width value as
described below.
FI~URE 2 shows a portion of a faceplate panel 40
15having ideally spaced and shaped matrix stripes 42 thereon.
Matrix stripe width is designated Ao~ the spacing between
stripes 42 is designated Bo and the repeat distance, which is
the same as Ao plus Bo~ is designated SO. Unfortunately, the
ideal is never achieved; rather,the matrix stripe pattern
20varies both in spacing and in stripe width as shown in the
actual embodiment of FIGURE 3. In this embodiment, the average
widths of the matrix stripes 44 on the faceplate panel 46 are
designated Al, A2 and A3, the average spacings between stripes
are designated Bl, B2 and B3 and the average matrix spacings
2~easured from the center of one open space to the next are
designated Sl, S2 and S3.
A major source of these stripe width and stripe
spacing variations is error in the setting or exposure of the
three separate light-houses which are used to form each of the
3~hree aperture fields in the matrix of stripes. Details of a
method of making a cathode-ray tube screen having a matrix
background of light-absorbing areas may be found in U.S. Patent
3,558,310 issued to E. E. Mayaud on January 26, 1971. Since
the variations in each set of stripes formed with the same
3~ighthouse are similar (each lighthouse forming every third
stripe), it is useful to group together measurements of the
widths of openings in a given field even though these measure-
ments are taken on different stripes. The method, therefore,
is to aggregate aIl measurements o~ Al, all of Bl, etc., taken
4~uring the camera sweep. This aggregate is then scaled to give
:,~
~72~83
1 -5- RCA 73,558
a read-out in volts corresponding to the average width, e.g.,
in mm.
The parameter of display chosen is not the average
5 matrix stripe spacing B but the avexage repeat distance S,
which can be more directly interpreted in terms of lighthouse
maladjustment. Derivation of S values from A and B is
accomplished by the circuit shown in FIGURE 4, wherein inputs
A2, Bl and B2 are used to obtain output S2. In addition, the
10 width integrators are initialized at the (negative) desired
or target value, so that the final values obtained represent
deviations from the target values. Target values are appro-
priately preset into the circuitry.
FIGYRE 5 shows a small portion 50 of a tube face-
' 15 plate panel having a light absorbing matrix stripe pattern 52,but no phosphor line,s, thereon. The smaller shaded area 54
represents the area to be measured by the stripe width reader.
; In the figure, the scan direction of the photodiode array is
left to right and the sweep direction of the camera is from
20 top to bottom. Each scan begins with a pause before informa-
tion is gathered. Following the pause, information is not
used until the leading edge 56 of the second stripe within the
measured area i5 reached. Information to be processed is
obtained during the scan over the next six stripes, and the en-
5 able cycle ends at the leading edge 58 of the seventh stripe.
FIGURES 6 and 7 show the details of the camera unit
.
22. Three seating pads 60 extend from a base 62 of the unit22 to contact the surface of a aceplate. Principal elements
of the unit include a lens 64 for focusing the light from the
30 light source 34 and a focus motor 66 for moving the lens 64
-to its focused location. Light passing through the lens 64
is reflected by two mirrors 68 and 70 onto a detector array
- 74. The first of the two mirrors 68 is attached to a sweep
mirror drive 76 which rotates the mirror 68 thereby moving
35 the line viewed by the scanning array in the sweep direction
shown in FIGU~E 5.
The electrical connections between the various
components of the novel stripe width reader are shown in
FIGURE 8. The camera unit 22 is controlled by the camera
40 control circuit B2. The first function of this circuit 82 is
to focus the lens 64 after the camera unit 22 has been placed
-
7383
1 -6- RCA 73,55~
on a faceplate panel. The focus motor 66 is provided with two
speeds. One is a fast focus speed which continues until the
lens 64 is nearly in focus. The other is a slower speed used
5 to bring the lens into final focus. To perform this focusing,
a video processing circuit 84 generates a signal from the
detector array 74 output which is proportionaltothe sharpness
of the video signal. This signal then is fed to the camera
control circuit 82.
Once the camera unit 22 is focused,thedetectorarray
74 scans the faceplate panel and the output is senttoa data
processing circuit 86 via the video processing circuit 84.
During these scans, a mirror deflection power supply 88 feeds
a constantly increasing current to the sweep mirror drive 76
lS causing the mirror 68 to rotate continuously so that each scan
is slightly displaced from the previous one.This procedure
continues until information is obtained from the entire
measured area 54 previously sho~n in FIGURE 5.
The data processing circuit 86 provides six outputs
20 sequenced in pairs,Snand Bn,to two meters 90 and 92. The meters
90 and 92 contain analog-~o-digital converters, the outputs
from which are fed to a printer 94. Two digits and a sign
for each output give sufficiently accurate indication of the
deviation from the desired or target values of S and B. A
25 third meter 96 indicates target values of Bo amd S~. Printer
operation i5 sequenced by a printer control circuit 98.
The primary function of the video processing circuit
84 is to take the charge pulses from the detector array 74 and
convert them into a quantized signal. The top waveform 100 of
30 FIGURE 9 shows this quantized signal.The next functicnperformed
by the video processing circuit8~istDgate the combinedwaveform
toobtainquantizedsignals ~or the various stripes,Al,A2and A3
and openings Bl,B2 and B3,as shown by the six lower waveforms
of FIGURE 9. These individual waveforms are used as inputs to
35 the data processing circuit 86 shown in detail in FIGURE 10.
The data processing circuit 86 of FIGURE 10 comprises
four basic parts: a switch section, an integrator section, a
reference section, and a spacing calculation section. The
- switch section includes an AND gate 102, a counter 104 and a
40 NOR gate 106 which provide one of the inputs to each of six
~ 173~3
1 -7- RCA 73,558
AND gates 108, 110, 112, 114, 116 and 118. The other inputs
to these six AND gates are the stripe width waveforms Al, A2
and A3 and the opening wavefo~ms Bl, B2 and B3 from the video
5 processing circuit 84. Individual pulse trains of each
waveform next are summed in the integrator section comprising
six inte~rators 120, 122, 124, 126, 128 and 130 including
their associated capacitors and resistors. The aiming voltage
of the integrators is arranged so that a pulse length
10 corresponding to a specific stripe width (e.g., 0.25mm) gives
a specific change on the integrator te.g., 1 volt).
Individual variations in the capacitors are compensated by
trimming resistors. Outputs from the integrators 120, 122
and 124 are averages ~1~ A2 and A3 of the stripe width values,
15 and outputsfrom the integrators 126, 128 and 130 are averages
Bl, B2 and B3 of the opening widths. Alternatively, as shown,
the outputs may be preset to the desired values of stripe
width Ao and opening width Bo. Thus, after summing, the
outputs of the integrators will be the deviations of the
20 measured average values from the deslred or target values,
which is a more convenient form of output. The target values
~re supplied by the reference section comprising the t~;o ampli~
fiers 132 and 134. Finally, the values of A - Ao and B - Bo
are combined in the spacing calculation section comprising
25 the three networks 136, 138 and 140, essentially as previously
described with respèct to FIGURE 4, to give S - SO values on
the outputs from the networks. The outputs ~rom the three
networks 136, 138 and 140 are thereafter sequentially ap~lied
to the meter 90.
4~
