U.S. patent number 3,640,625 [Application Number 04/735,336] was granted by the patent office on 1972-02-08 for multiplex spectrometer.
This patent grant is currently assigned to National Research Development Corporation. Invention is credited to David Aspinall, Roland Norman Ibbett.
United States Patent |
3,640,625 |
Ibbett , et al. |
February 8, 1972 |
MULTIPLEX SPECTROMETER
Abstract
A multiplex spectrometer has its spectral line information
encoded by transmitting the line through rows of apertures
presented in sequence. The apertures occur in a pattern derived
from a matrix of binary numbers. Decoding is achieved by adding or
subtracting the output in a store containing addresses equal to the
number of columns in the matrix.
Inventors: |
Ibbett; Roland Norman (Cheadle,
EN), Aspinall; David (St. Anne's-on-Sea,
EN) |
Assignee: |
National Research Development
Corporation (London, EN)
|
Family
ID: |
10246185 |
Appl.
No.: |
04/735,336 |
Filed: |
June 7, 1968 |
Foreign Application Priority Data
|
|
|
|
|
Jun 8, 1967 [GB] |
|
|
26,601/67 |
|
Current U.S.
Class: |
356/310; 356/326;
250/233 |
Current CPC
Class: |
G01J
3/2846 (20130101) |
Current International
Class: |
G01J
3/28 (20060101); G01j 003/00 (); G01j 003/02 () |
Field of
Search: |
;356/81,86,87,51,96-101
;250/237,233,226,232 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Primary Examiner: Wibert; Ronald L.
Assistant Examiner: Evans; F. L.
Claims
We claim:
1. Spectrometer apparatus comprising:
means for producing a spectrum,
a single photodetector for producing an electrical output provides
in response to radiation incident thereon,
means for focusing said spectrum onto said photodetector,
a movable coding matrix device positioned between said spectrum
producing means and said focusing means for encoding said spectrum
by defining a plurality of channels spaced along said spectrum and
alternately passing and blocking transmission of said spectrum
through each of said plurality of channels in accordance with
respective time varying coding patterns for each channel,
a plurality of storage registers, one for each channel, coupled to
receive the output of said photodetector,
means coupled between said photodetector and said registers for
either accumulatively adding or subtracting each of the successive
outputs from the photodetector in each register in an individual
sequence based on a received decoding signal in such a manner that
after the sequence for each register is completed the information
from the photodetector not derived from that channel which
corresponds to that particular register cancels and the information
derived from that channel which corresponds to that particular
register only remains and accordingly each register contains
information derived solely from its corresponding channel, and
decoding means synchronized with said coding matrix device for
producing a decoding signal determining the sequence in which the
outputs are added and subtracted for each register so that for each
of the other channels the number of outputs added equals the number
of outputs subtracted and applying said determining signal to the
adding or subtracting means.
2. Spectrometer apparatus as in claim 1 in which the coding device
carries a plurality of apertures at selected points in a matrix of
rows and columns.
3. Spectrometer apparatus as in claim 2 in which the coding device
comprises a disc having apertures in the matrix the rows of which
are defined by radial lines and the columns of which are defined by
rings of different radii.
4. Spectrometer apparatus as in claim 2 in which the number trial
apertures in every row of the matrix is constant.
5. Spectrometer apparatus as in claim 2 in which the coding device
has the matrix repeated twice and one of the matrices is used for
controlling said adding and subtracting means.
Description
This invention relates to spectrometers. It is an object of the
invention to provide a spectrometer with the capability of
multiplex operation, by which all the spectral information is
encoded and passed through a single detector and then decoded.
Accordingly, the present invention comprises a spectrometer in
which the spectral line output is focused on to a single
photodetector through a coding device which alternately passes and
blocks transmission through a plurality of channels spaced along
the spectral line in different time-varying coding patterns in each
channel and in which the output of the detector is applied to a
plurality of storage registers corresponding in number to the
number of channels, the successive inputs to each channel being
cumulatively added or subtracted in an individual sequence in each
channel based on the coding patterns of the coding device in such a
manner that each register finally contains information derived
solely from the corresponding channel.
The coding device may comprise a disc of a radius at least equal to
the length of the spectral line of the spectrometer and carrying a
plurality of apertures at positions in a matrix the rows of which
are defined by radial lines and the columns of which are defined by
rings of different radii. Apertures are provided at selected points
of the matrix so formed in a pattern that will be more fully
described below. Alternatively, in place of a disc, a drum may be
used or else a plurality of fixed apertures controlled by shutters
may be provided which are successively opened and closed in
time-varying patterns.
The requirement for the coding is that the coding of each column of
the matrix is mutually orthogonal and preferably the ratio open
times to closed times in each column should be the same. It is also
advantageous if the number of apertures in every row is constant,
since this minimizes the dynamic range of the signal at the
detector. In fact the dynamic range of the output signal is then
determined only by the dynamic range of the spectrum itself.
A set of codes possessing the required properties can be obtained
for 2.sup.n -1 channels by performing a matrix multiplication on
the matrix formed by writing the binary numbers 1 to 2.sup.n and
its transpose, and multiplying these together with modulo 2
addition of the subproducts which form an element. In the resulting
matrix an "0" indicates a closed aperture and a "1" indicates an
open aperture. Each column of the matrix gives the respective code
for a channel and the number of channels contributing to the signal
at the detector at any one instant is determined by the number of
"1's" in a row.
It is possible to provide a code for situations where the number of
channels is not equal to 2.sup.n -1. In such cases columns, but not
rows, of the next higher matrix can be eliminated without affecting
the uniqueness of the decoding. However, the number of apertures in
every row is no longer constant.
The decoding operation that is required in the respective storage
registers is obtained from the matrix by adding the output of the
detector to a storage register when the aperture of the
corresponding channel is open (i.e., has a value "1") and
subtracting the output of the detector from a storage register when
the aperture of the corresponding channel is closed (i.e., has the
value "0"). It can be shown that this computation will always give
the required result that the cumulative total standing in any
register is derived solely from the corresponding channel.
Where the coding device comprises a disc, control of addition to or
subtraction from the storage registers may be obtained by encoding
the desired patterns in half the disc and then duplicating the
patterns in the other half. A row of photocells positioned along a
radial line then reads the coded patterns as the disc rotates. Each
photocell controls a corresponding storage register so as to cause
addition when the cell is energized and subtraction when the cell
is not energized.
In order that the invention may be more fully understood reference
will now be made to the accompanying drawing in which FIG. 1
illustrates an embodiment thereof, and FIG. 2 shows a modification
and FIG. 3 illustrates in frontal view a suitable apertured
disc.
In FIG. 1 a spectrometer 1 provides a line spectrum, which may be
in any waveband, and its spectrum is focused by means of a lens
system 2 to a single photodetector 3. The output of detector 3 is
amplified in an amplifier 4 and then applied to a sampling circuit
5. The output of sampling circuit 5 is applied to an
analogue-to-digital converter 6 and thence as one input to an
adder/subtractor unit 7. The other input of the unit 7 is obtained
from a register in a bank of registers 8 and the output of the unit
is returned to the same register in the bank 8.
A bank 9 of photocells energized from a lamp 10 is used to control
unit 7 to determine whether at any instant the output of converter
6 is to be added in unit 7 or subtracted therein. The photocells
are selected in sequence through a gating array 12. An additional
photocell 11 provides reference pulses to control the operation of
sampling circuit 5. These reference pulses can also be used to
provide a timing pulse train which selects each register of the
bank of registers 8 in sequence and in synchronism therewith
operates gating array 12 to ensure that the appropriate photocell
of bank 9 controls unit 7 at the time when the output of a
corresponding register from bank 8 is being fed thereto.
The line spectrum from spectrometer 1 is encoded by means of a disc
13 which is interposed between spectrometer 1 and the lines system
2 and between lamp 10 and photocells 9. Disc 13 carries patterns of
apertures arranged in a matrix with the rows lying along radial
lines and the columns on circumferential rings. For a seven column
matrix with the columns labeled A-G inclusive, a suitable pattern
of codes is as follows:
A B C D E F G
__________________________________________________________________________
1 0 1 0 1 0 1 0 1 1 0 0 1 1 1 1 0 0 1 1 0 0 0 0 1 1 1 1 1 0 1 1 0 1
0 0 1 1 1 1 0 0 1 1 0 1 0 0 1
__________________________________________________________________________
in the above encoding pattern a "1" indicates the presence of an
aperture and an "0" the absence of an aperture. It will be seen
that every column has four "1's" or transmitting sections and
similarly every row has four "1's." Thus, at any instant detector 3
receives the output of four channels and each channel contributes
four samples. The above code is encoded in half of the disc and a
similar code is repeated for the other half of the disc. The second
code serves to appropriately energize the bank 9 of seven
photocells to control the decoding process. In addition an eighth
column of apertures is provided for enabling the reference pulses
to be produced.
FIG. 3 illustrates in frontal view the apertured disc 13. The
apertures are arranged in a matrix of rows and columns with the
rows lying along radial lines and the columns on circumferential
rings. The columns are labeled A to G inclusive and the pattern of
apertures is as set forth in the table above. In addition disc 13
has an inner ring of apertures provided in each row for the purpose
of energizing the photocell 11 providing the reference pulses. As
mentioned above, the entire pattern is repeated twice to enable
encoding and decoding to take place simultaneously.
It can be shown that if the output from the detector, to which all
of a row contributes at any instant is added to a particular
register when the associated channel contributes to the output at
that instant and is subtracted from the associated register when
the associated channel does not contribute to the output, then the
resulting content of a register when the disc has rotated half a
revolution through all the rows of the matrix contains information
derived only from the associated channel. As an example channel A
and channel B is decoded below writing a plus sign (+) for addition
and a minus sign (-) for subtraction: ##SPC1##
Thus, the register relating to channel A contains contributions
from channel A only and similarly the register associated with
channel B contains contributions from channel B only. This is the
required result and is true for all the channels.
In operation of the apparatus disc 13 is rotated and during the
time that each row of apertures in the disc is passing a signal to
detector 3, each cell of bank 9 is selected in sequence by gating
array 12 and a series of add or subtract signals is provided. In
synchronism with the selection of the photocells 9, the
corresponding register of bank 8 is selected and its contents fed
to unit 7 where the output of converter 6 is added or
subtracted.
In place of the decoding arrangement described above photocell bank
9, lamp 10 and gating array 12 can be replaced by the decoding
arrangement illustrated in FIG. 2. In this arrangement two
registers 20 and 21 are provided. Register 20 is a binary counter
which is set initially to zero and which is incremental along one
each time a measurement is made. Thus the binary number in register
20 will correspond to the numerical value of the row of the
encoding matrix being used. In the above example register 20 will
increment each time the disc 13 is stepped by one row. Register 21
contains a binary representation of the register of the bank 8
currently being incremented. In the above example there are seven
registers in bank 8 and hence register 21 will increment from 1 to
7 each time that the count in register 20 is increased by one.
Corresponding pairs of digits in registers 20 and 21 are fed to two
input AND gates such as gate 22 and the outputs of all the AND
gates are fed to a NOT EQUIVALENT circuit 23 to produce the
add/subtract signal which controls unit 7.
While the encoding device has been shown as a disc, which disc also
control the decoding of the system, it will be appreciated that
either or both of these functions can be controlled by alternative
arrangements. Thus, encoding can be achieved by apertures having
shutters controlled in accordance with a program generating the
appropriate code and similarly the decoding arrangements can be
controlled by the same program.
In operation the measurements can be continued for a number of
cycles to improve the signal-to-noise ratio of the results. This is
preferable to attempting to improve the signal-to-noise ratio by
lengthening the time of each measurement.
* * * * *