U.S. patent number 3,784,875 [Application Number 05/251,272] was granted by the patent office on 1974-01-08 for stage lighting control units.
This patent grant is currently assigned to The Rank Organisation, Limited. Invention is credited to David Keith Baker, Ronald Charles Eason, Richard Melville Gascoigne, Andrew Schryver.
United States Patent |
3,784,875 |
Baker , et al. |
January 8, 1974 |
STAGE LIGHTING CONTROL UNITS
Abstract
A lighting control apparatus for stage lighting and the like in
which lighting values are stored and handled entirely by digital
techniques, the apparatus being based on a general-purpose digital
computer. The invention also provides a novel method of controlling
crossfades from one lighting state to another which is economical
in its demands on data handling capacity, since it operates by
defining the required changes as a given number of increments thus
permitting further control to be by means of addition and
subtraction only. A novel clock circuit for use in this incremental
cross-fading is described.
Inventors: |
Baker; David Keith (Farnham,
EN), Eason; Ronald Charles (East Ewell,
EN), Gascoigne; Richard Melville
(Kingston-upon-Thames, EN), Schryver; Andrew (Hartley
Wintney, EN) |
Assignee: |
The Rank Organisation, Limited
(London, EN)
|
Family
ID: |
10028663 |
Appl.
No.: |
05/251,272 |
Filed: |
May 8, 1972 |
Foreign Application Priority Data
|
|
|
|
|
May 7, 1971 [GB] |
|
|
13748/71 |
|
Current U.S.
Class: |
315/294; 315/312;
315/316 |
Current CPC
Class: |
G05B
19/075 (20130101); H05B 47/155 (20200101) |
Current International
Class: |
G05B
19/04 (20060101); G05B 19/07 (20060101); H05B
37/02 (20060101); H05b 037/00 () |
Field of
Search: |
;315/292,294,312,316 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Demeo; Palmer C.
Attorney, Agent or Firm: Joseph F. Brisebois et al.
Claims
We claim:
1. A lighting control apparatus for controlling dimmers in a
plurality of channels, comprising:
A. a digital lighting-cue memory means for recording a plurality of
series of dimmer control signals, each such series constituting a
lighting cue,
B. manual control means responsive to a manual input for selecting
any given location in the lighting-cue memory means by providing
electrical output signals to the dimmers to cause the dimmers to
produce a lighting effect corresponding to the lighting cue stored
in that location, and
C. digital computation means responsive to operation of said manual
control means for digitally computing a succession of incremental
electrical signals which act upon said electrical output signals to
cause a value of the output signals to move over a period of time
from their levels which existed before said manual input to the
levels recalled from the memory location by the operation of the
manual control means.
2. Apparatus according to claim 1, in which the digital computation
means performs an initial computation identifying those channels in
which said output signal value must change from an existing level
to a different level required by the new cue and thereby compiles a
"movement list" identifying these channels and the sense and
magnitude of the necessary changes in value, the movement list
being subsequently used to compute said succession of outputs.
3. Apparatus according to claim 2, in which the digital computation
means defines said magnitude as a predetermined number of equal
increments.
4. A lighting control apparatus for controlling dimmers in a
plurality of channels, comprising:
A. a digital lighting-cue memory means for recording a plurality of
series of dimmer control signals, each such series constituting a
lighting cue,
B. manual control means responsive to a manual input for selecting
any given location in the lighting-cue memory means by providing
electrical output signals to the dimmers to produce a lighting
effect corresponding to the lighting cue stored in that
location,
C. a manually operable control member for setting the desired speed
of cross-fading from one cue to a following cue selected by said
control means,
D. a clock responsive to said control member to produce electrical
pulses at a rate determined by the setting of said control
member,
E. a resettable counter connected to receive said electrical pulses
from the clock, and
F. digital computation means responsive to the operation of said
manual control means for computing a list defining the sense of
change of illumination in each channel and the magnitude of each of
a number of predetermined equal increments into which the necessary
change is divided, and also responsive to the counter to
subsequently increase the value of said electrical output signals
of the channels sequentially by adding a respective number of said
increments equal in number to the count held in the counter, the
counter being reset each time all channels have been so
incremented.
5. Apparatus according to claim 4, including further manually
operable control means to provide signals representing desired
levels of illumination of selected lighting channels as inputs to
the lighting-cue memory means.
6. Apparatus according to claim 4, in which the clock comprises a
resistor and capacitor arranged in series across a voltage source,
a computer having one input connected to receive the voltage on the
capacitor and its other input connected to receive a voltage
determined by said control member, a monostable circuit triggered
by the comparator output changing sign, and means responsive to the
output of the monostable circuit for discharging the capacitor.
7. Apparatus according to claim 6, in which the voltage determined
by the control member is derived from the slider of a linear
potentiometer, the potentiometer being connected across said
voltage source and its slider being mechanically coupled to said
control member.
8. Apparatus according to claim 6, in which said means for
discharging the capacitor comprises a transistor whose emitter -
collector path is connected in shunt across the capacitor and whose
base is connected to receive the output of the monostable
circuit.
9. Apparatus according to claim 4, in which the digital computation
means includes a binary shift register for computing the magnitude
of said increments by dividing the required change by a constant
which is a power of 2.
Description
This invention relates to stage lighting control units. Such units
are used to control the illumination produced on a set in
theatrical, television or similar fields, and are required to
control the illumination provided by a number of lighting channels
both individually and in some situations en bloc.
The development of stage lighting control units has been directed
from the beginning towards greater flexibility of operation, with a
corresponding increase in the freedom of the operator from
restrictions in technique imposed by limitations of the mechanical
and electrical systems at his disposal. More recently, the trend of
development has been towards units which are capable of recording
entire lighting cues in some form of memory (originally mechanical,
but later magnetic) so that a set of lighting cues constituting a
lighting plot can be stored and the stored plot subsequently
re-played without the need for manual presetting of the dimmer
controls. Changes of cue in this type of unit are effected by
cross-fading from one cue memory to the next.
An example of this type of system is illustrated by the embodiment
described in our U.K. Patent Specification No. 1,220,815. FIG. 1 of
the accompanying drawings depicts a simplified schematic of this
arrangement. During playback (performance) cues are read back
alternately into Stores C and D. A modulating device allows the
outputs from these two stores to be multiplied by any factor
between 0 and 1 by the setting of corresponding "mastering"
controls. The setting of these controls can therefore be used to
mix the outputs from the two stores, perform crossfades, etc.
However, since the introduction of the equipment described in the
aforementioned U.K. Patent, a number of features have been
scrutinised with a view to achieving improved performance/cost
ratios.
One of the first areas for scrutiny was the channel fader lever
which has conventionally been used on manual preset controls and
has been similarly adopted in systems of the type described in U.K.
Patent Specification No. 1,220,815. As a means of temporarily
setting and storing a dimmer level, the fader lever is many more
times expensive than alternative digital solutions. An all digital
approach would avoid the necessity for costly analogue to digital
interface circuits associated with each fader lever. The fader
lever also presents difficulties from the operational point of
view. The live performance is rarely completely predictable, being
subjected as it is to continuing artistic improvement and
modification as well as to human errors. The lighting operator must
therefore be able to quickly modify a previously recorded stage
lighting level without attracting the attention of the audience.
Using the fader lever this entails careful matching of the fader
lever output to the previously recorded dimmer level before the
change can be executed. One solution to this problem has already
been described in U.K. Patent Specification No. 1,083,408 namely
the use of a rocker switch which can either be used as a command
control to raise or lower a particular channel level at a pre-set
rate or alternatively as a means of channel identification operated
in association with a single level setting means or common command
buttons which can instruct the circuit to be flashed on or off for
identification purposes during rehearsal. The operation of this
control in one embodiment of the system described in this
specification is outlined later. However, the principal feature of
the rocker control is that it provides essentially digital
information which can be interpreted by a digital system without
the necessity for costly interface circuits.
A more extensive use of digital techniques is embodied in a system
known as Q File invented by A.A. Isaacs described in U.K. Patent
Specification No. 1,171,914. A multiplexed arrangement is used
throughout which enables a common set of basic logic circuits to
carry out similar functions in turn to each channel. Thus a number
of recorded lighting states can be logically added or subtracted to
produce a composite scene made up from a number of sub-groups of
lighting. Dimmer levels are set by means of a single multiplexed
fader lever which is assigned to one channel by means of a
numerical keyboard. Furthermore, during a fade, output levels are
changed by means of computed increments or decrements to the
original lighting state. This computation is performed by a
combination of analogue and digital techniques. A further advantage
is that if the fade process is interrupted then the digital levels
at the time of interruption can be immediately used as the starting
point for the computation of a new fade. This permits processional
fades to be carried out i.e. a fade to a new cue can be started
before the previous one has been completed.
The embodiment described in U.K. Patent Specification No. 1,171,914
is based on an arrangement of digital and analogue circuits
designed specifically for the purpose required. In other words, it
is a system based on hardware logic.
Such a system has obvious drawbacks in that the use of hardward
logic makes it difficult to facilitate the incorporation of
additional or modified operational facilities.
It has been proposed to overcome this problem by the use of a
stored program computer in a lighting control. At least one such
system has been built in which a computer retrieves and files
lighting data. However, crossfading from one cue to the next is
still, in this prior system, handled by conventional analogue
techniques.
An object of the present invention is therefore to enable the use
of a digital computer to control all aspects of a lighting system,
including the dynamic requirements of crossfading.
Accordingly, one aspect of the present invention provides a
lighting control apparatus for controlling dimmers of a plurality
of lighting channels, comprising a digital computer including a
lighting cue memory for recording series of dimmer control signals,
each such series constituting a lighting cue, manual control means
for selecting any given cue memory location for providing output
signals to the dimmers to produce a lighting effect corresponding
to the lighting cue in that location, the computer being arranged
to compute, when said means is operated to recall a new lighting
cue, a succession of outputs which cause said output signals to
move over a period of time from their existing levels to the levels
recalled by said operation.
According to a further aspect of the invention, there is provided
apparatus for controlling dimmers of a number of lighting channels,
comprising a digital computer including a lighting cue memory for
recording series of dimmer control signals, each such series
constituting a lighting cue, manual control means for selecting any
given cue memory location for providing output signals to the
dimmers to produce a lighting effect corresponding to the lighting
cue in that location, a manually operable control member for
setting the speed of crossfading from one cue to a following cue
selected by said control means, a clock arranged to produce pulses
at a rate determined by the setting of said control member, and a
resettable counter arranged to receive pulse from the clock, the
computer being arranged to operate, when said means is operated to
recall a new lighting cue, to initially compute a list defining the
sense of change of illumination in each channel and the magnitude
of each of a predetermined number of equal increments into which
the necessary change is divided, and thereafter to increase the
output signals of the channels sequentially by adding a respective
number of increments equal in number to the count held in the
counter, the counter being reset on completion of each cycle of
computation.
An embodiment of the present invention will now be described, by
way of example, with reference to the accompanying drawings, in
which:
FIG. 1 relates to the prior art, as has been described;
FIG. 2 is a block diagram illustrating the components of the
apparatus and their relationship to each other;
FIG. 3 shows the operator's control desk;
FIG. 4 is a flow chart of the computer main program loop;
FIG. 5 depicts the organisation of part of the stored data; and
FIG. 6 is a circuit diagram of a clock and counter arrangement used
in this embodiment.
As is well known in the stage lighting art, individual light
sources such as spots and floods are patched together to form a
number of channels each controlled by a dimmer, usually a thyristor
arrangement. A channel may have from one to a large number of light
sources, and a given light source may form part of more than one
channel. By suitably controlling the dimmers, a given lighting
state or effect, also referred to as a lighting cue, is
achieved.
Referring now to FIG. 2, the present embodiment includes a control
desk 10 (which will be described in more detail below) connected to
a computer 12. The computer 12 is a small general-purpose computer,
specifically a PDP 11 by Digital Equipment Corporation, and
comprises a central processor 14, a local store 16 and a program
store 18 and is connected to a ferrite core main store 20, as is
all well known in the art. Also connected to the computer 12 is a
magnetic tape read/write unit 22 which acts as a dump store. The
computer 12 controls the dimmers (not shown) via channel circuit
cards 24.
The electrical connections between these various units are by means
of a conventional address and data line structure such as is used
to provide the standard means of interconnection between a computer
and its peripherals. Operationally, the system hardware can be
considered as performing the functions of a very fast telephone
exchange. However, in this case there is a simplification in that
only the computer can do the "ringing up" of numbers, or addresses
as they are normally called. When an address has been called up,
transfer of data may take place either from the computer to the
hardware or vice versa. This method of operation is basically an
extension of the internal operation of the computer. In order to
allow for varying time delays which can occur dependent on the
length of interconnecting cables, an electrical "handshake" system
is used to secure the transfer of data before the computer moves on
to the next address. Again this principal is a conventionally
accepted computer technique.
Thus in operation, the computer controls the high speed telephone
exchange and can manipulate and process the data it receives.
The control desk 10 of FIG. 2 comprises a main desk shown in FIG. 3
and a rocker panel (not shown) which carries a number of rocker
switches or rockers, one for each channel. Each rocker switch can
be depressed at one end to raise the lighting level of the channel,
and at the other to lower it, and the rockers act in conjunction
with controls 30, 32 on the master desk which determine the speed
of this movement and its upper limit. A number of cues (in this
case 250) can be built up in this way and each recorded in the main
store 20 at a location identified by a one number zero to 299
entered by a keyboard 34. The remaining controls may be generally
designated "action" controls since they determine record, playback
and crossfade actions. These actions are all controlled by the
computer 12 which generally recycles on a program loop shown in
FIG. 4. When any of the action controls or rockers is operated,
this is detected during the appropriate scan period and the
computer branches to a subroutine to give effect to it. These
routines are straightforward and it is easily within the
capabilities of a person skilled in the computer art to arrange the
recording and recall of cue data, and the like.
The essence of the invention lies in solving the problem of
handling a cross-fade (i.e. a move from one cue to another in a
given time) digitally, and the solution according to the invention
will now be described.
In order to respond to any of the Rockers or Action controls, the
computer is programmed to interrogate each in turn. This is
accomplished by addressing contacts in blocks of ten, whence 0/1
data representing the ON/OFF conditions of the controls is
transmitted back to the computer along the data bus.
As far as the operator is concerned, response to the operation of a
control must appear to be almost instantaneous, which means that
any control must be interrogated at least once every 50
milliseconds or less. During a digital crossfade, it is important
to minimise the size of the increments of decrements applied to
each dimmer control signal output, otherwise the lighting
discontinuities will be apparent to the eye. The applicants have
found that for crossfades of longer than 10 seconds, output
increments of dimmer control voltage must be restricted to 1/256 of
full output. This conclusion is based on the theatre requirement to
operate over a very wide range of illumination levels and assumes
that the dimmer control characteristic is also suited to this
application.
For crossfades of less than 10 seconds, larger discontinuities can
be tolerated provided that outputs are updated at least once every
30 milliseconds. This ensures that the response time of the lamp
(typically 100-200 milliseconds) and the eye's persistence of
vision have an overall "smoothing" effect upon the control signal
discontinuities.
As the result of this analysis, it was concluded that a main
programme loop of the form shown in FIG. 4 should be adopted.
Furthermore, in order to satisfy all of the foregoing requirements,
the programme loop must on average be completed in less than 30
milliseconds.
The crossfade requires each channel to be updated on the basis of
the following equation.
Output at time t = X + t/T(Y-X)
Where
X = dimmer lever at the start of a crossfade.
Y = required dimmer level at the end of the crossfade.
T = the set crossfade time.
t = elapsed time.
The more obvious approach to this computation would involve
multiplication. However, with the constraints placed on programme
cycle time, this could only be achieved by using an expensive
computer equipped with a fast multiplication facility.
Thus although (as has been previously mentioned), a computer has
been used previously to handle lighting control cues, the apparent
complexity and expense has so far deterred the incorporation of a
digital crossfade.
However, a much simpler solution was achieved as the result of a
novel approach to the crossfade computation which will now be
described. The computation is carried out in two parts. In the
first, the normal program cycle is interrupted whilst information
is computed and tabulated. The program then returns to its normal
cycle during which the tabular information is used to update
channels during the crossfade period. It is more important to
minimise the time for the latter computation rather than for the
initial tabulation since a slight delay in starting an action is
more acceptable than discontinuities appearing during the updating
of channel levels.
Whilst Rocker and Action controls are inactive, the program cycles
quiescently through the routine outlined in FIG. 4, cycle time for
a 360 channel system being in the region of 20 milliseconds.
Immediately a crossfade action button is operated, the quiescent
program cycle is interrupted. The required end of fade conditions
are placed in an active computer store labelled the Destination
Store. The Destination level for each channel is then compared with
the current output and the difference (.DELTA.) is determined for
each moving channel. If we now divide each fade into 256 equal time
intervals, then the fade can be progressed by adding (or
subtracting) at each time interval, the appropriate fade increment
(.delta.=.DELTA./256 etc.). In general, it will be different for
each moving channel.
In practice, the division by 256 is carried out in a single
computer operation cycle by exchanging the more significant (8 bit)
byte in a 16 bit word for the less significant byte. All the fade
increments are then placed in an active computer store (Increment
Store) as are the numbers of all those channels required to move
during the fade (Channel Number Store). The general stacking
organisation is depicted in FIG. 5 from which it can be seen that
increasing channels are segregated from decreasing channels. These
stored tabulations are sometimes referred to as the Movement
List.
Having prepared this information which may take any-thing up to 60
ms for a 360 channel system, the computer can now return to the
main program loop of FIG. 4.
The system is equipped with four fade time levers 36 to enable
RAISE and DIM fade times to be set independently on each of the two
play back controls (see FIG. 3). Each fade time lever is coupled to
a clock which produces 256 pulses during the time period set on the
scale of the respective fade time lever 36. Each clock output is
connected to the input of a respective elapsed time counter, the
output of which can be addressed, read and cleared to zero by the
computer. Once the movement list has been prepared, the computer
returns to the main program loop. Each time the crossfade updating
routine is entered, the incremental counters are in turn addressed,
read and set to zero. The number D in the DIM's counter is read
first and this establishes the decrement to be applied to each of
the corresponding channels namely D .times. .delta.. The DIM
counter is set at zero immediately after it has been read and
number D noted. The up-dating of the channel level is then
performed by D successive subtractions of the stored decrement .
The RAISE counter is similarly read and increments added to those
channels being raised. Crossfade updating is carried out with
reference to the movement list (see FIG. 5) and only the moving
channels listed are processed. Furthermore the separate stacking of
positive and negative increment permits each group to be processed
independently in conjunction with the appropriate incremental
counter output.
The system has two independent sets of playback controls, known as
GREEN and RED respectively. These have identical sets of action
controls on the master desk, designated 38 and 40 in FIG. 3. The
Green crossfade updating routine is carried out first followed
immediately by RED crossfade updating.
Each time the computer program enters the crossfade updating
routine, the count read from each elapsed time counter will depend
not only on the fade time set but on the time taken for the program
to complete the loop. The average program cycle is never longer
than 30 milliseconds so that for a fade time in excess of 10
seconds, an incremental counter reading of either 1 or 0 will be
received at each crossfade update. For crossfade times of less than
10 seconds, with a 360 way system, each counter output will be
greater than 1 since the average program cycle time may be as long
as 30 milliseconds particularly if a large number of channels is
involved in the movement list. Furthermore the incremental counter
has a maximum capacity of 8 which facilitates the selection of a
crossfade time as short as 1 second without overloading the
counter. Nevertheless, the asynchronous nature of the design allows
the system to compensate for these varying conditions. In
particular, the arrangement permits the simple interleving of four
independent fade computations i.e. the RAISE and DIM fades of both
RED and GREEN playbacks. The absence of complex interrupt or
synchronisation facilities makes for a simple, flexible design
which is readily adapted to meet the varying tasks which the
computer is required to undertake. Thus if additional Rocker or
Action controls are required, they can be readily serviced in the
main program loop, there being no critical constraints on timing or
synchronisation.
A number of different types of crossfade are possible viz:-
1. CROSSFADE, in which all channels move to the levels recorded in
the next cue.
2. MOVE, in which all channels move to the new level except those
for which the new level is set at zero; these channels stay
constant. This simplifies a change where only a small number of
existing channels are altered.
3. DIM, in which the channels identified in the new cue are reduced
to zero.
4. ALL DIM, in which all channels move to zero.
5. REVERSE, in which all channels move from their existing state to
the last cue.
6. INSTANTANEOUS -- when operated with any of (1) to (5), the
required change takes place instantaneously.
It will be appreciated that (2) to (6) above are special cases of
CROSSFADE which are provided simply for operational
convenience.
From the programming point of view, the main differences between
these actions occur in the initial setting up routine which
interprets the contents of NEXT and DESTINATION stores according to
the particular action called for. Otherwise, the determination of
fade increments and the updating routine remains as previously
described. The REVERSE action only involves changing the contents
of the destination store during its initial routine. At each
crossfade update routine, the signs of the increments are inverted
so previous increments are now subtracted etc.
The crossfade computation is the most important of the system
functions since the method chosen establishes the flexible
asynchronous approach to executing all the system functions. The
novel design of the associated clock and incremental counter will
now be described.
Referring to FIG. 6, the appropriate manual control 36 (FIG. 3)
moves the slider 60 of a linear potentiometer 62 connected across a
supply voltage Vs, to give a variable control voltage Vp. The same
supply voltage Vs is applied across a resistor 64 and capacitor 66
in series. The voltage Vc across the capacitor C follows the
law
Vc = Vs (1-e.sup. .sup.-t/RC)
where
t is the time elapsed from Vc = 0
R is the value of the resistor 64 and
C is the value of the capacitor 66
The capacitor voltage Vc is fed to one input of a comparator 68
whose other input receives the control voltage Vp. When Vc reaches
the value of Vp, the comparator output goes positive and triggers a
monostable circuit 70. The resulting output pulse from the
monostable circuit 70 passes to a counter 72 via a gate 74 and to
the base of a transistor 76 shunting the capacitor 66 to earth. The
counter 72 therefore counts up, while the capacitor 66 is
discharged to Vc = 0, and the cycle is restarted.
The time for each cycle is the width of the pulse from the
monostable circuit 70 plus the charging time of Vc and Vp.
Rearranging the expression above for Vc thus produces a logarithmic
function for cycle time. Resistors 74 and 76 are small fixed
resistors and prevent the invalid conditions of the capacitor
charging in zero and infinite times respectively. The speed control
lever 36 has two logarithmic scales calibrated respectively in
seconds and minutes (fade time). The component values are so chosen
that 256 clock pulses are generated in the set fade time.
The binary coded output of the counter 72 is interrogated at
intervals by the computer 12. The time taken by the computer 12 to
do this is short in comparison with the cycle time of the clock
circuit. However, to prevent the computer 12 accessing during a
pulse from the monostable circuit 70 and thus losing a pulse or
producing two pulses, extra logic circuits are provided. The output
from the counter 72 on output lines 78 is connected to data lines
80 to the computer via a data gating circuit 82. An output signal
on line 84 from the computer 12 causes a control circuit 86 to
generate an enabling gate pulse on line 88 to gate the counter
output to the data lines 80, immediately followed by a reset signal
on line 90 to the counter 72. During these gate and reset periods
the gate 74 on the input to the counter 72 is inhibited by the
control circuit 86.
An advantage of the above timing circuit or clock is that very few
components are required to control the accuracy of the timing.
These are the resistor 64 and capacitor 66 which must be of high
stability and the potentiometer 62 whose ratio must be stable.
Different time ranges may be obtained by switching in different
values for resistor 64.
From the embodiment described, it will be apparent that the
invention provides an elegant solution to the problem of providing
a digital lighting control apparatus in which the entire control
including crossfading is handled by an on line digital computer.
This solution gives great flexibility and permits many easy
modifications to the embodiment described.
For example, in a theatre the control desk is typically in the
backstage area, but it is frequently desired to have additional
"stalls" control in the audience area. The present invention allows
a stalls control to be simply added, the computer then merely
scanning both control desks during its main program loop.
Another modification made possible by the invention replaces the
many (typically 240 ) rockers to be replaced by a simpler control
layout. The control desk is provided with a keyboard by means of
which the operator can select any channel, and with a handwheel
driving a rotary digital encoder. To alter a channel lighting
level, the operator selects that channel on the keyboard and then
moves the handwheel in one direction to raise the light level and
in the other to lower it. The digital encoder outputs a pulse for
each given angle of rotation with an indication of the sense of the
rotation and these pulses are simply added to or subtracted from
the stored level for that channel. Digital encoders of this type
are well known in many fields, for example in the numerical control
of machine tools. With such an arrangement the program for the
action controls is unchanged -- only the program for level setting
requires to be rewritten.
Other modifications are of course possible within the scope of the
invention as defined in the appended claims.
* * * * *