Pulse or digital communications – Pulse width modulation
Reexamination Certificate
1998-07-07
2001-12-11
Chin, Stephen (Department: 2634)
Pulse or digital communications
Pulse width modulation
C356S370000
Reexamination Certificate
active
06330279
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to optical position sensing and more specifically to an improved signal amplifier for a solid-state position sensitive detector.
2. Description of the Background Art
Accurate position sensing is needed in such diverse fields as robotics and disk drives. Typically the problem arises in automatic control of structures, when an element of the structure is moved by some kind of motor. Servo systems compare the desired position of the structure with the measured position of the structure, and using this difference information supply more or less power to the motor. For this kind of servo system to work the measured position must be known to great accuracy.
Optical measuring systems are attractive in servo controlled systems because they do not introduce friction into the systems. Such friction could negate carefully designed critically-damped systems and cause either slow response or oscillation about the desired position. A typical prior art optical measuring system is shown in FIG.
1
. Here the rotational position of arm
110
about pivot
112
is to be measured by light supplied by light source
100
. Light source
100
may be a laser or some other collimated light source. The incident beam
104
from light source
100
is reflected by reflector
114
. Reflector
114
is shown as a mirror but alternatively may be a beam splitter. As arm
110
pivots about pivot
112
, angle A changes and the reflected beam
106
traverses a series of photodiodes
120
through
136
. Depending upon which photodiode
120
-
136
is illuminated by reflected beam
106
, the angle A of arm
110
is approximately known.
The device shown in
FIG. 1
has the drawback of low positional resolution. The position of arm
110
is known only to a resolution depending upon the size and spacing of the photodiodes
120
-
136
. Smaller photodiodes which are more closely spaced will yield higher resolution, but there is a limit to the practical size and spacing of discrete photodiodes. In addition, each photodiode has an anode and a cathode lead. Biasing and sensing a large number of individual photodiodes adds unwelcome complexity to the device.
A special kind of photodiode called a position sensitive detector (PSD) offers improved resolution and accuracy over the use of many discrete photodiodes.
FIG. 2
shows the device of
FIG. 1
where the individual photodiodes
120
-
136
have been replaced by a single PSD
210
. The PSD
210
is a photodiode with an anode of width L. The PSD
210
has a common cathode
212
and a pair of anode connections, anode A
214
and anode B
216
, attached at opposite ends of the anode of width L. The distance x from the center of PSD
210
of an illuminating spot produced by reflected beam
106
may be calculated by measuring the relative currents flowing in anode A
214
and anode B
216
. The continuous anode of the PSD allows measuring resolution and accuracy to 1 part in 10,000 if coupled to a sensing amplifier of sufficient accuracy.
FIG. 3
shows a schematic symbol for a PSD. PSD
300
comprises a common cathode attachment
310
and a pair of anode attachments, anode A
312
and anode B
314
. The schematic symbol for the PSD
300
also shows a schematic representation of incident light
316
. When the PSD is reverse biased with voltage Vcc, a current Io flows depending upon the intensity of the incident light.
PSD
300
has the property that Io is dependent only on the intensity of the incident light and not on its position along the long anode. Currents I
A
and I
B
flow in anode A
312
and anode B
314
, respectively. By current junction law, I
O
=I
A
+I
B
, and therefore the sum (I
A
+I
B
) is also dependent only on the intensity of the incident light and not on its position.
In
FIG. 4
a schematic diagram for a prior art sensing amplifier and servo driver circuit using Gilbert cells is shown. The use of the 2-quadrant Gilbert cell for performing analog multiplications and divisions is well known in the art. In the
FIG. 4
schematic, PSD is connected to an integrated circuit model AD880 (
402
), containing the Gilbert cells, manufactured by Analog Devices, Inc. The AD880 (
402
) has a sum output node
404
which may be used for laser power control, and a normalized difference servo output node
406
which yields the relative distance from the center of PSD
400
(as shown for PSD
210
in FIG.
2
). In the
FIG. 4
application, the position information is used for servo control over that position. The desired position is entered as digital data
412
into an inverted-output digital-to-analog converter (DAC)
410
. The inverted analog output
414
of the DAC
410
is added to the non-inverted signal from the normalized servo output node
406
by lead/lag compensation circuit
420
. Lead/lag compensation circuit
420
contains a summing operational amplifier (op amp)
422
whose output is zero if the measured position is the same as the desired position, and gives a correction signal otherwise. The output of lead/lag compensation circuit
420
is the input of servo control circuit
440
. Servo control circuit
440
drives the arm control motor
450
in proportion to the correction signal from lead/lag compensation circuit
420
. Servo control circuit
440
contains a current source op amp
442
and a current sink op amp
444
whose outputs at the current source node
446
and current sink node
448
send currents through the windings of arm control motor
450
, keeping the arm in the desired position which was entered as digital data
412
.
The primary shortcoming of the prior art circuit of
FIG. 4
is the error induced by the analog divisions performed by AD880 (
402
). The observed error with this circuit is 1 part in 100, far below the 1 part in 10,000 intrinsic to the PSD
400
.
In
FIG. 5
a schematic diagram for a prior art sensing amplifier using gain control on the incident laser power is shown. In the
FIG. 5
circuit the need for analog division is removed by controlling the illumination intensity from the laser
502
incident upon PSD
500
. If the illumination intensity incident upon PSD
500
is a constant, then the sum of the anode currents (I
A
+I
B
) will be a constant, eliminating the need for normalization and the analog division errors induced thereby. The
FIG. 5
circuit converts the PSD
500
anode currents into voltages with A buffer op amp
510
and B buffer op amp
512
. The signals at voltage A node
514
and voltage B node
516
are added with analog adder
520
to yield a signal at A+B node
532
proportional to the incident intensity on PSD
500
. Using this signal on A+B node
532
, automatic gain control (AGC) circuit
534
sends a signal on AGC node
536
which adjusts the laser power controller
504
and thereby the laser
502
power output.
In other aspects, the circuit of
FIG. 5
is equivalent to that of FIG.
4
. The desired position is entered as digital input
528
to DAC
526
, producing a desired position voltage on the analog output
524
of DAC
526
. The A and B signal voltages are subtracted in analog subtractor
518
and compared with the desired position voltage using analog subtractor
530
. The output of subtractor
530
is the input to the servo control circuit (not shown) which is identical to the servo control circuit
440
of FIG.
4
.
The AGC circuit
534
control over laser
502
power output holds the incident light intensity on PSD
500
constant so that error-inducing normalization by division is not required. The circuit of
FIG. 5
eliminates much of the error induced in the circuit of
FIG. 4
, but with the newly added limitation that the laser
502
power cannot be varied for other system requirements. In many applications, the laser's power needs to vary over a wide range, such as in the case of a read/write optical or magneto-optical disk drive. In such a case the positions of head positioning arms or fibers in a fiber-optic switch need to be controlled to high accuracy, and prefer
Artusy Max
Belser Karl A.
Hrinya Stephen J.
Cesari Kirk A.
Chin Stephen
Dempster Shawn B.
Fan Chieh M.
Olson Jonathan E.
LandOfFree
System and method of correcting gain and offset error in a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method of correcting gain and offset error in a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method of correcting gain and offset error in a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2563662