Newport 290 Series User manual

US ER ’S G UI D E
Quadcell Photoreceivers
Model 290X
phone: [phone] e-mail: [email protected] www.newport.com
290x quadcell rcvr RevA.fm Page 1 Monday, November 15, 2004 3:39 PM

Warranty
Newport Corporation guarantees its products to be free of defects for one year
from the date of shipment. This warranty is in lieu of all other guarantees,
expressed or implied, and does not cover incidental or consequential loss.
Information in this document is subject to change without notice.
Copyright 2001-1998, 2013, Newport Corporation. All rights reserved.
The New Focus logo and symbol are registered trademarks of Newport
Corporation.
Document Number 200316 Rev. B
290x quadcell rcvr RevA.fm Page 2 Monday, November 15, 2004 3:39 PM

Model 290X Contents • 3
Contents
Introduction 5
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Tips for Best Operation . . . . . . . . . . . . . . . . . . . . . 5
Operation 7
Setting up the Receiver . . . . . . . . . . . . . . . . . . . . . . 7
Checking and Replacing the Battery . . . . . . . . . . 8
Using an External Power Supply . . . . . . . . . . . . . 9
General Features and Principles 11
Amplifier Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Gain and Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . 13
Normalization of Outputs . . . . . . . . . . . . . . . . . . . 14
Position Transfer Function . . . . . . . . . . . . . . . . . . 14
Selecting the Optimal Gain Setting . . . . . . . . . . . 16
Frequency Response and Noise . . . . . . . . . . . . . . 16
Characteristics 23
Physical Specifications . . . . . . . . . . . . . . . . . . . . . . 23
Operating Specifications . . . . . . . . . . . . . . . . . . . . 24
Customer Service 25
Technical Support . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
290x quadcell rcvr RevA.fm Page 3 Monday, November 15, 2004 3:39 PM

4 • Contents
290x quadcell rcvr RevA.fm Page 4 Monday, November 15, 2004 3:39 PM

Model 290X Introduction • 5
Introduction
Overview
The Model 290X family of photoreceivers use a
combination of a quadrant cell photodiode and signal-
processing amplifiers to generate position-sensitive
signals from an incident light source.
Versions of the photoreceiver are available for visible
(Model 2901) or infrared (Model 2903) beams. The
photoreceiver can be operated from battery or a ±15 V
DC power supply (we recommend the low-noise
Newport Model 0901 power supply for best
performance).
Each model has a photodiode with a 3-mm diameter
active region.
Tips for Best Operation
Avoid Room Lights
The 290X quadrant cell photoreceiver has a large active
area. For a typically illuminated room of 1–10 µW/
mm2, the SUM output can reach 1 V on the 30-µW
saturation gain setting. The 120-Hz component of
fluorescent room lights is extreme, and will almost
certainly cause measurement difficulties.
290x quadcell rcvr RevA.fm Page 5 Monday, November 15, 2004 3:39 PM

6 • Introduction
The best solution is to shield the receiver as much as
possible from room lights. A Newport Model 1280
can be installed into the threads on the detector
housing to hold filters or a light shield.
Limiting Measurement Bandwidth
As seen in the discussion “Impact of Noise on Position
Measurement” on page 19, the measurement
bandwidth plays an important role in the noise of your
system. Select a measurement bandwidth consistent
with your measurement goals. For instance,
observation of thermal drifts in the sub-Hz range will
benefit greatly from signal averaging or other
bandwidth-limiting actions.
Using Optical Filters
Narrow-band optical filters, which transmit your
laser’s wavelength while blocking other light sources,
can be installed in a Newport Model 1280 1" filter
holder. Many filters are commercially available.
When using optical filters during position
measurement, please consider the effects of multiple
surface reflections. The ideal filter has a wedge shape,
and good anti-reflection coating on one surface to
ensure that the optical beam incident on the receiver
does not contain interference fringes.
290x quadcell rcvr RevA.fm Page 6 Monday, November 15, 2004 3:39 PM

Model 290X Operation • 7
Operation
Setting up the Receiver
1. Supply power. The Model 290X is powered
either by a single 9-V alkaline battery, or by a +/-
15 V DC power supply. If both battery and exter-
nal power are present, an internal relay connects
the external power to the amplifier circuitry. To
check the battery condition, push the red power
switch to the BATT CHK position. If the green
LED lights up, the battery is in good condition; if
the LED does not light, the battery needs to be
replaced (see page 8). Adjust the power switch to
ON.
Remember to switch the power OFF when not in use to
maximize battery life.
2. Mount the photoreceiver. Use the 8-32 thread
(M4 for metric versions) on the bottom of the
casing to mount the photoreceiver to a post or
pedestal.
Be careful not to over-tighten when attaching to a post or
pedestal, or the threaded insert can strip out of the plastic
pad.
3. Connect the receiver output. Connect your volt-
meter, oscilloscope, or other instrument to the
X, Y and SUM connectors on the back of the
receiver.
The threading is seated in a non-conductive plastic pad to
reduce the electrical noise associated with ground loops.
Note:
Note:
Note:
Note:
290x quadcell rcvr RevA.fm Page 7 Monday, November 15, 2004 3:39 PM

8 • Operation
Electrical ground is only introduced to the housing through
the BNC outer shells, and the power-supply ground pin. If
you use a floating measurement instrument or a floating
power supply, the photoreceiver may not be electrically
grounded.
4. Align an optical beam onto the detector. Be
careful to keep the optical power below the
stated limits to avoid damage to the photore-
ceiver and amplifier circuitry.
5. Adjust the gain. Use the rotary switch on top of
the receiver to set the gain. The table on page 13
gives a detailed description of the different chan-
nel gains and bandwidths associated with each
switch setting.
Checking and Replacing the Battery
The Model 290X is powered by a single, standard 9-volt
alkaline battery. Under normal operating conditions
with low light levels and a high impedance load
attached to the BNC connectors, the photoreceiver
draws about 25 mA from the battery, and the battery
lifetime will be approximately 40 hours.
To check the battery condition, push the power switch
to the BATT CHK position. If the green LED lights up,
the battery is in good condition.
When the battery voltage falls below about 6.5 volts,
the green LED will not light, and the battery should be
replaced.
1. Switch off the power and disconnect any exter-
nal power supply.
2. Remove the screw in the back of the photore-
ceiver casing and remove the back cover.
3. Unplug the old battery by rotating it away from
the circuit board about the snap-on terminal
contacts.
290x quadcell rcvr RevA.fm Page 8 Monday, November 15, 2004 3:39 PM

Model 290X Operation • 9
4. Install a new 9-volt alkaline battery.
5. Reinstall the back cover and screw. Take care not
to pinch the power wires from the external
power connector.
6. Test the new battery by pushing the power
switch to the BATT CHK position.
Using an External Power Supply
The Model 290X can be operated from any power
supply which can supply 50 mA of ±15V DC power.
Newport recommends use of the Model 0901 Power
Supply, which features three pairs of low-noise linear
±15 V DC outputs.
If using the Model 0901 Power Supply, connect the
Model 0923 M8B-to-M8B cable from the power supply
to the jack on the back of the Model 290X.
If using a third-party power supply, connect the Model
9024 banana-plug-to-M8B cable from the power
supply to the jack on the back of the Model 290X. The
red jack should be connected to +15 V DC, the black to
-15 V DC, both withrespect to the green jack, which
should be connected to the power supply COMMON
terminal. It is recommended, but not required to use
the green wire asan earth ground.
290x quadcell rcvr RevA.fm Page 9 Monday, November 15, 2004 3:39 PM

10 • Operation
290x quadcell rcvr RevA.fm Page 10 Monday, November 15, 2004 3:39 PM

Model 290X General Features and Principles • 11
General Features and Principles
Amplifier Circuitry
The circuitry inside of the Model 290X consists of a
two-stage amplifier, with parallel gain paths for X, Y,
and SUM outputs. Each stage has an adjustable gain
and bandwidth. Figures 1 through 3 illustrate the basic
circuit design, and the table on page 13 shows the gain
and bandwidth of each amplifier stage for each
possible setting of the rotary switch.
Each photocurrent from the quadrant photodiode is
amplified at a first-stage amplifier. The value of R1
depends on the gain setting, and can vary from 200 Ω
to 200 kΩ.
Figure 1: Each
photocurrent is
amplified at a
first-stage
amplifier
The signals from each quadrant are differenced and
amplified at a second-stage amplifier. The value of G2
AB
CD
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
R
1
-I
A.
R
1
-I
B.
R
1
-I
C.
R
1
-I
D.
R
1
290x quadcell rcvr RevA.fm Page 11 Monday, November 15, 2004 3:39 PM

12 • General Features and Principles
depends on the setting of the gain switch, and can vary
from 1 to 10. The bandwidth of this gain stage also
varies, as indicated in the table on page 13.
Figure 2:
Signals from
each quadrant
are differenced
and amplified at
a second stage
amplifier
The signals from each quadrant are summed at a
second-stage amplifier. The value of G3is fixed at
unity. The bandwidth of this gain stage is fixed at 30
Hz.
Figure 3:
Signals from
each quadrant
are summed at a
second-stage
amplifier
-
+
-
+
G
2
-I
A.
R
1
-I
B.
R
1
-I
C.
R
1
-I
D.
R
1
G
2.
R
1.
(I
A
+I
B
-I
C
-I
D
) “DELTA Y”
-
+
-
+
G
2
-I
B.
R
1
-I
C.
R
1
-I
A.
R
1
-I
D.
R
1
G
2.
R
1.
(I
B
+I
C
-I
A
-I
D
) “DELTA X”
1000
1000
-
+
-
+
G3
-IA.R1
-IB.R1
-IC.R1
-ID.R1
G3.R1.(IA+IB+IC+ID) “SUM”
100
290x quadcell rcvr RevA.fm Page 12 Monday, November 15, 2004 3:39 PM

Model 290X General Features and Principles • 13
Gain and Bandwidth
The following table shows the gain and bandwidth of
each amplifier stage as a function of switch setting.
Switch
Position
R1Gain
(V/A)
1st-Stage
BW (kHz)
G2Gain
(V/V)
∆-StageBW
(kHz)
approx.
G3Gain
(V/V)
Sum-Stage
BW (kHz)
30 mW (1) 200 100 1 100 1 0.03
30 mW (3) 200 100 3 100 1 0.03
30 mW (10) 200 100 10 100 1 0.03
30 mW (10) 200 100 10 0.03 10.03
3 mW (1) 2000 100 1 100 1 0.03
3 mW (3) 2000 100 3 100 1 0.03
3 mW (10) 2000 100 10 100 1 0.03
3 mW (10) 2000 100 10 0.03 10.03
0.3 mW (1) 20,000 100 1 100 1 0.03
0.3 mW (3) 20,000 100 3 100 1 0.03
0.3 mW (10) 20,000 100 10 100 1 0.03
0.3 mW (10) 20,000 100 10 0.03 10.03
30 µW (1) 200,000 100 1 100 1 0.03
30 µW (3) 200,000 100 3 100 1 0.03
30 µW (10) 200,000 100 10 100 1 0.03
30 µW (10) 200,000 100 10 0.03 10.03
290x quadcell rcvr RevA.fm Page 13 Monday, November 15, 2004 3:39 PM

14 • General Features and Principles
Normalization of Outputs
The signals X and Y vary inresponse to changes of the
light position relative tothe center of the quadrant
photodiode. Because X and Y also vary in response to
changes of the total light power, the SUM signal is
provided as a means of normalization.
The bandwidth of the SUM signal is limited to 30 Hz.
Therefore, the normalization cannot be performed for rapid
power fluctuations characteristic ofsome laser beams.
Position Transfer Function
When normalized, the signals X/SUM and Y/SUM can
be used to measure position. For clarity, we will use the
dimensionless units V/V for the normalized signals
when describing the measured position transfer
function. Because the signals are normalized, position
output does not depend on R1gain setting, only on G2.
Figures 4 and 5 illustrate the output response as a
function of input position for an ideal Gaussian laser
beam input. The nearly linear portion of the output
response has a slope of 0.65 radii/V/V. For example, a
normalized output of 0.5 V/V would occur for a
Gaussian beam displaced 0.5 V/V * 0.65 radii/V/V =
0.32 radii from the center of the quadrant photodiode.
If the beam radius in the above example were 1 mm,
the normalized output of 0.5 V/V would occur at a
displacement of 0.32 mm from the photodiode's
center.
Note:
Note:
290x quadcell rcvr RevA.fm Page 14 Monday, November 15, 2004 3:39 PM

Model 290X General Features and Principles • 15
If the optical beam under measurement is not
Gaussian, the above relationship will not hold. For
example, if the beam is elliptical, the transfer function
will be different along the major and minor axes. You
can calibrate the position transfer function effectively
by placing the photoreceiver on a two-axis translation
stage and measuring output as a function of position.
Figure 4:
Normalized
position output
transfer
function for
G2=1 settings
assuming
Gaussian beam
input
Figure 5:
Normalized
position output
transfer
function for
G2=10 settings
assuming
Gaussian beam
input
Output Transfer Function (x1 settings)
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
-1.50E+00 -1.00E+00 -5.00E-01 0.00E+00 5.00E-01 1.00E+00 1.50E+00
Location (beam radii)
Normalized output (V/V)
Output Transfer Function (x10 settings)
-5.0
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
-0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4
Location (beam radii)
Normalized output (V/V)
290x quadcell rcvr RevA.fm Page 15 Monday, November 15, 2004 3:39 PM

16 • General Features and Principles
Selecting the Optimal Gain Setting
In order to maximize signal fidelity, you should select
the maximum gain setting possible, but not saturate
the amplifier. The first and second stage gains can
saturate independently, so you must take care to select
an optimal value for each.
It is easy to identify if the first stage is saturated. The
SUM output will reach its “rail” voltage of 2.5 V. Any
time the SUM output exceeds 2 V, it is a good idea to
reduce the gain setting by reducing R1by one decade.
Saturation of the second stage depends on the position
of your optical beam relative to the photodiode center.
X and Y can reach a maximum of ±2.5 V or G2* SUM,
whichever is smaller. Test for saturation at your desired
gain setting by creating the maximum expected beam
displacement and observing the X and Y outputs. If the
signal saturates, reduce the value of G2by selecting a
different gain switch setting.
Frequency Response and Noise
Measuring Bandwidth
The frequency response and noise characteristics of
the quadrant cell photoreceiver depend onthe selected
gain. Figures 6–10 onthe following pages give the
typical frequency response andnoise behavior for the
photoreceivers at each of the 16possible gain settings.
The frequency response of the receiver inFigure 6 is
plotted using the expression
20*log[Vout(f)/Vout(DC)]
where Vout(f)is the output from the X, Y, or SUM
output of the receiver. Newport uses the -3 dB
electrical definition for bandwidth.
290x quadcell rcvr RevA.fm Page 16 Monday, November 15, 2004 3:39 PM

Model 290X General Features and Principles • 17
Figure 6:
Frequency
response for X,
Y and SUM
outputs with
30-µW
saturation
setting
(R1= 200 ohm)
Figure 7:
Output voltage
noise spectral
density for X, Y
outputs with
30-µW
saturation
setting
(R1= 200 kΩ)
Figure 8:
Output voltage
noise spectral
density for X, Y
outputs with
0.3-mW
saturation
setting
(R1= 20 kΩ)
2901 Frequency Response (30 µW Settings)
-30
-27
-24
-21
-18
-15
-12
-9
-6
-3
0
1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06
Frequency (Hz)
Frequency Response (20*log(Vout(f)/Vout(DC))
XY1
XY3
XY10
XY10
SUM
2901 Noise (30 µW Settings)
1.E-08
1.E-07
1.E-06
1.E-05
0 102030405060708090100
Frequency (kHz)
Noise (Vrms/sqrt(Hz))
X1
X3
X10
X10
2901 Noise (0.3 mW Settings)
1.E-08
1.E-07
1.E-06
1.E-05
0 102030405060708090100
Frequency (kHz)
Noise (Vrms/sqrt(Hz))
X1
X3
X10
X10
290x quadcell rcvr RevA.fm Page 17 Monday, November 15, 2004 3:39 PM

18 • General Features and Principles
Figure 9:
Output voltage
noise spectral
density for X, Y
outputs with
3-mW
saturation
setting
(R1= 2 kΩ)
Figure 10:
Output voltage
noise spectral
density for X, Y
outputs with
30-mW
saturation
setting
(R1= 200 ohm)
Measuring Noise
The photoreceiver noise (Vn) is measured as a noise
spectral density [Vrms/]. The noise spectral density
at each of the gain settings is provided in Figures 7–10
on the previous pages. Depending on your
measurement system, you may utilize only a fraction
of the receiver's available bandwidth. The noise in your
measurement will be the integral of the receiver's noise
2901 Noise (3 mW Settings)
1.E-08
1.E-07
1.E-06
1.E-05
0 102030405060708090100
Frequency (kHz)
Noise (Vrms/sqrt(Hz))
X1
X3
X10
X10
2901 Noise (30 mW Settings)
1.E-08
1.E-07
1.E-06
1.E-05
0 102030405060708090100
Frequency (kHz)
Noise (Vrms/sqrt(Hz))
X1
X3
X10
X10
Hz
290x quadcell rcvr RevA.fm Page 18 Monday, November 15, 2004 3:39 PM

Model 290X General Features and Principles • 19
spectral density over the frequency range (fmin to fmax)
observed by your measurement system:
For a flat noise spectrum in a DC-coupled
measurement system with a 3-dB bandwidth of f3dB,
the measured rms noise will be .
Impact of Noise on Position Measurement
Unlike photoreceivers which are often used to
measure optical power, the quadrant cell
photoreceiver is a position transducer. Therefore, the
receiver's output noise is observed as position
measurement noise. Although a real measurement
system may also have noise due to laser power
instability, or undesired mechanical vibrations, the
receiver's output noise sets a lower limit on position
sensitivity.
Recall that the position transfer function for a
Gaussian beam is 0.65 radii/V/V of normalized output.
For other beam shapes, the position transfer function
will differ. Considering that the electrical noise of the
receiver effects both numerator and denominator of
the normalized output, you can approximate the
expected noise in the normalized output signal as
Vn
2df
fmin
fmax
∫
Vnf3dB
2π
4
-------
⋅⋅
290x quadcell rcvr RevA.fm Page 19 Monday, November 15, 2004 3:39 PM

20 • General Features and Principles
The bandwidth of the SUM output is intentionally
limited to 30 Hz, which reduces the contribution of
Vn_SUM to negligible in most cases. Therefore, the
position uncertainty is simply equal to the output
voltage noise divided by the normalized SUM output,
multiplied by the position transfer function.
If you take the worst case, which shows up for the
30-µW settings (200 kΩR1) and 10 voltage gain second
stage, the Y output noise level peaks at about
2 µV(rms)/ . All other gain settings have lower
noise.
Over the 100 kHz bandwidth of the receiver, this noise
integrates to about 0.5 mV(rms). If your voltage
converter, such as an oscilloscope has wider
bandwidth, you may observe somewhat more voltage
noise. If you reduce the second stage gain to 1, the
integrated noise in 100 kHz drops to about
0.1 mV(rms).
As an example we can approximate the position
measurement noise you might observe in a typical
application. If you have about 1 mW of HeNe power,
you will use the 3 mW setting (2 kΩR1). If you are
trying to observe small, high frequency position
changes, the 10 setting for G2will give you a voltage
swing in the X/SUM or Y/SUM calculated output from
-1.56 to +1.56 when the beam travels over the range of
-0.1*radius to +0.1*radius. The position transfer
function is therefore given by X/SUM/1.56 = x/
0.1*radius, or x = 6.5*radius*X/SUM. The SUM output
Xactual
SUMactual
----------------------------
Vnx
SUMactual
----------------------------+≈
SIGNALactual ∆SIGNAL+
Xactual V+ nx
SUMactual VnSUM +
-----------------------------------------------------=
Hz
290x quadcell rcvr RevA.fm Page 20 Monday, November 15, 2004 3:39 PM