Wideband Transimpedance Amplifiers for
Optoelectronics: Applications to Dynamic
Interferometry
Lucas M. Riob
´
o
1∗†
, Francisco E. Veiras
∗†
,Mar
´
ıa T. Garea
∗
and Patricio A. Sorichetti
∗
∗
Universidad de Buenos Aires. Facultad de Ingenier
´
ıa. Departamento de F
´
ısica. GLOmAe.
Av. Paseo Col
´
on 850, CABA C1063ACV, Argentina
†
CONICET
Godoy Cruz 2290 (C1425FQB) CABA - Rep
´
ublica Argentina
1
lriobo@fi.uba.ar
Abstract—This paper describes the design and performance
of transimpedance amplifiers (TIAs) for optoelectronic
systems, optimized for optical dynamic interferometry. In
order to perform the analysis of the amplifiers, we show some
experimental schemes, where different topologies were used.
We describe the relevance of each TIA implemented and the
corresponding guidelines and design considerations.
Resumen— Este art
´
ıculo describe el dise
˜
no y desempe
˜
no
de amplificadores de transimpedancia (TIAs) para
optoelectr
´
onica, optimizados para su uso en aplicaciones
de interferometr
´
ıa
´
optica din
´
amica. Para desarrollar el
an
´
alisis de los amplificadores, mostramos algunos esquemas
experimentales donde se utilizan distintas topolog
´
ıas. En cada
esquema desarrollado, describiremos la relevancia de cada
TIA, mencionando lineamientos y consideraciones de dise
˜
no.
I. INTRODUCTION: DETECTION OF OPTICAL
INTERFEROMETRIC SIGNALS
In optical sensing systems, the current generated from
photodetectors is generally small. Moreover, most of the
subsequent processing occurs in the voltage domain. Thus, a
conversion from current to voltage is required. A current to
voltage converter is also called a transimpedance amplifier
(TIA). TIAs are commonly used in many applications that
require high dynamic range, speed and sensivity, such as
optical communications and optical metrology systems [1].
It is known that optical interferometric methods allow
the measurement of sub-nanometer dynamic displacements
[2], [3]. These techniques have been applied to a variety
of determinations [4], [5] and are particularly interesting in
the case of electromechanical properties of materials [6],
single nanoparticle detection in fluids [7] or observation of
gravitational waves [8]. Since the light used for sensing
usually does not alter the sample under study, optical in-
terferometric techniques also present significant advantages
for non-destructive testing (NDT) [9].
A general expression of a two-beam interferometric signal
is shown on (14):
I(~r, t) = A(~r, t) + B(~r, t) cos θ(~r, t) (1)
where A(~r, t) is the background intensity, B(~r, t) is the
fringe contrast and θ(~r, t) is the phase difference between
the interfering waves. These parameters can be both space
(~r) and time (t) dependent. Usually, the information of
interest is encoded on θ(~r, t). Thus, we need to detect and
process I(~r, t) without any kind of distortion.
An interferometry system can be divided in three func-
tional blocks. The first block is the experimental scheme
on which the device under test (DUT) is placed and the
interfering beams are generated and combined. The second
block is the detection block whose front-end can be either
a camera or single photodetectors, or both. The third block
is the processing block that is used to extract the required
information.
Due to the limited capturing rate of the camera, initially
camera-based interferometers were mainly used for quasi
static measurements. For high-frequency vibration measure-
ment, there are various techniques that allow steady-state
surface vibration mode measurements, but these can only
be applied in principle on periodical disturbances [10].
Non periodical measurements can be made by high speed
cameras, having frame rates in excess of 250 frames per
second. However, they can be very expensive in comparison
to high speed single photodetectors. Moreover, they may not
be capable of working with low intensity signals. In addition,
if low-speed CCDs are used, the time for acquiring the inter-
ferograms may be quite long and this makes the technique
poorly suited to be used in an industrial environment [11].
In this work, we are interested in the detection of time
varying interferometric signals. For this purpose we use
photodiodes as photodetectors. A photodiode converts the
photon energy of light into an electrical signal by releas-
ing and accelerating current-conducting carriers within the
semiconductor. In this case, the diode junction is customized
improving their spectral response and efficiency as in PIN
and avalanche alternatives Also, multiple-element and lateral
photodiodes provide position-sensing through the relative
magnitudes of multiple output currents. By itself, a photo-
diode can produce a voltage output as required for most
electronic instrumentation. However, this operating mode
(photovoltaic mode) produces a highly nonlinear response
and a very restricted bandwidth. Instead, accepting the diode
output as a current (photoconductive mode) and performing
a current-to-voltage conversion dramatically improves its
performance [1].
In the following sections, we present three different inter-
ferometric schemes, each of them using a specific detection
Revista elektron, Vol. 1, No. 1, pp. 16-22 (2017)
ISSN 2525-0159
16
Recibido: 23/06/17; Aceptado:14/07/17
setup. The first setup is the simplest form of current to
voltage conversion. We take advantage of its simplicity to
study the detection process of photodiodes and also to define
our design considerations in optoelectronic measurements.
In the second setup we present a photodiode transimpedance
amplifier. We describe the critical parameters involved in the
design of a transimpedance amplifier destined to measure
interferometric signals with low modulation depth. The last
setup is a wideband photodiode transimpedance amplifier
which can detect high speed optical signals. Finally, the
results of this work are discussed.
II. FIRST APPROACH: DETECTION OF LOW SPEED
PHENOMENA
Consider the birefringent phase demodulator shown in
Fig.1, detailed in [12].
Polarization modulated light
Grounded interface
Quartz crystal (ϴ = 45º)
Polarizer
Ω
Phase modulator (DUT)
He-Ne laser
Photodiodes
DAQ
Phase demodulator
Interference plane
z
y
PC
Interference plane
Photodiode
V
BIAS
R
To DAQ
V
OUT
I
T
(B)
(A)
Fig. 1. Birefringent phase demodulator. (A) An array of photodiodes
is placed in the interference plane. The photocurrent I
T
is converted to
a voltage signal (V
OU T
) by a resistor R and then sampled by a data
acquisition system (DAQ) and processed by a computer (PC) [13]. (B)
Distribution of the photodiodes and connection scheme.
A spatiotemporal interferogram is produced by the system
itself and detected by an array of photodiodes. Each photo-
diode is placed in a position ~r
i
= (y
i
, z
i
) where the intensity
is
I(~r
i
, t) = A(~r
i
) + B(~r
i
) cos [∆φ
o−e
(~r
i
) + ∆φ
v−h
(t)] (2)
Here ∆φ
o−e
(~r
i
) is a fixed spatial phase and ∆φ
v−h
(t) is
the phase produced by the device under test. This phase can
be recovered by applying
b
∆φ
v−h
(t) = arctan
"
P
N
i=1
n
i
˜
I(~r
i
, t)
P
N
i=1
d
i
˜
I(~r
i
, t)
#
(3)
where
b
∆φ
v−h
(t) is the estimated phase and N is the total
number of photodetectors. Each intensity is converted into a
photocurrent and then converted again into a voltage signal
by a resistor R. Both weighted summations of photodiode
signals in the numerator and denominator can be performed
electronically. Therefore, in a post processing stage, the
arctangent can be calculated.
The interferogram provided by the system in Fig.1 can be
very diffuse, depending on the optical components used. In
addition, the interferogram has a large background intensity.
The active surface area of the photodiodes plays a key role
on the performance of the detector. If it is very small, the
effect of speckle in the interferogram will decrease the signal
to noise ratio (SNR) of the detected signal. If the area
of the photodiode is too large, the SNR will also decay
because we lose spatial resolution [13]. Since the position
of the photodiodes can be adjusted, we can use large area
detectors in order to measure the intensity signals (Fig. 1)
without saturation and loss of spatial resolution in expense
of signal strength. Therefore, the current to voltage converter
needs to be sensitive to the intensity variations produced by
∆φ
v−h
(t) under low spatial modulation depth.
A. Characteristics of the current to voltage converter
As we can see in (3), to effectively retrieve ∆φ
v−h
(t),
the current to voltage conversion needs to be linear. The
photodiode response is linear when we sample the pho-
tocurrent under the conditions V
BIAS
≥ 0 and V
BIAS
>
V
OU T
= I
T
R (Fig.1 (B)). The total current provided by
the photodiode is I
T
= I
P H
+ I
Dark
, where I
P H
is the
photocurrent and I
Dark
is the dark current. The dark current
is the current that will flow through the photodiode under no-
light conditions and it increases with V
BIAS
. The bandwidth
(BW ) of the system is determined by the sample resistor
R and the junction capacitance of the photodiode C
P H
,
which decreases with V
BIAS
: BW = 1/(2πRC
P H
). The
noise sources are mainly: shot noise in the photodiodes,
thermal noise in the sample resistor R, and external coupled
interfering signals. A complete study of different noise
contributions on the current to voltage converter is detailed
in [14].
In our experimental setup we used four photodiodes
Vishay BPW43 [15] reverse biased at 9 V with a battery,
each with a sample resistor R = 1 MΩ. The measured SNR
is approximately 80 dB on a 10 kHz bandwidth. A picture
of the detector is shown on Fig.2.
Fig. 2. Detector implemented. The photodiodes are placed into a metal
box to prevent external noise sources from entering the system. [13]
This current to voltage converter has the advantage of be-
ing simple to construct and understand. The main limitation
of this configuration is the bandwidth, which turns out to
be a critical issue when detecting low intensity signals. The
bandwidth can be increased by means of a reduction of the
Revista elektron, Vol. 1, No. 1, pp. 16-22 (2017)
ISSN 2525-0159
17
http://elektron.fi.uba.ar
value of R at the expense of reducing the SNR. Another way
of widening the bandwidth is by increasing the reverse bias
of the photodiodes. However, this increases the dark current
and consequently reduces the SNR of the detector. In the
next section, we analyze the transimpedance amplifier (TIA)
based on operational amplifiers. The TIA is the topology
most frequently used to detect optical signals. Compared
to the current to voltage converter previously described, it
improves the linearity of the photodiode, the SNR and the
bandwidth. However, the analysis of the TIA is far more
complex: the inclusion of an active element and a feedback
network requires careful attention to stability issues and
noise contributors of the operational amplifier.
III. SECOND APPROACH: DETECTION OF HIGH SPEED
PHENOMENA
Consider the experimental setup of Fig. 3. The detailed
description of the system and the operation principles are in
[6].
QWP
1
QWP
2
HWP
PBS
Detection System
He-Ne
632.8 nm
MO
DUT
Mirror
d(t)
Fig. 3. A Polarization Michelson interferometer. HWP: half waveplate,
PBS: polarized beam splitter, QWP
1,2
: quarter waveplates, MO: micro-
scope objective. d(t) is the off-plane displacement of the Device Under
Test (DUT) to be measured. [6]
The physical displacement of the DUT modulates the
polarization state of the light reaching the detector. If there
is a single polarizer in the detection block, we have an
interferogram I(t),
I(t) = A + B cos
4π
λ
d(t) + θ(t)
(4)
where λ is the wavelength of the laser source, d(t) is
the displacement of the DUT and θ(t) is a random phase
difference between the branches of the interferometer. If the
displacement d(t) is less than λ/2, the detected intensity is
I(t) = A + B
cos θ(t) −
4π
λ
d(t) sin θ(t)
(5)
On high speed applications, such as in ultrasonics, d(t)
varies much faster than θ(t). Then we can write I(t) as the
sum of two terms,
I(t) = I
DC
+ I
AC
(6)
where,
I
DC
= A + B cos θ(t) (7)
I
AC
= −B
4π
λ
d(t) sin θ(t) (8)
The displacement d(t) is encoded in the I
AC
component
of the interferometric signal (Eq.(8)) and θ modulates both
I
AC
and I
DC
components. The I
DC
component (Eq.(7))
limits the performance of the detector when measuring high
speed signals by reducing the dynamic range of the sensor,
as well as its effective bandwidth [1], [16], [17]. In summary,
the interferogram consists of a small amplitude, high speed
signal buried in a high amplitude, near-constant background
intensity. Since θ(t) varies randomly, it is difficult to extract
d(t) when measuring small displacements [9]. In conse-
quence, it is necessary to improve the ratio between I
AC
and
I
DC
. For this purpose, we designed the detection scheme
showed in Fig.4.
QWP
I
1
I
2
I
4
I
3
To Acquisition
and Processing system
TIA
Photodiodes
and Polarizer Fil
ms
Lens
Interferometer
output beams
TIA
Fig. 4. Detection Block. The interferograms are produced by polarized
films and detected with photodiodes. Dotted lines indicate the direction of
the transmission axis of the films. The current provided by the photodiodes
is converted into voltage signals v
I,Q
(t) by Transimpedance Amplifiers
(TIAs). [6]
The detector produces four phase shifted interferograms
(similar to the ones in Eq. (2)) that are converted into
voltage signals by TIAs as we show in Fig. 4. The TIAs
are made by operational amplifiers in differential mode. The
output signals of the TIAs are voltage signals that depend
on the difference in the currents of the photodiodes (i.e., the
difference of the interferograms). The output signals v
I
and
v
Q
are, respectively,
v
I
(t) = R
λ
[I
1
(t) − I
3
(t)]R (9)
v
Q
(t) = R
λ
[I
2
(t) − I
4
(t)]R (10)
where the identical resistors R determine the transimpedance
gain of each TIA at the low frequency limit. The photodetec-
tors (assumed identical) have responsivities R
λ
. This way,
the products R
λ
I
N
(t) represent the currents produced by
the photodiodes.
A. Sensor implementation and characterization
The sensor consists of four photodiodes BPW34s [18]
and two FET-input operational amplifiers OPA657 [19] in
transimpedance configuration. Its topology is shown in Fig.
5. The transimpedance amplifier was designed to achieve
maximum transimpedance gain in a 10 MHz bandwidth
and to minimize the total input noise current I
EQ
(i.e.
maximizing the SNR). I
EQ
is measured in A/
√
Hz and
represents the minimum input current, integrated over the
bandwidth of the amplifier, which yielded a SNR of 0 dB
Revista elektron, Vol. 1, No. 1, pp. 16-22 (2017)
ISSN 2525-0159
18
http://elektron.fi.uba.ar
up to the cutoff frequency. In order to achieve a maximally
flat second order Butterworth frequency response, we must
fulfil the following relation:
1
2πR
F
C
F
=
r
GBP
4πR
F
C
D
(11)
where R
F
is the feedback resistance (transimpedance gain at
low frequency) and C
F
is the feedback capacitance. Parallel
to R
F
and C
D
is the total input capacitance at the inverter
pin of the operational amplifier. This capacitance includes
both the photodiode capacitance and the capacitance of the
operational amplifier inputs. GBP is the gain-bandwidth
product of the amplifier. R
F
is calculated to achieve a
bandwidth of 10 MHz using,
f
−3dB
=
r
GBP
2πR
F
C
D
(12)
Instead of introducing an extra component, we take ad-
vantage of the parasitic capacitance of the resistor as the
compensation capacitor C
F
to avoid instability issues. The
total input noise current of the amplifier is,
I
EQ
=
s
I
2
N
+
4kT
R
F
2
+
(e
N
2πC
D
f
−3dB
)
2
3
(13)
where I
N
is the input current noise for the operational
amplifier inverting input and e
N
is the input voltage noise.
Full calculations and design considerations for Eqs. (11),
(12), and (13) are found in [1], [14], [19], [20].
Fig. 5. Electrical implementation of the Quad Optical Sensor. (A)
Schematic Diagram. (B) PCB Diagram. Red color is for top copper, blue
is for bottom copper, vias are green, and the component mask is grey.
The photodiodes are very close to the inverter pin of the
operational amplifiers and the decoupling capacitors are very
close to their power supply. In this work, the transimpedance
is set R = 100 kΩ to maximize the SNR in a bandwidth of
10 MHz. The printed circuit board (PCB) is etched on a FR4
substrate and has ground planes over both sides of the PCB
to reduce electrical noise and interference through ground
loops and to prevent crosstalk between adjacent circuit
traces. The output paths of each TIA (one for photodiodes 1
and 3 and the other for photodiodes 2 and 4) are impedance-
matched to 50 Ω, have equal lengths, and are terminated
with female BNC connectors. The voltage supply pins of
each TIA are decoupled to ground by capacitors. The bias
voltage of the photodiodes is decoupled by a second-order
passive filter (cutoff frequency below 40 kHz). The size of
the PCB is 40 mm × 30 mm and the effective illuminated
area (which includes the four photodiodes) is 7 mm × 7
mm. The sensor developed is shown in Fig. 6.
Fig. 6. Physical implementation of the Quad Optical Sensor. [6]
A separate test setup (Fig. 7) is used to characterize the
electro-optical response of the sensor in the frequency range
from 100 kHz to 10 MHz. A LED light source with
sinusoidal intensity modulation illuminates the four photo-
diodes uniformly. The modulation amplitude and linearity is
verified independently using a ThorLabs PDA 155 optical
detector [21].
Sensor
Modulated
Light Source
Signal Generator
HP 8648A
Oscilloscope
Tektronix TDS210
CH-A
CH-B
Fig. 7. Experimental scheme used for the characterization of the sensor.
[6]
Each photodiode is characterized separately, covering the
others with an opaque screen. At the low frequency limit
(100 kHz), the output signals from the four photodiodes
have the same amplitude within the measurement uncer-
tainty (±0.1 dB). The frequency of the intensity modulation
is swept linearly. The output signal is captured with an
oscilloscope and then processed using a Fourier transform-
based algorithm. A good match between the overall response
of photodiodes and TIAs is achieved along the full frequency
range, as shown in Fig. 8.
The relative amplitude response of the four photodiodes is
shown in Fig. 8 (A). The -3 dB frequency is approximately
8 MHz. There is also a good match of the phase response
and group delay of the photodiodes and amplifiers, as shown
in Fig. 8 (B) and Fig. 8 (C). The phase response is approx-
imately linear up to 3.7 MHz with a slope of 0.5 µrad/s.
In summary, each photodetector and TIA has a flat relative
amplitude and a linear phase response up to 3 MHz. These
results show that we can use the sensor to analyze amplitude
and phase responses of sub-nanometric DMI systems up to
this frequency. In practice, if amplitude response is required
Revista elektron, Vol. 1, No. 1, pp. 16-22 (2017)
ISSN 2525-0159
19
http://elektron.fi.uba.ar
1 10
-8
-6
-4
-2
0
2
1 10
0
1 10
−2.5
−2
−1.5
−1
1
2
3
Frequency [MHz]
Phase [rad]
Group Delay [rad*s]
Relative Amplitude [dB]
Frequency [MHz]
Frequency [MHz]
(B)
(C)
x10
-5
-0.5
-0.6
4
-
1
3
2
4
1
2
3 4
Fig. 8. Measured frequency response of the sensor. (A) Relative amplitude,
(B) Phase, (C) Group Delay. Measurement uncertainties are smaller than
the thickness of the dotted lines. [6]
only, the frequency range may be extended up to 8 MHz.
The measured response agrees with the design simulations.
The differential nature of the circuit (Fig. 5) reduces
all DC components and therefore increases the dynamic
range of the amplifiers without compromising their gain-
bandwidth product [1]. We increase the common mode
rejection ratio (CMRR) of the system by biasing both
photodiodes to the same bias voltage V
bias
. In this case,
the impact on the performance of the detector due to a
high I
DC
component is reduced, since we have a distributed
photodetector configuration where each photodiode receives
nearly a quarter of the total optical power delivered by the
interferometer output. Since the outputs of the photodiodes
are balanced, the output signals of the TIAs are practically
insensitive to stray incoherent light and to intensity modula-
tion of the laser source. This improves the overall robustness
of the system and the signal to noise ratio (SNR) of the
output signal of the TIAs [22].
In the next section, we analyze a design optimized for
wideband applications, such as those found in heterodyne
interferometry.
IV. THIRD APPROACH: WIDEBAND OPTICAL DETECTION
An interferometric signal I(t) is produced by the het-
erodyne polarization Michelson interferometer described in
Fig. 9. The polarization plane of a linearly polarized He-Ne
laser is adjusted by means of a half-wave plate (HWP) to
ensure that only the polarization normal to the plane of the
experiment enters an acousto–optic modulator (AOM). The
AOM is fed by an amplified radio–frequency (RF) generator
which inputs a sinusoid with a frequency f
AOM
= 70
MHz. The AOM is oriented at the Bragg angle to improve
the diffraction efficiency to the first order. The first-order
diffracted beam passes through a Polarization Beam Splitter
(PBS) using a fixed mirror, and serves as the reference beam.
The test beam is focused on the Device Under Test (DUT)
with a microscope objective (MO), passing twice through a
quarter waveplate (QWP) oriented at 45
◦
from the horizontal
plane of the experiment, rotating the polarization plane of
QWP
PBS
He-Ne
632.8 nm
MO
DUT
Mirror
d(t)
AOM
Zero Order
+1 Order
RF Generator
@70 MHz
P
Acquisition and Processing
Fig. 9. A heterodyne Polarization Michelson interferometer. HWP: half
waveplate, AOM: Acousto-Optic Modulator, PBS: polarized beam splitter,
QWP: quarter waveplate, MO: microscope objective. P: Polarizer. d(t) is
the off-plane displacement of the device under test (DUT) to be measured.
the beam in order to reflect into the PBS, recombining with
the reference beam. Using a polarizer (P), we force the
interference between the beams
I(t) = A + B cos
2πf
AOM
t +
4π
λ
d(t) + θ(t)
(14)
Heterodyne techniques offer the advantage of reducing the
number of optical components and noise. The introduction
of a carrier signal f
AOM
allows us to neglect the influence
of 1/f noise and reach the shot noise limit detection
[23]. However, demodulation electronics are more complex
and the bandwidth of the front ends are quite broader in
comparison to the ones in homodyne systems.
A. Front-end design and characterization
An electrical implementation of the front-end is shown in
Fig.10. The PCB was carefully designed to reduce parasitic
capacitance in the sensitive parts of the circuit, such as the
inverting pin of the operational amplifier and the feedback
resistor.
Fig. 10. Front end electrical implementation. (A) Schematic Design.
(B) PCB. Design. The ground plane at the bottom under the operational
amplifier is removed to reduce parasitic capacitance in sensitive pins of the
amplifier and feedback resistor.
To test the TIA, the photodiode (OPF482 OPTEK Tech-
nology [24]) was replaced by its Thevenin electrical rep-
resentation: a highly valued resitor R
P H
and its junction
Revista elektron, Vol. 1, No. 1, pp. 16-22 (2017)
ISSN 2525-0159
20
http://elektron.fi.uba.ar
capacitance C
P H
at the corresponding reverse voltage. The
test setup is shown in Fig. 11.
Signal Generator
HP 8648A
R
PH
TIA
R
F
C
F
C
PH
Oscilloscope
Tektronix TDS2024B
CH A CH B
Photodiode
Thevenin model
Fig. 11. Test setup for wideband TIA characterization. R
F
and C
F
are
the feedback elements of the operational amplifier.
In Fig.12 we show the measured transfer function of two
implemented TIAs. One TIA uses the OPA657 operational
amplifier from Texas Instruments [19] and the other one
uses the LTC6268-10 from Linear Technology [25]. The
photodiode parameters are R
P H
= 22 kΩ and C
P H
= 2
pF in both cases.
10
6
10
7
10
8
79
79.5
80
80.5
81
81.5
82
10
6
10
7
10
8
2.4
2.6
2.8
3
10
6
10
7
10
8
75
76
77
78
79
10
7
10
8
1.5
2
2.5
3
Frequency [Hz]
Magnitude [dB]
Phase [rad]
Frequency [Hz]
Magnitude [dB]
Phase [rad]
Frequency [Hz] Frequency [Hz]
OPA657 - RF: 10KΩ - CF: 300 fF LTC6268-10 - RF: 22KΩ - CF: 100 fF
Fig. 12. Transfer function measurement of developed TIAs.
The measured −3 dB frequency in the first TIA is 140
MHz and has a linear phase response up to 20 MHz. In
the second case, the −3 dB frequency is 320 MHz approx-
imately and has a linear phase response up to 40 MHz.
The LTC6268-10 has a broader frequency response and has
less noise than the OPA657. These characteristics arise from
the fact that the LTC6268-10 has less input capacitance.
However, the OPA657 has a flatter magnitude response. The
peak observed around 160 MHz in the transfer function of
the LTC6268-10 is also seen in [25]. An implementation of
the photodiode TIA using the OPA657 is shown on Fig.13.
Fig. 13. Implementation of a photodiode transimpedance amplifier using
an OPA657 operational amplifier.
A metal enclosure isolates the circuit from external noise
sources. The photodiode is coupled to a multimode fiber by
an ST connector. The output connector of the TIA is a 50Ω
male SMA connector.
V. CONCLUSIONS
We described the design and performance of three tran-
simpedance amplifiers optimized for optical dynamic inter-
ferometry. In the first case, the most significant charace-
teristics of the detector arise from the photodiode param-
eters. In the second case, we present the typical topology
of the operational amplifier TIA, for medium frequency
applications. Moreover, we present an approach to balanced
detection to improve the quality of the measurement and
the resolution of the entire interferometric scheme. Finally,
we present a wideband TIA implementation for heterodyne
optoelectronics. The design considerations are similar to
the second case: the gain and bandwidth must be set to
maximize the SNR in the band of interest, but good practices
in high speed PCB layout must be applied to overcome
parasitics. As a prospective, we will analyze the design of
ultra high speed transimpedance amplifiers for optical pulse
measurements.
VI. ACKNOWLEDGEMENTS
This work is supported by four UBACYT
grants from Universidad de Buenos Aires (2014-
2017 UBACYT 20020130100346BA, UBACYT
2016-2017 20020150200143BA, UBACYT 2017-
2019 20020160100042BA, UBACYT 2017-2019
20020160100052BA). The corresponding author
acknowledges a doctoral scholarship from CONICET.
REFERENCES
[1] J. Graeme, Photodiode Amplifiers: Op Amp Solutions, ser.
Gain technology. McGraw Hill, 1996. [Online]. Available:
https://books.google.co.uk/books?id=sHV0c5hBW4QC
[2] D. Guo, M. Wang, and H. Hao, “Displacement measurement using a
laser feedback grating interferometer,” Applied optics, vol. 54, no. 31,
pp. 9320–9325, 2015.
[3] N. E. Myridis, “Fundamentals of picoscience, edited by klaus d.
sattler,” Contemporary Physics, vol. 56, no. 1, pp. 95–95, 2015.
[Online]. Available: http://dx.doi.org/10.1080/00107514.2014.962096
[4] L. Miccio, P. Memmolo, F. Merola, S. Fusco, V. Embrione,
A. Paciello, M. Ventre, P. A. Netti, and P. Ferraro, “Particle tracking
by full-field complex wavefront subtraction in digital holography
microscopy,” Lab Chip, vol. 14, pp. 1129–1134, 2014. [Online].
Available: http://dx.doi.org/10.1039/C3LC51104A
[5] A. Rosenthal, S. Kellnberger, D. Bozhko, A. Chekkoury, M. Omar,
D. Razansky, and V. Ntziachristos, “Sensitive interferometric detec-
tion of ultrasound for minimally invasive clinical imaging applica-
tions,” Laser & Photonics Reviews, vol. 8, no. 3, pp. 450–457, 2014.
[6] L. M. Riobo, F. E. Veiras, P. A. Sorichetti, and M. T. Garea,
“Wideband quad optical sensor for high-speed sub-nanometer
interferometry,” Appl. Opt., vol. 56, no. 3, pp. 397–403, Jan 2017.
[Online]. Available: http://ao.osa.org/abstract.cfm?URI=ao-56-3-397
[7] S. Kurmulis and L. Novotny, “Common path hetero-
dyne interferometer for single nanoparticle detection in
fluids,” in Frontiers in Optics 2016. Optical Society
of America, 2016, p. FTu2F.5. [Online]. Available:
http://www.osapublishing.org/abstract.cfm?URI=FiO-2016-FTu2F.5
[8] B. P. Abbott et al., “Observation of gravitational waves
from a binary black hole merger,” Phys. Rev. Lett.,
vol. 116, p. 061102, Feb 2016. [Online]. Available:
https://link.aps.org/doi/10.1103/PhysRevLett.116.061102
[9] N. Alvarez, M. T. Garea, and F. P. Quinti
´
an, “Ultrasonic lamb
waves detection using an unstabilized optical interferometer with
uncalibrated phase shifts,” Journal of Modern Optics, vol. 62, no. 7,
pp. 556–559, 2015.
[10] P. K. Rastogi, Digital Speckle Pattern Interferometry & Related
Techniques. New York, NY, USA: John Wiley & Sons, Inc., 2000.
[11] G. P. Yu Fu and X. Li, “Interferometric dynamic measurement:
Techniques based on high-speed imaging or a single photodetector,”
The Scientific World Journal, vol. 2014, pp. 1–14, 2014. [Online].
Available: https://www.hindawi.com/journals/tswj/2014/232906/cta/
Revista elektron, Vol. 1, No. 1, pp. 16-22 (2017)
ISSN 2525-0159
21
http://elektron.fi.uba.ar
[12] F. E. Veiras, L. M. Riob
´
o, C. L. Matteo, L. I. Perez, and M. T.
Garea, “Birefringent phase demodulator: application to wave plate
characterization,” Appl. Opt., vol. 54, no. 9, pp. 2326–2333, Mar
2015. [Online]. Available: http://ao.osa.org/abstract.cfm?URI=ao-54-
9-2326
[13] L. M. Riob
´
o, Dispositivos demoduladores de fase de senales
´
opticas
con cristales birrefringentes. Facultad de Ingenier
´
ıa, Universidad
de Buenos Aires, CABA, Sep 2014, Tesis de grado Ingenier
´
ıa
Electr
´
onica.
[14] M. Johnson, Photodetection and Measurement: Making Effective
Optical Measurements for an Acceptable Cost, ser. Professional
Engineering. McGraw-Hill Education, 2003. [Online]. Available:
https://books.google.com.ar/books?id=K5IKWDwUYZAC
[15] Vishay Semiconductors, “BPW43,”
https://www.vishay.com/docs/81523/81523.pdf, Silicon PIN
Photodiode.
[16] P. S. Matavulj, D. M. Gvozdi
´
c, J. B. Radunovi
´
c, and J. M. Elazar,
“Nonlinear pulse response of p-i-n photodiode caused by the change
of the bias voltage,” International Journal of Infrared and Millimeter
Waves, vol. 17, no. 9, pp. 1519–1528, 1996. [Online]. Available:
http://dx.doi.org/10.1007/BF02088505
[17] M. Dentan and B. de Cremoux, “Numerical simulation of the nonlin-
ear response of a p-i-n photodiode under high illumination,” Journal
of Lightwave Technology, vol. 8, no. 8, pp. 1137–1144, Aug 1990.
[18] Vishay Semiconductors, “BPW34, BPW34s,”
http://www.vishay.com/docs/81521/bpw34.pdf, silicon PIN
Photodiode.
[19] Texas Instruments, “OPA657,” http://www.ti.com/lit/ds/symlink/opa657.pdf,
OPA657 1.6-GHz, Low-Noise, FET-Input Operational Amplifier.
[20] P. Staric and E. Margan, Wideband amplifiers. Springer, 2006.
[21] Thorlabs Inc., “PDA155 - Amplified Si Photodetector,”
https://www.thorlabs.de/drawings/6fd7b2edbf13b66c-3F32185B-
5056-2306-D9F732FD98FB4D18/PDA155-Manual.pdf, pDA155 -
Amplified Si Photodetector.
[22] C. Scruby and L. Drain, Laser Ultrasonics Techniques and
Applications. Taylor & Francis, 1990. [Online]. Available:
https://books.google.com.ar/books?id=KgXPqx9ST-wC
[23] E. Leirset, H. E. Engan, and A. Aksnes, “Heterodyne
interferometer for absolute amplitude vibration measurements
with femtometer sensitivity,” Opt. Express, vol. 21,
no. 17, pp. 19 900–19 921, Aug 2013. [Online]. Available:
http://www.opticsexpress.org/abstract.cfm?URI=oe-21-17-19900
[24] OPTEK Technology, “OPF482,” http://uk.farnell.com/optek-
technology/opf482/receiver-fibre-optic-to-18-st/dp/1497962, OPF482
Fiber Optic Detector.
[25] Linear Technology, “LTC6268-10,”
http://www.linear.com/product/LTC6268-10, LTC6268-10 - 4GHz
Ultra-Low Bias Current FET Input Op Amp.
Revista elektron, Vol. 1, No. 1, pp. 16-22 (2017)
ISSN 2525-0159
22
http://elektron.fi.uba.ar

Enlaces de Referencia

  • Por el momento, no existen enlaces de referencia


Copyright (c) 2017 Lucas Matías Riobó, Francisco Ezequiel Veiras, María Teresa Garea, Patricio Anibal Sorichetti

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Revista elektron,  ISSN-L 2525-0159
Facultad de Ingeniería. Universidad de Buenos Aires 
Paseo Colón 850, 3er piso
C1063ACV - Buenos Aires - Argentina
revista.elektron@fi.uba.ar
+54 (11) 528-50889