ETC AN1016/D

Infrared Sensing
and Data Transmission
Prepared by: Dave Hyder
ON Semiconductor
Field Applications Engineer
Many applications today benefit greatly from electrical
isolation of assemblies, require remote control, or need to
sense a position or presence. Infrared light is an excellent
solution for these situations due to low cost, ease of use,
ready availability of components, and freedom from
licensing requirements or interference concerns that may
be required by RF techniques. Construction of these
systems is not difficult, but many designers are not familiar
with the principles involved. The purpose of this
application note is to present a “primer” on those
techniques and thus speed their implementation.
in the form of the 50 or 60 Hz power frequency. Also, recall
that the sensitivity of silicon photo detectors extends well
into the visible range. This sensitivity, albeit reduced,
causes severe interference since the sources in this region
are often of significant power, e.g., incandescent lighting
and sunlight. In addition to the visible component, both
produce large amounts of infrared energy, especially
Some IR applications are not exposed to this
competition, and for them dc excitation of the source may
be adequate. These include some position sensing areas
and slow data links over short distances.
But the bulk of IR needs require a distance greater than
30 cm, speeds greater than 300 baud, and exposure to
interfering elements. For these needs high–frequency
excitation of the source is necessary. This ac drive permits
much easier amplification of the detected signal, filtering
of lower frequency components, and is not difficult to
produce at the driving end. Optical filtering for removal of
the visible spectrum is usually required in addition to the
electrical, but this too is quite simple.
Figure 1 represents a generalized IR system. The
transmitting portion presents by far the simplest hurdle. All
that needs to be accomplished is to drive the light source
such that sufficient power is launched at the intended
frequency to produce adequate reception. This is quite easy
to do, and specific circuits will be presented later.
Usually Infrared
and Filtering
Data Separation
Figure 2 shows the three basic detection schemes: a
phototransistor, a Darlington phototransistor, and a
photodiode. All three produce hole–electron pairs in
response to photons striking a junction. This is seen as a
current when they are swept across the junction by the bias
voltage, but they differ greatly in other respects.
The most sensitive is the Darlington. The penalties are
temperature drift, very–low tolerance to saturation, and
speeds, limited to about 5 kHz (usually much less). Next is
the single transistor, having similar penalties (but to a lesser
degree), with speeds limited to less than 10 kHz. Typically,
they are limited to less than half that number. These two
detectors normally find their use in enclosed environments,
where ample source intensity is available to provide large
voltage outputs without much additional circuitry (their
prime advantage). Their detection area is almost never
exposed to ambient light.
Figure 1. Simplified IR Sensing/Data Transmission
The bulk of the challenge lies in the receiving area, with
several factors to consider. The ambient light environment
is a primary concern. Competing with the feeble IR
transmitted signal are light sources or relatively high
power, such as local incandescent sources, fluorescent
lighting, and sunlight. These contribute to the problem in
two ways. First, they produce an ambient level of
stimulation to the detector that appears as a dc bias which
can cause decreased sensitivity and, worst of all, saturation
in some types of detectors. Second, they provide a noise
level often 60 dB greater than the desired signal, especially
This document may contain references to devices which are no
longer offered. Please contact your ON Semiconductor representative for information on possible replacement devices.
 Semiconductor Components Industries, LLC, 2002
July, 2002 – Rev. 1
Publication Order Number:
a. Phototransistor
b. Darlington Phototransistor
c. Photodiode
Figure 2. The Basic Detectors for IR Photosensing
For the general case of remote control or sensing at
distances greater than 30 cm, the vast majority of
applications utilize an LED source switched at a carrier
frequency of 20 kHz to 50 kHz and a diode detector coupled
to ac band–limited amplifiers. Although certainly more
complex than the simpler short–distance sensors, today’s
product offerings make it an easy task to achieve 10 meters
with a data rate of around 5,000 baud at very modest cost.
The transmission end is easily configured. Figure 3 shows
a simple IR source capable of 50 kHz transmission. Note
that no special techniques are needed to switch the diode at
these frequencies. A burst of high frequency is created for
each bit time in the data being sent. This mode of gating a
carrier on and off is known as continuous wave (CW).
In virtually all remote–control applications (implying
distance), the diode is the detector of choice. This is due
primarily to its near–freedom from saturation, even in most
sunlit environments. The penalty is sensitivity, often in the
nanoamp or low microamp region, but balanced by
response speed in the nanosecond range. This permits
transmission frequencies in the 50–100 kHz area,
providing ample data rates, inexpensive amplification, and
easy filtering of noise.
Many applications in position sensing lend themselves
well to the sensitive, if slow, nature of phototransistors.
When a go, no–go situation exists, these provide a simple
solution provided that ambient light is not present at the
detector. The designer must ensure that the system operates
even if this portion of the equipment is exposed, as by
opening a hatch during servicing or final adjustment during
production. This is often achieved via covers, tubes
limiting light paths, or that enough directionality exists in
the basic device construction to provide the needed
isolation. Also available for this application are logic–level
output devices, usually of the open–collector type, making
processor or logic interfacing convenient.
The light source for these uses is chosen primarily by the
distance needed. LEDs work well up to about 5 cm. Above
this, incandescents are often used due to their high output
and ease of drive with low–voltage ac. Fluorescent sources
are seldom adequate due to their “cool” color temperature
compared to incandescent. That is, not enough output in the
near–infrared or infrared portion of the spectrum.
Data can be transmitted in these short distance situations,
provided the speeds required are not great. An example is
the electrical isolation of two adjacent PC boards in a rack,
with IR elements facing each other across the short space.
Here the data can be used to drive the LED directly;
modulating a high frequency is not necessary.
Speed and sensitivity are the tradeoffs. The resistor used
to develop a voltage can be made larger to provide
increased sensitivity, but speed suffers and tendency
toward saturation increases. Values of 50–200 Ω are
common, but can be higher.
Data To Be Sent
50 kHz Square Wave
Figure 3. Basic IR Source Drive for CW Operation
The main areas of interest are the switch device and the
diode current. Today’s infrared emitting diodes (IREDs)
are generally capable of around one ampere peak currents,
but applications typically limit this to half that value. Most
designs that use a 50 percent duty cycle square wave
switching waveform have diode currents in the 100–500 mA
range. It is important to realize that although IRED output
increases linearly with drive current, it drops rapidly with
increasing temperature. Therefore, reliability is not the
only reason for resisting the temptation to increase range by
driving the IRED harder. A diode with a 100 mA
continuous rating can be reliably driven with a 200 mA
square wave, and so on. It is quite common to use more than
one IRED in series for increasing output and range,
lowering the current requirements, and increasing
reliability of the diodes.
The driver device can be a bipolar transistor or a FET.
The bipolar works fine, but requires enough base current
for saturation that the driving circuitry often must provide
10–20 mA or more. This may not be available directly from
CMOS devices. Darlingtons solve this problem, but are
usually much too slow. Another solution is an inexpensive
logic–level FET such as the MTP3055EL, its physically
smaller cousin, the MTD3055EL, or an MTP4N06L. This
provides plenty of speed while being driven directly from
any CMOS device, with absolute minimum parts count. A
resistor (50–500 Ω) is sometimes used in series with the
gate to moderate the very–high switching speed and noise
from high frequency oscillations. The resistor is usually not
needed if the gate is driven from a medium–speed CMOS
gate such as the MC14081B or MC14011UB.
methods. First, coupling capacitance values are judiciously
chosen to begin rolloff just below the transmitted
frequency. This is quite effective since the area of interest
is usually about a factor of 103, or some 9 to 10 octaves
above the power–line frequencies. The second method is to
use explicit high–pass filter circuitry, but in practice this is
seldom needed due to the effectiveness of the other
techniques. A third option is to use a bandpass amplifier,
usually with an LC tank. More discussion of this later.
After the signal is brought up to a level sufficient for
detection, some method must be employed to extract the
data. Most common is a simple peak detector. This detects
the presence of the high–frequency pulses, charging a
capacitor up to a threshold in a few cycles, at which point
a comparator signals the new level. In the absence of a
signal (the carrier), the capacitor discharges until the
comparator ’s lower threshold is reached, signifying the
opposite logic level. Other techniques are also available,
such as the phase–locked loop, whose lock–detect output
can be used as the recovered logic–level data.
The Receiving Process
At the receiving end, the first item encountered is an IR
optical filter as shown in Figure 4. This serves the sole
purpose of attenuating the visible portion of the spectrum
while leaving the IR intact. It can be a material specifically
designed for the purpose, such as the Kodak filter series,
but is usually an inexpensive acrylic plastic. This is almost
any readily–available red, non–opaque plastic. Suitability
is easily proven by inserting a sample between an emitter
and detector while observing the detector output. The IR
signal should be minimally altered. This filter may be
incorporated into the system as a unique piece of the
material in front of the detector, or the entire front panel of
the product may be made of this plastic. Sometimes lenses
are actually molded from it (discussed in a later section).
More on Receiving Circuits
Two general methods are used to begin the amplification.
First the diode light current (a few microamps or less) may
be used to develop a voltage across a series resistance,
which is then capacitively coupled to the amplifier using
the rolloff of low frequencies mentioned above, as shown
in Figure 5a. Second, the current may be driven directly
into the amplifier, as in Figure 5b, where the photo current
is summed with the feedback current at the amplifier input.
Note that in these and other figures, the amplifier symbol
does not necessarily denote an actual integrated
operational amplifier, but may symbolize a discrete
IR Filter
Amplifier &
HP or BP Filtering
Figure 4. Basic IR Receiver
The detector diode behind the filter is usually
constructed as a large–geometry device specifically
designed for IR remote control, and presents a large area
simply for more IR energy absorption or increased
aperture. It is not unusual to find the material used for
encapsulation to be red or black, and apparently opaque.
The encapsulation serves as an IR filter, as in the case of the
MRD821. Even so, an additional one is usually employed
as mentioned above, often for the cosmetics of the product.
In addition to visible–light filtering mentioned above,
electrical filtering must be applied to greatly attenuate the
low–frequency interference present in both the visible
spectrum and the IR. This is accomplished by three
a. Capacitively–Coupled Front End
b. Direct–Coupled Diode Front End
Figure 5. Front–End Amplifier Options
1.0 mH*
0.01 µF
common mode range includes ground, permitting the diode
or the other amplifier input to be referenced there. If greater
gains are needed, and higher supply rails are available, the
MC34082 series provides slew rates of 25 V/µs, or twice
that of the MC34083. These operational amplifiers in
general do not have the low–noise performance of discrete
versions, with the above devices being in the 30 nV/ Hz
region. However, the MC33077 provides excellent noise
performance of about 4.5 nV/ Hz at a similar slew rate on
a 5 volt supply, although its common mode range does not
include ground. A simple discrete amplifier example is
shown in Figure 7.
Another option that should be considered for data
reception is the MC3373 (Figure 8), which integrates many
of the functions already described. This device contains the
front–end amplifier, a negative–peak detector with
comparator, and requires only a few external components.
The amplifier may have the diode directly connected to it,
or ac coupled for purposes of rolloff. A tuned circuit can be
used for the better noise performance of a band–limited
system. Some words of caution: supply bypassing close to
the device, particularly at the gain–determining impedance
(resistance or tuned circuit), is critical. Without proper
bypassing, gain and range suffer. Also, a higher supply
voltage of around 12 volts or so assists in greater range
The vast majority of IR links in consumer products
(VCRs, TVs) use an LC tank. The inductor is a shielded,
adjustable slug type in the 1–5 mH range. Shielding in the
form of a metal can usually encloses the entire
subassembly, and the designer should expect to employ
such shielding in most applications requiring moderate or
long distance operation.
Note that in Figures 7, 8, and 9 the bias supply to the
receiving diode is heavily decoupled from the supply via an
RC. Any noise present at this point directly impacts system
noise and sensitivity. Bandwidth is also often limited at the
upper end as an aid in overall noise performance as seen in
Figures 7 and 9. These amplifiers use small capacitors (33 pF,
10 pF, 100 pF) to roll off frequencies above 100 kHz.
*Toko type 10 PA or equivalent. Available from Digi–Key
Corporation, phone 800–344–4539.
Figure 6. Amplifier Chain Showing 50 kHz Bandpass
Filter Second Stage
+5 V
1.0 k
4.7 k
1.0 µF
100 k
1.0 k
1.0 µF
300 k
33 pF
Figure 7. Simple Discrete Front End with Op Amp
Figure 6 shows an amplifier system coupled to a
bandpass amplifier centered about 50 kHz. Here the front
end is actually an operational amplifier, used in the mode
of Figure 5b. Various choices for operational amplifiers
exist; perhaps the first hinges on the supply voltage. Some
recent advances in the technology have greatly increased
slew rates and gain–bandwidth products. This has
permitted devices that are capable of operation on a single
5 volt supply, yet can be used in the 50 kHz range. An
example of this is the MC34072 series, whose input
+12 V
4.7 µF
1.0 k
+5 V
1.0 mH
0.01 µF
82 k
4.7 µF
18 k
5 V Logic Out
100 k
7 6
0.001 µF
4.7 µF
Figure 8. IR Receiver Using the Integrated MC3373
10 k
+12 V
4.7 µF
20 k
100 k
10 k
100 µF
10 pF
6.8 k
2.2 k
100 pF
10 µF
10 k
Figure 9. High–Performance Discrete Front–End Amplifier with Special Attention Paid to Noise
When the distance to be covered extends beyond 10
meters or so, other methods must be considered. The
methods described below have resulted in ranges of 100
meters or more.
At the transmitting end, most of the options available
center on increasing the power output. One way is to
increase the IRED current, but this is subject to limits as
perviously discussed. Another solution is to use multiple
diodes in series, often three. Note that this does not require
additional supply current. Multiple diodes also provide one
solution to those applications requiring less directionality,
with the IREDs being slightly misaligned from one another.
The diodes can also be driven much harder, and produce
proportionally higher instantaneous power, if they are
pulsed with a very short duty cycle. Currents of about an
ampere are common, but for only a few microseconds and
with a duty cycle of 5 percent or less. This also requires
modified receiving techniques.
At the receiving end, most solutions center on increasing
the aperture of the system such that simply more energy is
gathered. Multiple receiver diodes can be connected in
parallel, adding their currents, with the additional
possibility of reducing directionality if needed. Another
technique is to add a lens, with the diode being placed at the
focal point. In higher volume production, this is often
molded into a front panel and is usually of the red filtering
plastic mentioned earlier. Some systems make use of a flat
Fresnel lens, being somewhat more difficult to mount but
very effective. They can also be hidden behind a plastic
Front–end amplifiers superior to the simple operational
amplifier or discrete versions already mentioned may be
found in these highest–performance situations. Such an
amplifier is shown in Figure 9, where low–noise transistors
are used in a circuit designed specifically for low–noise
When pulsed sources are used, some encoding scheme is
normally used to transmit the data. One common technique
is to use a single pulse for one edge of a data bit, and two
or more closely spaced pulses to signal the opposite edge.
These are simply differentiated by some flip–flops and a
small amount of timing circuitry. Other schemes use
multiple pulses at close intervals to indicate one logic level,
and a differing number to denote the other.
One last option is sometimes seen at the end of the
amplifier chain and used for the data detection. An analog
phase–locked loop circuit can be used to pull a signal from
noise and lock to it if appropriate. This lock signal is then
used as the recovered data stream. One such device, shown
in Figure 10, is the EXAR XR567, a small 8–pin tone
decoder with both Type I and Type II phase detectors. It is
capable of locking to analog signals in the 25 mV range,
and makes/breaks lock at a rate sufficient for about 5,000
baud with 50–100 kHz inputs. The device can be operated
up to about 500 kHz.
+5 V
Analog In From
0.01 µF
4.7 k
0.01 µF
0.1 µF
20 k
Lock Signal Used
As Recovered Data
1000 pF
Figure 10. PLL Tone Decoder Used to Recover Data From Analog
As can be seen from the above discussion, IR links have
become quite easy to implement. With the basic principles
in mind, the designer should be able to adapt the techniques
mentioned here to his specific system needs.
An advantage of the all–analog system is that the signal
never needs to be amplified to the point of rail–to–rail
limiting. Thus, system–wide noise potential is decreased.
Back–to–back diodes or similar methods are normally
employed ahead of the loop input to hold the signal within
a few hundred millivolts to protect against overdrive at
close ranges.
Figure 11. Utilizing ON Semiconductor’s Encoders and Decoders
ON Semiconductor and
are registered trademarks of Semiconductor Components Industries, LLC (SCILLC). SCILLC reserves the right to make
changes without further notice to any products herein. SCILLC makes no warranty, representation or guarantee regarding the suitability of its products for any
particular purpose, nor does SCILLC assume any liability arising out of the application or use of any product or circuit, and specifically disclaims any and all
liability, including without limitation special, consequential or incidental damages. “Typical” parameters which may be provided in SCILLC data sheets and/or
specifications can and do vary in different applications and actual performance may vary over time. All operating parameters, including “Typicals” must be
validated for each customer application by customer’s technical experts. SCILLC does not convey any license under its patent rights nor the rights of others.
SCILLC products are not designed, intended, or authorized for use as components in systems intended for surgical implant into the body, or other applications
intended to support or sustain life, or for any other application in which the failure of the SCILLC product could create a situation where personal injury or death
may occur. Should Buyer purchase or use SCILLC products for any such unintended or unauthorized application, Buyer shall indem nify and hold SCILLC and
its officers, employees, subsidiaries, affiliates, and distributors harmless against all claims, costs, damages, and expenses, and reasonable attorney fees
arising out of, directly or indirectly, any claim of personal injury or death associated with such unintended or unauthorized use, even if such claim alleges that
SCILLC was negligent regarding the design or manufacture of the part. SCILLC is an Equal Opportunity/Affirmative Action Employer.
Literature Fulfillment:
Literature Distribution Center for ON Semiconductor
P.O. Box 5163, Denver, Colorado 80217 USA
Phone: 303–675–2175 or 800–344–3860 Toll Free USA/Canada
Fax: 303–675–2176 or 800–344–3867 Toll Free USA/Canada
Email: [email protected]
JAPAN: ON Semiconductor, Japan Customer Focus Center
4–32–1 Nishi–Gotanda, Shinagawa–ku, Tokyo, Japan 141–0031
Phone: 81–3–5740–2700
Email: [email protected]
ON Semiconductor Website:
For additional information, please contact your local
Sales Representative.
N. American Technical Support: 800–282–9855 Toll Free USA/Canada