Infrared Audio Link - Engineering Seminar

Design Project

Goal: Design, simulate, and build a transmitter and receiver for an infrared audio remoting unit. The transmitter will have as its input a signal with an amplitude .1v to 1v peak and a frequency from 150Hz to 5KHz. The input will be converted to a pulse width modulated signal which will then drive an infrared LED. The transmitter will illuminate a phototransistor in the receiver. The received signal will then be demodulated and amplified so as to be capable of driving an 8 ohm speaker to at least 20mv peak. The maximum range of the system must be at least 1.5 meters. The demodulated signal amplitude must be independent of range up to the maximum range limit. The transmitter and receiver will use no more DC power than 600 mw each.

Overview 

Cordless headphones and TV remote controls often make use of infrared (IR) light to make the desired links. Infrared light emitting diodes (LEDs) and phototransistors of the type to be used in this project are the most inexpensive type. They operate at a wavelength of 940 nanometers. Higher performance fiber optic local area networks generally use devices that operate at wavelengths of around 840 nanometers. The highest performance links, such as fiber optic telephone links use lasers and photo diodes with operating wavelengths of 1300 and 1550 nanometers. By contrast visible light has wavelengths from 450 to 750 nanometers. So you will not be able to see the light emitted by the devices used in this project.

The LED voltage/current relationship is just like any other diode. However, when forward biased, it emits light. The more forward biased it is, the more light it emits. The design in this project will vary the current going through the diode and thus vary the light output. In this case the diode will be modulated (by turning it on and off) by a series of pulses with varying widths. The LED has a maximum forward current that you will not want to exceed. It also emits its maximum intensity in one direction.

A phototransistor is just a regular transistor except that instead of the collector current being controlled by the base current , the collector current is controlled by the amount of light incident on the base. See reference for more background. If one visualizes the typical transistor curves (Ic vs. Vce for a given IB), the same set of curves occurs for a phototransistor except that each of the curves is parameterized by incident light power instead of base current. Most phototransistors will respond to light with wavelengths which span a fairly broad range (750 nm to 980 nm for example) and therefore the phototransistor in your receiver will respond to ambient light in addition to the light emitted from your transmitter. This extra light just adds noise to the receiver and is thus undesirable. The sun, for example, is one source of such noise. Fluorescent lights also emit significant 950 nanometer light which is modulated around 100 Hz. As was the case for the LED, the phototransistor is directional.

A pulse width modulation scheme is to be used in this project. Refer to references for a more detailed description. Basically a pulse width modulation system converts an input signal to a train of pulses with equal amplitude, but with pulse widths that vary in proportion to the input signal amplitude. A detailed analysis of the frequency spectrum of the resulting signal is quite complicated, but it turns out that a simple low pass filter will demodulate such a signal.

Design Procedure

Figure 1 shows a block diagram of the transmitter/receiver. This project can be divided into three sections: (i) the pulse width modulator; (ii) the LED transmitter and phototransistor receiver; and (ii) the lowpass filter/speaker driver.

(1) The pulse width modulator consists of a triangle generator and a comparator. An oscillator that will deliver an approximate triangle wave can be designed using a single op-amp. An oscillation frequency of roughly 20 KHz is suggested. The oscillator plus the comparator is the general topology for creating the desired modulation. However, you will need to determine the design details such that your design will cover the specified frequency and amplitude range.

(2) The transmitter part of the transmitter/receiver is relatively simple. All that is needed is a BJT (or FET) switch that will switch current through the LED when the pulse is high. You must be careful not to exceed the power rating of the transistor when it is on. You also must not exceed the maximum current allowed in the LED. Depending on how much current your comparator will source, you may need an additional stage of amplification.

The receiver part of the transmitter/receiver consists of a phototransistor plus possibly some gain stages. The phototransistor is a somewhat slow device so it should be connected like a common collector amplifier stage in order to maximize its frequency response. To keep from loading the phototransistor too much, it should be followed by a high input impedance gain stage. It is important to keep the frequency response as high as possible. Otherwise the sides of the received pulse will start to slope and this will distort the pulse width. The gain stages can be used to reshape the pulses such that pulse amplitude is independent of the distance between the receiver and the transmitter. How much extra gain will be necessary to meet the 1.5 meter range specification will probably need to be determined empirically.

 (3) The low pass filter can be a two to four pole filter that starts cutting off at 5 to 10 KHz. One way to achieve this is with an op-amp circuit, but you may want to use discrete components to give you a reduction in power drain. The speaker will be a 0.2 watt 8 ohm speaker. Your design should drive the signal voltage at the speaker to a 20 mv level when the transmitter and receiver are separated by the full 1.5 meter distance and the input signal is at 0.1v.

(4) According to the specifications, your design should be such that the minimum detectable input signal (the sensitivity) is 0.1v. Signals smaller than that are buried in the unwanted noise generated in the receiver, by ambient light, or other sources. We will define the minimum detectable noise by saying that when the minimum detectable signal is applied, the signal to noise ratio at the output is no worse than 1.0. Normally one would use a noise meter to determine the sensitivity, but you can estimate the signal to noise ratio by putting a 0.1v peak sinusoid in the input and observing the signal at the speaker terminals with an oscilloscope. If the average amplitude of the noise is about the same as the RMS amplitude of the signal, the signal to noise ratio is very roughly one.

The criterion used to determine the maximum signal level varies depending on the application. For this project we specify that your design should show clipping of no more than 20% of the peak output voltage for input signal levels up to 1.0 v. The ratio of the maximum to minimum signal levels is called the dynamic range. Therefore the specifications on this project include a sensitivity of 0.1v and a dynamic range of 20 db.

(5) Your report should include: a brief discussion of your design procedure, selected PSPICE simulation results (put the net lists in an appendix), a neatly drawn circuit diagram, a parts list and cost estimate, plots of measurements that demonstrate receiver operation over specified parameter range (speaker voltage amplitude versus frequency over the specified range), discussion of cost/power/complexity trade-off between discrete BJTs, FETs and op-amps.