Infrared Remote Receiver Circuit Design

An infrared remote link is cheap, simple and remarkably robust, provided the receiver is treated as the analogue device it really is. The module on the board contains a photodiode, an amplifier and a demodulator, and it outputs a clean logic signal only when the incoming light is modulated in the way it expects. Everything else it sees is noise that it has to reject.

How an Infrared Remote Link Works

The transmitter drives an infrared emitter at a carrier frequency, usually between thirty and fifty-six kilohertz, and switches that carrier on and off in bursts that encode the command. The receiver looks for the carrier, strips it off and reproduces the burst pattern as a logic signal for the controller to decode.

Modulation is what makes the link work in daylight. A continuous infrared source such as the sun or an incandescent lamp produces a large but slowly varying photocurrent, while the wanted signal is a burst at tens of kilohertz. The receiver amplifier is tuned to that band, so the steady component is largely rejected before it reaches the detector.

The emitter and the receiver are usually a matched pair in terms of carrier frequency, and a mismatch of a few kilohertz reduces the usable range noticeably. Ordinary remote controls use a small set of standard frequencies, and the receiver is chosen to match the transmitter that will be used with the product.

The Receiver Module

The module contains a photodiode, a bandpass amplifier, a gain control stage and a demodulator, all in a package with an infrared filter and a lens. The filter blocks visible light while passing the infrared band, and the lens concentrates the incoming light on the diode to extend the range.

The output is an open collector transistor that pulls low while a burst is being received, so the logic is inverted relative to the light. It requires a pull up resistor, and the value affects the rise time; a large resistor with a long trace produces a slow edge that the controller may interpret as more than one transition.

A metal shield around the module reduces the effect of electrical interference, and the module should be mounted so that its face is exposed to the incoming light without obstruction. Recessing it behind a dark plastic window is normal, but the window material has to pass infrared light efficiently and must not filter the wavelength the emitter uses.

Carrier Frequency and Modulation

carrier frequency is the frequency of the modulation applied to the emitter, and the receiver is tuned to it with a bandpass characteristic a few kilohertz wide. A transmitter at a different frequency produces a signal at the edge of the passband, which reduces the range and makes the link sensitive to interference.

The duty cycle of the carrier affects the range as well. A fifty percent duty cycle delivers more average optical power than a narrow pulse of the same peak current, but it also draws more average current from the battery. Most transmitters use a duty cycle between a quarter and a half, and the receiver is specified for the resulting pulse width.

modulation also determines how the data is encoded. The common schemes differ in how a zero and a one are represented, and the receiver does not care which is used as long as the burst lengths fall within the range it can handle. The decoding is done in the controller, and the timing tolerance there is usually generous compared with the receiver bandwidth.

Automatic Gain Control and Burst Length

automatic gain control adjusts the amplifier gain according to the average received signal, which is what allows a receiver to work both at close range and at several metres. At close range the signal is large and the gain is reduced, while at long range the gain is increased to recover a small signal.

The control loop has a time constant, and that is why the receiver ignores long bursts. A continuous carrier looks like ambient light to the control loop, which increases the gain until the signal is clipped and then suppresses it. The maximum burst length the module accepts is typically a few hundred microseconds to a millisecond, and the protocol has to keep its longest burst below that value.

The same mechanism explains the blanking time between bursts. After a burst ends, the gain control needs to recover before it can respond to the next one, and a protocol with very short gaps can produce a signal that the receiver merges into one long burst. The transmitter timing has to respect that recovery period.

Ambient Light and Interference

Sunlight contains a large infrared component, and a receiver facing the sun can be saturated even when the transmitted signal is strong. Fluorescent lamps and some LED lighting produce a modulated component at twice the mains frequency, which can fall inside the receiver passband and appear as data.

An infrared filter in the module removes most of the visible light, and the bandpass amplifier removes most of the low frequency component, but neither is perfect. The practical remedies are mechanical: shielding the receiver from direct sunlight with a small hood or a recess, and orienting it so that it does not face a window or a strong lamp.

Electrical interference also matters. A receiver module near a switching supply or a motor drive picks up noise on its supply and on its output, and the noise appears as spurious edges. The methods for keeping that kind of noise out of a sensitive analogue input are described in EMI immunity in mixed signal design.

Layout, Supply and Decoupling

The receiver module draws current in short pulses at the carrier frequency, so it needs a local decoupling capacitor, and it needs to be placed close to that capacitor. A module with a distant capacitor sees a supply that dips on every burst, which reduces the sensitivity and produces a marginal range.

The supply should be quiet and separate from any switching regulator. A small series resistor with a capacitor, or a ferrite bead, forms a filter that keeps the switching ripple out of the module supply, and the selection of that component is described in ferrite bead selection. The module is a sensitive analogue circuit and belongs on the quiet side of the board, as described in mixed signal board design.

Keep the output trace short and away from switching nodes, and place the pull up resistor close to the controller rather than at the module. Where the module is mounted away from the board, for example behind a front panel, the cable carries both the supply and the output, and a shielded or twisted pair reduces the pickup on the output line.

Verification and Common Faults

Verify the link by measuring the range at which the output still shows clean bursts, using the actual transmitter and the actual window material. Measuring the module output with a scope rather than relying on the controller is important, because a receiver that produces marginal edges may still work intermittently and the fault will be attributed to the software.

Measure the supply at the module terminals while it is receiving. A supply that dips on each burst indicates that the decoupling is inadequate or that the supply trace is too thin, and the resulting loss of sensitivity appears as a reduced range that varies with the battery voltage.

A receiver that works with one remote control and not another is usually a carrier frequency mismatch. Comparing the carrier frequency of the two transmitters, or simply testing with several receivers, identifies the cause. A receiver that works in the dark and fails in daylight has an ambient light problem, and the answer is mechanical rather than electrical.

FAQ

Why does my infrared link fail in sunlight? The receiver is saturated by the infrared component of daylight. Shield the module from direct sun and check that the window material passes infrared light.

What is the maximum burst length a receiver accepts? Typically a few hundred microseconds to about a millisecond. Longer bursts are treated as ambient light and suppressed by the gain control loop.

Do I need a shielded cable to a remote receiver? Where the module is on a panel away from the board, yes. The output line is a logic signal with fast edges, and it is easy for it to pick up noise on an unshielded cable.