TSOP IR receiver

Question

I've been working with TSOP IR receivers and emitters on robots since the past year. With multiple robots communication we have signal collision algorithms and such. We still faced a lot of interference from the fluorescent bulbs. We have tried covering the receiver from the top. I need a deeper understanding of the datasheet of Vishay TSOP 34156 receiver. This is the datasheet: https://www.pololu.com/file/0J19/TSOP34156.pdf

These are my questions :

1. What does transmission distance mean in terms of the receiver? As far as I know it doesn't transmit anything.

2. What is directivity? Angle of half transmission distance?

3. Figure 9 : max envelope duty cycle vs burst length?

4. "Burst length should be 6 cycles/burst or longer. After each burst which is between 6 cycles and 70 cycles a gap time of at least 10 cycles is necessary. For each burst which is longer than 1.8 ms a corresponding gap time is necessary at some time in the data stream. This gap time should have at least same length as the burst. Up to 2200 short bursts per second can be received continuously." What is burst length exactly?What exactly do you mean by a cycle? I have tried to find the information on cycles unsuccessfully. This is how my communication protocol looks like.

Legend for the image : signature bit(high) : 1200 uS interval(low) : 300uS zero (high) : 300uS one (high) : 600uS I transmit total of 32 bits consisting of 1s and 0s and intervals in between them excluding the signature wave.

In terms of cycles what does this mean? Where do I stand? within the limits or I have to send shorter bursts of data? Is the TSOP34156 good for a swarm robotic purpose or are there better IR receivers out there that I do not know of?

Answer 1

You asked too many questions in one, so I'll just briefly make a few key points.

Think of the problem these devices have. The dynamic range of light levels they have to work with is enormous, and the actual signal can be small relative to the ambient light level. One way they deal with this is to high pass filter the received light level. That essentially subtracts off the ambient level, hopefully leaving the small variations at the carrier frequency that are the signal.

However, when the light level changes abruptly, is that signal or a change in ambient level? How long is long enough to declare the current state the new normal? The answer is these devices have some sophisticated filtering to try to detect the weak carrier signal superimposed on what could possibly be a much stronger background. This is one reason each device variant is specified for a particular carrier frequency, usually around 40 kHz.

The device therefore is looking for changes from the recent average condition. This means any data you send needs to be in bursts that can't last too long, and must have gaps between the bursts that can't last too long either. The bursts also need to be long enough so that short spikes can be rejected as not being a valid carrier.

The datasheet is telling you that you must send bursts that are at least 6 carrier cycles long, but not longer than 70 cycles, and that you need at least 10 cycles gap between any two bursts. I usually use 10/10 burst/gap minimum.

That still leaves you lots of ways of encoding data. You usually leave the gaps fixed at 10 cycles, and encode data in the 10-70 cycles you get for a burst. Keep in mind that the output won't resolve burst length down to a cycle or two, but it should be quite possible to detect the difference between 10 and 20 cycle bursts, for example.

There are many schemes for encoding data using this low level mechanism, but that's beyond the scope of this answer.

Added in response to comments

One cycle refers to one carrier cycle. Since you are using the version that works with 56 kHz carrier, the easiest way to make the carrier is to drive the transmitting IR LED with a 56 kHz square wave. 1 / 56 kHz = 17.9 µs, which is the period of one carrier cycle. For a square wave, that means the LED will be on half that time and off half that time, or 8.93 µs for each phase.

A burst is a sequence of carrier cycles in a row. A 10-cycle burst is therefore ten 8.93 µs IR pulses in a row, separated by 8.93 µs of no light emitted between each pulse. For the gap between pulses, you need to leave the equivalent of 10 carrier cycles with no light emitted, which means a 179 µs gap.

The way the reciever is built, it is more sensitive to signals received straight into the front. Sensitivity falls off as the IR comes in more from the side. The datasheet tells you the angle at which the effective transmission distance goes is down by a factor of 2. Note that since light level falls off with the square of the distance, this is the angle at which the received signal strength is 1/4 of what it would be if the signal was coming straight face-on.