0
\$\begingroup\$

We are trying to have a moving IR receiver pickup a specific IR-led for a range of IR-LEDs.

The options we're looking at :

  • Send and unique pwm signal from each different led, the moving object knows then at which led it is by receiving the id
  • light up each subsquent led exactly once at a timebased interval, the moving object and ledstrip are synced in time so that we know ata given point in time at which led the moving object is

Questions we have about these two methods:

  • is it possible have a ledstrip of reasonable length which would be able to produce these patterns
  • is there anything that can drive this ledstrip for both methods
  • what would be the max length we can have for such a ledstrip
  • which type of IR Leds should we use to transmit to a length of up to 7meters
  • what kind of IR receiver is best suited for this
\$\endgroup\$
9
  • \$\begingroup\$ You have leds, IR leds and led strips. I'm confused as to which you may be referring to. How about a picture? \$\endgroup\$
    – Andy aka
    Commented Mar 16, 2015 at 10:46
  • \$\begingroup\$ We only have IR leds and IR receivers basically, the ledstrip should consist of IR leds. This is purely theoretical, are these options viable and what should we look for if we try and make such a setup \$\endgroup\$
    – mesger
    Commented Mar 16, 2015 at 10:59
  • \$\begingroup\$ Why don't you simply lit up the whole led strip? If I understand correctly, you want to transmit a IR signal / data to a moving object (circling ?) You want to optimize the transmission, by only lighting up those LEDs, that are pointing towards the transmitter at any given point of time? \$\endgroup\$
    – DThought
    Commented Mar 16, 2015 at 11:15
  • \$\begingroup\$ the object needs to know which ir-led its at, so we have thought of 2 possibilities : time based (each ir-led lits up once at a time and based on that timeinterval we know which led it's at) and unique id's sent by every ir-led. \$\endgroup\$
    – mesger
    Commented Mar 16, 2015 at 11:20
  • \$\begingroup\$ Is the movement linear & static? then i'd try the time-based aproach. Now your PWM makes sense, as well.. LED 0 will be 1% on, LED 360 will be 99% on... The receiver simply needs to measure who long it is lit and has an approximate location? The problem with that approach is that you won't get signal of 1 led, but of many, while you're moving... \$\endgroup\$
    – DThought
    Commented Mar 16, 2015 at 11:25

1 Answer 1

Reset to default
1
\$\begingroup\$

It is customary to ask one question per question, but as your questions are related to the overall design I'll try to answer them more or less in combination.

What you want is certainly possible, but don't expect any off-the-shelf LED assembly to fit your project: you'll have build it yourself, including the system (probably a microcontroller) that drives the LEDs.

7 meter distance is possible with normal IR LED and normal TV-remote style IR receiver modules (google for instance TSOP IR receiver), but it probably requires a good driver circuit for the IR LED.

Assuming a 7 meter window and 20m/s means that you have ~ 350 ms within the window. That is enough to transfer a few bytes of data (an RC5 message is ~ 25 ms), but reliability will be far from 100%.

You will have to consider spillover from nearby windows, much like cellphones have to do. This will eat up some of your bandwidth: the signals from IR LEDs that are next to each other must not overlap, and there must still be enough silent time to satisfy the IR receiver's requirements.

Another approach would be to use IrDA-style send/receivers, which would enable a much higher bandwidth/resolaution, but I doubt 7 meter would be achievable.

\$\endgroup\$
5
  • \$\begingroup\$ I might not have been entirely clear. The distance forward a led should reach is about 7 meters but there is only 2m of sideways distance between the leds. Maybe we can help guide it with a small cap around them. As for not overlapping signals do you mean certain pwm signals that easily mix or just the physical boundary of the beams ? \$\endgroup\$
    – mesger
    Commented Mar 16, 2015 at 17:10
  • \$\begingroup\$ You want to get 2 m resolution at 7 m distance? Might not be impossible, but my gut feeling is that it won't be easy. \$\endgroup\$
    – Wouter van Ooijen
    Commented Mar 16, 2015 at 18:29
  • \$\begingroup\$ we can play with the 7m distance maybe 2m is enough aswell, all depends on where the IR leds are positioned. \$\endgroup\$
    – mesger
    Commented Mar 17, 2015 at 11:32
  • \$\begingroup\$ Then my gut feeling is that it could work. But you won't get a detailed design from this forum, you will need to start reading up on IR remote protocols, and IR LED and IR receiver datasheets, do some calculations, learn to program a microcontroller, and start building a prototype. Or get (=pay) someone to do this for you. \$\endgroup\$
    – Wouter van Ooijen
    Commented Mar 17, 2015 at 11:36
  • \$\begingroup\$ We already made a test with 4 different arduino's, 3 are used to each transmit a unique id and one is used to receive. We could get it to work at around 12m/s with a timing of 40ms between messages from the transmitters, we adjusted the timings and message protocol to around 5ms between each message and still need to test it at greater speeds. What question I still have is, is it possible to have some driver multiplex those signals to each different ir led there are about 600 ir leds in total.what about cablelength vs timing of the signals can they transmit by multiplexing on a 5ms basis? \$\endgroup\$
    – mesger
    Commented Mar 17, 2015 at 12:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.