Using TSOP31240 in parallel

[Tvmender]

Hi All

Firstly, thank you for having me on the group. I am a relative newbie to PICAXE although I did once do a full qualification in MPASM many years ago. I am currently working on a few PICAXE projects.

My question is: Can I use the TSOP31240 directly in parallel (I need 3 this way) or do they need separate input pins on the PICAXE or buffering circuitry between them?

Many thanks!

 

[erco]

I have not used that exact sensor but others have used a buffer, like this AND gate application: https://teslaandi.wordpress.com/2014/01/18/using-multiple-ir-receivers-with-arduino-to-increase-receiver-coverage/

 

[AllyCat]

Hi,

Welcome to the forum. IMHO a (N)AND gate is unlikely to do anything "useful". That sensor appears to be intended for optional (Active Low) "Wired OR" connection, i.e. either "sensor1" OR "sensor2" OR "sensor3" can pull the (linked) output pins Low. A (N)AND gate will behave in exactly the same way, i.e. any input going Low can "Activate" its output (or Deactivate it, according to your point of view). The main difference is that a NAND gate will give an "Active High" output (i.e. "True" = binary "1").

Presumably the reason for using multiple sensors is to increase the "field of view" (it's unlikely to increase the sensitivity for a "direct" IR signal)? The only issue might be if any (one) of the sensors is subject to "interference" from another IR source. That could potentially block reception of the "desired" signal, whether you use a Wired-OR or a "real" OR/AND gate. In that case you would need separate inputs from each sensor, and separate decoding software inside the PICaxe for each channel, which could be quite problematic, not just for a PICaxe.

Cheers, Alan.

 

[inglewoodpete]

Yes, you can use three (or more) IR receivers wired in parallel and mounted back-to-back to make a 360-degree receiver (on, say a remote controlled toy vehicle or robot).

Use a single pull-up resistor to give an idle-high output. When one or more IR receivers detect a modulated IR signal they will pull the output line low while the signal is present.

 

[hippy]

The TSOP3120 datasheet shows the outputs are open-collector with an internal 30K pull-up to +V - https://www.vishay.com/docs/82492/tsop312.pdf - so parallel wiring is possible as inglewoodpete describes.

If you had a lot of sensors and the paralleled internal 30K's were being reduced too much one could diode mix the outputs -

        -.------------.-----------.--- +V
 .-.     |    .-.     |          .|.
 | |--.  |    | |--.  |          |_|
(| |--|--'   (| |--|--'           |
 | |--|--.    | |--|------|<|--.--^--> Input Pin
 `-'  |  |    `-'  |           |
      |  `---------|------|<|--'
      |            |
     -^------------^------------------ 0V

Yay! ASCII Art is back

 

[WhiteSpace]

This is interesting - I have been using parallel TSOPs for remote control for my vehicle project, but I had assumed that if they all fed into the same Picaxe input pin, there would be a risk of bits arriving separately from the two TSOPs - perhaps with partial interruptions, or bounced off the other side of the room - and being read as part of the same byte and therefore causing errors. I have therefore had them come into separate pins, and then used the Picaxe to look for a good byte first on one pin and then if not found, on the other. It sounds as though that might not be necessary?

 

[inglewoodpete]

This is interesting - I have been using parallel TSOPs for remote control for my vehicle project, but I had assumed that if they all fed into the same Picaxe input pin, there would be a risk of bits arriving separately from the two TSOPs - perhaps with partial interruptions, or bounced off the other side of the room - and being read as part of the same byte and therefore causing errors. I have therefore had them come into separate pins, and then used the Picaxe to look for a good byte first on one pin and then if not found, on the other. It sounds as though that might not be necessary?

The IR is 'light', so travelling at about 300,000km/second. So, even with a signal bouncing around a room would probably arrive picoseconds after the original, incident signal. Also, I understand that some IR receiver/demodulators have a full open collector/drain output, so there is no need for diodes to separate multiple outputs.

 

[hippy]

Corruption from reflections is more likely if using IR sensors of different models or manufacture. Their AGC response and signal latency is likely to have more adverse effect than bounced light. More likely a problem if busking it with random sensors rather than buying however many you need from the same supplier.

 

[Tvmender]

Hi All

Many thanks for your replies! I am going to have a go at paralleling three together and see what I get.

I did some further research after my post and did stumble across a diagram showing them connected in parallel but it took some searching.

Again, many thanks for your replies and assistance!