Hello,

I have some questions regarding my implementation of IR LED detection with a phototransistor.

Idea
The main idea of the circuit is to make a circuit with incorporating a phototransistor, which will create a 0-5V signal on its output whether it detect IR LED presence. IR LED will be pulsed at approximately 1kHz square wave.

Requirements
The main requirement is that this circuit needs to have SMD phototransistor. I cannot incorporate a solution from Vishay (TSOP series) as my IR reception board is already made.

Implementation
My current implementation consists of a following setup:
1. Phototransistor is wired on a SIGNAL net with a 47k pull-up resistor. The output is then passed through a circuit, which effectively AC couples the signal (removes DC voltage presence - ambient light). Note, the capacitor was replaced with 10u.
2. The AC component on a SIGNAL net has 50mV peak-to-peak value measured on oscilloscope when IR LED is 5cm away from phototransistor. Through an amplifier circuit we amplify input signal and pass it to the INPUT net.
3. ENABLE net is to 'enable' the INPUT signal to be passed on to the next stage. It works with inverted logic.



Phototransistor used: Datasheet
IR LED used: Datasheet

Issue
The main problem I am having is that the distance of where phototransistor starts detecting IR LED presence is too small.
1. The desirable distance of IR LED detection should be 1m. Are my goals possible with components selected above?
2. Is there any way to increase phototransistor sensitivity? I am wondering, how can TV remotes achieve such long reception distance? Is the main reason the high-power LED in the remote or is it something else?

Thanks to anyone who will take a little time to help my research.

Best regards,
Marcel

 

The second stage of the amp needs to be AC coupled as well and both stages need to be biased properly to get good sensitivity. It also needs a 1KHz bandpass filter so, for example, LED or fluorescent lighting does not trigger it.

The TSOP modules do all of that and more.

If you must DIY it, use opamps for amplification and filtering. And use a higher frequency.

 

I would drop the LED modulation Frequency down to maybe ~100hz,
there appears to be no reason for it to be so high.

I'm guessing that You are depending upon the IR-Light to be reflected
off of the objects that You would like to detect and avoid.

The following Schematic is what I came up with for reliably detecting the illumination of
an Amber/Yellow LED "Flood-Light" style Bulb that is roughly ~40-feet from the detector.

The "Detector" consists of 6 ridiculously expensive,
high-output, Red-LEDs, with a 5-degree Beam-Angle,
that are used as Photo-Diodes.

You may be able to get some ideas from it.

It is simply 3 very high-Gain-Filters that pass only high-Frequencies,
and ignore slower changes in Light-Intensity.

You may be able to use just the "first-stage",
I wanted a long Output-Pulse for my purposes.
.
.
.
.

.

 

TV remotes will generally use an IR detector module that detects 38kHz modulated IR signal.
https://www.hellasdigital.gr/electr...frared-module-38khz-vs1838-for-arduino/?sl=en



https://www.digikey.com/en/maker/bl...-receive-data-over-ir-signals-with-an-arduino

 

Thanks everyone for your replies.

@MrChips Unfortunately, as I stated in the initial post, I cannot use this part because the sensing board is already done with phototransistors. The space plays a critical role here, therefore I cannot use those components. Only SMD footprints are acceptable.

@LowQCab The final product will be a phototransistor - IR LED grid. I am looking for possible cheaper solution as for every LED/PT pair there would need to be mentioned op-amp circuit and that could increase the overall cost a little bit too much. Still, I will try to consider your design if there is no other alternative. However, can I achieve better sensitivity with using cheaper LEDs and PTs? I can 'abuse' used IR LED to drive them with 35mA in pulses. Maybe this can increase distance because the LED should be brighter? Am I thinking in the right direction?

@BobTPH I will try to mix your idea with LowQCab's about amplifying the signal. On which side (LED or phototransistor) is more room for improvements regarding the reception distance sensing?

 

If your circuit board is already made to the schematic in post #1 then I don't see a way to solve your problem. You can find ready made Modulated IR beam units on ebay. ( I did buy one of these and it worked quite well but it only uses a single beam. I use a double beam to avoid it being triggered by an insect breaking a single beam. To make one your self I suggest using op-amps in the receiver. You will not easily get performance as good as the TSOP devices as they have automatic gain control and filtering built in. If you use TSOP devices the transmitter must produce a pulse stream that matches the specification of the TSOP device. I find using a small (8 pin) microcontroller is the easiest way to produce the pulse stream.

Les.

 

Only SMD footprints are acceptable.

That's not a problem. Here's one that will work as low as 20kHz, if you can get that frequency on your source LED. If you can get even higher modulation frequencies, there are more choices.

https://www.mouser.com/datasheet/2/427/tsmp6000-1767176.pdf

 

Switch to one of those cheap red laser diodes. That way the emitter's energy will be concentrated on the photo transistor die rather being spread out in space.

 

I would drop the LED modulation Frequency down to maybe ~100hz,
there appears to be no reason for it to be so high.

At this low frequency, there are many optical noise sources - ripple from mains powered lighting- 50/60 and 100/120Hz depending on where you live.

I would choose a much higher frequency to stay away from this noise pollution.

 

That's one of the reasons why I made the Filter-Circuit attached in a previous post.
It's not actually necessary to Modulate the LED at all with
an Infa-Red Light-Filter in front of the receiver.
.
.
.

 

Hello,

I have some questions regarding my implementation of IR LED detection with a phototransistor.

Thanks to anyone who will take a little time to help my research.

Hello, welcome to AAC.

Is this a school project?

 

Hello all,

This is not a school project. It is a hobby project where I want to create a basic IR grid where I can detect object 2D position based on beam interruption. Each LED would light up sequentally and its reciever phototransistor pair will detect its presence. If an object is in the midway, it will not detect the IR light source.

Currently I leaning towards the op-amp implementation, where the received input signal is amplified to a 0-5V level. I have made some simulation and it is looking promising. The TX side (IR LED) still needs to be somehow amplified so that it will produce 20mV signal on receiving side. A band-pass filter also needs to be added to allow only specific frequency to be amplified in that stage.

Will keep you posted when new findings are made. Thanks everyone.



Best regards,
Marcel

 

"" I want to create a basic IR grid where I can detect object 2D position based on beam interruption. ""

You have an incorrect concept of how this device might work.
It will be heavily affected by "Temperature-Gradients" between objects, and their distance.
Also, some objects will tend to absorb the projected IR-Light
while others will readily reflect it.
The relative Output-Voltage from each individual IR-Receiver is going to be
determined by the reflective qualities, and the distance, from any object.

Ultra-Sonic, or Radar, may be better choices than Infa-Red.
.
.
.

 

1) Without of any phase capture circuit the distance will be damn small.
2) With the simple PLL circuit like 567 will work for ten...20, maximum 30 meters afar.
4046 may give a slight better result.
Better suited PLL with very high gain LNA in the middle between keys may work up to kilometer in sunshine and several kilometers in night, however fog is crushing it to few meters, and fog is rarely absent in nighttime. In russian times we produced such on basis of 140YD13 in pair with 140YD20. Very much means a right vawelength choice, some IR frequencies are ignorant about water vapor, near 1.6-1.7 and better 3.5-3.7 um. Precize figures please check yourself, because I wrote it by memory.