IR instead of laser

[SimonTHK]

Hello smart people.
So Ive made this circuit with a photo transistor and a laser, to see a really fast object at 4 cm wide, passing by. (though it actually work with small object´s under 5mm wide.I have done this with a comparator, and comparing voltages. Sooo great so far!

Now I want to try and make this same circuit, with IR light instead of a laser light. The problems are it is placed outside (will the sun interfere with IR light?). And are there specific phototransistors, that only "see" IR light, not sunlight.

The reason for this, is that right now, I need that laser to be completely be aligned (pointing on) with the phototransistor that is 3 mm wide. I want the circuit to work in a more versatile invironment.

Ive put an image.
Thanks in advance
Best regards SImon

 

[Pommie]

Normal IR transmitterz use a frequency (normally 38kHz) so the reciever can filter out the effect of the sun.

Mike.

 

[SimonTHK]

Normal IR transmitterz use a frequency (normally 38kHz) so the reciever can filter out the effect of the sun.

Mike.

What does that mean?

 

[Pommie]

Try google. If you don't understand what I said then you need to study.

Mike.

 

[SimonTHK]

Yeah ok so I need to make 38 khz signal for transmitter and the receiver has to also read the 38 khz signal? Right now I just measure light potency on phototransistor, getting either less than or above 2,5v, and use a comparator to tell me whenever the lightsource has been cut off and become less than 2,5 v.

I dont think I need 38 khz signals to do this. Why would I?

 

[Pommie]

So the sun becomes irrelevant. Your telly remote uses 38kHz so it still works in day light. But your way will work, go for it.

Mike.

 

[JimB]

If you modulate the IR source at 38kHz (or any other frequency of your choice), at the receiver you can detect that 38kHz and know that there is an unobstructed path from source to detector.

When the path is obstructed by the object that you wish to detect, the 38kHz signal will be absent.

Problems:
Depending on the environment,
it could be that the modulated IR signal is scattered from other objects nearby and reaches the detector,
even though the object that you wish to detect is obstructing the direct path,
sufficient modulated IR reaches the detector so that the obstruction is not detected.

Strong sunlight my saturate the the detector, so that the modulated IR is not detected.

JimB

 

[AnalogKid]

What does that mean?

He is talking about remote controls for TV's, cable boxes, etc. They use IR modulated with a 38 kHz squarewave, that pulses on and off in several bursts per second with a different pattern for each button. This works very well in a family room, but is *way* too slow to detect a "really fast object". The problem is that the standard remote control receiver part, which is an IR phototransistor, amplitude detector, bandpass filter, envelope detector, and output driver all in one little module, is purpose-built for one job that is not speed-dependent. However, if you replicate the parts you need, tweaked for detection speed, you can get what you want. A bullet chronograph can detect speeds above Mach 2 with optical sensors. Not right for your job, but it indicates that what you want to do is possible.

Give us more details. Object size and speed. distance from transmitter to detector, power supplies available, etc. The basic approach is the same - modulate the IR beam so the detector can differentiate it from sunlight, ambient light variations from wind in trees, etc; bandpass filter the signal from the phototransistor so the receiver is seeing energy only from your transmitter; amplitude-detect that signal to detect a dip caused by the object; use that signal for whatever. With a higher modulation freq, such as 100-200 kHz, the delay through the receiver electronics should be short enough not to mess up the timing.

ak

 

[rjenkinsgb]

I dont think I need 38 khz signals to do this. Why would I?

So it becomes a "carrier wave" type system - like radio - rather than trying to directly detect the data content.

With the simple "beak beam" detector, you cannot amplify the "wanted" signal without also amplifying any other signal or interference in the same frequency range.

By using a carrier wave system, you can selectively filter or amplify the one spot frequency used as the carrier, without amplifying any baseband noise or interference.
The level of background light from the sun or artificial sources like lamps flickering at 100 or 120Hz becomes irrelevant, as long as the sensor is designed so it does not saturate.

The making and breaking of the beam can be considered amplitude modulation, the wanted data is the amplitude of the carrier frequency.

The higher the carrier frequency, the better response time you can get from the demodulator, but you may need better rated optical components.


Or, if the whole thing has a somewhat enclosed optical path with some level of shading & there are no fast changes of unwanted illumination? You may get it to work with some slow DC feedback at the detector, eg. an automatic bias circuit that keeps the DC level out of the detector and preamp constant.

That needs a fast enough response to track such as clouds blocking the sun, but also slow enough to not change too much in the time the beam is blocked by a wanted object detection.

That will not work if eg. it's in bright daylight and people walking past can cause solid shadows; only slow changes.
You need a carrier system if fast ambient changes must not have any effect

 

[ClydeCrashKop]

If the distance isn’t too far, you can put an infrared photo transistor inside a tube to reduce ambient light. But it has to be aimed at the IR LED.

 

[SimonTHK]

He is talking about remote controls for TV's, cable boxes, etc. They use IR modulated with a 38 kHz squarewave, that pulses on and off in several bursts per second with a different pattern for each button. This works very well in a family room, but is *way* too slow to detect a "really fast object". The problem is that the standard remote control receiver part, which is an IR phototransistor, amplitude detector, bandpass filter, envelope detector, and output driver all in one little module, is purpose-built for one job that is not speed-dependent. However, if you replicate the parts you need, tweaked for detection speed, you can get what you want. A bullet chronograph can detect speeds above Mach 2 with optical sensors. Not right for your job, but it indicates that what you want to do is possible.

Give us more details. Object size and speed. distance from transmitter to detector, power supplies available, etc. The basic approach is the same - modulate the IR beam so the detector can differentiate it from sunlight, ambient light variations from wind in trees, etc; bandpass filter the signal from the phototransistor so the receiver is seeing energy only from your transmitter; amplitude-detect that signal to detect a dip caused by the object; use that signal for whatever. With a higher modulation freq, such as 100-200 kHz, the delay through the receiver electronics should be short enough not to mess up the timing.

ak

Ya ok I will get back to you. Right now it works. I just thought that I could somehow change the light source and receiver and make it easier to install so I dont need to have the installation so precise.
Also it is a problem that I dont own a oscilloscope, I just cant test well enough.
Distance is 25 cm.
All power supplies available.
Object size 3,4 cm wide cirkle.
Speed unknown.

 

[AnalogKid]

So, a 10" gap and a 1" to 1.5" object. Yes, an IR or any color visible LED and a suitable phototransistor will do this. 10" is not very far; depending on the environment, you might not need beam modulation. LEDs come with various beam angles - how wide a beam is emitted. Many "super-bright" and "daylight-visible" LEDs appear so bright because the beam angle is only 10 or 20 degrees. The beam angle usually is defined as the included angle at which the intensity drops by 50% at the edges. A 20 degree LED paints a width of about 3.5" at a 10" distance, or 1.75" in the middle of the gap.

First, I would try it without modulation. Please post your schematic of what works now.

ak

 

[gophert]

If the distance isn’t too far, you can put an infrared photo transistor inside a tube to reduce ambient light. But it has to be aimed at the IR LED.

the tube idea works pretty well - painting the inside of the tube black really helps for sunlight attenuation.

 

[ChrisP58]

Be aware that there is a lot of IR energy in sunlight.

The modulated carrier schemes described above are probably the best way to distinguish between your IR source and any other IR source,

 

[gophert]

Most lidar systems run at 905nm and a growing number at 1550nm specifically because those wavelengths correspond to dips in the solar spectrum. 1550nm is getting more popular because of growing concern for eye danger of 905nm lasers.