Optical Laser to refract/reflect water vapor and smoke.... angles?

[MattLogue9]

Building visibiity sensor...

What is best angle to detect water vapor reflection from laser, and refraction?

I plan to have two open cylindrical containers painted flat black and put inside each other so the overlap is about 0.5-1" (adj to limit ambient light vs airflow).

Laser is cheap red laser diode. dime a dozen deal. Laser points straight thru a plywood hole to target. I plan on using red LEDs to sense since I don't have photodiodes. It goes to Mux, then a LM324 set to ouput 0-1V, then ESP32 MCU ADC.

I figure I can work with ratio of fore and aft LED V to determine air contents to some degree. Water refracts of course, dust, smoke, pollen reflect.

Precision is NOT critical. Accuracy somewhat. I just want to be able to say, real foggy, foggy, dusty, or clear. Thats it.

It will be paired with Temp, Hum, Rainfall, RGB UV Sunlight load, Soil Moisture, Wind Vector, Lightning detectors.

It may also be paired with MOx gas sensors...

1 So what angles should I use?
2 What is good method to block out ambient light yet maintain airflow and not condense the water vapor?
2a Pulsed laser and differential signalling?
2b Overlapping cans like [> ?
2bb Orientation for device [] V or = H
3 Should i expect to get a viable signal from R LED and a LM324?Part of larger project: Wx Station.

 

[Baluncore]

Water refracts of course, dust, smoke, pollen reflect.

Water vapour is transparent to visible light. Liquid water forms a fog. Fog scatters light in all directions. Dust also scatters light but less efficiently as it has colour sensitivity.

You need to modulate the laser light source. Then detect that AC signal to eliminate background illumination.

You need a multi-path instrument with a detector near the laser to verify brightness. The laser brightness will fall as it ages. LEDs are temperature sensitive and also age.

 

[MattLogue9]

So I generate a pulse say 50% duty 145hz - how would one process the signal? I'll use a ESP32 microcontroller. I could use another LED in separated chamber with same light as a bias. Would a IR LED pick up the red laser well enough to read by its 12bit ADC? - fraid I don't have a tunneled detector or photodiode on hand right now.

 

[chemisttree]

I would go for the same angle seen in a rainbow for the color of the laser you are using. For red that’s about 42 degrees. Pulsing is optional. Take a zero reading before or after taking a reading and subtract it from the reading. Laser on a momentary switch to read the fog state.

 

[Baluncore]

Those LM324 chips are real cheap. I think you are on a budget and want to keep the price down. The best way to solve the problem reliably, and for the minimum cost, is to use a trick called engineering. That involves thinking carefully about what you need to do, and the different ways you might do it.

The next thing to think about is photodetectors.
https://en.wikipedia.org/wiki/Photodiode
https://en.wikipedia.org/wiki/LED_circuit#LED_as_light_sensor

Depending on the circuit, photodiodes can generate a voltage or a current. If you use them in a linear mode, measuring a current proportional to light intensity, then you must convert the current to a voltage while operating them at a fixed voltage. Look at the circuits in this note.
https://www.hamamatsu.com/resources/pdf/ssd/e02_handbook_si_photodiode.pdf