Sensing in Darkness... through river water !
VonSzarvas
Posts: 3,450
Hello all... Seeking advice time.... !
I have been having a little experiment with some photoresistor and photodiode sensors to sense _very low_ light levels. It seems to me there must be a better way....
To explain with a simple example:
Imagine a sealed box with all dimensions 1meter.
The light sensor is at one end of the box, the openings are at the other end (so ~1 meter away). The openings are a row of clip-lids, each about the size of a lead-acid-battery refill cap.
When the lids are on, the inside of the box is ~100% dark.
The box is filled with river water. Might be clear-ish, might be greenish (algae).
If a lid is removed (obviously depending on the ambient light outside the box), then some light should pass into the box, and I want to sense that light; signal strength is not so important- rather any presence, so an on/off condition is fine. If one or more lids removed then ON, otherwise all lids are closed so OFF.
It might be an option to sense non-visible light, either higher or lower frequency than the human eye can manage, although when I scientifically "stick my head in said dark box", I can "see" the light at the opening when it is removed. Only my sensor is not always as sensitive as my eyes, especially when outside the box is low light.
It leads me to wonder if I need a special directional sensor which points approximately at the light sources? I tried experimenting with some silver foil and even a mirrored convex piece from an old hand-torch, but I do not get as good as a reading as my eye can manage!
Is there any such sensor "directional / super sensitive" in existence...?
Caveat 1: Hobby box price, so using a video camera not an option!
Caveat 2: Maybe the simple photodiode should be enough for the job, and I need to think about my interface electronics to make it more sensitive?? I have tried many resistor divider combinations, but not tried any type of op-amps yet... maybe that is my mistake, although with the device attached to a Propeller, I would hope to avoid too many external components.
I have been having a little experiment with some photoresistor and photodiode sensors to sense _very low_ light levels. It seems to me there must be a better way....
To explain with a simple example:
Imagine a sealed box with all dimensions 1meter.
The light sensor is at one end of the box, the openings are at the other end (so ~1 meter away). The openings are a row of clip-lids, each about the size of a lead-acid-battery refill cap.
When the lids are on, the inside of the box is ~100% dark.
The box is filled with river water. Might be clear-ish, might be greenish (algae).
If a lid is removed (obviously depending on the ambient light outside the box), then some light should pass into the box, and I want to sense that light; signal strength is not so important- rather any presence, so an on/off condition is fine. If one or more lids removed then ON, otherwise all lids are closed so OFF.
It might be an option to sense non-visible light, either higher or lower frequency than the human eye can manage, although when I scientifically "stick my head in said dark box", I can "see" the light at the opening when it is removed. Only my sensor is not always as sensitive as my eyes, especially when outside the box is low light.
It leads me to wonder if I need a special directional sensor which points approximately at the light sources? I tried experimenting with some silver foil and even a mirrored convex piece from an old hand-torch, but I do not get as good as a reading as my eye can manage!
Is there any such sensor "directional / super sensitive" in existence...?
Caveat 1: Hobby box price, so using a video camera not an option!
Caveat 2: Maybe the simple photodiode should be enough for the job, and I need to think about my interface electronics to make it more sensitive?? I have tried many resistor divider combinations, but not tried any type of op-amps yet... maybe that is my mistake, although with the device attached to a Propeller, I would hope to avoid too many external components.
Comments
On the figure below the absorption dips around green light. Note that 1 micrometer = 1000 nm, which is near IR.
If you're using ambient (sunlight) then your only option is to somehow make the sensors more sensitive or focus the light using a lens or mirror or something. If you can provide a light source, then perhaps consider using something in the center of the visible spectrum, such as greenish LEDs.
You might check out the circuit examples here:
http://sales.hamamatsu.com/assets/applications/SSD/si_pd_circuit_examples.pdf
I hope that helps get you started solving your problem. Feel free to provide more details about what you're doing and why, and ask the forum for more detailed questions.
I'm not sure this would apply, but perhaps you could line the inside of the caps with reflective tape. A light source would then be placed next to the sensor. This light source would beam toward the location of the caps, so when the caps are in place, light is bounced back by the reflective tape and is detected by your sensor. Of course, if the water has a lot of turbidity (cloudiness), it might bounce this emitted light back toward the sensor even when the caps are off, in which case this idea would not work. Just a thought, though.
However, human scotopic vision (mediated exclusively by rod cells) applies for illumination levels from 0.001 and down, maybe to a threshold 0.000001 lux, and the sensitivity peak for human vision shifts toward blue. That would be down in the noise level of most photosensors. Of course, photomultiplier tubes are exquisitely sensitive at the photon level.
A reverse biased green LED can make a pretty sensitive photodetector for low light, in an RC or RCTIME oscillator circuit. The thermal leakage current is very low. Are you using a BASIC Stamp or Propeller in this project?
At first I tried an LDR from my spares box, but quickly moved onto an SFH300 photodiode (because the LDR I had was not water-friendly!).
The datasheet is here:
http://www.datasheetcatalog.org/datasheets/270/486514_DS.pdf
Anyhow, I am really encouraged by your comments. I don't think I can provide a source of green light (this is a kind of alarm on a fish tank which is submerged in a river or pond). But that might be useful for testing, and I will try that- and also think about the possibility to swap out sunlight for greenlight !!
BTW, the logic is that if someone opens the box, it will either be during the day (so sunlight is the trigger), or during the night (so the bad-guys torch will be the trigger). The opening catch is fairly tricky to operate, and so likely as not the bad-guy would require a light source of some kind whilst working at night. It might be a torch or even a box of matches, so of course we cannot catch everyone, but it should help with some decent percent.
So, I will study the suggested circuits and try to improve the sensitivity pre-Propeller, which will be logging the activity with a timestamp. I feel better about the direction I am heading now - it all seemed a bit hopeless for a moment!!
Thank you for the lux comparison. It makes good sense, and perhaps I do need a device that will detect down to 0.001 lux.
I am using the Propeller and so had hoped to use an RCTime circuit to save on external components. I have been working with the parallax educational docs on the matter.
Do you think a reversed biased green led might be able to detect light down to (or close to) 0.001 lux ?
How about putting (battery-powered?) LEDs on the caps so they emit a very fast pulse every 0.5 second or so, and if the detector does not see a pulse within a few seconds, then you know somebody has lifted the cap from the tank?
Just thinking out loud.
Like electricAye said, maybe there is an easier way to detect/deter intruders! I'm curious about the purpose of this box... let's see, if I pull off this little cap that is marked, "danger, high voltage experiment"...
LOL. Tracy might be on to something. Put a tiny, feverishly blinking red LED on the caps, make the caps look like a dangerous electrical component, maybe nobody would even think of touching them!
The PIR concept outside the tank would not work in this instance, but thanks again for thinking about this ElectricAye!
Reason being that the boxes are full of fish and delivered/removed to/from various locations (often remote, but not always). It would not be practical to secure the surroundings, which will often change (I mean different locations), and are not owned by the tank owners anyhow. And also, then the security is obvious, so the intruders will just work out another way....
So my idea was to stick by magnet a little box at the bottom of each tank with a light sensor on top. It will be out-of-reach for anyone that might otherwise want to tamper. I wanted to avoid a sensor or pulsing led on the actual lids- and also the headache of multiple power sources, and the point that the lids are not uniform sizes/shapes, etc.. might make that difficult to install anyway.
When the tank returns to base, we can use another magnet on the outside of the box to slide the circuit out of the tank, then download the recorded data. If it turns out that people have been opening the tank at times not in tune with the paperwork, then there will be some evidence. Whilst the tanks are in remote locations, they are "supposed" to be under control of a specified guard or responsible person. If this covert device shows that 1 or 2 locations are responsible for the missing fish, then someone can go and watch next time the tank goes to that place! A guard to watch the guard!
To be fair, this could perhaps be done some other ways, but I like to set myself little challenges/learning-exercises which I might solve with parallax parts, and its a kind of exercise for my day job at which I am supposed to be developing something a little more serious with a propeller!!. I think it was PJ-Allens old avatar which gave me the idea of "sensing in darkness" !! ...so I might be a little narrow in trying to solve the matter this way. We are only talking about 6 or 7 tanks, so perhaps it could be solved by watching them all for a week, but that would not be such fun!!
Thank you both again. I will experiment with the Green leds as suggested and write again with my results.
... except the guards know there are very valuable fish inside, and it won't take them long to get the electrical expert around who will open the thing anyhow!!!
We don't have great safety standards here...
once a guy came to change one of our 3-phase fuses which had blown out. He did the work live (just ignored the big rcd cut-off switch he perhaps should have used), and in doing so managed to touch the phase and neutral together (or was it neutral and earth.... cannot recall)... anyhow, EVERY electrical item in the house smoked. I mean literally everything electrical had to either have the transformer replaced or be dumped and bought new. Every laptop charger and laptops, printers, TV's, the gate opening and garage door motors, the washing machine, the ice maker, the home cinema, the mobile phone chargers, the pool pump, my daughters beloved stereo which must play strange modern dance tunes every-night or hell breaks.... everything.... aarrrghhh,, Its amazing how much electrical stuff we have in our lives... Goodness, the burglar alarm, the phone system, the wireless phones, the oven, ... gosh, the internet modem, <breath deep, count to 10, think about clear blue skies/>
IT was a nightmare for a few days!!! I asked the guy about it and of course he denied it, then broke down in tears and begged me not to report him.. Hopefully he learned his lesson that day (if he didn't I wasted a lot of money on his education). Sadly, there are many more cavalier sparkies around to take his place. So I don't think "high voltage" is of big concern, although perhaps a sticker with a crocodile on.....
As for the fish, mainly a special variety of African catfish (no less!!), Telapia and Carp in the main rivers. Along with various "prototype" breeds hidden in boxes !! Its funny that a land-locked country should have such an internationally well respected fish research institute! As a non-profit org. I like to help out where I can; and of course I can learn some new things too.
Perhaps "very valuable" was too broad a brush... Some of the fish are more valuable than others (both in terms of sale value but also development time). Of course the hungry (Hungarian!) can only see food, not value.