An underwater touch sensor for small aquarium fish?

Hello all, does anyone have any suggestions for a touch sensor that will function underwater to detect bumps by small fish? Is there an uber-sensitive force-resistive sensor or resistive touchscreen? Am I missing some obvious solutions? Proximity-based solutions might work but I'm hoping to nail down a more discrete stimulus response.

You are going to struggle with that one - what is the project there may be other ways to achieve what you want ( eg break beam sensor )

hammy:
You are going to struggle with that one - what is the project there may be other ways to achieve what you want ( eg break beam sensor )

This is an academic research project to create a response mechanism that can fit into a small aquarium tank and a fish can learn to operate. I have worked out a beam-break sensor solution to this problem, but I'd like to expand it into a more general solution, especially a touch-screen for simultaneous display of multiple "targets" and sensing the fish's physical contact with each target. Mixing electronics and water is one of the stumbling blocks for me, along with finding a sensor sensitive enough to detect a small fish's nose-bump.

Touch screens such as used for mobile phones don’t work under water.

A mesh of break beam sensors placed right in front of the screen may be your best bet. It sounds like you have the underwater optics under control having done such a sensor, as that’s what I expect to be an issue.

This should be quite easy to implement, indeed the watery environment will be your largest obstacle. I’ve heard of this technique having been used for early “touch” screens. Of course it doesn’t sense actual touch, but a <1 mm proximity to a specific spot of your screen is probably good enough for your experiment.

I have no idea at all if it would work but, maybe a rain sensor, or the concept, could be adapted.

wvmarle:
A mesh of break beam sensors placed right in front of the screen may be your best bet.

This is actually a really good idea and something that I hadn't thought of before. I currently have two beam-break sensors configured in a vertical setup to detect two distinct "swim-through" responses. As you suggest, it may be easy to adapt this to a grid pattern for detecting proximity to multiple targets, though this solution may be a bit bulkier than what is ideal. I am hoping that someone knows of a more compact mechanism (i.e., the thickness of a touchscreen or force-resistive sensor).

cschmoutz:
This is actually a really good idea and something that I hadn't thought of before. I currently have two beam-break sensors configured in a vertical setup to detect two distinct "swim-through" responses. As you suggest, it may be easy to adapt this to a grid pattern for detecting proximity to multiple targets, though this solution may be a bit bulkier than what is ideal. I am hoping that someone knows of a more compact mechanism (i.e., the thickness of a touchscreen or force-resistive sensor).

Commercial fish counters use an inductive coil the fish has to move through. But these are probably much larger fish that you are working with. Perhaps you or someone you know can design a similar device for smaller fish.

Paul

If you will be training the fish to touch something, a small paddle on a lever arm should work, and the switch or sensor can be out of the water.

dougp:
a rain sensor, or the concept, could be adapted.

I assume you mean the automotive example on that Wikipedia page. Can you expand on that idea? Using a LED/photodiode pair is a good idea and currently I'm using an infrared through-beam setup to detect the fish. I know that other researchers have used a fiber optic reflective sensor (Keyence FU-23X) with an amplfier to detect the approach of a fish. This is an option and an array of them would be able to pinpoint a fish's position, however they are expensive compared to other components.

Paul_KD7HB:
Commercial fish counters use an inductive coil the fish has to move through.

Thanks, I like this idea. I have currently implemented my own sort of fish counter using an IR beam-break sensor from Adafruit. I put each side of the beam in a test tube and suspended the tubes in the water about 1.5" apart. It's cheap and from preliminary tests, the fish are able to learn to break the beam. Do you have any suggestions for something based on the fish's electric charge or is it as easy as a bare copper coil plus the right amplifier?

jremington:
a small paddle on a lever arm should work, and the switch or sensor can be out of the water.

This is also a good idea. A lever arm could be very sensitive and simple and the switches are cheap.

Thank you

Probably a bit of a weird suggestion but can a doppler shift principle be utilized

Magnetic switches can be waterproofed easily, and used fully under water. You may look at modifying existing float switches or water flow switches.

cschmoutz:
Hello all, does anyone have any suggestions for a touch sensor that will function underwater to detect bumps by small fish? Is there an uber-sensitive force-resistive sensor or resistive touchscreen? Am I missing some obvious solutions? Proximity-based solutions might work but I'm hoping to nail down a more discrete stimulus response.

Perhaps a piezo-disc heavily coated in conformal coating? Or perhaps better a stick glued to a piezo disc
pointing down into the water, so the sensor itself is mounted dry.

sensitive enough to detect a small fish's nose-bump.

Today I learnt that fish have real noses! Fascinating.