I have a project which will live in a weather-proof box (something like this) with a clear front, mounted to a post and exposed to the rain that will need to be able to respond to user input.
I initially was going to run with an iR remote, but that's not ideal since I'm forgetful So I was thinking of using a cheap membrane matrix keypad in addition to the iR receiver, but while that doesn't mind the rain it will mean cutting the case.
Before I go down that path is there a way of adding (I only need 3) momentary switches that can be externally controlled without cutting holes in the case? Marine/waterproof switches all appear to be expensive overkill for this.
With a magnet kept on a string on the pole nearby - could be just the ticket.[quote author=Nick Gammon link=topic=112017.msg841588#msg841588 date=1340862725]
Capacitive sensor? Just inside the case.
[/quote]And that's also a great idea and simple too - will have to do some testing. [quote author=From the playground page on the subject:]An Arduino and the library may be used to sense human touch through more than a quarter of an inch of plastic, wood, ceramic or other insulating material (not any kind of metal though), enabling the sensor to be completely visually concealed.[/quote]So either way I could put markings on the opaque side of the case for my 3 buttons.
PeterH:
Are capacitive sensors OK to use on a surface that gets rained on?
In this case I'd plan to have them mounted inside the box so they'll remain dry. Since I'm unlikely to be using them in the rain the only other concern is potentially raindrops will be detected as a touch? I don't know but it will be something I'll check out.
I've been looking at weather-protected switches as an aside as I have another use that's marine where exposure to salty air will also be a corrosion issue...and it appears that some cheaper vendors think IP67 is an adjective describing the colour or something.
PeterH:
Are capacitive sensors OK to use on a surface that gets rained on?
I don't see why not, but you could experiment. I think the capacitive switches work because your body has ... er ... electrical mass (I'm just making this up). But raindrops wouldn't. See:
The switch works using body capacitance, a property of the human body that gives it great electrical characteristics.
I don't know if it's the same for modern switches, but I had a TV with touch controls years ago when they were new and fashionable and that went completely pear-shaped after somebody tried to operate it with damp hands. It was OK after it had dried out again, but that's why I was wondering whether capacitive sensors in a rainy environment might not be reliable.
I like this idea, it would be easy to implement and you wouldn't have to worry about weather's effects on capacitive sensors (and you could wear gloves). Space 3 sets of 2 LEDs across the front plate. One would be illuminated, one would act as a light sensor. It seems like you'd be able to measure a brightness difference no matter what material you used to "press" the button. You could even use pulse the LEDs on and off at different rates to eliminate false positives.
In case you want to follow my suggestion and use LEDs for illumination, make sure that the detector LEDs have longer wavelength than the illuminators. Otherwise the will be blind at night. I strongly suggest to test such a setup before deploying it. However I am pretty sure that this will work since I have code to prove this
One problem with using LEDs and such outdoors is in really bright conditions, the background light will be too bright and the Arduino won't be able to see the light. I don't know if IR would be similar to visible light.
As others have mentioned, bluetooth is a possibility. You could always use velcro (tm) to attach the bluetooth keychain fob to the outside of the box, so you don't have to remember bringing it.
If you aren't going to be using the Arduino in the rain, but want it protected against rain, get an enclosure that has a lid that can open, and just have the use open the lid and use a normal physical button.
Wouldn't a simple light sensor do? You can't push buttons you can't see anyway, so maybe leave a torch on top for night use.
You could poll all (10?) buttons and if one is showing a lot less light than the average, that is the one a finger is in front of. Or an ant, I suppose.
Hehe, the illumination could be visible for a nice backlight of the buttons
Another option would be the use of tilt sensors instead of buttons. Tilt it left for function 1, up for function 2, right for function 3. Includes the option for a 4th function.
Still another option would be the use of an IR sensor + IR remote control. IR remotes are cheap and IR sensors are cheap as well. No clue if this would be a valid option for the desired application.
[quote author=Udo Klein link=topic=112017.msg846205#msg846205 date=1341230559]
Still another option would be the use of an IR sensor + IR remote control. IR remotes are cheap and IR sensors are cheap as well. No clue if this would be a valid option for the desired application.
[/quote]This project was initially going to have an iR remote, but since the ones I have aren't for outdoor conditions (and because I know I can/will lose or misplace them) I was looking for an alternative way of controlling it. Now I'm thinking I might do without it.
This thread has certainly given me lots of options to try. Thanks everyone !
Geoff
I have one more: put a microphone + dmtf decode inside. Then you would need something that creates dmtf tones. Dmtf generators are very cheap and I would bet that most smart phones will have an app to generate them.
So I was thinking of using a cheap membrane matrix keypad in addition to the iR receiver, but while that doesn't mind the rain it will mean cutting the case.