Proximity detection hardware suggestions needed

Hello Everybody,

I need suggestions for a cheap method for detecting the state (is it pressed down, or not) of a valve, See picture:

cca. 1.5Cm diameter and 10cm high metal tube, with a shiny metal valve in it. the valves default position is on the picture, and it can move downwards ~1.5cm, still leaving about 0.5cm place unter it. It closed from both up and down, so there is dark in the casing.

I need the cheapest and fastest (Should be digital, because rading the analog inputs need time) method for sensing even if the valve moved a little bit (altough it has only 2 states, i would like to know its being pressed before it is pressed totally)

I can only access the tubing from down.

What i can think about is using two led diodes. One emits while the other is tied between ground and the base of an NPN BJT such that incident light upon the photodiode produces voltage on the BJT’s base. Because the valves down is shiny, i think that the voltage will alter if the valve gets closer to this ledpair sensor.

Any other , possibly better ideas?

Perhaps using an infrared emitter / detector pair would be cheaper / easier. Since it's enclosed, ambient light is a non-issue.

Here's the schematic...

Are you planning on doing a little machine work to the cylinder cap to mount whatever you decide to do, or what?


The cylinder cap is removable, i will make custom cylinder caps with the detectors built in. The infrared looks good, but has it any advantages over the simple LED diodes?

I'm not real familiar with the little infrared sender/receivers. I have a few, but only know what I've read. Like I mentioned, the only thing I've read that they're bad about is sunlight affecting them. I have a phototach that I use to measure the rpm's of engines off the harmonic balancer. If it's a sunny day, especially if there's a shiny spot on the balancer, it will make trouble for the phototach, which I believe is just a IR sender/receiver and a timing circuit in the inside.

However, I don't see how this would be a problem for you. It's dark in there. Plus, it's hard to beat the $2 price on the one I gave the link to. You could rig that whole deal up f/ less than $5.

However, sparkfun also provides a setup already on a breakout board (grunt work already done)

Read the datasheet and see if it will work

What is the application ? Are you trying to automate your tuba ?

You could mount a very small switch at the top of the cylinder, which is impacted by the underside of the key when you push it down.

You could also mount a very small magnet and a hall effect sensor or magnetic reed switch.

What is the application ? Are you trying to automate your tuba ?

Yes, i'm trying to sense the fingering of a trumpet. Yout magnetic idea is interesting!

Yes, i'm trying to sense the fingering of a trumpet.

It's helpful when you let us know what you're working with.

I'm with Michinyon. I'd recommend hall effect over reed switch; no moving parts = never wear out. Mount hall effect in cap, fasten magnet to bottom of valve. Proximity in which the sensor will "see" the magnet is entirely dependent on the magnetic strength of the magnet itself. I buy my neodymium (rare earth) magnets from K&J magnetics, and they supply the gauss of each magnet on the website. Compare to hall effect datasheet to see which magnet you need.

I use these hall effects from digikey...

If you go this route, the sensor only reads the south pole of a magnet. Also, the data sheet is full of different applications these devices can be used for. Really neat little device.