Recommendation for a hall effect sensor for linear position sensing

[EEBlackSheep]

I am adapting a throttle position sensor to a motorcycle slide carb.  I would like to use a hall effect sensor and neo magnet, as it is compact and easy to install.  The distance I would like to measure is about 30mm, which is quite a bit for a hall effect sensor.  I am looking at various datasheets, and find that this one from TI give me -90mV/mT, which hopefully should work for my application.

http://www.ti.com/lit/ds/symlink/drv5053.pdf

Do you think this will be suitable for my application?

thanks for the info.

[EEBlackSheep]

Another friend suggested a magnometer - what d'yall think?

[max_torque]

Given the environment, you'd have to be relatively crazy to not use some sort of off-the-shelf linear potentiometer or similar non-contact version.  Anything you make yourself is going to have to be very robust to all sorts of wildly varrying outside effects (temprature, vibration, magnetic fields, humidity etc etc)........

If you don't like the cost of that (linear pots are expensive) then just use a standard (Automotive rated) rotary pot driven by a bell crank and linearise it in software   

[cdev]

I know Inkscape has an extension for rotary encoders.

http://wiki.inkscape.org/wiki/index.php/Inkscape_Extensions

I would spend some quality time searching all the possible key words to see if somebody may have already written a script to make and print your own (linear) optical encoders.

Why re-invent the (non) wheel?

[janoc]

Magnetometer - definitely no, those things will pick up all sorts of interference like there is no tomorrow, including Earth's magnetic field. Oh and the field changes depending on the direction the sensor is facing so you would get different values depending on in which direction you are riding, including up and down hill, so you would have to compensate for that. And you need to calibrate them before use.

Hall sensors are not good for distance sensing, most cheap and common ones are designed for on/off operation (e.g. as an end stop), even if you could find one that would trigger at 3cm range. Also the magnetic flux is highly non-linear with distance, so this would require quite a bit of math and/or calibration.

You could use an industrial inductive sensor, but you likely don't have the required voltage on your bike - most need at least 12V.

Alternatively I would go for either a mechanical position sensor or an optical encoder. If it is for a motocycle, you want something that is robust and idiot-proof, so that a random interference from the ignition or weather won't throw your system off (and potentially cause you to crash, depending on what exactly are you trying to use the sensor for).

[mikerj]

I wouldn't even think about knocking up DIY automotive sensors when commercial parts are readily available for low lost.  It's a very harsh environment and especially on a bike the consequences of failure could be nasty.

Why are you against using a potentiometer for throttle angle sensing? OEM sensors (rather than cheap chinese knock offs) are simple and robust as proven by the hundreds of millions of vehicles driving around using them.