On Thu, 15 Jan 2009 03:10:00 -0500, Wes wrote:

Don Foreman wrote:

I think we're making some progress here.

Magnetic field sensor done today, on the bench, electronics for it
half done that I'll finish tomorrow. I now have some real specs, I
have Fitch's datalogging scope-meter, we'll see if I can skin this
kitty or whut.


Keep us posted Don. Who knows, some day years later, your words may speak to someone
facing the same thing. After all Tee Nut lives on.

How does your sensor work? The only hal effect stuff I'm familar with is solid state
sensor looking for a magnetic piston in pneumatic cylinder.

Wes

A Hall effect chip is a bit of silicon that produces a voltage
proportional to the strength of the magnetic field passing thru it.
Imagine a rectangle with contacts on all four sides. Apply bias
voltage from top to bottom. Electrons then flow from top to bottom --
but if there is a magnetic field normal to (thru) the rectangle
they'll tend to drift sideways. This will produce a potential
difference from left to right.

Modern sensors integrate a bunch of electronic trickery to improve
temperature stability and reduce offset and drift, but that's the
basic idea.

Many industrial sensors integrate this function with a circuit that
switches at some level of field intensity, providing a binary or
"on-off" signal. I'm using a linear hall sensor that provides a
voltage proportional to field strength, about 4.25 millivolts per
gauss. It's the HAL400C from Micronas. Current product would be the
HAL401:

http://www.micronas.com/automotive_a...ion/index.html

Why that one? Because I had 4 of them in my junkbox. Allegro also
makes Hall sensors, and Honeywell used to, probably still does. I
like the differential output on the Micronas for rejection of noise. 4
millivolts isn't much signal when 60 Hz is in the passband. You
probably know what happens when the ground comes loose on a microphone
or magnetic phono jack: HUMMMM! I want this sensor to sense magnetic
field but ignore E field. I'll measure that separately.

It's differential output will drive an instrumentation opamp
(Burr-Brown/TI INA121) with gain of 100, then a 2-pole lowpass filter
with corner frequency of 200 Hz. I should have that all going today.