Guy Lavoie
Active Member
Here is something I found a week ago that might be very useful:
http://www.allegromicro.com/sf/1321
This is a linear hall effect sensor. What this does is put out a voltage that is proportional to the strength of a static magnetic field perpendicular to it. I ordered two sample units which arrived today. I took the A1321EUA, which has the most gain.
The application I'm going to use it for is to measure the level of fuel oil in my 275 gal. tank. The tank has a plastic dome about one inch in diameter and 4 inches high on top. In it there is a round indicator disk connected to a float mechanism, to give you an idea of the amount of oil in the tank. The total travel from empty to full is about 3 inches. Having a non contact sensor like this is important for something like an oil tank, since any sparks or contact closures could provoke an explosion (although this is very unlikely for fuel oil). Also, the plastic must remain hermetically sealed because any opening will let the fuel oil evaporate and the smell is quite strong.
So I hooked up a +5 volt power supply to the chip and played with a moderately strong magnet to see the voltage change with distance. It works like a charm! The sensing range seems to be just about right too. To use with my tank, I needed a rather flat magnet to glue on top of the indicator disk. After a bit of thinking, I remembered the very flat, yet strong magnets used in hard disk drives for the head positioning mechanism. Disk drives have two sets of flat magnets between which there is a "voice-coil" that is driven like a speaker cone to position the heads. I yanked apart an old 500 meg drive that gave me 4 such magnets.
Hall effect sensors are polarity sensitive and behave like a bridge circuit. With no magnetic field, it puts out 2.5 volts. When the North end of a magnet approaches, the voltage goes one way while a South end makes it go the other way. I arranged mine to range from 2.5 down to 0 volts, to make any mathematical gain adjustments easy. The range is non-linear, with the largest voltage swing occuring when the magnet is close. This can be compensated for mathematically too in the device interpreting the reading. My intention is to connect it to a SECU16 analog input. It seems to drive a 1k resistor quite easily, so the 10k input impedance of an analog input should not be a problem.
http://www.allegromicro.com/sf/1321
This is a linear hall effect sensor. What this does is put out a voltage that is proportional to the strength of a static magnetic field perpendicular to it. I ordered two sample units which arrived today. I took the A1321EUA, which has the most gain.
The application I'm going to use it for is to measure the level of fuel oil in my 275 gal. tank. The tank has a plastic dome about one inch in diameter and 4 inches high on top. In it there is a round indicator disk connected to a float mechanism, to give you an idea of the amount of oil in the tank. The total travel from empty to full is about 3 inches. Having a non contact sensor like this is important for something like an oil tank, since any sparks or contact closures could provoke an explosion (although this is very unlikely for fuel oil). Also, the plastic must remain hermetically sealed because any opening will let the fuel oil evaporate and the smell is quite strong.
So I hooked up a +5 volt power supply to the chip and played with a moderately strong magnet to see the voltage change with distance. It works like a charm! The sensing range seems to be just about right too. To use with my tank, I needed a rather flat magnet to glue on top of the indicator disk. After a bit of thinking, I remembered the very flat, yet strong magnets used in hard disk drives for the head positioning mechanism. Disk drives have two sets of flat magnets between which there is a "voice-coil" that is driven like a speaker cone to position the heads. I yanked apart an old 500 meg drive that gave me 4 such magnets.
Hall effect sensors are polarity sensitive and behave like a bridge circuit. With no magnetic field, it puts out 2.5 volts. When the North end of a magnet approaches, the voltage goes one way while a South end makes it go the other way. I arranged mine to range from 2.5 down to 0 volts, to make any mathematical gain adjustments easy. The range is non-linear, with the largest voltage swing occuring when the magnet is close. This can be compensated for mathematically too in the device interpreting the reading. My intention is to connect it to a SECU16 analog input. It seems to drive a 1k resistor quite easily, so the 10k input impedance of an analog input should not be a problem.
