Hall sensor quiescent voltage to zero

[m718]

I have a Hall sensor that outputs about 2.45 volts with no magnetic field applied and then the voltage goes higher or lower depending on the magnetic field applied(north or south).
I need to measure the voltage from zero. How can I do this?

 

[vk6kro]

I would use a Picaxe chip. Probably a Picaxe 14. They cost about US$3.45 plus postage.
http://www.phanderson.com/picaxe/

This has several 10 bit A to D converters in it which will measure from 0 to 5 volts in 1024 steps of about 5 mV each.
2.45 volts would be perfect.
You would just subtract 2.45 from the reading you get and display the result on a computer or a serial LCD. If it was less than 2.45 you would get a negative result.

 

[Bob S]

I have had the same problem. The Hall Effect probe is a 4-wire device vaguely similar to a Wheatstone bridge in its hookup. Voltage is applied to the top and bottom of the "diamond" and the signal is the difference of the left and right outputs. You can zero-reference these outputs if you have access to +/- 5 or +/- 15 volts for an opamp Vcc and Vee. You have to get an opamp and set up as a differential amplifier. I guess you will want 10k ohm resistors from the left and right sides of the diamond to the + and - inputs of the opamp, and a 100k ohm from the + input to ground (assuming your opamp is tied to +/- 5 or 15 V), and a 100 k ohm resistor from the - input to the output. This symmetric layout rejects the offset, and the opamp output will be single-ended referenced to ground rather than floating differential. Using 10k and 100k resistors gives you a signal gain of 10. If you want a very high input impedance, you will need to use two more opamps on the inputs as gain-of-one voltage followers.

 

[vk6kro]

It might be like a UGN3503U which has an amplifier built into the package to amplify the otherwise very weak signal from the Hall Probe.

This device looks like a small transistor with 3 wires and only one output.
This might show a picture of one:
http://www.dse.com.au/cgi-bin/dse.storefront/4a28856d0eb8cfb4273fc0a87e01070e/Product/View/Z2620

 

[m718]

I have had the same problem. The Hall Effect probe is a 4-wire device vaguely similar to a Wheatstone bridge in its hookup. Voltage is applied to the top and bottom of the "diamond" and the signal is the difference of the left and right outputs. You can zero-reference these outputs if you have access to +/- 5 or +/- 15 volts for an opamp Vcc and Vee. You have to get an opamp and set up as a differential amplifier. I guess you will want 10k ohm resistors from the left and right sides of the diamond to the + and - inputs of the opamp, and a 100k ohm from the + input to ground (assuming your opamp is tied to +/- 5 or 15 V), and a 100 k ohm resistor from the - input to the output. This symmetric layout rejects the offset, and the opamp output will be single-ended referenced to ground rather than floating differential. Using 10k and 100k resistors gives you a signal gain of 10. If you want a very high input impedance, you will need to use two more opamps on the inputs as gain-of-one voltage followers.

I will use this method because I need analog resolution on the output but I have a separate amplifier that I want to use so can you tell me exact values of resisters and where to put them I will be using OP07 op amp with either 5 or 9 volts +/- and my hall sensor only has Vcc GROUND and OUT so where would I connect these.

thank you.

 

[vk6kro]

try this:

http://en.wikipedia.org/wiki/File:Op-Amp_Non-Inverting_Amplifier.svg

http://en.wikipedia.org/wiki/File:Op-Amp_Inverting_Amplifier.svg

You would connect the output of your hall device directly to the input of one of the above.

The gain is given by the ratio of the two resistors in the second case and almost the same in the first.

If you had a voltage change of 0.2 volts in each direction and wanted 1.5 volts in each direction, you would want a gain of 1.5 / .2 or 7.5.
So, you might choose 10 K for the input resistor (or R1) and two 150 Ks in parallel for the feedback resistor (or R2). You need a plus and minus power supply for this op amp and also for the Hall device. The Hall device would require +/- 2.5 volts. That is 4 power sources.

I'd still use a Picaxe chip. You can have PWM out of the same picaxe to give an analog readout if you wanted.

 

[m718]

try this:

http://en.wikipedia.org/wiki/File:Op-Amp_Non-Inverting_Amplifier.svg

http://en.wikipedia.org/wiki/File:Op-Amp_Inverting_Amplifier.svg

You would connect the output of your hall device directly to the input of one of the above.

The gain is given by the ratio of the two resistors in the second case and almost the same in the first.

If you had a voltage change of 0.2 volts in each direction and wanted 1.5 volts in each direction, you would want a gain of 1.5 / .2 or 7.5.
So, you might choose 10 K for the input resistor (or R1) and two 150 Ks in parallel for the feedback resistor (or R2). You need a plus and minus power supply for this op amp and also for the Hall device. The Hall device would require +/- 2.5 volts. That is 4 power sources.

I'd still use a Picaxe chip. You can have PWM out of the same picaxe to give an analog readout if you wanted.

The part number is AD22151 .
the reason I can't use a microcontroller is because I will need microvolt resolution.
also I only know how to blink LEDs with Pic MCUs.

 

[vk6kro]

I hope you don't really need microvolt resolution. If you do, there might be problems with mains voltage hum on the signal.
Anywhere near a city will give you big problems at that resolution.

This is how you flash a LED light once a second with a picaxe:
High 2, pause 500, low 2, pause 500, high 2, pause 500... (500 is in milliseconds)
or even
start:
toggle 2
pause 500
goto start

What would you use as a readout? Would it go to a meter or to a computer?

 

[m718]

I hope you don't really need microvolt resolution. If you do, there might be problems with mains voltage hum on the signal.
Anywhere near a city will give you big problems at that resolution.

This is how you flash a LED light once a second with a picaxe:
High 2, pause 500, low 2, pause 500, high 2, pause 500... (500 is in milliseconds)
or even
start:
toggle 2
pause 500
goto start

What would you use as a readout? Would it go to a meter or to a computer?

I'm planning to amplify it with a gaussmeter amplifier basically I will replace the sensor it has which is a coil type with a hall type, which is why the high quiesent was a problem and then I'll take the output from the gaussmeter and send it to a Pic MCU.I saw on NVE site that for very low field detection the noise is the limiting factor in their sensors and that some kind of AC modulation/demodulation should be used. Do you know how to implement something like that?

 

[vk6kro]

No. Not really. I've heard of it but never tried it.

I would think that the Earth's magnetic field and the pickup of mains hum would be more serious problems than the noise of the device itself.

Why not build up a separate unit before you try pulling the Gaussmeter to bits. At least you can then get an idea of how serious the noise problems are likely to be.

 

[m718]

No. Not really. I've heard of it but never tried it.

I would think that the Earth's magnetic field and the pickup of mains hum would be more serious problems than the noise of the device itself.

Why not build up a separate unit before you try pulling the Gaussmeter to bits. At least you can then get an idea of how serious the noise problems are likely to be.

I have no idea how to start building an amplifier as good as this. I think this uses AC modulation/demodulation and I'm not really taking it apart just replacing the sensor that's all I made a backup one just in case.

 

[Bob S]

Here is some information regarding using the Analog Devices AD22151 and OP07. This will shift your reference voltage to zero.

AD22151=================
Vcc pin 8 + 5 volts
gnd pin 4
R1 15k pin 2 to 3
R2 10k pin 7 to 6
R3 10k pin 6 to 5
Pin 7 reference output to OP07 -input thru a 10k resistor

Pin 5 sensor output to OP07 +input thru a 10k resistor

OP07 ================
Vcc +5 volts pin 7
Vee -5 volts pin 4
put 10k resistor from +input to ground
put 10k resistor from output to -input

This will give you 0.8 mV/gauss out of AD22151 (see their eq 1.)
The gain of the OP07 is +1 with the 4 10k resistors
Output signal now ground referenced.
Look at the following for differential opamp circuit
http://www.ecircuitcenter.com/Circuits/opdif/opdif.htm
I hope this helps
Bob S