I'm using the RS 304 267 hall effect current sensor, inputting the voltage output of it into an A2D converter.
The A2D converter uses a reference voltage to determine the smallest voltage change required to result in a change of 1 bit of the output. The A2DC outputs 8 bits and therefore the reference voltage is divided by 256 to determine the smallest input voltage change that results in a change in bit output. A display then uses these 8 bits to display a value.
The data sheet for the RS 304 267 shows a relationship between Gauss(which it uses as 'input') and voltage(output). This is tricky to use since I want to use the bits that is the output of the A2DC to display current. This means i have to find a relationship between gauss and current so i know what current corresponds to the output voltage. Is there any way to use this same current sensor so i know what current corresponds to the voltage being outputted? Or perhaps, another current sensor that actually takes current as an input? I would prefer an arrangement that takes an input current and gives an output voltage depending on the input.

 

You could measure the voltage across a resistor.
I=E/R. If the resistor is 1 Ohm, then a 1A current through it will produce a difference of 1V across it.

This is safer than using an ammeter. It's easy to blow the fuse on an ammeter if you make a mistake. If you burn up a $1 resistor, nobody will care.

 

You can experiment with the output. Simply use an ammeter to measure the current as you examine the sensor output. The data sheet I have is a bit vague about the inner workings of the sensor, but it should be simple enough to observe what changes with current. Feed it 5 volts and ground, and look at the other two pins. If the output voltage is a bit puny, amplify it with an op amp before the converter.

 

What can i use to check the voltage across the resistor continuously if neither end is at potential 0? Comparator?

Do you think the relationship between the input and output of the sensor would be approximately linear?

 

What can i use to check the voltage across the resistor continuously if neither end is at potential 0? Comparator?

It does not matter what the absolute value of the voltage is on either side of the resistor; only the difference in the voltage across it.

A digital voltmeter could be used to measure the voltage drop.

You CAN use an ammeter in series, but you'll need a load somewhere. Otherwise, even a small amount of voltage difference across the wire will result in a large current flow.

Do you think the relationship between the input and output of the sensor would be approximately linear?

You would have to determine that by looking at the datasheet, and experimentation.

The conductor should be in a fixed position on the Hall-effect sensor. If it moves even a small amount, your readings can vary widely.

I've attached the datasheet for your Hall-effect device.

 

A digital voltmeter could be used to measure the voltage drop.

I want to continuously monitor the circuit. When you say voltmeter i'm thinking of the one you use just for measurement when you connect the leads.... Is there a component like a voltmeter that i can place in my circuit and also connect the voltage that the voltmeter is reading to my A2DC. I don't want to see the voltage. I just want to use it as input to the A2DC.

 

We are suggesting that you "see" the voltage for long enough to discover the magnitude of the sensor output and the relationship to the current. After you have discovered that, then tailoring the output to the ADC and converting the numbers to engineering units will be easy. And you only have 8 bits of resolution - you're going to want to use all of them.

 

Okay. I understand. But if the relationship is non linear then i can't use it. If this happens what do you suggest? Another sensor?

I was thinking about the same 1 ohm resistor connected to inputs of an op amp.

 

If it is non-linear, as long as it is repeatable, you can use a look up table to translate the reading from the A2D to the corresponding value of current.

hgmjr

 

I would rather not get into making LUTs.. I already have a LUT I'm supposed to use where the relationship is approximately linear.