I’ve been working with a couple of these boards, and I’m not sure if what I’m seeing is a limitation in the Hall-effect IC or something else. The one I’m having problems with is producing 2.5V out at 0A, which my Arduino Mega 2560 using a 4.096v precision analog reference says is 625.74 “bits” (it’s an average of 105 ADC readings taken at ~10ms intervals, throwing away the first 5, thus the decimal point), while a Fluke meter says 2.535v, which is pretty close. Reading current in an ~70VDC circuit in series with the anode of a deuterium lamp, I get 651.57 “bits”, minus 624.74 offset = 27.16 bits. Multiplying by the ADC resolution (4.096V/1024 bits=.0040 v/bit) gives me 0.109v, times the ACS723 sensitivity of 1.0A/0.400V gives 272.5 mA. A Fluke 289 meter in series with the ACS723 reads 293.5mA (which is very close to the ideal specification for the circuit, or ~300mA). The other ACS723, which is in the 24VDC supply line to the board running the lamp, is reading very close to what a meter indicates. Any suggestions? Anyone see errors in my methods?
-Bill
Thanks for reaching out to us on this.
It looks like the ACS723 reports 0A at 1/2 VCC. We carry the 5 volt version so that would be 2.5 volts.
For every amp of current you run through the board, the output voltage will change by 400mV (0.4V)
Running 1 amp of current through the board in on direction would net an output of 2.9V, 1 amp in the other direction will net 2.1 volts.
Sounds like one board is reading correctly, and one is not? If that is the case, please ensure there isn’t ANYTHING near the set-up of the one giving you problems (even a nearby screwdriver, wire, metal bit, etc) can distort readings via inductive interference.
If it ends up being determined that the sensor is faulty (try swapping them?), simply fill out this form https://www.sparkfun.com/returns and we’ll get you squared away.
Have a great day!
I tried replacing the board seemingly giving an incorrect reading with another one I had around. It reads very close to the same as the original. I’m not entirely clear on what you mean by the “set-up” of the board; it’s connected to power, ground, and signal by wires, as well as being inserted into the circuit where the current of interest flows. It’s difficult to keep wires away from it when they’re needed to connect it. I’ve made sure there aren’t any wires other than those necessary near it, and it’s reading the same. Any other ideas?
If higher accuracy at low currents is needed, maybe this will help: https://www.sparkfun.com/products/14544
It might help if you could snap/share some photos of your set-up. The datasheet offers plenty of info about the inherent accuracy of the component https://cdn.sparkfun.com/assets/d/0/0/5 … asheet.pdf - page 11 shows a few ranges
It also could prove useful to measure the output voltage of the sensor with his meter when he applies a_ known_ current.