ASC712 - measuring current shows only default value

Hello everyone,

So I got ASC712 sensor and installed it on the device (coffee heater). Since it is 5A sensor I did not attach it to the wires leading up to the heater itself, as I believe it can draw over 20A of current, I connected it to an indicator lamp connected in parallel to the heater. This way I think it will not draw that much current.

Anyway, it is connected and working, by that I mean I can read a nominal value of 2.5V on the measurement pin with multimeter. I can also measure that with the Arduino.

The problem is that it doesn't matter if heater is on or off, it is always 2.5V (give or take 0.02V shift). I tried switching the wires to the sensor but that should and did not give any effect. Any other idea why it is not measuring anything?

Best regards, D.

That’s because you now only measure the indicator light and it probably draws very very little current…

If you just want to know if the heater is on or off it’s not really the right device.

Thanks for the answer, I was afraid it is like that. So what would be the right device for that? One idea is light sensor for indicator light but am looking some more direct way.

Do you want an real easy option or a neat option?

Real easy option Grab a USB charger, connect it in parallel with the heater. Connect the GND of the charger to the Arduino GND. Connect the 5V of the charger through a 10k resistor to a pin.

Neat option Use an opto coupler. But you need to add a diode (bridge) in front of it and you need to limit the current. To do that with less heat and because you only want on/off you can use a capacitive dropper. But I would use a bridge and the circuit does need more modifications / other values. So it's more advanced and a wayy bigger shock / BANG risk.

Thank you for the help and ideas!

You must be careful that if you have a bidirectional current sensor the zero would be Vcc/2. What you reading it is the zero of the ACS with not current. To get a reading you must subtract 2.5 from the reading. A negative reading means current go in in or a positive means it is out. Checked page 6 of the data sheet. You would see that the zero it is at 2.5volts. By looking at the graphic 5 amps = 2.5 -3.40 = .9 volts
Read the ACS voltage Vcc /2 and subtract it from the reading.

@tauro0221, that's exactly what he's saying, he reads Vcc/2 all the time. Not that weird because he only tries to measure the current to a (neon?) indicator light...

Hi, I just missed reading which current to read. I was thinking he wanted to read the heater current. The ACS712 it will not read the indicator lamp current. It is too small. Why not change the ACS512-5 amps for the ACS712-30 amps and connect it directly to the heater. I know that it is easy to say than do it. But that's what he should do? Not?

But why read the current if he's only interested in the state (on or off)? ;) That's like exactly measuring the level of liquid while you're only interested if you have a cup or not :D

Hi, Since he is turning on the heater he knows when it is was turned on. Just use another port and turn on the indicator light. He does not need to read the indicator light current with a ACS712 sensor to find out if it is On or Off. No?

Yes, he turn on the heater. But he never writes he does that via a uC ;) As far as I know he just wants to see if it's (the heater that is) turned on by some external device.

Hi, In that case your suggestion of using the opto coupler it is the best solution.

The default sensitivity of the ACS712 is 66mv per amp so its not suited to measuring very small currents. A really simple solution would be a LDR glued onto the indicator light. Just connect a resistor in series with the LDR and read the voltage change when the indicator light is on.

But if that indicator light is a neon you just opened up a new can of worms...

I had successfully use the ASC712 to monitor a minimum 7W Led lamp with it.My solution has using an opamp with a comparator reference of 2.5V to supress the sensor output and amplify the difference +- 100 times.Then measure that signal in arduino analog input ...

But I doubt (and hope) an indicator will not even use 1W, let alone 7W...