# Current Sensing

I am interested in measuring current in the range of 0 - 5 A with some degree of reliability. Does anyone know of a calibrated chip that will allow me to do this? What I have found is that there are some great sensors out there ( such as the Allegro Microsystems ACS750SCA-050 ) but the current range is so high the resolution is low.

For example, the above chip has a +/- 50 A range. At 40mV/A that translates into about 125 mA / step using the Arduino 10 bit analog input. At 12 V that has a sensitivity of about 1.4 W / step. I'd like something with about 10x that range but I can't find anything other than simple hall effect transistors that would require s/w calibration ( I think ).

I have no experience with this type of sensor for any help greatly appreciated.

current sense transformer

I have one that puts out 0.10vdc per amp.
http://www.pearsonelectronics.com/current-monitor-products/standard-current-monitor.htm

Or the easier way for small currents is to make a shunt.

Java applet
http://www.phy.hk/wiki/englishhtm/Meter.htm

Note: You'd want to feed the lower shunt voltage to an optocouple then to the arduino. I had a meter blown up using a shunt to measure 3kv when a connection came loose.

Wow, thanks - exactly what I was looking for. Hopefully this post will be helpful for others traveling down this path.

no problem

You didn't say if it's AC or DC. In either case offer the arudino some respite and try and isolate it from the current side.

If it's AC and you use a current transformer that will do the isolation.

If it's a shunt then you'd want the optocouple. Calibration will be a challenge since you're adding parts.

Good luck.

Originally I went off the shunt method because of voltage loss, but after looking around I'm leaning back in that direction, particularly sinceI'm measuring DC. Specifically, I am making a remote monitor for a solar powered wireless access point. I intend on using the Arduino to measure/ send out serial V/A measurements to a wrap board that is runing a Linux AP. In this way I can get an nice graph of solar, battery, and power consumption over time, as well as a watchdog on the system.

One thing I was considering was using a 1/4W 0.1 ohm resistor, which at full load for my test ( 1A ), would produce a voltage drop of 0.1V ( power of 1/10 W ). The thing that I think might be cool about that is once I measure the current, I can then have the Arduino offset the voltage loss when measuring load voltage. To get the sensitivity up from ( 0 - 0.1 V ) I would use a simple op amp ( like a 741 ) at 100x gain to give my a 0-10V range.

This is a sub \$1 hardware implementation, but I'm not sure if there are any real world problems implementing it that way. Does anyone have any experience witha simple (and obvious) shunt / op amp circuit like that?