Fork me on GitHub

Project Notes

#572 PowerMeter

A voltage, current and power meter using the INA226, with an ATtiny85 driving the process and displaying measurements to a 128X32 OLED display.



I played around with INA226 high-side power monitoring in LEAP#504. It is a great little chip with an I²C interface.

This project puts it together with an ATtiny85 and 128X32 OLED display to act as an inline power meter.

Designing for Load Power

The INA226 is programmed with a calibration factor (CAL), a function of the shunt resistance and the current register LSB (Current_LSB).

The current range that can be handled is fundamentally determined by the full-scale range of the shunt voltage register: ±81.92 mV (2s complement FFFF-7FFF) i.e. LSB representing 2.5 μV. Having this fixed resolution of 2^15 parts means of course that the higher the current we design for, the lower the resolution.

For this build I’m going to design for a relatively modest maximum current of 500mA, which would mean using a shunt resistor of 0.16Ω for full scale deflection. If I replace that with a standard 0.2Ω value, it reduces the full scale current range to 409mA, but with a resolution of 12.5µA.

At maximum current, the shunt resistor would only dissipate 33mW, so 1/4 watt resistors are fine.

The shunt resistance and the maximum design current are configured in the PowerMeter.ino sketch, for the calibration factor (CAL) to be automatically calculated.

Two things are obvious and important:

  • handling a different current range means re-compiling the source (this could be avoided by adding a range selector input to the project)
  • the actual value of the shunt resitance really maters! Either measured with a very accurate mΩ meter, or calibrated in-circuit. Especially for breadboard tests, I calibrated in-circuit, as by the time breadboard traces, mounting pins etc are taken into account, an R020 0.02Ω resistor was actually more like 0.19Ω in total.

When run with an Arduino Uno, the internal settings are reporting over the serial port, for example:

INA226 Power Meter
Manufacturer ID : 5449
Die ID          : 2260
Calibrating for shunt resistance of 0.200Ω and maximum current of 0.500A
  * current_LSB : 0.006554mA
  * power_LSB   : 0.163838mW
  * CAL         : 3906

Arduino Breadboard Prototype

To verify the basics, I first setup the project on a breadboard, and ran it with an Arduino Uno.

When compiled for an Uno, the PowerMeter.ino sketch uses the standard Wire library for communications with the INA226 and OLED, and the u8g2lib library for driving the OLED display.



Running the Arduino prototype on a breadboard:


ATtiny85 Breadboard Prototype

For the final build, I was keen to see if an ATtiny85 would be perfectly capable of controlling the project. And it is!

When compiled for an ATtiny85, the PowerMeter.ino sketch uses the Adafruit-TinyWireM library for communications with the INA226 and OLED, and the Tiny4kOLED library for driving the OLED display.



Running the ATtiny85 prototype on a breadboard (with an Arduino UNO for programming with the Arduino ISP sketch):


Checking that it can ever run from a coin cell:


Power Switching for the Project

The power meter is independently powered. While it would be possible to tap the power supply under test, this approach introduces additional constraints and easiest to avoid.

I’m using a 3V CR2032 cell for powering the circuit - just enough jiuce it seems for the ATtiny, OLED and INA226.

So how to switch the power to the INA226 monitoring circuit? Three options:

  • just add a switch!
  • find a DC switched jack with a normally open (“make”) 3rd pin - so that plugging in the load supply would also switch on the INA226.
  • use a DC switched jack with a normally closed (“break”) 3rd pin, with a FET switch to invert the switching logic.

I like the idea of avoiding a manual switch. The best solution would be DC switched jack with a normally open 3rd pin, but while these are theoretically a thing, I’ve not been able to find any.

Considering a DC switched jack with a normally closed 3rd pin (this is ~100% of commonly available DC barrel jacks)… It is not ideal however: since I’m already operating at 3V, there’s not much headroom for any additional voltage drop across a FET, so a very low Rds(on) is essential. With a normally-closed switch, a low-side n-channel MOSFET switch configuration makes sense, but even with a high value pullup (1MΩ works) there will be a small but continuous current draw of 3µA to accept. There is perhaps a bigger issue that the low-side FET switch adds an “unknowable” voltage drop that means that bus voltage sensing will always be off.

I did try this on the breadboard and it does work:

  • n-channel MOSFET connected with drain on the low-side of the IN226 power meter and source to ground.
  • FET gate pulled down by the nomally closed DC jack 3rd pin
  • a 1MΩ pull-up resistor on the gate to ensure the FET switches on when a plug is inserted and breaks the 3rd pin connection to ground
  • tested a few FETS that work: 2N7000 (barely), IRFZ44N, and AO3400.




So while the low-side auto-switching does work, I’m not particularly happy with it.

ATtiny85 Version in a Project Box

Until I can find a DC switched jack with a normally open 3rd pin, I think I’m back to just using a manual switch!

If this works out, I might make a PCB, but for now I’m mounting some SMD components on protoboard for a prototype build. Here’s a sketch of the layout:


And all installed in a small project box:



Next Steps - Improvements?

A few ideas:

  • add auto-ranging (adjust the calibration when readings are heading under/over range)
  • design a PCB for the circiut
  • find a DC switched jack with a normally open 3rd pin?

Credits and References

Project Source on GitHub Project Gallery Return to the LEAP Catalog

This page is a web-friendly rendering of my project notes shared in the LEAP GitHub repository.

LEAP is just my personal collection of projects. Two main themes have emerged in recent years, sometimes combined:

  • electronics - usually involving an Arduino or other microprocessor in one way or another. Some are full-blown projects, while many are trivial breadboard experiments, intended to learn and explore something interesting
  • scale modelling - I caught the bug after deciding to build a Harrier during covid to demonstrate an electronic jet engine simulation. Let the fun begin..
To be honest, I haven't quite figured out if these two interests belong in the same GitHub repo or not. But for now - they are all here!

Projects are often inspired by things found wild on the net, or ideas from the many great electronics and scale modelling podcasts and YouTube channels. Feel free to borrow liberally, and if you spot any issues do let me know (or send a PR!). See the individual projects for credits where due.