Good morning all,

As part of my masters dissertation, it is going to be necessary to measure the energy consumption (quite accurately) of some LoRaWAN radio devices. At this moment in time, I am thinking to measure the consumption in terms of milliamp hours. However, the problem is: How to measure the consumption?

I have been searching online for 'milliamp hour meter' and similar search expressions, but all I really uncover are cheap Chinese 'energy meters' designed for measuring energy consumption in domestic mains-powered AC devices.

I'd appreciate some advice on any appropriate meters/other hardware that can measure energy consumption, preferably in milliamp or microamp hours (amp hours will not give enough resolution - I'd have to run the devices for tens of months or even years to consume an amphour unit of energy).

Failing that, is there a circuit I can construct and (for example) connect to the ADC of a microcontroller in order to build a 'home-rolled' meter? Maybe a simple in-line resistor to measure voltage drop (and therefore infer current flow) and integrate it over a period of time? However (and this is why I'm looking a meter) the current draw is (mostly) mostly very low - tens of microamps. However, when these devices transmit they typically draw ~40-50mA, and typically ~10mA in receive mode - so the range that I need to measure over is huge. I presume a simple in-line resistor connected to an op-amp or something is ruled out by the large range that needs to be covered? I'm a software person and this isn't really my area! Hence my request.

I'd appreciate any advice from those in the know!

Kind regards

Mark

 

Hi Mark. I think your microcontroller idea would be the first thing I would attempt. However, before I went through the trouble building/programming, I would investigate the wide variety of voltage and current data loggers out there which do just that. It might get your project up and running a lot faster.

 

If you find that you have to build your own monitoring/measurement circuit, I would recommend you consider an AD637 as the front-end. This device has a very wide measurement range (60dB) and has the advantage of changing the AC input signal to a positive signal proportional to the rms, mean, or absolute value - which will make logging simpler.

The datasheet is available here:-
https://pdf1.alldatasheet.com/datasheet-pdf/view/48103/AD/AD637.html

Ebay sellers offer a ready-made module that might be useful (but not cheap).
https://www.ebay.co.uk/itm/1Pcs-AD6...111730&hash=item1a48b781d5:g:fFsAAOSwj3xbPkYi

One important thing to bear in mind is that the IC does not provide any isolation from the measurement circuit (which could be in the mains supply).

The device below might be good enough to provide a useful signal in the 10s of µA range – feeding its output to the AD637.
https://www.ebay.co.uk/itm/ZMCT103C...var=671096078507&_trksid=p2060353.m1438.l2649

 

There are quite a few power supplies intended for use in design lab operations that provide data outputs, both voltage and current. These could be connected to a computer, probably with a USB hookup, and suitabkle logging software could provide you with records accurate within 1%. AND it is often possible to rent such equipment at a rate much cheaper than purchasing it, if your need is only for a semester or two.

 

This is what I use for recording such things:


A lot of work to make one but it can store and play back either through a serial interface or its DAC a waveform over seconds to weeks, up to 128 kB. You would need a simple circuit to convert the current to a voltage since this only records voltage.

The output is a CSV file you can capture via a terminal emulator and then analyze that with a spreasheet.

https://bit.ly/2BMCmUe

 

The following appears a ridicules idea – but it might just work; certainly all the parts could be purchased for less than £10.

In the neutral line of the circuit that you want to measure the current flow, place a bridge-rectifier. Connect the output of the bridge-rectifier to one of these (observing the polarity), with the output of the device fed into a 5.1V zener diode. You may need to add a capacitor across the zener to smooth the voltage. If the zener diode has a 5W power rating, the circuit should be able to handle up to 1A.

The bridge rectifier diodes should be capable of handling the applied reverse voltage (1N4007 types).

https://www.ebay.co.uk/itm/USB-Volt...747723?hash=item523498ce0b:g:8aQAAOSwHkhesU4c

To connect the zener to the output of the above device you could consider this:-

https://www.ebay.co.uk/itm/Double-e...684832?hash=item3b28dda520:g:FUoAAOSwuHdcRuUL

Remember that the complete circuit has no isolation from the mains, so you should take suitable precautions in use.

 

While the circuit in post #5 looks like it would work, the effort to duplicate that package and make it accurate by calibration will be large. Renting a system designed for the exact task at hand would not only be faster and cheaper, it would also provide a paper trail back to calibration. Knowing that the data is correct may have some value in the future.