Archived Forum

Hi Pat,

Both the schematics of the emonTH and emonV3 are up on solder pad: http://solderpad.com/openenergymon/

The issue is that on the emonTH the battery voltage is lower then the supply voltage since a boost converter is used, this makes it easy to simple measure the battery voltage through an ADC input.

In the emonTx V3 with battery voltage is higher than the supply voltage (3x AA's). I resistor voltager divider will need to be used to reduce the voltage.

A divider with 47K on top and 100K on bottom or 470K on top and 1M ohm on bottom should do the trick. The latter will consume less power but might be a bit noisy since the values are so high. The voltage from the middle of the divider can then be sampled into an ADC input like on the emonTH.

emonth.battery=int(analogRead(BATT_ADC)*0.03225806);                    //read battery voltage, convert ADC to volts x10

I've got this on the list as possible additions to next PCB revision.

Glyn, you put a small value capacitor to ground to reduce the noise. Then the resistor value isn't in the equation.

See http://www.dsplog.com/2012/05/06/thermal-noise-rc-low-pass-filter/

There was also some good discussion of this over @ JeeLabs. Without getting into detail, he ended up using a FET to control whether current flowed through the divider. Of course, that requires an additional IO pin, not sure if that's available on emontx.

Here's the link for reference: http://jeelabs.org/2013/05/18/zero-power-measurement-part-2/

That circuit does not bypass the divider resistors, so even though the resistance is about one 50th of ours, it might actually be noisier. According to the way I do the sums, "our" input circuit has a noise bandwidth of 0.06 Hz and the rms noise amplitude is about 5 nV. I think the software high pass filter should be able to remove any residual thermal noise that gets through the RC filter.

Thanks, somehow missed the schematics there. Sorry. I'll give this a shot over Xmas and report back. 

Pat

No, there is no bypass circuitry there. The intention in posting it was a circuit that allows monitoring battery voltage without constant power draw. It does nothing to improve accuracy, but looked like you've got that covered already.

Please make this easy for me. I notice the latest payload from the emonTX V3 includes "battery" information. I assume I need to modify the hardware somewhere to enable this but don't know how. Can someone please let me know what needs to be done.

Thanks in advance,

Geoff

You need to do as the post by glyn.hudson on Wed, 18/12/2013 - 11:19 above.