Choose power supply for long time outdoor using

[TomWS]

I have 10 thermosensors ds18b20 which measure in this 3 sec window and they powered from digital pin of MC.
so maybe consumption grows because of these sensors.
<snip...>

The DS18B20 sensors take some time to power up enough to provide a reading (750mS IIRC).  Their current isn't very high, but, when you wake up and take a reading from your sensors, keep your radio asleep until you're ready to send the entire packet.  I think your power profile will be a lot lower if you do.  Single digit mA until you turn the radio on and then, as WhiteHare suggested, TX only, don't bother waiting for Ack.  It's a waste of power if you're sampling so often.

Tom

[d00m178]

ok
thank you all for advices..

but seems I have new issue.

My results of measurement of power consumption which I have (work mode ~50mA and sleep mode ~0.05mA) I did when my device was powered via USB FTDI adapter, which power MC via VIN pin I suppose.
So 5V from USB goes through onboard stabilizer and then convert to 3V.
But when I power device direct to VCC pin from my LiPo battery my multimeter shows different result for sleep mode - about 0.7mA ! for work mode result the same - ~50mA.
I decided to power device from battery via VIN and it works ok, but when I connect my multimeter into the circuit it goes crazy and blink with different values and device doesn't work. Seems it is not enough power for device when multimeter connected.
 
So I'm stuck - I can't check how many power my device consume when it powered via VCC from battery..
Any help and explanations will be very helpful.
Thank you in advance.

[WhiteHare]

It may be that the voltage drop across your ammeter (aka "burden voltage") is so high that your circuit gets inadequate voltage to operate.  Try using your multimeter as a voltmeter instead and measure the voltage drop across a 1ohm resistor as a proxy for a current reading.  If you want higher accuracy than that, consider buying a uCurrent Gold from Dave Jones (at EEVblog).

[d00m178]

Yes, Im also thinking about voltage drop.

But I'm totally don't understand why power consumption grows to 0.7mA if I power device via VCC from 3v battery?
Is there some explanation for such effect ?

Or I just need power it via VIN and hope that consumption will be the same as when I power it via VIN from USB adapter?
Actually I have 3.8v on battery contacts which shows me my multimeter, so after VIN it should be 3.3 which means that stabilizer works and drain battery.

[joelucid]

The DS18B20 sensors take some time to power up enough to provide a reading (750mS IIRC).

BTW, one issue I initially overlooked with my sensors is that the 328p should be powered down while sensors are powering up. Otherwise you waste a lot of energy just waiting for the sensors to get ready. A battery powered node should not contain any invocations of the function delay() for this reason unless delays <50ms are required (which aren't available via the WDT). Use LowPower.powerDown instead.

[TomWS]

BTW, one issue I initially overlooked with my sensors is that the 328p should be powered down while sensors are powering up.

Does this mean even GPIO powered sensors?  I hadn't thought of that. Sequence would be:

• Wake up from listen mode burst.
• Set the GPIOs that power the sensors.
• Send the reset, SKIP_ROM sequence to the one wire sensors.
• Lowpower.Sleep for 750mS.
• Do the readings.
• Wake up the radio, send the info, and go back to sleep.

Cool...  Ya know, Joe, one of these days you just might amount to something!   

Tom

[joelucid]

Yup - that's how I do it 

[damonb]

...with RFM12B radio I have temperature measuring nodes that run almost 3 years on a single AA alkaline cell via a 3.3V boost chip, while sending a measurement every minute.

@damonb That impressive!  Which 3.3v boost chip are you using?  I ask because I was under the impression that the boost circuits added considerably to the current drain, even if the Moteino is asleep.  If that's not true, then I would maybe like to do the same.

@WhiteHare: My sensor network is built on hardware and libraries from JeeLabs. The long-lived temperature node is a JeeNode v6 with a single DS18B20 sensor powered by 3.3V (always on) and the whole thing is powered by a AA Power board. The boost regulator chip is a  LTC3525. The LDO 3.3V regulator on the JeeNode is left in place i.e. not bypassed. I haven't measured the actual current drain but the battery life speaks for itself.

[WhiteHare]

...with RFM12B radio I have temperature measuring nodes that run almost 3 years on a single AA alkaline cell via a 3.3V boost chip, while sending a measurement every minute.

@damonb That impressive!  Which 3.3v boost chip are you using?  I ask because I was under the impression that the boost circuits added considerably to the current drain, even if the Moteino is asleep.  If that's not true, then I would maybe like to do the same.

@WhiteHare: My sensor network is built on hardware and libraries from JeeLabs. The long-lived temperature node is a JeeNode v6 with a single DS18B20 sensor powered by 3.3V (always on) and the whole thing is powered by a AA Power board. The boost regulator chip is a  LTC3525. The LDO 3.3V regulator on the JeeNode is left in place i.e. not bypassed. I haven't measured the actual current drain but the battery life speaks for itself.

Thanks!  That's a nice chip.  You may also be interested in a variant of that chip, which allows for a passthrough:  http://cds.linear.com/docs/en/datasheet/3525d33fb.pdf  I presume that way you could power-up everything (so you don't have to worry about backfeeding) to 3.3V prior to using the DS18B20 and then switch back to pass-through afterward.  The passthrough is controlled by a single pin, so switching between 3.3V and passthrough would be easy to implement.