Power Shield High Current Draw [SOLVED]

[NixHydra]

Hi Felix,

   Recently I purchased one of your Power Shield boards, as I wanted to try and run a Moteino R4 from a
1.2V NiMH battery instead of a 9V alkaline battery. I have played with the MCP1640BT and LTC3525-3.3
boost chips previously, so was curious to see how the TPS61220 would perform.

   Just soldered two 13 pin headers to the Power Shield and left the default output voltage set for 5V. This
way when the Moteino R4 ( sleep mode Node sketch flashed) board (with RFM69HW wireless module) is
plugged into the Power Shield the Moteino Vin is fed 5V. Powered up the Power Shield and the Moteino
and measured the current draw from the 1.2V NiMH battery. Checked the Power Shield output voltage
with a multimeter and found it to be 5.11V, with the single NiMH battery attached. Also checked the Moteino
Vcc, which was found to be 3.30V. As you can see from the attached photo, the current draw is a MASSIVE
4.4mA. No good for any prolonged battery use. Same Moteino, running same sketch, when run from a 9V
battery draws 15uA. Any ideas?

                                  cheers,

[Felix]

Since you're only powering the Moteino I would suggest switching to 3.3V (cut 5V jumper trace, resolder 3.3 side) and see how that performs, I would expect a significant improvement. Assuming you keep Moteino asleep most of time.

The higher difference between input and output voltage on such a booster the higher the inneficiency. The booster will work very hard to create your 5V so if you look at the DS efficiency curve that's towards the bottom efficiency.

I will need to make more of these tests myself though and evaluate what is the sweet spot for these.

[NixHydra]

Hi Felix,

   Thanks for getting back to me. As you suggested, I tried setting the Power Shield output voltage to 3.3V instead of 5.0V and repeated the previous experiment. On attaching ONE NiMH battery, I measured the boost output voltage as 3.32V and the Vcc as 3.30V with a multimeter and a current draw of 2.2mA. Please see fig(i) attachment. On attaching TWO NiMH batteries in series, I measured the boost output voltage as 3.32V and the Vcc as 3.30V with a multimeter and a current draw of 0.9mA. Please see fig(ii) attachment. Both scenarios have too high a current draw to run the Moteino for any time from NiMH battery/batteries. Next I intend to remove the Moteino MCP1703CB LDO regulator and feed the Power Shield output voltage of 3.3V directly to the Moteino Vcc and check the current draw with both ONE and TWO NiMH batteries again.

   I know there are at least 27 Power Shields "out there" and was wondering if others who have bought Power Shields have had any luck running a Moteino/Power Shield combo with ONE NiMH battery and are getting a current draw in the 7 to 15 uA range when sleeping? Anyone got any ideas?

                                 cheers,

[NixHydra]

   Just a follow-up on checking current draw for the Power Shield/Moteino R4 combo. Removed the MCP1703CB LDO regulator from the Moteino R4 board and soldered  a wire link between the Moteino R4 board Vin and Vcc pins. Measured the Vcc with a multimeter to be 3.35V. With ONE NiMH battery connected, measured the current draw as 3.1mA. Please see fig(iii) attachment. With TWO NiMH batteries connected, measured the current draw as 0.8mA. please see fig(iv) attachment. These high current draws make running this combo, from battery/batteries, for any reasonable period of time not practical.

   Unless a solution can be found for this, it would seem that the Power Shield can't be used to run a Moteino in sleep mode from a single NiMH battery and expect reasonable battery life. Any input, suggestions, user experience with the Power Shield appreciated.

                                        cheers,

[Felix]

One thought is that a non true RMS multimeter might give a false reading.

[NixHydra]

Hi Felix,

   Once again, thanks for getting back to me. Am surprised have not heard back from someone else who has purchased the Power Shield, considering there would have to be a few out there by now. Sorry to be a pest but is there any chance you could attach one of your Moteinos to a Power Shield and power it from a single 1.2V NiMH battery and measure the current draw to see what you get? It's probably something I'm screwing up but it would be good to know for sure, so I can use my Power Shield combo and power it up for a prolonged period of time with just the ONE NiMH AA battery.

                               cheers,

[NixHydra]

Hi Felix,

   Just checked out the shop and noticed you still have SIX only of the Power Shields for sale. Any chance you have had an
opportunity to confirm that the current draw for a Moteino/Power Shield combo, when sleeping, is in the 7 to 15uA range?

                                   cheers,

[john k2ox]

I believe Felix is correct about your meter.  The chip is doing something akin to PWM.  You are measuring the peak current, not the RMS or average current.

Power is voltage times current.  Every time the voltage is reduced by two the current goes up by two.

Energy = time * power. 

[NixHydra]

Hi john k2ox,

   Thanks for the reply. Any chance you have purchased one of Felix's little Power Shields and had the opportunity to measure the current draw when it is used in combination with a Moteino R4, spending most of it's existence in sleep mode? Really had expected, given the number of these little guys out there, to have had some more feed-back ... Ah well!

                                            cheers,

[john k2ox]

No I haven't.     Most 'high tech' batteries tend to supply a constant voltage then fall off the cliff.  Carbon zinc tend to slowly head south.  I like to choose enough cells to get the job done without adding any additional circuitry.  I like doing the most with the least.

I do think that the power shield is most likely very efficient.  If it isn't getting warm it isn't eating power.

Maybe someone with a scope can take a look.

Nix, you're sounding more like an engineer than a hacker. 

[Felix]

Ok let's wrap this one up already.
Here are the results of some tests with my true-RMS DMM, a Moteino with radio put to sleep and sketch that sleeps for 8seconds then wakes up and blinks the LED for 5ms, then back to sleep:

- using 2xAA (pretty depleted ones, total about 2.55V) I get 58uA draw from the battery while Moteino sleeps
- using 1XAA (about 1.3V) I get about 110uA draw from the battery when Moteino sleeps
- checking current between PowerShield and Moteino I get the same 6.6uA while Moteino sleeps, and the same if tested from a FTDI adapter

Conclusion:

1) Don't use a $5 DMM to test for current
2) Don't use a $5 DMM at all
3) The TPS62222 is very efficient in my opinion and does a great job of giving your 5V/3.3V output
4) No real concern about "High Power Draw", the extra current that the shield draws is inherent to the booster operation, ie it is working hard to provide the voltage difference
5) Revisit 1) and 2) again from time to time 

[NixHydra]

Hi Felix,

   Thanks for taking the time to actually check the current draw when using the Moteino R4 and Power Shield combo.
Empirical data of 58uA at approximately 2.55V and 110uA at approximately 1.3V, think I will give the boost approach
a big miss and stick with a 15uA current draw using a KISS 9V battery.

   John, hacker/engineer ... not mutually exclusive!!

                                             cheers,

[kobuki]

Just to note:

9.0  *  15 uA = 135 uW
2.55 *  58 uA = 147.9 uW (+9.6%)
1.3  * 110 uA = 143 uW (+5.9%)

So they're not at all very different. 5..10% is still a difference, though. For cheapo batteries, there might not be any significant difference in battery life at all. Just my 2¢.

[Felix]

Empirical data of 58uA at approximately 2.55V and 110uA at approximately 1.3V, think I will give the boost approach
a big miss and stick with a 15uA current draw using a KISS 9V battery.

Not sure what you mean. But as kobuki noted the total power draw is not significantly different, I think it's very reasonable.
When you try to revive a "dead" battery it might give you a very linear response, so it may be different than an equivalent voltage but constant current source. That's why you can't possibly compare a full 9V battery and a "dead" AA that reads 1.3V at no load. It's much easier to go from 9V to 5 or 3.3V (and waste some heat in the process) than to boost 1.3V to 5V and still be very efficient.

[NixHydra]

Hi Felix,

   Sorry for the confusion but simply meant that I will keep powering the Moteino really simple and not use a boost chip.
This way it cuts down on the circuit complexity as well as cost. As I said, I'm pretty happy with just running a Moteino
from a 9V battery with a 15uA current draw when sleeping.

   kobuki, thanks for the 2 cents worth and got to love that "wife detector" anecdote of yours.

                                        cheers,