Hello everyone,
I am trying to figure out the basics on what would be the best way to power a battery Moteino and thought that maybe others could share their experience.
My objective is to define the most suitable way for each of these cases:
(always to power a 3.3v moteino + RFM69w or hw)
1) LiPo battery (4,2v to 3.6v) to Moteino (3.3v)
Given that the LiPo Battery voltage will vary from 4,2v to around 3.6v during it useful voltage range, what would be the simplest most efficient way of powering the 3.3v moteino?
Please suggestions on this one cause although it seems the easiest I cant find an easy solution.
Update (thanks TomWS)
The LiPo battery can be directly connected to the Moteino Vin. Given the LiPo discharge curve, and the low voltage drop of this regulator, I guess that we only need to consider LiPo voltages from 4.2v to around 3.4 or 3.5v as valid inputs, which will mean mostly all the battery capacity. Great!
(https://lowpowerlab.com/forum/index.php?action=dlattach;topic=1425.0;attach=1066;image)
2) LiPo battery (4,2v to 3.6v) to Moteino (3.3v) + LiPo recharge circuit
The idea is to be able to recharge the lipo battery by just connecting a usb cable to it and therefore you dont need to change battery, or leave the node disconnected during recharge.
Possible solution 1?
Based on Adafruit new Feather board, they are using a MBR120 low forward voltage (0,35v) to allow powering via USB or LiPo into a SPX3819-3.3 regulator.
This setup uses a MCP73831 as LiPo charger when the USB is connected.
Power supply comes from the SPX3819-3.3 regulator.
Schematic: https://learn.adafruit.com/system/assets/assets/000/028/445/original/microcomputers_feathershc.png?1446678127
Hopefully with this setup there wont be a massive lost of power in the schottky diode + regulator. Any thoughts?
(https://lowpowerlab.com/forum/index.php?action=dlattach;topic=1425.0;attach=1060;image)
Possible solution 2?
Similar setup is Moteino Power Shield
https://github.com/LowPowerLab/PowerShield/blob/master/Schematic_Rev2.pdf
however, please confirm I am reading this correctly. The USB will power the charger MCP73831 and the charger will charge the LiPo. This output supply will arrive to
a boost converter (TPS61220) that will output 5V. This 5V will finally arrive to Vin of the Moteino which has a low voltage drop regulator to 3.3v
How is efficiency wise with so many steps?
(https://lowpowerlab.com/forum/index.php?action=dlattach;topic=1425.0;attach=1062;image)
3) LiPo battery (4,2v to 3.6v) to Moteino (3.3v) + LiPo recharge circuit using solar panel
The idea is to have a Moteino which cant be easily recharged, or one which due to its nature has a quite high current consumption and requires solar recharging.
I am used to the TP4056 lipo charger with a solar panel. Dont know if there are better options. My panel is a 18V panel so I first thought of dropping it down using a buck converter to 5.2V
and then feeding that via the usb of any of the previous options.
I have seen similar questions to this around internet, but no real solution to these specific conditions, maybe we can create a thread gathering information for anyone interested.
My ultimate goal would be to see if it is possible to design a circuit that willallow the Moteino to:
- power through LiPo battery
- allow LiPo recharge by connecting usb cable
- allow to connect solar panel
- power through USB if connected
This way, one design could fit may situations.
Thank you all.
Interesting parts I have found that could be useful when designing any of the circuits above:
Low voltage drop regulator (120mV) and 700mA ouput: http://uk.farnell.com/torex/xc6222d331mr-g/ic-ldo-700ma-3-3v-sot-25/dp/1830952 (http://uk.farnell.com/torex/xc6222d331mr-g/ic-ldo-700ma-3-3v-sot-25/dp/1830952)
Great topic! Re: #1, AFAIK, the moteino itself can handle up to 5.5v no problem. The radio though probably can't go that high, so you should indicate which one you're using. IIRC, Rtfm69 stated limit is 3.6v, but maybe it can tolerate higher for short durations? Anyone know?
@Sergegsx, without going through all of your details, it seems to me that you missed the point that the Moteino already has a voltage regulator on it so you can provide anywhere from 3.4 to 16 volts on the Vin pin. Hook the LiPo directly to this. Re charging, I prefer the MCP73812 because you can have a load while you charge. The MCP73831 is fussy about that.
Tom
Quote from: WhiteHare on November 21, 2015, 01:04:31 PM
Great topic! Re: #1, AFAIK, the moteino itself can handle up to 5.5v no problem. The radio though probably can't go that high, so you should indicate which one you're using. IIRC, Rtfm69 stated limit is 3.6v, but maybe it can tolerate higher for short durations? Anyone know?
@sergegsx Just to clarify, Re: #1, I gave the above info because usually the advice for "the simplest most efficient way" is direct connection to the battery with no regulator. So, in that case, you'd be connecting to the 3.3v pin rather than Vin, even if the actual voltage you're supplying is higher than 3.3v (e.g. 3.6v). Some alternative battery chemistries (e.g. LiFe) may be more compatible with that approach. Not sure (?), but I suspect that in such cases removing the Moteino's regulator is recommended.
Quote from: TomWS on November 21, 2015, 02:51:35 PM
@Sergegsx, without going through all of your details, it seems to me that you missed the point that the Moteino already has a voltage regulator on it so you can provide anywhere from 3.4 to 16 volts on the Vin pin. Hook the LiPo directly to this. Re charging, I prefer the MCP73812 because you can have a load while you charge. The MCP73831 is fussy about that.
Tom
You are completely right Tom, I had not seen that the MCP1703 of the Moteino will accept such low voltages.
Input Operating Voltage Range: 2.7V to16.0VTherefore, I can update the first section by saying that the LiPo battery can be directly connected to the Moteino Vin. Given the LiPo discharge curve, and the low voltage drop of this regulator, I guess that we only need to consider LiPo voltages from 4.2v to around 3.4 or 3.5v as valid inputs, which will mean mostly all the battery capacity. Great!
Regarding the charging circuit, nice information about the mcp73812. I am taking note of it and will update the first post once I have a good example of such usage.
Thanks !!
Quote from: TomWS on November 21, 2015, 02:51:35 PM
@Sergegsx, without going through all of your details, it seems to me that you missed the point that the Moteino already has a voltage regulator on it so you can provide anywhere from 3.4 to 16 volts on the Vin pin. Hook the LiPo directly to this. Re charging, I prefer the MCP73812 because you can have a load while you charge. The MCP73831 is fussy about that.
Tom
Checking the different options, I cant find what you mention about the charge with load, do you have any info?
I see something interestring about the MCP73832, it can automatically recharge the battery, maybe this is useful when adding a solar panel. see attached.
My idea regarding recharge so far is to include 3 connectors:
- usb in
- solar panel in
- lipo in
Based on previously mentioned examples, by adding a schottky diode MBR120 (0,35v drop) to each input, we could obtain an all in one solution where lipo would be charged either by solar o usb connector. The node would be powered by the input with highest voltage into the MCP1703 regulator.
something similar to this.
https://www.sparkfun.com/products/12711 (https://www.sparkfun.com/products/12711)
see picture attached
Quote from: Sergegsx on November 21, 2015, 05:31:40 PM
Quote from: TomWS on November 21, 2015, 02:51:35 PM
Re charging, I prefer the MCP73812 because you can have a load while you charge. The MCP73831 is fussy about that.
Checking the different options, I cant find what you mention about the charge with load, do you have any info?
Somewhere on this forum Felix posted a link to a great article describing the Microchip chargers and their behavior if there is a load in addition to the battery itself. You'll have to search for it...
Quote from: Sergegsx on November 21, 2015, 11:32:49 AMMy panel is a 18V panel....
Maybe I missed how you plan to address it, but isn't that voltage too high for this chip, where Vin (max) = 6v?
Hi, I did say I was thinking of using a buck converted to 5.2v
Maybe there are better options...
Quote from: TomWS on November 21, 2015, 06:10:28 PM
Quote from: Sergegsx on November 21, 2015, 05:31:40 PM
Quote from: TomWS on November 21, 2015, 02:51:35 PM
Re charging, I prefer the MCP73812 because you can have a load while you charge. The MCP73831 is fussy about that.
Checking the different options, I cant find what you mention about the charge with load, do you have any info?
Somewhere on this forum Felix posted a link to a great article describing the Microchip chargers and their behavior if there is a load in addition to the battery itself. You'll have to search for it...
Thanks again tom. I am going to look for it!!
At present I have built/tested nothing but am thinking about remote sensor power options.
I have been considering using energy harvesting from a 20x10mm PV cell in room lighting to maintain charge in a Li-Ion AA or AAA cell (I may even try the little postage-stamp sized cells used in the tiniest quadcopters - about 150 mAH), with reporting of battery state so that the central system can advise external recharging if energy harvesting should fail.
I am dithering between running the system from the unregulated Li-Ion cell and limiting the charge so that the cell voltage never exceeds 3.75V (absolute max for the RF modules is 3.9V) or using the 3.3V micro-power regulator. Using maximum cell voltage of 3.75V reduces the capacity by 60-70% but gives almost unlimited recharges (Li-Ion batteries degrade during the last part of their charge - if they're only charged to 4.0V or less their total life (measured in AH delivered, not number of charge/discharge cycles) increases by a factor of 10-40. Using the regulator allows the full battery capacity to be used, but it is still necessary to have hardware to prevent over-charging..
Quote from: JamesBryant on November 30, 2015, 12:56:03 PM
At present I have built/tested nothing but am thinking about remote sensor power options.
I have been considering using energy harvesting from a 20x10mm PV cell in room lighting to maintain charge in a Li-Ion AA or AAA cell (I may even try the little postage-stamp sized cells used in the tiniest quadcopters - about 150 mAH), with reporting of battery state so that the central system can advise external recharging if energy harvesting should fail.
I am dithering between running the system from the unregulated Li-Ion cell and limiting the charge so that the cell voltage never exceeds 3.75V (absolute max for the RF modules is 3.9V) or using the 3.3V micro-power regulator. Using maximum cell voltage of 3.75V reduces the capacity by 60-70% but gives almost unlimited recharges (Li-Ion batteries degrade during the last part of their charge - if they're only charged to 4.0V or less their total life (measured in AH delivered, not number of charge/discharge cycles) increases by a factor of 10-40. Using the regulator allows the full battery capacity to be used, but it is still necessary to have hardware to prevent over-charging..
James,
Milliamp Hours are Milliamp Hours, no matter what the supply voltage. If you have a low power LDO VR then your load (in mA) is lower since the voltage at the load is lower. As long as the quiescent power of the LDO VR is low (MCP1703 is a good example), then you'll get the most 'radio' time with a LiPo using a VR. Trickle charging from an Energy Harvester into a small LiPo is actually an excellent solution. The LiPo can handle the current surges of the radio and 150mAH will provide plenty of backup should you lose your energy source for brief periods of time. With this solution you can still limit your charging to 4.0V. This, of course, will limit the capacity if recharging energy disappears.
Tom
Quote from: JamesBryant on November 30, 2015, 12:56:03 PM
Using maximum cell voltage of 3.75V reduces the capacity by 60-70% but gives almost unlimited recharges (Li-Ion batteries degrade during the last part of their charge - if they're only charged to 4.0V or less their total life (measured in AH delivered, not number of charge/discharge cycles) increases by a factor of 10-40.
That would be awesome. I pretty much never read about it though. Can anyone else here confirm whether this is true?
Quote from: WhiteHare on December 01, 2015, 06:22:01 PM
Quote from: JamesBryant on November 30, 2015, 12:56:03 PM
Using maximum cell voltage of 3.75V reduces the capacity by 60-70% but gives almost unlimited recharges (Li-Ion batteries degrade during the last part of their charge - if they're only charged to 4.0V or less their total life (measured in AH delivered, not number of charge/discharge cycles) increases by a factor of 10-40.
That would be awesome. I pretty much never read about it though. Can anyone else here confirm whether this is true?
I can say that feels very true although I cannot confirm the exact numbers themselves, but certainly the LiPo will last a very long time at the 10-20% top voltage then start to drop faster and faster.
Quote from: Felix on December 02, 2015, 07:58:25 AM
I can say that feels very true although I cannot confirm the exact numbers themselves, but certainly the LiPo will last a very long time at the 10-20% top voltage then start to drop faster and faster.
Felix, I'm not sure what end points you're referencing, however, a LiPo discharged (and not recharged) in a lowpower environment like Moteino will spend the bulk of its time between 4.0-3.7 volts. Once you get to 3.65, game's over, you're 'about to'
1 fall off the cliff. If a charger can keep the battery in the upper half of this 4.0-3.7 volt range, you should get the longest life out of your battery, but not the longest capacity in a discharge only situation.
I will agree that I find the 10-40X number surprising, however, and it definitely would be worthwhile exploring.
UPDATE: From http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries
QuoteMost Li-ions charge to 4.20V/cell and every reduction in peak charge voltage of 0.10V/cell is said to double cycle life. For example, a lithium-ion cell charged to 4.20V/cell typically delivers 300–500 cycles. If charged to only 4.10V/cell, the life can be prolonged to 600–1,000 cycles; 4.00V/cell should deliver 1,200–2,000 and 3.90V/cell 2,400–4,000 cycles.
On the negative, a lower peak charge voltage reduces the capacity the battery stores. As a simple guideline, every 70mV reduction in charge voltage will lower the overall capacity by 10 percent. Applying the peak charge voltage on a subsequent charge will restore the full capacity.
In terms of longevity, the optimal charge voltage is 3.92V/cell. Battery experts believe that this threshold eliminates all voltage-related stresses; going lower may not gain further benefits but induce other symptoms. Table 4 summarizes the capacity as function of charge levels. All values are estimated.
Tom
Note 1: 'About to' in this sense means that 95% of your capacity is used. In a lowpower environment you may still have months of sufficient power left.
Quote from: TomWS on December 02, 2015, 10:06:49 AM
Quote from: Felix on December 02, 2015, 07:58:25 AM
I can say that feels very true although I cannot confirm the exact numbers themselves, but certainly the LiPo will last a very long time at the 10-20% top voltage then start to drop faster and faster.
Felix, I'm not sure what end points you're referencing, however, a LiPo discharged (and not recharged) in a lowpower environment like Moteino will spend the bulk of its time between 4.0-3.7 volts. Once you get to 3.65, game's over, you're 'about to'1 fall off the cliff.
It's just from my general use of LiPos in my own nodes. I think we're in agreement, maybe the way I said it wasn't very clear :)
Thanks, Tom, for the references. They made for good reading: in addition to the cut-off voltage, the other two significant factors would seem to be depth of discharge and temperature (Tables 2 and 3). The table you highlighted appears to assume a Depth of Discharge of 100%. It looks as though limiting DoD to 10% would give another big boost of longevity. So, I can believe JamesBryant now that by conservatively managing the three factors (cut-off voltage, Depth of Dischard, and temperature) could give >10x improvement in the useful life of a LiPo. Impressive!
Provided you have the space, using LiFeO4 as the rechargeable battery chemistry would appear to help with the heat issue. As near as I can tell, LiFePO4 seems to actually perform *better* at higher temperatures, whereas regular LiPo's perform worse. Also, LiFeO4 would give a fairly flat voltage curve between 3.2 and 3.0 volts, which would also seem to be a good match for the Moteino and these HopeRF radios (which can tolerate at most 3.6v max), as you wouldn't need a voltage regulator.
If the magic number is 3.92v for regular LiPo batteries, what should it be for a LiFePO4 battery? This series of charts suggests it maybe (?) is between 3.3v and 3.4v: https://www.powerstream.com/lithium-phosphate-charge-voltage.htm
Interestingly. his measurements showed that one of the allegedly LiFePO4 batteries was spiked with Cobalt. I do wonder whether that spiking carries with it possible negative safety implications, as for me the better safety profile would be one of the reasons for picking LiFePO4 in the first place.