The Twine forums are now archived. If you need help, please visit http://help.supermechanical.com

Rechargeable Battery for TWINE2

It seems that with most products going to Li-ion or Li-po batteries, it would be wise to integrate it with TWINE too. After all, a perfectly good charge socket is already there (Micro USB), and there's still plenty of room on the PCB to include a charge chip. A Li-ion battery of a similar size to 2xAAA's would have oodles of power to run it for a considerable length of time.

Whether to have the battery permanent or changeable is the question though. Permanent is simpler, cheaper and could be more powerful for its size, but it has the difficulty that you will need to charge it where it's located which could be awkward for some people. However changeable allows a user to go and swap the battery at an instant when it's getting low.

I might have a look at testing the idea anyway, a simple charge circuit would be easy to incorporate into it from the USB socket, and as long as TWINE doesn't get upset at having 4.2v at its battery terminals, it should work just fine.

[Edit] Just found out recently that most of the chips are rated up to 3.6v, so there may be issues going up to 4.2 at full charge.

Answers

  • 4 Comments sorted by Votes Date Added
  • *** Disclaimer first... I take no responsibility for anyone who damages their TWINE due to reproducing these experiments ***

    Ok I've had a play and found my TWINE worked fine at up to 4.4v on the battery terminals, I didn't take it any higher as it wasn't necessary. The LED got really bright at this voltage, so I've changed it's resistor from 75ohm to 560ohm, this will alleviate the going blind problem, save some current draw and generally be easier to look at, even at normal battery voltage.

    There was no noticeable increase in current draw, so I'm making the assumption that I'm not stressing the components unduly.

    This means once my components come in (eBay + Christmas could slow it down), It should be no problem to have a charger IC, 3 other components around it, and 2x 600mAh 3.7v Li-Ion AAA's in parallel running my TWINE. Initially I'll probably run a single battery, and bypass the second battery position with a link. Later I'll swap the lower battery terminals to match the upper ones, re-label the orientation on the PCB, and make the appropriate track changes underneath (1 track cut and 2 links).

    I'll solder the IC (BQ24080) to a small piece of brass about the same size as the Wi-Fi module to be used as a heatsink, and mount it above using foam double sided tape. Only 3 wires are necessary to connect it to the TWINE, +in (on the input side of FB1), GND (on the -Batt terminal), and +Battery (on the + Batt terminal of course). I'm hoping to keep the current set resistor tucked away between pins 6 & 7. Decoupling capacitors aren't likely to be needed, as the TWINE has these already. I'll set the charge current at 300mA to start with to limit heat production, extend battery life, and aid compatibility with any USB plug pack, but the IC can go up to 1A.

    The issues I see that could come up are...
    a) Low battery voltage warning will not be correct, so by the time it sends the email, it's basically already flat. Li-Ion's end of life is about 3.0v.
    b) Extra heat created by the charging circuit may perhaps send the temperature measurement higher than ambient during charging, however this is a short term issue, 3-5 hours should be the maximum charging time needed. This should settle within 30 minutes after charging is complete.
    c) Warranty - of course this would likely void any warranty I may have.

    Further reports will come when I've played more.
  • Ok after a long break with too much other stuff going on to get back to this, I've now re-jigged my idea. It's actually simpler too! 

    The original IC was too small to solder to, so now I'll be using a MAX 11555EZK+T. It has 2 inputs, DC or USB, I'll be using the DC input. Why? Becuase the charge current will be 280mAh as opposed to 100mAh. This will mean a full charge on about 5 hours for the batteries I've selected. If I did this again, I'd also get a small PCB to mount the charge IC on, because the pins are quite delicate and easy to break off if bent much. 

    There's also no need for any external components. As mentioned before, the bypass capacitors are already on the TWINE PCB, and I know they should be located as close as possible, but it'll be fine for this application. However to be even safer, you could use some anyway. This IC also doesn't need any resistor to set the charge current, that's all pre-set

    There is one output other than the battery connection. Pin 3 pulls low when it's in normal charge mode, so I used an LED on this to show it's charging. 

    It's all working now. Unfortunately I was rushing so I managed to destroy FB1 in the process. It's just a fuse so I bypassed it. 

    I noticed that if it's powered up by USB, but no batteries are installed, the TWINE LED flashes madly. I can only assume this is the charge IC affecting the circuit. If you intend on running it with AND without batteries, you will need to install a switch to disable the charge IC when no batteries are installed. 

    The rechargeable Li-ion battery idea is good though, just be more careful than I was in doing the work.  

    Issue"b)" of overheating will be eliminated with this chip, as it has internal limiting. 
  • Note: Whilst charging, the internal temperature does go up considerably. Mine is measuring 30 Deg C when the room temperature is about 24 Deg C. 30 min's or so after disconnection should see it revert to ambient temperature. 

    This is normal. 
Sign In or Register to comment.