Shop OBEX P1 Docs P2 Docs Learn Events
PropBOE+Li-ion power pack+charge level — Parallax Forums

PropBOE+Li-ion power pack+charge level

RsadeikaRsadeika Posts: 3,837
edited 2012-10-26 18:48 in Propeller 1
I just picked up a PropBOE at Micro Center, and I am trying to figure out, the best way to read the charge levels of the Parallax Li-ion power pack/charger. I know that their have been discussions about voltage dividers, and such. But, it has never been discussed in terms of the PropBOE and Li-ion power pack.

I know that the PropBOE has an ADC, I am assuming it is for 5V, which simplifies it on that end, but what do you need for the Li-ion power pack side, like where would you hook up the wire(s) to get a reading while it was in charge mode, and when it is in use. Also, since the Li-ion power pack is rated at 7.5V, I have not seen any info as to what the fully charged rate would be. If the fully charged state is, let say, 7.9V, then how would you overcome this with a voltage divider setup, and still get useable readings? I guess basically I am asking if anybody has done this, and if the information is available?

Ray

Comments

  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-10-24 23:19
    Hi,
    Lithium ion batteries are a special problem for measuring battery capacity as the voltage curve is so flat in the middle that just reading voltage really isn't going to tell you much. You have to do something more sophisticated, and that generally means monitoring how long at a given voltage as well as the voltage.

    Are you up for the complexity, or were you just hoping for an easy study project?

    And BTW, the 5 volts is regulated - so that will not show anything about the state of the battery. You have to actually get your reading from the unregulated battery outputs (via a voltage divider).

    AT full charge, the per cell voltage can go as high a 4.2v, though Parallax may not actually have its charger go as high as that. With two cells, that would be 8.4v. Just make a voltage divider out of two 1/4 watt resistors of the same value - say 10K ohms each and the tap in the center would always be less than the +5 ADC limitation. 5volts/20,000ohms = ? (very little current drain on the batteries, BUT you should still have a disconnect when you are not using the power anyway.).

    These lithium ion cells should Never discharge below about 3.2v or they become damaged. So with a 50/50 voltage divider, your ADC range will roughly be between 4.2V high and 3.2V low and the cell will hang around 3.6 or 3.7 for most of its use.

    I think the Lithium Ion power pack will actually shut down around 3.2V to protect the cells. I have no idea about the exact details of how it manages charge and discharge. That can become quite complicated.
  • SRLMSRLM Posts: 5,045
    edited 2012-10-25 00:05
    I'd recommend a "fuel gauge" chip, such as the MAX17049 to measure and record the state of the cells.
  • RsadeikaRsadeika Posts: 3,837
    edited 2012-10-25 03:09
    In brief, the project, I have a 3.5" x 5" piece of cardboard, where the backside, holds the li-ion pack screwed down. On the front side, a GG prop platform board and a small breadboard which holds an XBee module, plus the Sensirion module. I call this my Mobile Temperature Unit, or on the PropGCC beta forum, slave0. I am considering switching out the breadboard and GG PP board for the PropBOE, because it has a breadboard and the ADC, plus the XBee connector, which would come in handy. Since one of the things that I want to do, data log a 24 hour period of temperature readings, I would like to have a good feel for what the battery usage is. The latest empirical data about the power pack, if I use my MTU for lets say about an hour, and then I plug it in to be charged, it goes into a charge session for at least an hour, give or take ? min. Actually I would like to have data better than that.

    If gathering charge level data for the li-ion pack is going to be that difficult, than maybe I should consider ni-cads? But the problem with that, I have not seen any battery holder units where it would hold 6 AA batteries plus have a built in charge unit that would have the same foot print as the li-ion power pack. I guess I would need at least 7.2V output with at least a 24 hour usage period, before a charge is necessary. I am open to some new ideas for this.

    Ray
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2012-10-26 09:54
    I am not sure that switching to NiCds is going to significantly improve data if you want to rely merely on voltage readings. You already invested in an excellent Lithium Ion unit, so it is a shame to not use it.

    The chip mentioned above might be gotten as a free sample and eliminates the need to measure current as well as voltage - so it really is the optimal way to go.

    The interface is some sort of SPI serial that is easily done by the Propeller. You will have to have a time base as well by either using one COG to provide a RTC or use an outboard RTC. Data might be stored in an EEPROM.
  • SRLMSRLM Posts: 5,045
    edited 2012-10-26 10:02
    Loopy Byteloose wrote: »
    The interface is some sort of SPI serial that is easily done by the Propeller. You will have to have a time base as well by either using one COG to provide a RTC or use an outboard RTC. Data might be stored in an EEPROM.

    Note quite. The chip is interfaced to via I2C, and it provides a charge level and charge rate. It's powered via the battery. So, you don't need an RTC or an EEPROM to use in conjunction with the MAX17049.
  • RsadeikaRsadeika Posts: 3,837
    edited 2012-10-26 10:43
    I just went to Mouser and they are selling them at $3.00 apiece, which does not sound like a bad deal. The problem is, I would need a lot of hand holding on this one, just to figure out how I would hook it up, let alone write the driver for it in C, and Spin. I guess I need someone to give me a little brief as to what would be involved in hooking this up to the li-ion pack. Who knows maybe somebody from Parallax will see this and add it to the li-ion power pack.

    Ray
  • SRLMSRLM Posts: 5,045
    edited 2012-10-26 18:48
    You don't want to solder these chips without a reflow oven. They are tiny! You can free sample some from Maxim, and they do have eval kits available (although, be aware that it will probably be ~$90 w/ s&h). You can always request a quote for that.

    Anyway, if you do get your hands on some then it's simple to hook up. Look on page 15 of the datasheet, and there is only one external component needed (a capacitor). I've got a board design that uses the MAX17048 (the single cell brother), so I'll be developing the code for that. Eventually it will be released as open source.

    I agree: Parallax should have incorporated some sort of charge level indicator onboard the power pack.
Sign In or Register to comment.