Rolf wrote: > Hi all. I am hoping to tap in to the collective experience on the list... > > I have a project ("advanced" shutter-release system for a camera - > time-lapse, long-exposure, etc.) that is currently prototyped on a > bread-board. The power will be derived from a single Li-Ion cell that > will need to be recharged on occasion. > > I have the following components: > The Li-Ion cell is 18650 sized 2000mAh. > The charging controller is the LTC1733 chip from Linear Tech. > I am using the LT1111-5 switching IC in DC-DC Boost-mode to get 5V for > the "real" circuit. > The "main" circuit contains an 18F4620 and an LCD 16x2 module as well > as op-amps, switches, opto-couplers, and some other minor components. > One of the critical components for this excercise is the REF194 chip > from Analog Devices which is a 4.5V Precision voltage reference I use > for all A/D conversions. Maximum draw of the whole "main" circuit is > 40mA@5V. Typical draw is 25mA@5V. I am hoping to trim that down some > more still. > > I have "everything working" in that when plugged in to the 5V wall-wart > the charging chip activates and starts charging the cell. Simultaneously > the power supply to the LT1111-5 chip gets swapped from the (now > charging) cell to a 3.3V LDO (MCP1700) that draws it's power from the 5V > wall supply. In other words, when plugged in, the circuit is powered > from the wall. When unplugged it is powered from the cell. I know there > is the inefficiency of the double-DC-DC conversion, but when using wall > power that is OK, right? > > I have it set up that the main circuit is always turned on when plugged > in at the wall. The 18F4620 always monitors the charging of the Li-Ion cell. > > Currently it measures the charging current (using output from the > LTC1733), as well as the charging voltage. The PIC keeps control of the > three status values the charging chip supplies (Charging status, Fault, > and ACPower). It also keeps track of secondary values like the number of > cell charges, the charge time, etc. > > The LTC1733 chip uses a Constant Power then Constant Current then > Constant Voltage strategy for charging. The constant power part is > because it is thermally limited internally to protect the IC, and this > causes the current to drop when the IC operates at it's thermal cut-off > temperature (which is at its worst case when the cell is at it's lowest > charge - the power is the difference between the wall power-supply > voltage (5V) and the cell voltage (2.5V) times the charging current > (1.5A) = 3.75W). As the cell charges the cell voltage increases causing > the power dissipation in the charging IC to decrease. Once the power > requirement is such that it falls within the thermal capacity of the IC, > it keeps a constant current mode... until the charging voltage hits 4.2V > at which point it maintains the 4.2V charging voltage, and the charging > current drops off steadily until the cell is charged. > > I have 3 questions... > 1. Is there some way I can calculate a "degree of 'chargedness'". I > want to get an idea of how close the cell is to being fully charged, and > preferably to indicate something like "35minutes charging still required". > 2. Is there a way to calculate the capacity of the cell during (or even > at the end of) the charge process? It would be nice to say "The cell has > been charged to 1800mAh capacity". This measure would be a good > indicator of the aging of the cell. The unit does not have to be mAh, > perhaps there is another more meaningful unit, something like mWh. There > would also have to be a definition of "empty". Right now my circuit will > automatically shut down if the cell voltage is less than 2.5V - unless > plugged in). Cell capacity is normally measured by draining the fully > charged cell at a constant current, then timing the discharge. Because > the cell supplies a 5V DC-DC converter, even if my circuit had a > constant 25mA drain, it would cause a variable drain on the cell as it's > voltage dropped off with it's state of discharge.... > 3. Should I even bother with the above challenges? It is secondary to > getting this project going, but, since this is a hobby for me, the fun > is in the learning, right? > > One thing I have learned is that the datasheets for Li-Ion cells all > indicate a few curves for the capacity/discharge current of the cell, > and the capacity decreases with an increase in the discharge current. > Thus, a 2200mAh cell is only 2200mAh for a given discharge rate..... The > rates given in the datasheets are normally much higher than the 30-50mA > that I anticipate drawing (depending on cell voltage). > Actually, all batteries exhibit this in some way. > Thanks in advance for any hints, pointers, references, or even "just > forget about it" type comments ;-) > > Rolf > > P.S. Apologies for the long content > > Just be very careful when charging Li-Ion cells. Use a quality self-contained chip to do it with; DON'T use a PIC to charge the battery. Proper charging requires a thermistor to measure battery temperature. My personal preference is to NOT use Li-Ion, but to use NiMH cells, which are quite safe and are being constantly improved. (The inherent instability of Li-Ion cells and the problem with Chinese manufacture of defective cells cause me not to use them.) In any case: Good luck. --Bob -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist