I have been testing a charger PCB based on the bq2005 for a couple of weeks now and cannot solve this final problem. The bq2005 successfully conducts the Fast Charge state and on dT/dt, stops Fast Charge. The charger is then quiet for a few seconds, before resuming charge - I assume this is the start of the Top-Off state. However, it only remains charging for 450ms before dropping to the Trickle Charge state.
My configuration is thus:
- V(tco) = 1.47
- T(dt) threshold = 1.12 degC/min
- Charge Rate = C/2
- TM1, TM2 currently set High, High, but same behaviour at Low, High and the recommended High, Float.
- Topology: as per "Application Example 3: P-Channel MOSFET Buck-Topology Switch-Mode Charger" in the UNITRODE U-507 "Using the bq2005 to Control Fast Charge" document.
- R(sns) = 0R18 (ie. average charge current = 1.25A)
I have adjusted V(tco) to ensure V(temp) does not fall below it but it makes no difference.
I have disabled -dV detection for now.
I have confirmed that the Fast Charge state can be terminated by dT/dt (using a heat gun or waiting for the pack to heat up), or by timeout (since the TM1, TM2 = High, High setting only takes 23 mins to timeout).
I have studied the oscilloscope to ensure the difference between V(ts) and V(sns) (which gives V(temp)) does not go below V(tco). Certainly it does not during the fast charge, because the Fast Charge continues. And during the transitions between states, I cannot see anything that would indicate V(temp)<V(tco).
The V(temp) signal does have some short, high frequency switching noise on it, but I assume it's not causing a problem because Fast Charge continues just fine.
The V(bat) signal stays well below V(mcv) the whole time.
The FCC signal goes low when Fast Charge stops (due to dT/dt) and stays low.
I have tried two different bq2005 chips.
Why would the bq2005 not stay in the Top-Off state?
