BQ24105 Charge Termination

I believe I have figured out the issue, but I wanted to confirm my conclusion. Does grounding the TTC pin mean that the STAT2 light will never come on because the charge is never terminated?

If so what are the advantages/disadvantages of taking TTC high/low? I realize that when pulled high it has a charge termination but timing is disabled. Does this just mean it will charge to full capacity then discharge until the threshold is reached and begin to charge again, whereas if TTC is grounded it will continuously charge the battery?

Are there any ill effects of continuously charging the battery on cycles and battery life?

Kevin

Terminatin and Safety timer are for two different purposes.

IF the TTC pin is fixed at any voltage it does not charge up and down for the counter to count and can't time out, which disables the safety timer.

The safety timer is an extra safety feature that can be used to shut off the charger if, by design, the battery should have completed charge and it has not.  Typically the charge time is 2 to 3 hours and the safety timer is set to 5 hours so something may be wrong after 5 hours.

Termination is disable if the TTC is held low.

Termination is slighly better on the cell because it allows the battery to relax and does not constantly stay a 4.2V.  The higher the voltage away from 3.7V the faster the Lithium decays.  This may be the difference in life of 6 months over three years if left at 4.2V for the whole time. Probably not a big deal on a device like a cell phone because it stays off the charger most of the time.  Note that if termination is disabled the charge current will go to zero as the battery reaches the regulation voltage and thus is not really charging at this time.

I will have to check what the status pin does since this indication is sometime slightly differnet on differnet part but if you test it, it should work the same each time.

My guess is that if the battery and source are removed and both inserted for the first time that the STAT1 will be low (LED on).  I would think STAT1 would stay on until termination, which is disable so it would not turn off.  It could be wired so the termiantion flag is set and the STAT1 turns off but the termintion is not acted (charger stay on).

I have since my original post switched the configuration so that TTC is now pulled high. As far as I can tell from the data sheet this should allow the charger to "complete" a charge cycle and the STAT1 and STAT2 pins to switch. However, still they indicate charge in progress. The charger doesn't charge the battery above the ~3.6V I desire, but I never get a charge complete output from the chip.

If I monitor the current coming off the supply voltage (12V battery) into the charge regulator I see that it is pulling about 500mA of current when the battery is charging and that this drops to about 40mA or so when the battery is fully charged.

What is causing this to occur and is it an issue? Is it because the voltage has been altered from 4.2 to 3.6 and the chip does not recognize this as a full charge. I wouldn't think so because it seems to be dependent on current (which is changing), but it was a thought I had.

Also,  I saw in another forum post that the following scheme is used to charge the battery depending on the voltage. However, this seems to be for the standard Li-Ion battery and not the LiFePO4 battery. I was wondering how these voltage values would change.

If battery voltage is 2V/cell to 3V/cell, use precharge current to charge, higher than 3V/cell but lower than 4.2V/cell, fast charing.

Also, what is the minimum voltage at which a battery can be recharged using this charger. (Yes I realize it is bad to drain these type of batteries below ~2V, but it does happen).

I think what is happening is that the termination is actually occuring and there is some load pulse that drops the battery voltage back down to the refresh thershold and a new charge is started.

Monitor the stat1 pin with a scope set for single seq trigger and see is this is happening.

I would try and eliminate variables to see what is happening.  disconnect the system load so all the charger is doing is charging the batttery.

The feedback regulates at 2.1V.

the iset2 pin sets the termination thrshold.  What is this set to?

If the battery drops below 2V this is considered a short circuit and the switching stops and only a 50mA current is delivered to recover the battery or reset the circuit once the battery is removed.

Precharge is at 3/4.2 = 71% of the regulation voltage, so if you have the regulation at 3.6V the precharge threshold is at 2.57V and only the precharge current is delivered at and below this voltage.

Please send waveforms of each pin on the IC during the operation you believe that is not correct and I will take a look at it.  Please place 2 to 4 waveforms on each plot so I can get some timing relationship on them.

Although I had originally tested this using a fully populated board that included a load, my recent tests have been performed on a board with only the charging circuit populated. Therefore, there should be nearly no load. Furthermore, I did design the circuit in accordance with the application note for LiFePO4 (SLUA443) Figure 4. This is supposed to lower the recharge voltage threshold so it doesn't continually think it needs to recharge. See the schematic I attached to the original post.

On a brighter note, I have tried a brand new battery and for the first time I have seen the status pins indicate that the charge was complete. I had not changed the circuit at all between the used battery and the new battery. Any thoughts as to why it may work with a new battery and not a used one. Also, note that although it did work for the new battery when I disconnect the battery and re-connect it, the output does not go to charge termination for quite some time, so I still feel like it isn't working as expected.

I do not currently have access to a scope to monitor the pin activity. I will try to get this done when I have a chance.

As you can also see in the schematic I have a resistor divider to FB that sets the charge voltage to ~3.5V based on the resistor divider.

The iset2 pin has a 20kΩ resistor to ground.

Let me know if you have any thoughts and I will work on trying to get some scope captures for you.