BQ40Z60EVM-578: Setting Full Charge and Full Discharge

I need to clarify this statement, where I misspoke: Currently, it is charging to 8200mV and discharging to 40mA.
It is not discharging to 40mA. The charge cutoff is occurring at 40mA rather than 45mA as input for the Charge Term Taper Current.

Thanks, TEC, for the quick feedback.

Charging Voltage = Minimum Output Voltage + Voltage Resolution × Voltage Register OR
CV = MOV + Vres *Reg
CV/2 = Per cell charging voltage for 2s and corresponds to value input for voltage in the temperature sections of the Advanced Charging Algorithm (but wait there is more)
Reg = (CV-MOV)/Vres
241 = (8074-4218)/16
This is true, but there are induced errors due to rounding as laid out in SLUA750, if I am reading and applying it correctly. While Vres is only accepted as an integer, it is actually not rounded when calculated, only when used. The true value is 16.477. Therefore,
234 = (8074-4218)/16.477
So the Reg value when charging is always 234 instead of 241. So plugging that back into the equation,
4218+16*234 = 7962 (this proves true in the log file for both the register and the charging voltage)
Going back to CV/2 = Per cell charging voltage for 2s and corresponds to value input for voltage in the temperature sections of the Advanced Charging Algorithm = 7962/2 = 3981mV (which is the value we input)
So this takes me back to, why does it then charge to 8200mV? If this is hysteresis and/or a limitation of the accuracy of the EVM, then I suppose I am left with trial and error to get this worked out. I would rather think that I am just too ignorant about how this works and need some instruction from a smarter individual than myself, as humbling as that may be. In reality, I thought I had it figured out, but the changes I made to TC and FC didn't do anything. RSOC has not come into play, yet, since I am trying to do this to get Chem ID and to do a Learning Cycle (aka RSOC is always 100).

As for stopping the charging current, the actual low point cutoff per the cell manufacturer is 40mA. I had read where there were two 40s intervals that had to be crossed to affect a cutoff (among other requirements). So I changed the Term Taper Current to 45mA based on review of the logs. But it didn't matter when tested because it continued to perform as before. That led me to believe that there is again something I am missing.

You might be on to something with your suggestion regarding the TCSETVCT/FCSETVCT in the Data Memory:Settings:Configuration:SOC Flag Config A area. Both of those are set now as well as TCCLEARRSOC and TDCLEARRSOC. I will look into that further.

Please attach the log file and a gg.csv file from the pack. That may help with the analysis.

Here is the gg.csv.  It wouldn't let me upload with .csv, so I changed the extension to .xls, but you will have to change it back to open it.  I am busy logging at the moment so I can't drop in a log showing cutoff with the latest settings.  Here is a graph showing the performance.170802 1537 TLIx2s2p Testing.gg.xls

I had not comprehended that you are using a 200 mohm sense resistor on the charger. I have not seen anyone try using a sense resistor that large and the charger was designed to use a 10 mohm sense resistor. It was also designed to charge in the 1+ A range and the current measurement accuracy is not great on the low end. You can pass the log file on when available.

I had seen a post from Thomas Cosby indicating that there were limitations on the low end for current sensing, so we have been expecting to fiddle with the Term Taper Current. As you noted, we changed out the R20 resistor to improve the current resolution. What caught my attention was that when we had it set at 40mA, it dropped to 36mA before cutting off which seemed reasonable due to the two 40s timer. When we bumped it up to 45mA, we saw no change in the behavior. I am less concerned about cutting off the charge too early than not getting the voltage within the manufacturer's top end spec. The way I thought I could bring it down didn't work. The change made did not have an affect. When I am done with this attempted Learning Cycle, I will post the log (tomorrow). Thanks for your help.

Here is the latest log.

170802 1445 TLIx2s2p Testing.zip

I have adjusted the Charging Voltage (by Temp) down in the Advanced Charging Algorithm. I was empirically testing the outcomes of various settings when I got your response. I have opted for the trial and error approach, but it will be difficult to explain to others how it works, since I won't know.
Yes, Termination Voltage was still at 9000 and FCC had consistently been showing 0 (and triggering an OC alarm). You suggest changing the Termination Voltage to 6000mV since I am using a 2s pack. It appears that this 6000 is a guess based on what you have assumed is my minimum allowed pack voltage. I should use whatever I want to designate as the minimum that I want to allow the pack to reach. I have opted for 5600mV. If I am misunderstanding something here, please let me know. Thank you for your insight.

TI typically recommends 6000mV for a 2S pack, because this is the lower cut-off voltage for most Li-Ion cells. You can set it in accordance with the specification for your cells.