First, before going on with the subject of the topic, you realize that if you are afraid of limping home due to approaching the lvc, you just need to slow down... I remind you that wind resistance energy loss works with the square of speed. With modern CA firmware, you can see your instant Wh/Km ratio and it really helps to "back off" the throttle a bit. Before long (and flat) journeys, I calculate my not-to-exceed "Wh/Km" value, and monitor it while I ride. Very simple to do.
Now, with the subject of the topic.
I tend to believe (with no science to back me up too, just some logic), that the max healthy C-rate (discharge OR charge) is a function of the remaining %SOC (to reach 0% or 100%, respectively), declining linearly (or maybe exponentially?).
If we assume that when a cell approach one %SOC extreme the Ions start to compete and slow down the reaction, then this makes a lot of sense. It's possible to measure this by temperature rise, but I think that in terms of electricity this would be reflected as a rise in IR as you approach either edge of the %SOC. So, if you suddenly changed from discharge to charge at 20% SOC, for example, you would measure a sharp drop in the IR.
I can verify the increased IR at the bottom end of the discharge from my small A123 26650 packs. I noticed they get significantly WARMER when being discharged at the lower %SOC spectrum. I was actually surprised at start, because the discharge currents were the same all over the spectrum, but now it makes a lot of sense!
I would hope, therefore, that the IR of each cell should be rated at %50 SOC, which is exactly in the middle, but more likely it's being given for the beginning of discharge at 100% SOC.
Keeping on with the logic, we can assume that at 0% SOC the initial charge current can be the same as the initial discharge current at 100% SOC. I still wouldn't be brave enough to test this on gunpowder-cells (LiPo's) which are rated for 90C discharge....
I can assume that cells are rated so purely with their charge rating for the following reasons, which take into account the "Average Joe":
1. Charging is usually being done while being stationary, meaning no heat convection from air flow. Pack will get hotter than the same discharge rate while having cooling air flow. (assuming for the sake of the example that we test this at 50% SOC)
2. Overcharge is more dangerous than under-discharge due to the more stored electrical energy, especially if you use LiPo.
3. Continuing with 2, the probability for meltdowns (over time) goes lower with lower charge rates. A manufacturer would be more lawyer-proof if he rates his cells for lower charge rates than discharges.
4. The average-joe has enough hard time with understanding the difference between "voltage" and "power", let alone to understand that the max C-rate is a function of %SOC, as noted earlier. It's easier to put just one value for each direction, and you would get less complains and less lawyers.
And so I would like to finalize with what I think determines the overall kWh you can draw from a Li-Ion cell:
1. Heat - The above theory about variable C-rates vs your remaining %SOC (in either direction) can maintain a cool battery.
2. Heat - Sustained high than the recommended C-rates on 1#. Pulse rates, for example during regen, should pose no problem.
3. Reaching the 4.2V mark. (even for brief periods - the cells just don't like it and it can be proofed by the official specs which shows what happens when you fully charge/discharge cells with relation to the life cycle. They don't spend time on the 4.2V. They just reach it and start discharging again )
4. Spending time on the 4.2V mark. (the aging effect)
5. Limiting your max %SOC to 4.05V (80%) and discharging down to 20% SOC seem to maximize the total kWh you can draw over the life of the cell, while still making sense economically and not carrying too much unused battery weight). Of course, extra battery storage is available at any time by topping up to 4.1V or 4.2V and fully discharging.
6. With lower temperature, the variable C-rates vs %SOC which were discussed earlier should be lowered linearly according to the battery spec discharge graphs vs temperature.