A bit late to this thread, but.....There are two triggers for the turtle mode light.
The normal one seen is that if the estimated state of charge goes below about 10-12% (as reported by Canion) the turtle light will come on and not go out again until the car is charged back up above the SoC threshold. The load on the battery doesn't matter for this trigger it's purely SoC.
The second trigger is the turtle light will come on if any individual cell goes below 3.0 volts, even momentarily. Normally this will never be hit in practice because even under full load at 10% SoC the cell voltage will not drop as low as 3.0 volts, as the open circuit voltage will still be about 3.6 volts.
However if you have a faulty cell that has high internal resistance the voltage of the cell may drop below 3.0 volts under full load even at a relatively high SoC. In this case the turtle light will come on when accelerating and then go out as soon as you ease off the accelerator.
So if this is happening with a reported 50% SoC the issue is almost certainly a faulty cell or cells with high internal resistance, and this would be easily confirmed by monitoring the individual cell voltages with Canion while the car is driven under load, as one or more cells will drop well below the voltage of other cells, and below 3.0 volts if the turtle light is coming on, something that should never normally happen.
One other thing that could cause it is if the BMS's estimation of Ah capacity of the battery is grossly out - for example if the battery is really 32Ah but it thinks it is 45Ah it will keep driving the car well past the point where the SoC turtle threshold would normally kick in, as a result you will eventually get to the point where cells start droping below 3.0 volts under load as the cells are OK but just over discharged.
This happened to me when I did a BMS reset causing the BMS to go back to the default 45.8Ah while my pack was really around 33Ah - it kept driving the car well past where it normally should until the low voltage shutdown started triggering under load.
Unless someone has accidentally done a BMS reset, it seems unlikely that this is the cause, more likely there is an individual high resistance cell.
Diagnosing this problem should be relatively easy with Canion - first check the reported Ah capacity to make sure it is not unusually high for a car of the age/milage (to rule out the erroneous BMS reset scenario above) and assuming not, monitor cell voltages under load when the turtle light is flashing on to identify the faulty cell(s).
I have three high resistance cells in my car at 56k miles that I will be replacing soon - they're not yet bad enough to cause the symptoms described in this thread, however they are already affecting rapid charging speeds and regenerative braking.
As far as I can see DC/DC inverter is a red herring - if that wasn't working the 12v battery wouldn't charge and before long the car would stop working altogether.
The DC/DC inverter is easy to test - measure the 12v battery at the battery terminals while the car has been turned off and unplugged for an hour, it should be between about 12.2 and 12.6 volts. Then turn the car on to ready mode - if the voltage immediately goes up to 14.2 - 14.4 volts and stays there, the DC/DC inverter is working normally. You can turn high beams on for extra load and it should still stay in that range.
