Open MadsR opened 2 years ago
Most Likely Cause Both the initial heat and later following chemical smell stemmed from batteries overheating. This is most likely due to being at full capacity and the additional current from Solar Panels were higher than the batteries could withstand.
Conclusions Heating element was never the culprit as the heating loop was not enabled due to safety reason. Batteries overcharged and in return overheated.
Todo Check battery integrity.
Would an off the shelf solar charge controller prevent the issue in future?
Solar Charge Controller Any controller capable to handling "single cell" charging should suffice. Regualr LiPo chargers require individual cell measurements to charge.
Battery manufacturer stated that such a Battery Management System (BMS) would not at all be necessary, and could in worst case make the system perform worse (efficiency, etc.).
This primary error would not happen if the Rover was active (move until low battery -> halt until high -> move until low). This is only due to having full capactiy and continuously charging.
Potential Solution 1 Add external power draw (power resistor) once batteries reach 100%, to avoid overcharging batteries.
Notes Power Resistor has to be dimensioned sufficiently to not overheat. Can be used as a means of 'clearing snow' around Rover while idle.
Potential Solution 2 Add Solar Charge Controller..
Notes As stated can potentially reduce efficiency.
Potential Solution 3 Wire solar panels through ON/OFF switch, to avoid charging batteries while system is Idle
Notes Can cause solar panels to overheat as well if they have no way to dissipate generated power. (Add bypass Power Resistor when disconnected)
We configured a morningstar TS-45 to work with the NiMH voltages we used. This has resolved the issue. The configuration should be optimized for the cells.
Thoughts: The morningstar can be configured to dump excess solar energy into a dump load that could heat the interior. (We did not do this as it would probably get too hot unless there is some temperature control too.
Potential issue: NiMH cells have a very flat voltage vs charge% curve. This makes it more important to allow for temperature effects on all voltage thresholds. -Both those configured in the charge controller, but perhaps also those in the arduino code.
ArcticRover firmware version
0.97.037
Area(s) with issue?
Hardware Issue, Batteries
Steps to reproduce
Rover was hot, heating element was disconnected. Tried maneuvering rover with remote control, unresponsive, assumed low battery. Left Rover in the sun throughout day, to presumably charge batteries. Returned later in the day, batteries hot, and a chemical smell inside the box.
✔️ Expected Behavior
No response
❌ Actual Behavior
No response