The solar charge controller (truly battery charge controller) used for the solar charging by the Mayfly (unless I am missing something) is not rated for the potential input voltages it would see over the operational range of the solar panel/converter.
The battery charge controller currently in use is the MCP73831. Conventionally the Mayfly loggers are used with Voltaic's 3.5W solar panel, or similar 6v panel (as referenced in #23). The problem being that the max open circuit voltage generated by the panel is 7.7v, while the absolute maximum input to the MCP73831 is only 7v, meaning that in low load or high insolation conditions, the input voltage could rise above the absolute maximum rating and cause damage to the chip or system.
There is a series diode (B5819W) in the circuit between the panel and the charge controller to allow for input switching, however, at the very low loading situations that would cause a high solar input voltage, the forward diode voltage is only 200~300mV, still resulting in a potential input voltage well over the 7v absolute maximum (7.4 ~ 7.5v).
Note: This same issue would also be present in the MCP73871 suggested in #23
Potential solution:
Have not researched the potential linear solutions extensively, but perhaps something in the BQ2401x line (specifically BQ24013) or the BQ2406x, which would solve this issue, and also address the desire for a larger output charge current (#23)
For our newest series of logger, we are using the LT3652, I have researched this segment of true MPPT systems in detail and could not find a better performing unit for this power range. Especially with the addition of an external control network, can result in >75% efficiency over all regions of operation, even into the <100mA input region where parasitics can dominate. And nearer to 90% for the >500mA range. Also allows for up to 2A output current.
Issue:
The solar charge controller (truly battery charge controller) used for the solar charging by the Mayfly (unless I am missing something) is not rated for the potential input voltages it would see over the operational range of the solar panel/converter.
The battery charge controller currently in use is the MCP73831. Conventionally the Mayfly loggers are used with Voltaic's 3.5W solar panel, or similar 6v panel (as referenced in #23). The problem being that the max open circuit voltage generated by the panel is 7.7v, while the absolute maximum input to the MCP73831 is only 7v, meaning that in low load or high insolation conditions, the input voltage could rise above the absolute maximum rating and cause damage to the chip or system.
There is a series diode (B5819W) in the circuit between the panel and the charge controller to allow for input switching, however, at the very low loading situations that would cause a high solar input voltage, the forward diode voltage is only 200~300mV, still resulting in a potential input voltage well over the 7v absolute maximum (7.4 ~ 7.5v).
Note: This same issue would also be present in the MCP73871 suggested in #23
Potential solution:
Have not researched the potential linear solutions extensively, but perhaps something in the BQ2401x line (specifically BQ24013) or the BQ2406x, which would solve this issue, and also address the desire for a larger output charge current (#23)
For our newest series of logger, we are using the LT3652, I have researched this segment of true MPPT systems in detail and could not find a better performing unit for this power range. Especially with the addition of an external control network, can result in >75% efficiency over all regions of operation, even into the <100mA input region where parasitics can dominate. And nearer to 90% for the >500mA range. Also allows for up to 2A output current.