In my portable 12V solar charging system, my MPPT charge controller and LiFePO4 battery pack don't play together well at end of charge. The battery disconnects at some low charge rate, and when this happens, the controller sends panel Voc to the load terminals and signals an overvoltage fault. So, in order to prevent damage to the load, I never connect both panel and load to the controller; rather, I alternately charge and discharge the battery. I would like to be able to leave the panel and load connected to the controller at all times, so that ideally, the panel will supply power to the load and charge the battery, and the battery will take over powering the load as necessary.
Today, I placed a step-up/step-down voltage regulator between the controller's load terminals and the load to keep the output voltage at around 13V. For the very first time, I saw the controller alternate between three states (load powered directly from panel; load powered from battery; battery charging) with no overvoltage fault. The only negative is that the regulator doesn't output enough current.
(Neil_UK, if you jump in on this, I recall that in this thread, you recommended that an LFP battery be disconnected when fully charged. My guess is that the battery still disconnects at end of charge, and that it is disconnected while the panel powers the load.)
My questions are as follows:
1) I will be powering a radio transceiver, so low noise is a priority. Should I look for a particular type of regulator for this application? The specs of one unit that I'm looking at show that it uses synchronous rectification (it's my understanding that MOSFETs can produce noise) and an operating frequency of 200KHz. Should I avoid this one?
2) How will the circuit, with a regulator, behave in the absence of input power from the panel? I think that as the battery discharges and its voltage drops below the regulator's fixed output voltage, the load will see current decrease at a faster rate than it would without a regulator. Do I have this right? If so, then might it make sense to disable the regulator in the absence of panel input power?