I'm testing a solar charge controller for LiFePO4 that uses absorption and float stages. By design, power-on initiates the absorption stage, and a timer terminates absorption and initiates the float stage. If battery voltage drops to a reset value, the device cycles back to the absorption stage.
Unfortunately, the unit I purchased isn't behaving as designed. It's cycling to the absorption stage when it should, but also randomly when it shouldn't.
Tech support is testing the device using an SLA battery, supplemented by a second power supply that's set to a voltage between the float and reset voltages. With sufficient PV input, the controller floats the battery as expected. When they remove the supplemental power, battery voltage drops to the reset level and charging cycles to the absorption stage as expected. They are not able to reproduce my problem with this setup.
So it appears that they are able to successfully trigger the main branches of their firmware's logic by alternately adding and removing supplemental power. But are there characteristics of LFP whose absence might skew the results? My intuition tells me they should be using LFP to more closely model real life, but I can't justify this position with facts.