It is the thermal vs. chemical reaction that is at issue. And electrochemical reaction is not subject to electronics formulae. Sorry. This is where most pure electronics folks jump the rails. The equation is REACTIVE and impossible to explain without creating pages of abstract calculus. This is precisely why E=MC2 predictions fail miserably to describe or control the charging process -correctly-
It takes electrical plus chemical engineering to configure a charger design OR
Much practical experience. Practical experience is still necessary even when electrochemical engineering is employed fully. The school of hard knocks really tempered my cocky attitude when I first started out.
Using a group 27 a a GENERAL example, at ideal depletion level (maximum charge acceptance/lowest impedance, whatever you want to call it, forty-eight amperes would not be an unusual starting point to achieve 14.8 volts Vabs @ 20c.
The human eye and judgement is vital for this formula to work. "When" bubbling starts, cannot be easily established by machine. But permutation correlation to electrical values can indeed be employed once the parameters have been established.
If chemistry were not involved this would be a snap to resolve. But simple resistance, current, and voltage workup predetermination is utterly useless. Constant monitoring and adjustment would compensate, but this isn't a lab experiment, it is a distilled formula that most folks can understand and follow.
I wish there was an even easier way. But please keep in mind, this is SPECIALTY CHARGING FORMULA with compromises throughout. Limitations of generator run-time complicates not simplifies the issue.
The BEST WAY to judge the validity of what I recommend is to actually do it.
Again I emphasize! And STRONGLY! The 14.8 temp corrected E value is TIME SENSITIVE. Charging MUST CEASE (be reduced) when mild bubbling is observed. This isn't time bomb planning, a span of ten minutes or so is a window wide enough to allow perfection. An old fashioned transformer charger with a wind-up timer is INFINITELY more prone to error than this formula. And it is assumed the user is intelligent enough to to realize the regimen, according to minutes spent is repeatable, and predictable. No, a battery 10% depleted uses the same time quantity as a 50% depleted battery. This is common sense.
Finally, thermal incline is absolutely related to wattage. This is balanced by the time spent at 14.8 volts. It is no crime to elevate battery temps to 45c and this will not happen UNLESS ambient temperatures are very high to begin with. I assume few RVers will spend a vacation at Stovepipe Wells, in July.