12V Lead acid battery charger

[hamdi.tn]

Hi,
am trying to build a 12V lead acid charger , using an existing 100W SMPS based on UC3844, am thinking to get rid of the existing feed-back using TL1431 and use a more complicated feedback loop using a µC to set different charge current and different voltage, i drew a preliminary schematics (attachment) of what am thinking of, the idea is that µC set either current limit for tickle charge phase and the supply work as a constant current supply, or set a voltage limit for the floating and the supply work as a simple constant voltage supply, i tried to simulate that and it work but, does this make any sense in real, is it feasible or just me dreaming about writing super perfect charging algorithm 

[NiHaoMike]

Just limit the output voltage to 13.8-14.4V and limit the charging current. That can be done by clamping the voltage on pin 1 of the controller.

[Seekonk]

How can you hate a TL431?  You certainly made it a lot more complicated.  Keep the TL431 for the max voltage.  Then say you have a UNO etc.  PWM a little voltage to the 431 to give it a little bias.  Adjust that with current and voltage into the A/D.  Timers to include equalize function etc.  Frankly, most batteries survive nicely with only bulk charge.  For long term, a discharge function is just as important as anything else.

[hamdi.tn]

Thanks, so i keep the TL431 circuitry to limit the voltage to the floating voltage, and the UC3844 will handle current limit.. so nothing to change on this power supply , it will work as it is, i just have to fix the voltage limit to 13.8 and the current limit with the current limit resistor in the UC3844 and that's it ...
Bring an other question ... it's just to damn simple why a dedicated charging chip like the LTC4000 or UC3909 both cost 6 to 10 euro are needed for

[Paul Price]

I have my $.02 and it seems to me a few things don't add up in your pdf schematic:

27R and 12R driving the power MOSFET..just use one 22 ohm resistor and tie the two emitters together.
The collector resistor on the PNP MOSFET gate drive couldn't possibly work, 330k is much too high to work. You'd see smoke.

The 10R resistor feeding the LED drive of the optocoupler would likely blow out the LED, it much too low to limit current to a safe value..

[hamdi.tn]

Yeah am aware of that  i copy past component and i didn't re-edit all component with the correct value, the 330K is just 12R, the opto-coupler resistor should be around 100R. What am asking is if this feedback loop is possible, and if it's worth the complexity.
But since just giving a maximum charging current to the battery then holding floating voltage will not harm the battery and charge it just fine , i don't think it's a justified complexity any more.