LTC4227 operation with dual 2S battery

AlexisKyriacou said:

Did anyone face that before?  LTC4227 switch to IN2 when IN1 has 0.6V less than IN2?

The key is the 0.6V figure - It's a diode drop.
Q1 and Q2 form a diode-OR, where the higher supply is passed to the output.
Even if you use the D2ON# feature, the diode-OR behavior will still occur due to the body diodes of Q1/2.

I believe the intent of the pin is to disable VIN2's ideal diode and prioritize VIN1.
This works as long as the V_SD across Q2 is <0.6V.

AlexisKyriacou said:

Did anyone face that before?  LTC4227 switch to IN2 when IN1 has 0.6V less than IN2?

The key is the 0.6V figure - It's a diode drop.
Q1 and Q2 form a diode-OR, where the higher supply is passed to the output.
Even if you use the D2ON# feature, the diode-OR behavior will still occur due to the body diodes of Q1/2.

I believe the intent of the pin is to disable VIN2's ideal diode and prioritize VIN1.
This works as long as the V_SD across Q2 is <0.6V.

Ashapiro thank you for the reply. 

So, I have to connect the #D2ON to IN1 to keep it HIGH and when the battery from IN1 removed then it will switch to IN2. But if battery at IN1 hits ON threshold first the system will turn off.

AlexisKyriacou said:

So, I have to connect the #D2ON to IN1 to keep it HIGH and when the battery from IN1 removed then it will switch to IN2.

If IN1 goes from 12V to 0V, I agree.

AlexisKyriacou said:

But if battery at IN1 hits ON threshold first the system will turn off.

I don't think system would turn off.
In your schematic, ON is connected to the output of the OR'd supplies through a resistor divider.
Basically, ON will always be connected to the higher of IN1 or IN2.
So even if IN1 drops to 0V, ON will see IN2 (assuming IN2 is present).

Ashapiro said:

I don't think system would turn off.
In your schematic, ON is connected to the output of the OR'd supplies through a resistor divider.
Basically, ON will always be connected to the higher of IN1 or IN2.
So even if IN1 drops to 0V, ON will see IN2 (assuming IN2 is present).

correct.

So, based of my needs, the LTC227 or any other diode-OR is not suitable because i will always have the 0.6V diode drop accross Mosfets, correct?

AlexisKyriacou said:

So, based of my needs, the LTC227 or any other diode-OR is not suitable because i will always have the 0.6V diode drop accross Mosfets, correct?

The 0.6V drop occurs only when an ideal diode is off (Vgs=0V).
When the the IN1 or IN2 channel is enabled, the controller will regulate a 25mV forward voltage across the ideal diode.
Just understand the Q1 and Q2 are basically diodes, not switches.
I recommend that you play with the part in LTSpice to get a feel for part behavior.

Hello Ashapiro,

I did some simulation in LTspice as you propose and after many tests I add back-to-back mosfet and I get the needed result, what is your opinion about using back-to-back?

Thank you,

AlexisLTC4227-1_Nch.asc

AlexisKyriacou said:

I did some simulation in LTspice as you propose and after many tests I add back-to-back mosfet and I get the needed result, what is your opinion about using back-to-back?

If the circuit is doing what you want, that's great.

You basically built a prioritizer, with 3 NFETs in the powerpath; The extra I*R drop will hurt at heavier loads.
If you use a dedicated prioritizer like the LTC4418 or LTC4421, you'll only have 2 FETs in the powerpath.
I recommend the LTC4421, since it has active current limiting and uses NFETs.

Hello Ashapiro,

Is there any solution using passive components to achive the above result ?