Calling all battery charger experts -- what's better? Separate or shared charging paths?

I’ve been looking at battery chargers and I’ve noticed there are some that offer a connection path to the battery that is separate from the system load. Others have a path that’s shared. What’s the tradeoff? With a separate path it seems like the charger chip can charge the battery separately while supplying the system load with the charging source (ie. a USB plug). Whereas with the shared path it seems like the battery is always supplying the system load – is that bad for battery health? How does a HW designer decide which is better?

Here are some examples so it’s clear.

Separate paths
https://upverter.com/upn/bdcb77dece07982b/

Here’s a reference design that makes it clearer:
https://upverter.com/anandh/7fc81be663a6e9f0/BQ24072/

Shared paths
https://upverter.com/upn/03b31e9ad9e31bd2/

Here’s a reference design showing the shared path:
https://upverter.com/anandh/abdc7eb998729126/BQ24040/

1 Like

I believe the point of a power-path controller like this is that the chip decides where to allocate available input power, either to the system or to charge the battery. At the same time, the power controller can switch from input power to battery power when the input power becomes unavailable, offering a seamless switch of power source.

1 Like

From ShoulderChip on /r/electronics/

I think I would prefer the separate one. With a shared power output, the charge management IC can’t really know how much charging current is really going into the battery, and how much is going to the load. I researched the battery charger chips from Microchip last year, but their data sheets didn’t even give me a clear idea of how to connect them! I couldn’t figure out if the load was supposed to be shared with the battery, or what. At least the ones you are looking at include a reference design.

From Jeff F on e2e.ti.com

If the output of the charger ties to the battery and the system load, there is no way to know how much current is going to the load versus into the battery. Charger ICs with separate path, called Power Path, allow for the system to be powered on as soon as input power is applied to the charger, even if the battery is completely dead. Also, Power Path allows for more control over the battery current, including proper charge termination. If the application were a cradle charger, then you wouldn’t need power path necessarily so it is application dependent. Also, cost and size are factors to consider.