lt8708 variable current limit

Hello!

As for the first part, the basic behavior of the LT8708 is quite straight-forward: When the output voltage is below the set voltage, the LT8708 will deliver current as the smallest current limit permits. When the targeted voltage is reached, it depends on whether you are on the Vin or Vout side. As for the Vout side, the behavior is like in case of a normal switching regulator and the regulation will stop when the targeted output voltage has been reached. So when you are attaching another 48V source to Vout, the LT8708 is seeing that there's nothing to do and stops switching. As for the Vin side, the LT8708 continues to pump current as much as the smallest current limit permits from Vout to Vin even though the targeted voltage has been reached. In order to stop that, you can use VINHIMON or you can take some other actions such as deactivating SWEN or decreasing the current limits.

As for the second part/second question: When the LT8708 is hitting any current limit, it has to limit the current ;-) So assuming that you have got a certain load at Vout, the same amount of power (plus efficiency losses) has to be drawn from Vin as it is being output on Vout. So essentially Vout * Iout = Vin * Iin. When you are decreasing the forward input current (IMON_INP), the according limit will be hit, eventually. This means that there cannot be pumped as much current as needed towards Vout, so that the targeted output voltage can be established. Consequently the Voltage at Vout is dropping. In case you are using a simple resistive load there, the voltage at Vout will drop to a level so that the equation  Vout * Iout = Vin * Iin will still be satisfied.

As for the third question I'm not seeing a difference to the first question. Again: Assuming you connect the output side of the LT8708 to this 48V bus and you did set the output voltage at Vout to 48V as well, the LT8708 can sit back and relax when it is finding that the voltage is already at 48V. When the voltage is below 48V, the LT8708 will try to establish these 48V by pumping current as much as the smallest current limit permits.

Anyway, I'm having a counter-question for you since you are experimenting with the current limits of actual hardware. Did you check how precise the behavior is when you are limiting the current via IMON_INP? You wrote that you did inject 25uA and the current has been lowered "correctly as expected". How correctly did you check that? Background is that I'm also working on a design based on the LT8708. So far I'm doing more or less extensive evaluation through simulation. Thereby I found that the filter for CSIN/CSPIN that is suggested by the datasheet and that has also been implemented similarly on the DC2596A is causing an offset shift of the current limit. This is like as if there are not pumped 20uA out of IMON_INN and IMON_INP, but instead around 17.85uA for IMON_INP and 22.15uA for IMON_INP. This is causing a noticeable shift in the current limiting (at least in simulations). Probably this is caused by the asymmetric filtering around the shunt. What is kind of odd: At the Vout side the very same asymmetric filtering is used and the offset shift is not present. Probably I'm also opening another thread here describing that matter.

Greetings,

Mario