Hi,
We're designing a charger based on the PMP40766 reference design. Our design includes minimal changes around the CC and CV reference points to make them easier to adjust from an external MCU.
The prototype is working great in every way except one. And we're having some trouble figuring out what to do about it.
The problem we're facing is that when a (near) empty battery is connected, the charger operates in the 1/10 CC mode where the output voltage is lowered by the maximum amount. Now when the battery is suddenly disconnected in this mode, the demand signal to the LLC controller instantly jumps up to maximum because the CV loop is also clamped to its max level due to the low output voltage. At this point OCP1 is triggered (the signal at the ISNS pin indeed reaches levels of about 10 volts peak). It seems that we would need something like the soft-start function in this case.
I had hoped that the smoothing circuit comprising C40/DD13/D15 would counter this, but it doesn't really.
What we have tried to remedy this situation (each action performed independently) is:
- Verified C26/R38 are correct according to the design procedure.
- Increased C40 in steps up to 1uF.
- Increased C32 in steps up to 2.2uF
- Removed R41
- Increased R40 to 10K
- Added extra resistor between the FB pin and ground in order to limit output-power. We're able to find a proper value for this resistor by experiment, but the minimum value required also limits the charger from reaching its full output power of 500W. It seems it's not just the level of the feedback signal that causes the problem, but rather a combination of the level and its slew rate.
- Added extra resistor between the cathode of U3 and ground, again in an attempt to limit output-power. Same problem occurs as before that this also prevents the charger from reaching full output power.
Any thoughts you might have on this would be greatly appreciated.
Kind regards,
Geert Kooijmans
