Our design uses a MPM3690-50D module with an output of 0.85V. We noticed that the module draws an unusually high amount of power without any load in both FCCM mode and Pulse-Skipping mode compared to the eval board. In fact, there is no power savings in Pulse-Skipping mode at all unless the MFR_CTRL_COMP RAMP property is lowered to “Multi-phase 8.6mV" from the default of 27mV. Even then it’s only a partial savings compared to what is seen on the eval board under very light load. It appears that the issue has to do with the feedback control loop. So, the question at this point is: what can be done to fix the control loop? Usually, Modules will have COMP pins where you can add caps and a resistor to adjust the loop stability. This part does not have that, although something could be possibly inserted at the point of the FB pin. I have no knowledge of the type of loop filter used in this device, so I am not sure what to try at this point. Hope you can provide some direction.
Hi @kpriest ,
If the issue is not replicable on our EVB, it must be an issue with the schematic or layout.
Please submit a ticket for us to review your layout and schematic: MPSNOW Technical Support Link
The module is constant on time instead of current mode control, so there is no need for a traditional compensation network in a COMP pin.
Regards,
Jonathan
I did fill out the Technical Support Link, but nobody ever contacted me.
When was the technical support link ticket filled out? I can check with the team to see if we received it.
Also - did you try using the EVB? Is the issue replicable on there?