I have made a PCB with the LM5013 buck regulator, but can't seem to get the output voltage high enough. The goal is to be able to adjust the output voltage using the LOAD_REGULATE pin, for a variable input voltage, but for now I have LOAD_REGULATE=0 V and V_IN=40V. I have a 20 ohm, 300 W load resistor at the output.
According to the datasheet, LM5013 regulates the output voltage so that the voltage at the FB pin is equal to an internal reference of 1.2 V. I am measuring 15 V at the output, and a switching duty cycle of 38 % and frequency 148 kHz. However, from the resistor values of the FB pin voltage divider, I would expect a duty cycle of 100 % as V_FB should be lower than the 1.2 V reference:
V_FB = V_OUT * (R5||R7) / ((R5||R7)+R6) = 15 * 6667 / (6667 + 432000) = 0.23 V
If V_OUT=V_IN: V_FB = 40 * 6667 / (6667 + 432000) = 0.61 V
The actual V_FB that I am measuring is 1.96 V.
My questions are: How can V_FB be 1.96 V when V_OUT if 15 V? Where does the "extra" voltage come from? And why isn't the output voltage regulated such that V_FB=1.2 V?
Edit:
Added pictures of the layout and oscilloscope. Ch.3 (blue) is SW pin, ch.4 (red) is V_OUT, and ch.2 (green) is FB pin.
I tried to remove C6, making the circuit similar to type 1 in fig 7.1 in the datasheet. This resulted in a higher V_OUT (28 V), but the inductor made a lot of weird high-pitch sounds. The sound changes as I hover over or touch the FB pin with a probe or finger.