LM5117: VCC pin output is only 5.2V and IC temperature is rising

Here is the schematic. This project is for my undergraduate project.

There are few changes I made, the DEMB pin is now connected to VCC, instead of R10. I also removed the slew resistor and capacitor, R5 and C4. Feedback capacitors C6 and C7 are not yet connected because they are for the 2nd part of the project.

About the quickstart file, how can i send it to you?

As for the current in the VCC, I will look at it later but my setup uses VCC as a source.

EDIT:

additional question, is it possible that the LM5117 was broken because the temperature reached around 130 deg Celsius?

lm(2)5117_quick_start_hv_calculator V1.5.xls

Here's the quick start Excel file.

The output of the converter is 5V, 3A for input voltages of 42 to 54V. Will try changing the components later then update here if it worked.

For the VCC, I was following a previous project which also used LM5117 for the same application with different input voltage. In that project, the VCC was connected to bootstrap diode, bootstrap cap, and the DEMB pin (which will set DEMB high, thus turning off diode emulation). 

Hi! I was able to test just now. I changed the gate resistors to 0 ohms and the boot cap to .1uF but the redult is still the same. However, I probed the SW and LO pins and they are not what is to be expected.

I also observed that when I have diode emulation enabled, the output goes from around 3.8V down to 3V then rises a little to 3.2V then it settles at 688mV. But when diode emulation is disabled, the voltage is only at 7mV.

Will try doing that. But if I remove the CC circuit, how can I do a CC/CV charging? I am also using TPS25740B for the USB-C controller.

Hi! Just some updates. I removed the CC circuit and it gave me an output of 4.8V with DEMB floating (diode emu enabled) and when the diode emu is enabled, the ripple is high.

To add to that, the gate signal is still wrong.

Hi! Will recheck my circuit again. And also try to change mosfets. The problem is that I dont have other mosfets available. I checked the datasheet of my curent mosfet and its VGS(th) is 3 to 4.9V while the CSD18563Q5A has a 2V threshold only. I know that the LM5117 should be able to drive 4.9V threshold since the output of the low side should reach VCC which is about 7.6V.

I see. I will try to change mosfet if that's the case. But right now, I can output around 4.8V only. And my VCC pin drops from 7.6V to around 6V. This is the LO gate signal. I think that the mosfet shorts for a certain period of time. Also, there are times when I probe the LO, it shows me a pulses of around low kHz. When I zoom in, it shows the supposed 100kHz gate signal, but it is still erroneous.

What can you recommend I do? I am planning to redesign my board but the problem is my time constraints.