TPS40170: Overcurrent Fault and Short cut

[Ashwin G. (Firma: Light'n TEC) (ashwin_gawande)]

Hello,
I made some prototypes base on the controller above and as can be seen 
in the picture joined. By first powering the board with 24V, we do have 
overcurrent by soft start as Vss was switching between 2,5V and 300mV 
and I change the Rilim to allow 8A and the capacitor and resistor on the 
gate of low side transistor. and everything was working so far as 
expected.

After some powering some other board, they work for about 2min and the 
following happen:

on one board they were short on GND and 4,3V. Trying to troubleshoot by 
removing all the capacitors on the bank, the short was persistent. I 
remove the controller, inductor, and transistors, They were not short on 
the complete path anymore. but after soldering all the components, it 
works for a few min again and goes in short. => what could be the cause? 
joined are layout and schematic.
on another board, the short occurs between 24_filtered and sw. I measure 
the voltage on sw expecting a voltage above 100V as specified in the 
datasheet of the transistor, but it is maximum 30V.  there is no more 8V 
at Vbp and 3,3V at Vdd. implying no LDRV and HDRV PWM. These errors 
occur on most of the develop board. I have analyzed the board as my 
design was mostly taken from web bench. I decided to change the resistor 
on the enable pin to 2,6Mohm to limit the current to enable pin to 10µA 
but no change. => could somebody tell me if he does have the same error 
with the chip??
on another board, the board has been working for more than 3hours and 
short occur. my fuse just burns. after troubleshooting, I came out with 
short between GND and sw pin.
It comes out that I am having anyhow always short on different path of 
the dcdc controller. I would be delighted to get any help in 
troubleshooting or resolving this issue.

thanks for your time and waiting to hear tonight from you.