we use LPC54618 for our product. During production/test we write firmware into the device using the internal mass storage device (MSD). Recently some devices failed in the production.
Tests show that the USB bootloader does not enable the mass storage device, instead I see a virtual com port. Device-Manager of Win10 shows "NXP RDDRONE-MFUK66 (COM5)"
Maybe the serial com port is an alternative to the MSD? Is it still possible to get the MSD (maybe at this device a pin is stuck at the wrong level which starts the virtual com port instead of the MSD)? We do a boundary scan test of the cpu and do not see any problems (maybe the BS does not cover all pins? need to check.)
Marking of the CPU:
Regards, Gunnar Bohlen
we have manufactued ~250 modules using this type of CPU without a problem like this.
We made some changes to the PCB and made 10 new modules to verify the changes. Now 1 out of 10 modules shows this "strange" behaviour.
On a different PC the CPU changes to MSD mode as expected (name: CRP DIABLED), I can write firmware into it and the firmware runs as expected.
On the PC I used initially I still see no MSD, and I see the virtual COM port "RDDRONE-FMUK66"
I tried to connect to this COM-port with HTerm, but this fails...
problem solved :-)
I recently installed the Software package "Mission Planner" with is a tool related to UAVCAN. This tool also installs a NXP USB-Serial driver (maybe to be compatible to the NXP drone kit which uses UAVCAN...)
The drone kit form NXP uses a different microcontroller than I use, but the NXP serial driver accepts the LCP54618 in USB MSD bootloader mode as a compatible device for the serial-usb connection.
Uninstalling and removing the driver from the PC solves the problem.