Flashing an off-the-shelf STM32L031 with ST-LINK

Posted on by roblesg

This is a continuation of my former post NUCLEO-L031K6 and ST-LINK SWD. What on earth did I do? in which I tried to flash a standalone STM32L031 and found some serious obstacles. Now, I have solved all issues that I found and I have been able to flash the darned MCU. The idea was to use the ST-LINK in a discovery kit with STM32L476VG MCU connected to the STM32L031 with only two pins, SWDIO and SWDCLK in the connector CN4 of the kit. The wiring is straightforward when every is tied and connected correctly. And I am stressing this because you have to be sure that you have connected all pins that have to be connected. In concrete,  pin 5 in the STM32L031 labeled as VDDA, the reference for the analogue inputs, is NOT internally tied to VDD in pins (1 and 17) so it HAS to be connected to power the MCU otherwise it would be pretty dead. With this in mind, you only have to follow the next figures to connect the ST-LINK pins to the STM32L031. First, the power supply at 3.3 V and ground which are taken from the Discovery kit.

Then the connections from CN4 to the MCU. Notice that I decided not to use the Vapp nor the SWO pins since they are not strictly necessary to flash a program. However, it is mandatory to connect the ground pin of the ST-Link port even when it is internally connected to the STM32L476 GND pins in the Discovery board.

Finally, the pins I used in the STM32L031 (LQFP32) were:

Continue reading Flashing an off-the-shelf STM32L031 with ST-LINK

Partial discharges in an XLPE high-voltage cable

Posted on by roblesg

In the early works to determine the exact site where partial discharges (PD) are happening in a 12/20 kV XLPE high-voltage cable, we need to know what are the previous conditions of such conductor. We are using a cable that has been stripped in the ends to show all layers. The shielding has been removed and connected to ground in both ends and the internal conductor has been connected to the rated high voltage.

Two high-frequency current transformers (HFCT) were clamped to the grounded shieldings and the signals in the oscilloscope showed that we had a stable and important activity of partial discharges. However, we didn’t know exactly what type of PD we had since the outcome of the phase-resolved PD pattern was not clear: might be a mixture of corona and surface discharges.

The next photographs were taken to cast some light on what was happening in the cable at 12 kV. The first one was taken with an automatic exposure and artificial illumination and the second one was taken with an exposure of 60 s and automatic aperture without external light. It can be clearly seen the glow of the discharges and how some of them reached the surface of one of the layers.

 

 

 

 

 

 

This information helps in determining why the time-length of the pulses seen in the oscilloscope was so long: 2-3 microseconds till the oscillation was completely damped and the amplitud was much higher than expected. Without a previous calibration to picocoulombs we know that internal discharges peaks may reach 10-20 mV and we were having pulses with peaks in the 1000 mV range which can only be associated to high-energy surface discharges.

Now, knowing the previous condition of the cable we may start creating internal defects to determine the time-of-flight of the pulse and, hence, to find out the discharging site.