cutable pcb

[John Griswold]

Posted on September 19, 2017 at 16:00

I am using both Nucleo-64 and Nucleo-144 boards in my project. On the Nucleo-64, the cutable PCB has been removed, and signals from CN6 on the cutoff board were tied to the Morpho connector as outlined in the docs. (NRST, SWCLK, SWDIO, and ground). The board is powered via the +5 inputs.

I get an error message indicating that the device cannot be programmed - 'Cannot Load Flash Programming Algorithm'.

The wiring has been double checked, so we think that's OK. Is the ground possibly screwing this up?

Any help would be appreciated.

[Vangelis Fortounas]

Posted on September 19, 2017 at 16:49

Hello!

>>> The board is powered via the +5 inputs.

Is the LD3 red LED on?

[Tesla DeLorean]

Posted on September 19, 2017 at 17:03

Check algorithm/debugger selection.

Check solder-bridges related to the debugger pins.

[John Griswold]

Posted on September 19, 2017 at 16:52

LD3 is green, but it is lit up. LD1, on the ST-LINK board, is also lit, and toggles green/red as the PC tries to access the Nucleo-64.

[Vangelis Fortounas]

Posted on September 19, 2017 at 17:07

Hi again.

I  wrote about Nucleo 64   I have one of this in my hands.

LD3 at Nucleo64 is tied to 5v  just before 3.3v regulator.  It should be On.

PA14, PA13, GND and NRST pins are used to program the board.

[John Griswold]

Posted on September 19, 2017 at 17:20

Also writing about Nucleo-64. LD3 circled in red is on.Wires are connected to PA14, PA13, GND and NRST. 

[John Griswold]

Posted on September 19, 2017 at 17:23

I would imagine that the algorithm/debugger selection should be the same as it was before cutting the boards apart. Is this incorrect? If that is correct, do you have specific recommendations for which solder bridges to clear/close? I've been caught before, but I can find no mention of solder bridges related to programming using the cut off board.

[Tesla DeLorean]

Posted on September 19, 2017 at 18:15

For the NUCLEO-64

CN4.2 PA14 CN7.15 SWCLK

CN4.3 GND

CN4.4 PA13 CN7.13 SWDIO

CN4.5 NRST CN7.14 RESET

[John Griswold]

Posted on September 19, 2017 at 19:35

I'm using the Nucleo-l433RC-P, so the connector refs are different, but the signal names are the same:

SWCLK CN5 pin 17, SWDIO CN5 pin 15 and NRST CN5 pin 14

CN3.2 clock CN3.5 NRST CN3.4 SWDIOCN3.2 CN5.17 SWCLK PA14CN3.4 CN5.15 SWDIO PA13CN3.5 CN5.14 NRST NRST

We've gone so far as to take a fresh board out of factory packaging, cutting off the ST-LINK board, wiring up +5/GND to another board for power, then connecting SWCLK, NRST, GND, and SWDIO from the ST-LINK to the Nucleo - no other wires involved. JP4 (nRST) is removed. 

Just on a lark, I lowered the clock from 4MHz to a ridiculously-low 125KHz, and it started to program, seemingly appropriately, as the program works OK when attached to the debugger. Detaching the debugger brings up another set of issues to study.

[Tesla DeLorean]

Posted on September 19, 2017 at 22:35

Sorry, not got a NUCLEO-L433RC, pins look a little moved around

Does the SWD header go through the level-shifter, and the board level stuff not? Are you running the board at 1.8V or 3.0V ?