NRF52832 (on BMD-350 Module) SWD not connecting...

[GJChristman]

I'm having no luck in getting JLinkExe or nrfjprog to see my BMD-350 device on my custom PCB. 

I have read though all the threads here that reference similar issues and have verified all of the following:

* Connections to SWCLK and SWDIO are made properly and have no pullup/pulldown resistors.

* The device is remotely powered

* The signals on SWCLK and SWDIO are there and look to be the proper levels.

* The device itself works - it has factory firmware that advertises services (RigDFU etc.) on BLE

I have tried both a Segger JLink as well as the using the development board programmer via the 10

pin connector.

There has to be something basic I am missing.  Below are a
PDF of the device schematic and the log.log output from nrfjprog -f nrf52 --recover

I'm doing the development in Linux, and have had no problem in connecting to the development board.

I've also asked for help from the manufacturer, UBlox, but have heard nothing back from them.

Any pointers here would be greatly appreciated.

-Glenn

6648.log.logSG_Bluetooth_4-Layer.pdf

[Jared]

Hi there,

The schematics looks ok. 

1. Exactly what voltage do you measure on VDD, GND and RESET pin?
2. Have you tried omitting the regulator stage and power the board directly from a bench top supply? 
3. Do you see this issue on all of your samples? 
4. Thank you for including the log from nrfjprog --recover. But could you also specify what error message that is reported by the program when you just run nrfjprog --recover? 

regards

Jared 

[GJChristman]

Hi Jared,

Thanks for responding.

 - The voltages on the board are:

   * VDD - 3.28V

   * GND - .009V

   * nRESET - 3.28V

Under test, I am powering the board over the SWD cable with a bench power supply into a 3.3V regulator. 

I've tried a couple of the BMD-350s, and they all seem to give the same results

One thing I noticed when looking at a couple schematics I found for the Thingy devices is that you use a 4.6 uF capacitor after the regulator - I have just a 1uF on my board.  I was going to try putting a 4.7uF in its place but have not yet done so.

I've attached the output from nrfjprog --recover:

Fullscreen nrfjprog_out.txt Download

ERROR: Unable to connect to a debugger.
ERROR: JLinkARM DLL reported an error. Try again. If error condition
ERROR: persists, run the same command again with argument --log, contact Nordic
ERROR: Semiconductor and provide the generated log.log file to them.
NOTE: For additional output, try running again with logging enabled (--log).
NOTE: Any generated log error messages will be displayed.

[GJChristman]

No joy

I pulled R7 and was able to lift a trace to disconnect the OP-AMP - neither seemed to have any effect.

Stubborn little bugger.

[GJChristman]

Just to share the misery, here are photos of what I've been talking about:

The whole setup:

The interface breadboard:

and the target...(yellow wire is scope connection on SWDIO)

[hmolesworth]

Quite sure about the colours? I see from the photo:

White  1 VDD
Yellow 2 SWDIO
Red    3 nRESET
Blue   4 SWCLK
Black  5 GND
(n/c)  6 SWO

TC2030-IDC-NL-Datasheet-Rev-B.pdf

[GJChristman]

Yes, Those are the proper colors - signal mappings. 

On the 10-pin connector, 3,5 and 9 are all tied to ground.

[Jared]

Hi Glenn,

GJChristman said:

I re-double-checked the SWCLK and SWDIO and verified that they were going to the appropriate places.  Just to be totally sure, I swapped them anyway and ended up with the same results.  I then swapped them back to the original connections.

I put a scope on the SWCLK and SWDIO lines and again verified that I had 0 to ~= 3.3V signals when issuing commands from nrfjprog or JLinkExe.  The fact that I get appropriate level signals would seem to indicate that they are indeed being connected to the BMD-350, since the pull-up resistors are on the chip.

I also verified that SWO was not connected to GROUND (see my other reply).

Is there any chance that you could try to use a standalone debugger and connect it directly to the modules SWD interface without using the breadboard in-between? I just want to rule out any chance of the breadboard affecting the debugging process in some strange way. 

GJChristman said:

I'm going to build up another board or two and try those.

Ok, it would be great if you could do that. The issue might be related to the module itself. 

keep me updated,

regards

Jared 

[GJChristman]

OK,

I was able to connect to and program the board that has just the BMD-350 and VDD caps - so progress - whoo-hoo!

This at least verifies the SWD connections and interface.  I'll try adding support components to this board and see if I can get it to fail.  It's possible that there is still an issue of getting the BMD-350 properly soldered even though the ones I had pulled off looked pretty good..

I appreciate all your help, Jared et al!

-Glenn

[Jared]

Hi Glenn,

Ok, good to see some progress on this. What exactly is the difference though? Did you remove the regulator stage or the breadboard?

regards

Jared 

[GJChristman]

I built up a target board that had only the BMD-350 and the two caps on the 3.3V line.  The interface board (breadboard) was unchanged.

[Jared]

I see,

GJChristman said:

ll try adding support components to this board and see if I can get it to fail.  It's possible that there is still an issue of getting the BMD-350 properly soldered even though the ones I had pulled off looked pretty good..

Keep me updated then,

regards

Jared