DA14580 - no SWD connection

Learn MoreFAQsTutorials

5 posts / 0 new
Last post
Moterrola
Offline
Last seen:1 year 11 months ago
加入:2018-07-25 07:33
DA14580 - no SWD connection

早上好。
我'm developing my own board based on DA14580. When I'm trying to conect to the MCU through the SWD, the connection fails - it doesn't appear in Keil's options for target -> Debug -> settings. The power is OK, because DC-DC converter works as intended. I disconnected all the peripherials, but to no effect. Sometimes, once in a while, the device becomes discoverable, but only for a very short period of time. Is there something I have to do before programming the MCU for the first time? What am I doing wrong?

Device:
PM_Dialog
Offline
Last seen:18 hours 27 min ago
工作人员
加入:2018-02-08 11:03
Hi Moterrola,

Hi Moterrola,

Could you please clarify which version of JTAG are you using? Be aware that in our products we are using the J-Link lite ARM and we support only this version of JTAG. In fact that is a custom board, please check the connections between the DA14580 and the SWDIO pins are correct. Also, you should not do something before programming the MCU for the first time.

Thanks, PM_Dialog

Moterrola
Offline
Last seen:1 year 11 months ago
加入:2018-07-25 07:33
Hello PM_Dialog,

Hello PM_Dialog,
我've got 2 JTAGs: SEGGER J-Link Edu and SEGGER Atmel SAM-ICE. No matter which one I use, the effect is the same. Connections are correct, double-checked this one. I checked it with my oscilloscope and after touching the MCU's pins with probe, I was able to observe the clock signal on SWCLK and pulled-up data signal at SWDIO - I was even able to measure rise and fall time, it's about 200ns if I remember right - rather reasonable value for sure. There were no oscillations, too, so it can't be the inductive or capacitive character of tracks. Actually, earlier today, I managed to get it discovered in Keil uVision - I had to disconnect the power from the board for relatively long time (around 1 minute). Then I was able to detect it on my PC, but when I tried to start debugging, it threw away the errors about the CPU not being able to halt. After that, the MCU was no longer detectable, not until I disconnect the power again for ~1 minute. Looks like something's generating some noise, but I have no idea what it could be - a DC-DC converter integrated in the chip? All of the peripherials are disconnected from the supply.
UPDATE: I analyzed the work of DC-DC converter. Right after plugging in the voltage source, the ripples are around 60-70mVpp (not that bad, but still much more than in the datasheet). But after a short while, they jump up to like 700-800mV! What could be wrong?

Moterrola
Offline
Last seen:1 year 11 months ago
加入:2018-07-25 07:33
Sorry for making fuss - the

Sorry for making fuss - the reason was - as usually in such cases - trivial. 16MHz oscillator was soldered the wrong way. MCU working great now.
Thanks for the answer.

PM_Dialog
Offline
Last seen:18 hours 27 min ago
工作人员
加入:2018-02-08 11:03
Hi Moterrola,

Hi Moterrola,

我f the 16MHz crystal oscillator does not working, this means that the DA14580 is not working, so you are not able to detect the JTAG. Thanks for your indications. Glad you figured your issue out.

Thanks, PM_Dialog