Dragonboard 410c not booting - steady BT light

I’ve been using my DragonBoard 410c for a few days now, but today for some reason it stopped working. I unplugged the power after it stopped, and now whenever I plug it in, only the BT light comes on. I’ve tried all the various recovery methods. Is the board dead? What happened? Why just the BT light?

Which release are you running (Debian 19.01, RPB…)? IT always useful to get a log from the debug UART if possible (e.g. via audio or UART mezzzanine).

The blue led is supposed to come on when bluetooth controller is enabled, that would imply that your board is alive and that kernel is running, but maybe something wrong with display or storage prevent complete booting to UI. Is the activity LED sporadically blinking as well?

I’m not sure which release it is, when I first loaded it up it had some sort of Android UI on it. As for log data, I haven’t been able to access any, and no other lights other than the BT light come on. The power button does nothing as well to alter this.

I suggest you to install a newer software release (depending your needs) like Debian or RPB:

I tried both of those and still no change. Nothing on the monitor, no response to buttons, just the steady BT light.

You mean you have been able to reflash the board (with e.g. latest release) ? but still unable to see anythin g on your monitor?

I wrote the various .iso files to a MicroSD card and followed the steps to flash them onto the dragonboard, but none of them change the state of the board, just the illuminated BT light.

I suggest you to follow USB recovery method (EDL mode): DragonBoard 410c Board Recovery - 96Boards
Then fastboot to reflash.

I haven’t had any success with this method either. The furthest I can get is running qdl, and the terminal just sits there saying “Waiting for EDL device” despite the board being plugged in and all the instructions followed

Is there any message in your kernel log (dmesg) when connection board over USB in EDL mode?

No, no messages at all in the kernel log.

Well, double check if you correctly configured the switch (S6) with ‘USB_BOOT’ on, and you are correctly connected to micro USB of the DB410C (not to FTDI/Console USB on a mezzanine). Then connect and power on the board. You should see something like this on your Linux Host (dmesg):

[165799.253977] qcserial 1-1:1.0: Qualcomm USB modem converter detected
[165799.254194] usb 1-1: Qualcomm USB modem converter now attached to ttyUSB0

Also double check your USB cable is not a ‘charge-only’ cable (no data).

I doubled checked that I’ve got my S6 switches properly configured, and that I’m connected to the right port (don’t have the mezzanine so I’m probably not plugged into it). Connected the board and powered it, dmesg continued to show nothing. Used udevadm to watch for events when plugging it in, it also saw nothing. I’ve used several different cables, and upon connecting other devices I know they’re data capable.

Given the way the first post in this thread is written, it sounds as if it failed spontaneously, rather than as a result of you making any kind of changes or otherwise software bricking it.

If that is the case, then you are really limited in the scope of what could have failed. Yes, it is possible for there to be a spontaneous software failure, but I find that improbable.

More likely, either the board or power supply is toast, and you would be surprised at just how often power supplies fail. My router, for example, is on its 3rd power supply in under a year.

This type of board can do some strange things when they’re not getting power that they like. Have you tried the board with a different power supply?

And also note that I do have a DB410C with a fried PMIC. I shorted something out on it by misaligning a mezzanine. Now I get one LED going on, starting out dim and growing in intensity until I pull the plug. Its been quite a while since I’ve powered that board up (a few years, actually), but I think it was the blue light. The only reason I still have that board is because it makes a great template for assembling mezzanines.

Well that’s unfortunate, but at least I can take solace in the fact I probably didn’t cause this. I’ve got two different power supplies I’ve tested this board with, and given that neither is working, my guess is a failure in the PMIC. Is there any way to definitively test for this, with say my multi-meter?

I didn’t say it is the pmic. I said on mine it’s the pmic. Yours could be a completely different failure mode. For instance, does the blue light on yours gradually grow brighter?

With your multimeter, check the voltage rails on the low speed header. If you see a 5v there, then you know the 5v buck regulator is good. If you see a 1.8v, then at least part of the pmic is ok (although its absence doesn’t prove it to be bad). There is also a 3.7v buck that you can check with a multimeter, see page 15 of the schematic.

The blue light on mine doesn’t grow steadily brighter, when plugged in the blue BT light just comes on full power. Probing around I can see both 1.8v, 3.7v and 5.5v at various points. All the bucks appear to be working as well.

If the Blue LED comes up as soon as you plug in the power, most likely it has nothing to do with the firmware. Check the power rails on the J8 header, especially the 1.8V. This is the only power rail generated by the PMIC accessible on J8. All the other power rails (5V and 3.6V) are generated using discrete regulators. If you don’t see the 1.8v power rail, it is highly likely it is a hardware issue. Most likely a dead PMIC.