Possible Ways to Brick?

[Nick]

Hello,

I have a board with an RFDuino on it which uses a Nordic nRF51822 internally. We are seeing a few of them auto-brick when the input voltage dips too low.

My first guess is that it is doing an EEPROM write even when the voltage dips or something, but I am just wondering what can brick these? Any known errata for this?

Basically it is running off of a big battery and as long as it stays above 2.9-ish volts, it works forever it seems. When it goes to 2.8v or so, sometimes it stops responding and then never turns on again. I haven't written the software but I am trying to get the code to look at.

[Øyvind Karlsen]

Did you heed the following sentence from the MCP1802T datasheet?

The minimum VIN must meet two conditions: VIN>=2.0V and VIN >=(VR + 1.0V). Where VR is the output voltage.

I'm not sure of the workings of this particular LDO, but could it be possible that it enters an unspecified state when the input voltage gets low?

Could you try probing the output voltage from the MCP1802 device?

Best regards,

Øyvind

[Nick]

Yes I am fully aware of how linear regulators work. The nominal battery voltage is between 6v (when it is dead) and more than 8v when fully charged. There is always enough headroom for the 3v LDO and the 5v LDO and the variable 0v-5.5v buck.

I have also looked at it with a lab supply. When the input voltage is decreased, the LDO output voltage tracks downwards with the input until about a diode above ground where it just dies. All perfectly normal.

I also would have no problem with the Nordic chip not working properly when the input voltage gets too low. However, once normal voltage is applied again, it should work again. Right now, it seems like 20% of the time (4/20) the part stops functioning when the voltage gets low (that's OK) but then on the next power cycle when the input voltage is good, it never boots up.

That's what I am trying to figure out, is what could cause the part to not boot up, and what can I probe to see if it is booted. Is there a signal it puts out on the chip antenna to check connectivity that will show up with a scope? Does the reset pin or some other pin have some discernible pattern before, during, or after boot up? Test mode pins?

[Øyvind Karlsen]

Ok, sorry about the voltage confusion.

Normally I would read the SWD interface, but this does not seem to be routed out on the RFduino. The reset pin on nRF51 is active low, I am unsure what is the case for your module. Probably RFduino support will have a lot of these answers ready for you.

[Carl]

We're seeing similar problems and hopefully we can help each other. We're working with RFDigital's newer chip Simblee and have a similar problem bricking modules that you're seeing with the RFDuino. We have a few that appear to have the RFDigital bootloader corrupted. We can use Nordic tools to connect and program the chip, but can no longer use the arduino tools. We have another one that we can program with arduino tools just fine but it won't run any code. I don't know if the actually execute portion of that chip's bootloader was corrupted or maybe the soft device was corrupted? Are you able to update code on your bricked modules? If you're able to figure anything else out, we'd appreciate it if you could let us know. Thanks.

[mstrouse]

See the following forum post ... at the bottom of the first entry there should be a link to a PDF showing the JLink connections to an RFDuino or RFD22301 module. At least you can check to see if the module / nRF51 is still alive. Note that since the readback protection is probably set, you'll have to get creative in how far you can go to interrogate the part.

forum.rfduino.com/index.php