B-L072Z-LRWAN1 Not working on battery

[shorai]

I have an application taht works fine when I power the board from USB ST-Link, but it does not work when powered by battery. LD7 lights up, but no other LEDs. My app turns some on immediately.

When powerded by USB I get LD5 Red, LD7 Red and two user lightsLD3, LD4 on after a second or two (my app).

The batteries are new Duracell AAA and give 4.6V to the board

Is there a jumper setting to switch to pure battery power?

I have tried this on 3 boards MB1296 D-03 series 210800*** and 211200***

It looks like part of the board may not be getting power.

It also does not appear to try to join LoRa.

My jumpers are factory default

JP9 1-2

JP10 2-3

JP5 off

[Tesla DeLorean]

Been a while since I worked on this, but I suspect it is powering the module. You need to check the ST-LINK firmware version, and perhaps remove the solder bridge allowing the ST-LINK to hold the module in reset.​

[shorai]

Thats weird. It works fine from a power bank plugged in to the ST-Link socket. i.e. it only gets 5V from there.

I would suspect that the LDO is perhaps not good enough (1.3V should be enough) or more likely one of the multitude of power rails is incorrectlyy connected for this configuration.

I hoped ST would know about this or at least be able to confirm the correct settings.

[Tesla DeLorean]

Pretty sure I'm using some of the C series board with 3.7 - 4.2 V coming out of my Lithium Ion battery board.

There should be a solder bridge, and path to injecting 3.3V via your own LDO, and the Murata module should run in the 2.4 - 3.4 V range.

[ty.locke]

When using battery power only you need to physically de-solder the MCU reset jumper on the bottom of the board. This connects MCU reset to the ST-Link so if it is closed when powering from batteries the ST LINK will hold the pin in reset since it's not powered.

Kind of ridic ST didn't put a removable jumper on this pin to make switching back and forth easier.

I normally just leave the MCU reset jumper open and then use a standalone st link + uart serial cable connected to the board headers so I don't need to keep soldering and desoldering that jumper every time I want to reprogram.​

[Tesla DeLorean]

+1 on jumper, the thing is hard to regain control of for programming, especially will low power modes in effect.

The ST-LINK firmware updates seem inconsistent as to whether they want to use NRST or not.

This board also has that JP5 against a gating of the ST-LINK USB supply

Lots of solder bridges, lots of powering options, and diodes to allow multiple sources.

The VIN regulator has a higher drop-out, I think that needs to be at least 5.5V so the 3/3.3V doesn't brown-out on one of my use cases.

The Series C has a LED that drains the battery by default.

[shorai]

Thank you so much for this. I noticed the link but did not fully appreciate it's significance. From your reply it is obvious!

You have saved me a lot of searching 🙂

I'll probably go for an external USB power bank. But at least I know what to do when I develop my own boards.

Thank you also Tesla de Lorean for a few insights that will be most useful to me.

I'm not too phased by the current drain on this board since I am using it for quick early prototyping. I plan to use the STM32WL55s but have a few of these that can be used for field testing in the short term.

Almost all the ST disco boards have some ridiculous features. Like me, the devs need to get a test board out quickly and are then immediately stuck with pin assignments etc. Case in poit :- the number of times a fully feaatured pin is used as a GPIO output to control or read a pin on an auxilliary part.

The NRST conundrum is still preferable to the Arduino / ESP problems of switching between download and debug modes that sometimes seem to depend on the phase of the moon..

I used to get mad, now I just smile and move on, knowing how often I have been silly.

[Foued_KH]

Hello @shorai​,

Based on the description in the "8.3 Power supply" section in the RM2115, I recommend you remove the bridge at SB37.

Does the article Cant power B-L072Z-LRWAN1 with battery (st.com) help you?

Please keep me informed about your update on this issue.

Foued

[Tesla DeLorean]

The secondary effect of the primary USB supply being gated is perhaps more frustrating than SB37, which can make it difficult to recover the board.

So many jumpers and solder bridges. It gives flexibility, but needs quite a check list of changes to remind me what to mod. Great power, great responsibility..

The Micro-USB connectors are also a bad choice for these types of boards, just too delicate. Much prefer Mini-USB or USB-C at this point.

Anyway, I remove the 3x AAA battery pack, and use those points to connect an AdaFruit battery/charger (one that can do both concurrently) and some 3.7V parallel cells.

Recommend adding a switch in-line, and if current is a big issue lose the LED, but I generally like to see if its 'hot' or not, and I'm basically using the battery as a UPS, and something I can solar charge.

Consider also the Arduino MKR WAN 1300 as an alternative, it has the Murata Module, and another quite capable ATMEL/MICROCHIP Cortex-M0+. This is sourceable at half the prices of the LRWAN-DISCO

[shorai]

Thank you @Community member​ for an excellent answer and synopsis.

Like you, I find the micro USB connectors inappropriate for this type of board.

Thanks also @KHALSI_Foued​ for reminding me to RTFM. Just looking at the diagram above, on first glnce, it appears that any of the power provisioning methods are equivalent. One has to look elsewhere to find discrepancies. Any reasonable human can fall into this trap. If flexibility is the aim, this circuit does not provide it.

When breadboarding, I much prefer the USB-B as it fits a 2.54mm pitch. Having adapter boards all over the place is a pain, and they need to be sourced. They suffer mechanical damage in this environment

The myriad of jumpers is also quite a minefield.

The newer Nucleo-WL55JC1 is a bit better, though it also uses a ridiculous USB connector for this type of equipment. (c.f Raspberry Pi :-).

I have not got round to testing this board as suggested, since i simply replaced the batteries with a USB Battery bank and cell phone charger. This suits my needs admirably and does not suffer the problems you have noted. Plug and play.