good news on the current board, thank you for that. 2 out of the 3 electron boards are doing well. both weather stations. 1 in placitas,nm 1 in colmar, france. the 3rd i never built cause i could never get a stable connection in tucson but recently moved to a area that has been very stable both for a 2G electron and a boron so may build it soon and see how long 2G electron can work here. i’m thinking i’d like to motorize one of those grass seed spreaders that normally hang from neck and has a crank handle. if i can get it motorized i’d like to use it to spread bird seed from a post. right now there is too much competition between birds for a space at the feeder. anyway, thank you for your work.
Very cool and thank you for sharing your use case. I am glad you found the Electron version useful and will share the Boron project on OSHPark as soon as I verify it with the first few samples.
I find it helpful to put vias into the mounting pads of connectors. This makes is less likely that the pads lift from the FR4 substrate during rough handling of the connector. Of course it only works if you don’t need the other side for routing.
Thank you for the suggestion.
I am trying these surface mount connectors in the hope that I could eliminate through hole parts - and their cost from my contract manufacturer. However, my concern is over the reliability of the connector through continued use. I like your idea of putting vias in the pads and I wonder if it would still give a benefit if it was only every other pad with a via. I do like having the space to route on the back side of the board after all.
OK, Progress - and a snag which currently has me baffled.
For some reason, if the Boron is inserted into the carrier, it will not turn on. By process of elimination, I was able to track the course of this issue down to one pin - D5.
From the schematics, you can see this is the Done pin for the watchdog timer. I removed the solder jumper to de-power the watchdog but still D5 prevents the Boron from turning on. I have checked for shorts, loaded Tinker on the Boron and am currently scratching my head. Anyone have an idea?
Double-check and triple-check the pin layout for the watchdog chip and make sure it is soldered in the correct orientation. If you grabbed the footprint from a library, assume it is wrong and validate it against the datasheet.
hey Chip, out of curiosity, is this also happening if you insert an Argon or Xenon?
if you had previously used that boron to test an ethernet feartherwing the D5 pin is reserved for ethernet use in configuration flash.
Great question. Tried it with an Argon and it worked just fine.
Tried it with a different Boron - same D5 sensitivity. Actually, upon further testing, it is D5 and Reset. If either is not connected the Boron turns on. I checked to see if there was a short between D5 and reset.
Good suggestions. I have been using this exact circuit on a number of designs so I am fairly confident in it. There is something different about the way the Boron works. I have un-re soldered the watchdog in case it was a bad chip and tried a new Boron. Not sure what it is but will keep trying.
Anyone else have luck with the TPL5010 and the Boron?
i read about a D5 issue that may be relevant to the problem seen with the carrier,
so is D5 being high at startup perhaps a problem with this board?
Thanks for finding this. Yes, it does seem to be an issue. The Done pin is meant to be an output so, I guess the TPL5010 is behaving badly when it is receiving an input from the Boron.
I will move this connection to another pin and give that a try. Appreciate you finding this.
@chipmc, D5 is only an issue if the Boron was configured to use the Ethernet shield previously. You should be able to “factory reset” the Boron and reconfigure it so the Ethernet config is wiped IMO.
I tried this on a couple Borons neither of which were ever used with an Ethernet sheild. The second one was right out of the box and both before the setup process and after with “Tinker” I observed this behavior.
I moved the TPL5010’s Done pin from D5 to D7 as a quick test and that fixed the issue. Then, using Tinker, I configured D7 as an output - this reset the Boron. I know that D7 is used for the on-board LED so, I need to find another solution. Still, it shows that the TPL5010 does not like the Done pin attached to anything other than a pin configured for Input. Good to know and should be an easy fix.
I do think that D5 on the Boron is a bigger issue than just being associated with the Ethernet shield.
Initial project sharing. Generation 3 Particle Device Carrier for outdoor applications:
So far, so good on the testing after I moved the Done pin to A3. I will start sharing the design file and OSHPark project in case anyone wants to get started on their own. Note, the files below have the corrected wiring and don’t require the wire you see in the picture.
GitHub Repo for EAGLE Files:https://github.com/chipmc/Particle-Boron-Carrier-v1.0
OSHPark Project so you can order your own: https://oshpark.com/shared_projects/7uc29YnL
Enjoy and thanks for all the help / suggestions. Please let me know if you see anything I should change for v1.1
One thing I’m not clear is if the price includes the components, or you have to buy them on your own then solder them on your own.
Would you mind elaborating, please?
Oshpark.com charges US$5 per square inch (but you get 3 copies of the board). At $26, the price is only for the unpopulated boards. You have to source and solder components on your own. If you do use oshpark, make sure you source three times the BOM since you will get 3 boards for the price.
Chip, you’ve done an amazing job on this! I’ve been following your progress closely as much of what you’ve designed will be required for my own carrier board. This post has saved me countless hours and dollars already, and I’ve only just started designing my board. Great work! Thank you.
Very glad to hear this is useful to you. Look forward to seeing what you come up with.
Hey Chip - something I noticed from one of the pictures above - if there is space, put a hole corresponding to one (or more) of the modules corner mounting holes so you can use a nylon standoff to secure the module to the base board - I have had modules pop off baseboards over time (no known reason…) and esp. for a remote location, I would prefer everything staying in place as assembled.