[Matt Barney]

I understand. I assume that whatever is in the master branch is considered stable and OK to deploy in the field (while keeping in mind all standard caveats to software development and testing 🙂 ).

 

[Matt Barney]

Thanks Sara, that’s very helpful. -Matt

[Matt Barney]

Thank you Sara! Using my XBee TH board and XCTU, I got my FPLMN list cleared, and am still trying to force a Verizon connection. Nothing yet, but I will keep trying, power cycling, switching to Bypass and back to transparent, etc.

[Matt Barney]

Thanks, Sara; I’ll check out the FPLMN issue and keep trying with a modem at my desk.

What’s the easiest way to send AT commands directly to the modem? Load StreamDebugger on a Mayfly, as described here? https://www.envirodiy.org/topic/trouble-initializing-xbee3-lte-m/#post-12658

I can’t seem to get any response; i.e., nothing gets echoed to the serial monitor window in PlatformIO.

[Matt Barney]

Re: updating XBee and u-blox firmware, I just noticed the ‘force cell modem update’ checkbox in XCTU defaulted to unchecked. Hadn’t noticed that before… oops. I’ll enable this from now on when updating my modems.

-M

Attachments:

1. 

   XCTU.png
   

[Matt Barney]

OK, thank you, Sara! Bummer that there’s no clean solution yet for reliable LTE that’s carrier-agnostic. I suppose that’s asking a lot. 🙂

Matt

[Matt Barney]

I also meant to mention that one of the ‘Bug Fixes’ in the XBee release notes indicates that they enabled flow control between the XBee and the u-blox, which hopefully is a good thing.

[Matt Barney]

This is great news! Thank you @srgdamiano and @ckillen!! I’ve begun ordering parts for 9 additional stations and have been fearing that buying another 9 XBee3 LTE-M modems was going to lead us into a world of frustration with failing modems; hopefully this discovery prevents that.

Regarding XBee firmware: Just a week ago, we deployed a batch of XBee3’s which I’d just updated to the latest, 11413, so it caught my eye when Cal reported installing 11415, which was just released Feb 19th. (not sure what happened to v.11414). Reading the 11415 release notes, I noted the following, under Known Issues:

1. By-pass mode is now deprecated and is not recommended for new designs. [perhaps this is not new information, based on what @shicks has written previously]
2. Modules registering to the AT&T network have been observed infrequently to get stuck in the PSM
   dormant state (AI=0x2c). This state persists until the XBee is reset. [Could be related to Cal’s issue in post #13835?]

Anyway, my practice has been to update firmware, using XCTU and the XBee TH board, when I unwrap a new XBee and program a Mayfly for it, and I’ll continue to do so unless I see advice against it here.

Finally, do I(we) need to also be concerned with updating firmware on the uBlox, or does XCTU take care of that as well?

-Matt

[Matt Barney]

Wow – thanks, all, for the great information!

Because our problem site is relatively inconvenient to access, I think on our (@jlemontu-org ‘s) next visit, he’ll come armed with a number of options. We’ll pick up a Taoglas antenna, as well as bring along a spare Pulse antenna, in case the existing one is compromised. And we can experiment with different antenna positions relative to the other conductive components in the enclosure. Perhaps it makes a difference which plane the antenna is oriented in as well.

Really appreciate you all sharing your wisdom! (and I love the photo of your workbench, Sara!)

Best,
Matt

[Matt Barney]

Our Mayfly, XBee3, and antenna are inside a Pelican case, so yeah, I was wondering whether that inhibits signal, but I’m really out of my technical depth on these antenna and signal questions. Jake was able to get decent voice service on his Verizon phone at the site, and my understanding is that ModularSensors (or maybe it’s the XBee3’s firmware) will select between Verizon and AT&T based on signal strength, so I was hoping we’d see solid signal after I upgraded both.

[Author]

Posts