@reidfo I've put more time than I would like into the lte here in oz too Steve. Stumbled across some bizarre current draws after modem reset in my battery powered application. On the upside I can now finish an lte connection & get into micropower deepsleep in 3s whereas the disconnect/detach sequence used to take more like 7s. These gems are hard won by those of us who go down these rabbit holes & I'd be happy to share on a wiki. I also think pycom HQ could post some material too, like exactly what lte.init() does for instance.

@wghirakawa Related to a different question (https://forum.pycom.io/topic/4118/gpy-or-fipy-using-lte-in-the-us), could I please ask you to check the result of 'AT+SQNHPLMN?'. I am trying to find out if Sequans supports 6-digit PLMNs in the US.

@wghirakawa The code seems to be ok. I don't have access to Cat M1 network (only NB-IoT available here) so I cannot test with the same Sequans firmware. Are you able to perform communication on the same APN and with the same SIM using different hardware? @iwahdan could you please take a look?

@panda Thanks for the input. My first thought was Hologram due to lower cost to get started. Long term commitment to data plan is another issue which I am not sure of. At this point, I am going to be trying lots of different things to arrive at a configuration that works for me. In a perfect world (lots of time and money) I would try both with two different setups connecting to the same sensor network and using the same code. Since my world is far from perfect, I thought I would ask. Hearing your experience with Hologram helps... functionality is also high on the list, and it seems like it is working good for you.

I've got an update. Today, I tried updating to the latest Sequans firmware and experienced strange issues. I tried updating to the latest firmware, downgrading to earlier release firmware, still problematic. Leaving the modem at 1.17, I unplugged and left it alone for a while. A few hours later, reconnected and tried running the exact code in this thread. So, at this point, the FiPy firmware and modem firmware are identical to when I first experienced these issues. This time, everything worked. I could successfully open sockets, send receive data, etc. This allows me to close out my evaluation of the FiPy modem. It seems there are still some major functional and reliability issues to be resolved from all parties, Monarch/Sequans, Pycom, and Verizon. I'll revisit this in the future and in the interim, drop in on the forums from time to time to see what current user experience has been.

@jmarcelino Thank you for all your help. We had provided Verizon the ICCID and IMEI 10 days ago at the time of provisioning request, but for some reason the IMEI was mixed up....and lost in translation. It was rectified over a troubleshooting ticket with Verizon.

I think Hologram.io is compatible but I have not yet tested that.