Closure on LE910 Problems

[Aleksander Morgado]

On Thu, Mar 24, 2016 at 6:54 PM, Randy Yates
<yates at digitalsignallabs.com> wrote:
> I am so thankful for all your help and quick responses for the past two
> months or so. To show my thanks I wanted to feed back the solutions we
> found to our problems back to the list.
>
> 1. USB Babble Interrupt (when wired ethernet was NOT connected) on LE910
>
> Turns out this was a hardware problem completely unrelated to the LE910
> or the processor. Our hardware engineer used a switching regulator (the
> TI TPS63060) and had configured it for power-save mode. That regulator
> goes in and out of power save mode when the load crosses about 100 mA.
> It also produces some horrendous (and documented) voltage transitions
> when transitioning. The ethernet PHY was drawing about 30 mA. In power
> up conditions our average non-ethernet load current was 90 mA - close
> enough to allow transient currents to transition the regulator in and
> out of power-save mode, so we had huge voltage spikes without the
> ethernet plugged in. Plugging in the ethernet got us up to 120 mA which
> was sufficient to avoid most of the transitions.
>
> 2. Connection Issues at Certain Cell Sites
>
> This problem was due to a confluence of using an unsupported Telit
> module, the LE910-NVG, which was intended for Verizon, and an AT&T SIM.
> It appears these sites were trying to force us into LTE mode and that
> combination only worked with GSM/GPRS and WCDMA. Issuing a AT_WS46=29
> fixed it since that limits the allowed connections to GSM/GPRS and
> WCDMA.
>
> Thanks again for everyone's help.

Thanks to you for coming back and reporting how you solved your
issues. That isn't usually the case, and it is much appreciated.

Cheers!

-- 
Aleksander
https://aleksander.es