Noob needs help setting MVNO's APN for RM520N-GLAA to override SIM's APN

Haha, yeah, I am still confused, but luckily, two things are certain:

a) My current setup is stable, after buying these things in order: set of new antennae, higher-wattage power supply, rework.network board+enclosure. I did have to give up on 5G entirely, as the signal is so poor in my area that it was still falling out of 5G into 4G about once a minute.

b) The RM502N-GLAA itself is functional (for much of this time I was not able to rule out the possibility that the modem was bad and that the MCUZone board+enclosure was fine. This was ultimately the reason why I took a chance and bought the rework.network box even though its top rating is 10W max.

Let’s add a bit more confusion, shall we? After reading various hardware design guides for other 4G and 5G modems and their power characteristics, I am leaning towards the possibility that the failures were due to insufficient voltage, not insufficient current. My reasoning is this: none of the guides say that the modem will shut down itself with insufficient current, but all of them say the modem will shut down itself with insufficient voltage. I suspect that the instability I was experiencing was sharp, quick shutdowns followed by rapid reboots.

There’s no way to prove this without attaching a precise logging voltmeter capable of reading down to the thousandth of a volt at fractions of a second (with no or minimal averaging). Either that, or if the modem itself keeps logs internally, read the logs to see if it shut down due to insufficient voltage.

Then again, I did have to give up on 5G use with the rework.network 10W board. Perhaps 10W is just not enough to achieve a stable connection with the extremely poor 5G signal in my area. Put another way, perhaps the rework.network 10W box is in fact providing at most 10W, and I am indeed exceeding its power capabilities, but the failure is a graceful drop out of 5G into 4G, unlike with the MCUZone box that had very disgraceful drops out of 5G.

I wonder: would a board capable of supplying the full rated 18.5W give me stable 5G? I even went so far as to buy a SECOND rework.network box that claims to supply up to 20W to the modem. With this, I could transfer the RM520N-GLAA into it and find out whether 10W is not enough to attain stable 5G. I have the second rework.network box in hand but don’t have the correct power supply for it (it is 5V, 4A). Finding a high-quality, trustworthy 5V, 4A power supply is surprisingly difficult. The ones I am looking at are typically used to power Single Board Computers like the ODROID XU4.

So, I’m not quite done with my experimentation.
I am also considering buying a 4G CAT12 modem for $30 or less and putting it into the MCUZone unit to see if it works. If it does, then I will have proven my hypothesis to my satisfaction: Based on the number of capacitors in the power circuit on the MCUZone board, I suspect they used a reference design for 4G modems and didn’t update the design for 5G modems.

Thanks for the elaborate reply.

With the currently used provider, I’m for sure not able to fully utilize what the RM551-GL is able to do, but at least at about 200~250 MBps down and 30~40 Mbps up, the total power draw of the 5G2PHY stays below 6W (rather 5W only), so there is still some headroom left for those 10W for the module alone.

Guess you mean that one. What I don’t understand though, is that this manufacturer lists the RM551 in conjunction with the 5G2PHY as being compatible but not for the one offering 20W, which doesn’t make any sense to me, given that the RM551 is more powerful performance-wise (unless more efficient, who knows?).

As for the concrete power supply demands in terms of voltage and amperage, ideally Quectel would contribute here, knowing it best.

Yes, that is the exact board+enclosure that I bought. Other users on other forums confirm that it works great with the RM520N. I have not seen any confirmation that the RM551-GL works in it, but I would expect it to since the mfg. refers to the RM500 series in the description.

Re: typical power draw
It’s pretty clear from the hardware design guides from Fibocom, Quectel, and Sierra Wireless that under “good” (my adjective, not theirs) conditions the actual power draw is very low. It appears to be a selling point. I wasn’t able to figure out exactly what “good” means in the real world from the specs provided in the guides. We can probably assume “good” means: a) temperature of unit is in the optimal range between 30 deg C and 40 deg C, b) all signal stats are in their optimal ranges.

Put another way, it looked to me like the modem mfgs were largely bragging about having very low power consumption in the best case. Almost nothing concrete was said about the worst case, with a super-weak signal in a high-interference area or situations with extremely warm or extremely cold temperatures (if I do end up trying out vanlife, these would matter if driving out in the mountains during ski season or in the desert).

Re: ideally Quectel would contribute here

I totally agree and wish that all the mfgs would be very specific about power consumption at the extremes, but I doubt any will speak up about the limitations of their modems. They likely are not testing performance at the extremes at all and are just testing under optimal lab conditions.