Forum - Recent Threads

Converting The SENSECAP to standard Lora Wan Gateway

meera_hussien — Wed, 01 Apr 2026 02:47:29 GMT

I have a SenseCAP gateway that I used previously for Helium, but I am not using it anymore. I am planning to convert it into a standard LoRaWAN gateway. I searched online and found that it seems possible. Just wanted to ask if any members here have tried this before, and was it successful?

The link for this product is below

https://www.sensecapmx.com/docs/sensecap-m1/overview/

Industrial Internet of Things (IIoT) solutions

athang — Wed, 18 Feb 2026 08:41:31 GMT

I am here looking for IIoT solution for organization. I also searched this on many different website and found many solutions. We are bit confused to which one to choose. One of my friend suggest me this platform, and i am hoping i will get the solution from here. So can anyone please suggest the good solution platform?

RE: Industrial Internet of Things (IIoT) solutions

BigG — Mon, 23 Feb 2026 15:59:00 GMT

I too struggle with acronym's. It's a bit like suncream... You need to apply liberally when out there in the field... or at trade shows. Otherwise you get burnt.

For me, "Industrial" implies environmental conditions. So temperature rating is a big thing (to -40C to 105C etc). Other aspects that apply are voltages (typically can accept 24V) and then immunity to noise etc.

Arduino Opta is a great case study of how you take a consumer led IoT product to switch it over to industrial: https://www.arduino.cc/pro/hardware-arduino-opta/

RE: Industrial Internet of Things (IIoT) solutions

colporteur — Mon, 23 Feb 2026 04:04:00 GMT

You got all that out of I am here looking for IIoT solution for organization.". Wow!

RE: Industrial Internet of Things (IIoT) solutions

shabaz — Sun, 22 Feb 2026 23:16:48 GMT

Hi Sean,

I suspect IIoT is an informal acronym than the full phrase Industrial IoT. The trade shows mentioned are industry/business focused so would have been ideal for the OP as a first step, if they are serious about moving forward.

The industrial part trickles down to requirements such as not having access to the devices for upgrades (I.e. remote upgrades become more important than USB update for a consumer item), needing to have security usually (and proving it), ability to not need to manually configure every single device (because the particular industry/enterprise may be purchasing hundreds, versus a consumer who doesn't mind a slower onboarding per device), maybe far more expensive cloud data storage (might need to mine data later, or might need to process more in near-real-time), more power options perhaps, or run for the life of the product without needing to change batteries, and a bit more ruggedness perhaps, run in colder/hotter temperatures etc. There are all manner of sensors and actuators etc., but it's not a solution, just a bunch of components or at best systems, until the work is done to understand the specific business and technical requirements, and design a solution that provides whatever processing, storage, presentation, control, reports etc that are needed to meet those requirements (i.e. to me it's illogical to ask the original question 'which IoT solution to choose', because no IoT solution already exists for the OP; they have to do the hard work (or hire someone) to design that, and then implement that.

RE: Industrial Internet of Things (IIoT) solutions

colporteur — Sun, 22 Feb 2026 20:57:12 GMT

While IoT refers to consumer technology, IIoT specifically targets industrial applications. I didn't know that.

You didn't inquire about the industry and the question didn't provide that detail. It involves interconnected devices, such as sensors, actuators, and advanced analytics platforms. Predictive maintenance, which prevents equipment failure, as well as supply chain optimization and increased operational efficiency are some applications.

Does IIot match an industry or is it just a collection of data generating sensors and their collectors?

RE: Industrial Internet of Things (IIoT) solutions

shabaz — Wed, 18 Feb 2026 16:43:01 GMT

Hi,

(Long answer): Technically what you'll find are often components or bits of systems or SaaS, it's not a solution. The solution cannot exist until you (or someone on your behalf) has designed it, and then developed and deployed it, ideally to meet (amongst other things) your business needs.

If initial needs are very basic (e.g. for a simple initial trial) then you could ask any IoT device manufacturer (for say one of the types of things you wish to monitor or control) to set you up a simple demo, using whatever platform they use (might be their own, or one they are familiar with).

Or, even more basic, you could have a play with any IoT platform, most have demonstration software or products, e.g. Azure, AWS, even Arduino I think. It doesn't matter too much since the solution has not been designed at this point, and you're rather using the trial as part of your research to decide what you want in your actual solution.

If you really want to move forward and design and implement a solution, then you could hire someone to help you with this, but often you will also need at least a few people in your firm (particularly developers) to dedicate time to the project too. If you really do want to hire someone for a period of time to design the solution for you, if you place the job requirement on your work website or normal site you use for recruitment, or maybe provide the link to it or detail here, I am sure someone may reply (I might depending on the circumstances).

However, if you're not at that stage yet, or wish to see what others have done, and if you're completely new to this, it's best to visit a trade show (e.g. there was a free-to-visit large IoT show in London just a week or two ago - and there's a less focused, more generic but larger show in London Excel in two weeks time from now) where you will find hundreds of firms showing their components or SaaS and so on. They will all be able to tell you how their offerings were used in solutions, and it's easier then to see which ones align with your needs, and they may even point you to individuals and firms that they worked with to help build those solutions for the end customer. I'd strongly recommend it, because it's a very quick way to start seeing some of the possibilities, and roughly gauging what you will want or need. I don't know your location, but if you keep an eye out I am sure there will be similar shows at least in your region (and sometimes it is worth visiting trade shows for similar-ish industries too, to see what others may be doing if they have similar requirements.

Has anyone built a truly low-power camera IoT device on an off-the-shelf SoM?

inxnik — Tue, 18 Nov 2025 10:33:26 GMT

I’m trying to achieve Amazon Ring–level battery life — several months on a 5000 mAh battery — using non-custom SoMs. The device should stay almost at zero power between photo sessions.

So far, all tested platforms (ESP32-S3, ESP32-P4, AmebaPro2 RTL8735BDM, Luckfox RV1103) show deep-sleep consumption ≥ 70 mA, far above the advertised values, making them unsuitable.

We also tested a two-chip architecture (small MCU like ATtiny85 waking the main SoM), but the main issue is boot-to-first-photo time. It must be <1 second, which is hard with Linux-based SoMs — yet only Linux SoMs support proper MIPI sensors like IMX219/IMX708.

Questions:

– Has anyone achieved a low-power camera system (photo only) on an existing SoM, with real deep sleep or fast wake-up?

– Is designing a custom board around a SoC the only realistic solution?

– Or can something like Radxa CM3 + a highly optimized Linux stack (Buildroot, custom suspend modes) reach <1 s wake and low idle power?

– Has anyone successfully built a battery-powered MIPI-camera device with either deep sleep or a two-chip wake architecture?

RE: Has anyone built a truly low-power camera IoT device on an off-the-shelf SoM?

lichurbagan — Sat, 27 Dec 2025 15:47:12 GMT

What camera did you use? Was it the ESP32-CAM or something else?

RE: Has anyone built a truly low-power camera IoT device on an off-the-shelf SoM?

Workshopshed — Sat, 27 Dec 2025 13:26:07 GMT

I've used a simple transistor switch to control the power to things like servos and sensors when trying get the best battery life from a project.

Also used a lower power sensor to wake the board and camera before.

RE: Has anyone built a truly low-power camera IoT device on an off-the-shelf SoM?

lichurbagan — Sat, 27 Dec 2025 10:55:15 GMT

If you need months of battery life, MIPI camera and <1 s perceived response then custom hardware is unavoidable, even if you reuse an existing SoC.

RE: Has anyone built a truly low-power camera IoT device on an off-the-shelf SoM?

BigG — Tue, 18 Nov 2025 11:55:26 GMT

Why are you focusing just on the SoM? There are range of image sensors with MIPI interface out there, and I'm sure some are more power hungry than others when powering on/off and capturing a frame in different lighting conditions.

RE: Has anyone built a truly low-power camera IoT device on an off-the-shelf SoM?

obones — Tue, 18 Nov 2025 10:39:43 GMT

I'm surprised at the levels you see for ESP32.
Are you sure you disabled all peripherals and used deep sleep?

Here is what I do:

    esp_sleep_enable_ext0_wakeup(GPIO_NUM_9, 0);
    epd_poweroff_all();

    esp_deep_sleep_start();  // Sleep until we have received a GPIO interrupt

The epd_poweroff_all() call disables entirely the epaper display that I'm using. Without it, I don't reach the sub mA consumption. Are you sure you did not leave the camera powered on?

IoT

juanjm7 — Mon, 17 Nov 2025 08:36:57 GMT

Hello, my name is Juan Jacinto Montenegro, and I am an electronic engineer at Rother.
We have contacted STMicroelectronics to obtain recommendations regarding their products for our IoT project. They have suggested that we contact their distributors.
Our project focuses on the automotive sector, both in the civil and military fields. For this project, we need to meet a series of specifications:

-UART or USB2.0 Full-speed communication interfaces to communicate with our microcontroller, which will act as a host controlling the IoT modules.

-Wi-Fi and cellular (LTE 4G/5G) connection for sending data via HTTP and MQTT to a remote server.

-Sending SMS messages and voice calls

-Support for GNSS geolocation.

-Interface for connecting a MicroSD card and storing local data.

I would like to know if you have any IoT modules or boards that meet these requirements. Initial testing would serve to validate the results obtained and, if these were satisfactory, we would proceed to purchase the different products tested (cellular module, WI-FI module...) to design our own board.

Help identifying SoC used in Amazon Ring (2024, battery version) + advice on low-power IoT camera platform

inxnik — Fri, 24 Oct 2025 16:25:32 GMT

Dear engineers, hello everyone — I’d be very grateful for your help.

Is there any way to identify which SoC is used in the Amazon Ring (battery-powered video doorbell, 2024 version) based on the following chip markings?

АС1ООЗВЗЕ-FВ

UVHA 17005 .000

J-2436 20

Research through neural tools and tech insider databases didn’t produce results. It’s known only that this is not the Ambarella CV25A75 SoC used in the wired 2021 versions.

—

And one more question — I’d also appreciate your advice:

Which SoC/SOM would you recommend for building a low-cost, low-power IoT camera device, similar to a battery-powered video doorbell?

Priorities:

Power efficiency
MIPI CSI-2 interface for Sony IMX sensors
Wi-Fi + Bluetooth
No need for powerful video processing

Important: an available ecosystem (chips and/or modules, SDK, documentation), not something too rare. ESP32 (Espressif) isn’t powerful enough, while Ambarella H22 SoM feels like overkill — looking for something in between. Target pilot batch: ~1,000 units.

RE: Help identifying SoC used in Amazon Ring (2024, battery version) + advice on low-power IoT camera platform

dougw — Sat, 25 Oct 2025 00:01:36 GMT

Google Lens had some interesting guesses....

RE: Help identifying SoC used in Amazon Ring (2024, battery version) + advice on low-power IoT camera platform

shabaz — Fri, 24 Oct 2025 19:51:45 GMT

If it were me, I'd probably look at K230D (or one of the other Kendryte parts), since there are ready-made small boards from various sources, about the same price as a Pi for a board in single qty, but probably less power consumption (guessing), and relatively easy to use. Obviously a final design could use the chip on a custom PCB instead of a ready-made board, but I'd be tempted to use an existing board if possible.

RE: Help identifying SoC used in Amazon Ring (2024, battery version) + advice on low-power IoT camera platform

cstanton — Fri, 24 Oct 2025 17:17:30 GMT

[quote userid="516502" url="~/technologies/internet-of-things/f/forum/56328/help-identifying-soc-used-in-amazon-ring-2024-battery-version-advice-on-low-power-iot-camera-platform"]

Is there any way to identify which SoC is used in the Amazon Ring (battery-powered video doorbell, 2024 version) based on the following chip markings?

[/quote]

Probably a proprietary or custom ASIC / FPGA / SoC

RangePi

nilssonm — Fri, 12 Sep 2025 07:30:11 GMT

Hi everyone,

The board seems to be a nice little board for LoRa testing. I have purchased one but having big problems to program it. I have followed the instructions and pressed the BOOTSEL button when plugging it in and I only get a serial device and no mass storage.

Is there someone out there that is able to help me and guide me in the right way. I have tried on my Apple Silicon Mac with macOS and with Windows. I have also tried on an Intel Mac and on a Windows PC. I have tried with different cables and ports. But no luck in any case. I have also tries with another RangePi with same results.

Is it possible to re-flash the device in someway or can I do something else to be able to connect to it?

RE: RangePi

nilssonm — Wed, 24 Sep 2025 11:21:55 GMT

I have now done some more research around the product RangePi and come to one solution to activate the boot mode with the presented mass storage device. I have tried some different Python scripts since I am capable of communicating with the board using Thonny and found the possibility to restart the board in boot-sel mode using the Python function call ”machine.bootloader()”, which restarts the board and activates the mass storage device. I have also tried some of the other scripts in your repo among them the ones using the buttons with no luck. It seems that none of the buttons are working because I am not getting any response from them. My thought is that maybe the BOOT-SEL button also is non-responsive which could explain my other problem, getting the board to start-up in the mode to be able to program it. If the BOOT-SEL button is unresponsive makes me uncertain if I want to re-flash the device and maybe not able to reactivate the mass storage anymore.

Are there anyone knowing if I can get a schematic for the board that I could maybe get my hands on to be able to make some more research and understand why the buttons are non responsive, or if there is any other ideas for me to try in the MicroPython environment?

RE: RangePi

nilssonm — Mon, 15 Sep 2025 05:01:34 GMT

Yes, I think so too. I have now mailed the manufacturer a question and hope they give me an answer for this. It seems that the firmware maybe have the mass storage option disabled or something as you are talking about. Maybe I'll check out other products fort this purpose.

RE: RangePi

shabaz — Fri, 12 Sep 2025 18:32:57 GMT

Maybe worth asking the manufacturer; they may have made a mistake in the documentation, or possibly disabled the boot capability if users were accidentally pressing it when inserting it in. No idea, but the manufacturer will know for sure. The procedure for a typical RP2040 board has already been mentioned. Your board is already quite weird (strange combination, USB-A port so that it sticks out, plus a RP2040 which is technically unnecessary when a USB-UART could have been used), so who knows what other unusual things they did in the circuitry.

You could simply buy a Pi Pico board for a few $, it's cheaper, there is a lot more information on it, and the Bootloader button works on it : )

If it's the LoRa you're interested in, the LoRa module (E22-900T22S) is cheap and easy to obtain, simple to wire up and there's plenty of documentation for it.

RE: RangePi

nilssonm — Fri, 12 Sep 2025 17:41:59 GMT

Yes I have tried using Thonny now and after re-installing it I managed to connect and use the RangePi with python. That’s nice, but I really would like to be able to re-flash the device and maybe use it with C / C++. I have been trying to follow that process when plugging it in, but still no mass storage. 😩

RE: RangePi

shabaz — Fri, 12 Sep 2025 15:09:18 GMT

Hi,

Possibly, you may not need to perform this process, if it came pre-installed.

If you're trying to upgrade it, be aware that the firmeware on GitHub is several years old anyway.

If, when you plug it in, you hear the short tone (or whatever sound or alert your PC usually makes) that a USB device has been plugged in, then there's a chance it is functioning and you might simply be able to proceed directly to using Thonny to access it. You won't see any Mass Storage Device in this case.

That being said, if the boot button was held down while plugging in, and then released, then the mass storage device ought to appear so that you can upgrade it. There's a chance the button could be faulty (some of the cheap ones are fragile and damage if pressed off-center). Although the instructions state to hold down the BOOT button all the time, that's unusual for a RP2040 board. There's no circuit that I could find to check. The typical method is to first hold down the button, then insert into the PC, then release the button. The 1-minute video here would apply since your board also contains an RP2040: https://www.youtube.com/watch?v=os4mv_8jWfU

Anyway, as mentioned, it may be worth just attempting to use Thonny first.

RE: RangePi

nilssonm — Fri, 12 Sep 2025 15:09:01 GMT

Yes, I'm pressing the button all the time from unconnected to it is connected and even wait some seconds before releasing it.