If not from scratch atleast how to convert a normal to smart one... I know it's a very very big topic but sometimes things are easy if come from Top to Bottom. Like you are buying a car and slowly learn about it.

 

Define what it means to make a bulb “smart”, and what is a “normal” one?

 

"Alexa turn off the light" is smart.
"Alice. Please turn off the light dear"
If you get a voice feedback like "Honey. I am busy making dinner" is normal

 

Do you have $8?

Smart bulb

 

'Retired' Researching Professional Here. Lots of spare time

 

Easily done with an ESP8266 or ESP32, which are probably used in the majority of IoT devices. There are many examples on the web.

 

The electronics hardware to add any sort of "smart" function to an LED light bulb is rather complex. Thus most of them use an ASIC created for the application. It requires all of the circuitry to run code, PLUS the I/O to control multiple colors of LEDs at different levels.

 

You can do it. You will learn a great deal and it will cost you a great deal more than buying one.

 

There's a kick in being a creative and to be a creator... Agree?

 

Ummm. No. You gentlemen are somewhat out of date. As I stated above, most IoT devices are based on ESP8266 & ESP32, which are programmable using Arduino code.

Building a lightbulb around one of these controllers.would require some engineering but both the circuit and code are well developed.

Developing an in-line module is entirely practical, but economically it doesn't make a lot of sense.

I believe Tuya, a huge IoT supplier, will make their development tools available so your device can be compatible with their system without many hoops to jump through.

 

Ummm. No. You gentlemen are somewhat out of date........

I don't understand your point. Are you saying it can be done, or can't? Or shouldn't? It seems you agree with us that it is possible, instructive, but less than economic.

 

It can easily be done as an external module at low cost. A good learning experience. But such modules are plentiful and cheap – cheaper than you could duplicate one.

 

Haha - no one is suggesting starting by boiling down silicon and making pn junctions, all that is long gone*.

It would be an exploration of IoT architecture using appropriate modules.

* At school aged 15 I spent many months putting together FET junctions and other semiconductors inspired by Amateur Scientist June 1970. I don't recall with what success tbh but I got all sorts of cool chemicals and stuff from the goverment radiological labs.

 

Back in 1971 I took a 300 level class in "Physical Electronics", which covered semiconductor physics in great detail. We learned about band gaps and energy levels and how they related to semiconductor functioning and failure.
But at no time was there any mention of ever considering producing any of the devices.. The closest to that was learning about die-bonding and connections to the actual silicon. That allowed us to understand the limitations at the time. There is a world of difference between knowing and doing, which is the skill set that is required.

 

Ummm. No. You gentlemen are somewhat out of date. As I stated above, most IoT devices are based on ESP8266 & ESP32, which are programmable using Arduino code.
...

Ha ha, out of date, Arduino code. True, maybe at the basic functionality level a few years ago but the commercial IoT world has moved on with Matter and Thread based systems that actually have security.
https://www.espressif.com/en/solutions/device-connectivity/esp-matter-solution

 

It is not likely that any aduino package could be included in any "smart light bulb" assembly. The size, power supply requirements, Output control, and temperature withstand requirements join to make creating a controller into a serious design project.
So while an I.T. Professional could certainly create code that could provide the functions, and undoubtedly create a reasonable block diagram, producing a workable package is rather unlikely.

 

It is not likely that any aduino package could be included in any "smart light bulb" assembly. The size, power supply requirements, Output control, and temperature withstand requirements join to make creating a controller into a serious design project.
So while an I.T. Professional could certainly create code that could provide the functions, and undoubtedly create a reasonable block diagram, producing a workable package is rather unlikely.

If he really wants the works in a standard sized light bulb, buying one is the only option likely to succeed.

But I don’t think he is saying that.

 

“Smart Bulbs” can mean a lot of things. The Home Automation and IoT landscape is far from simple.

First, I don’t see any benefit in making a smart bulb. I would start with a smart outlet or switch. The space constraints and high temperatures in a light bulb are needless complications for a learing exercise, and there is not practical advantage to be had.

Second, there is a great deal to learn about the variety of IoT ecosystems before you can select the one you want to participate in. Things definitely seem to be converging on Thread and Matter, and also moving away from Wi-Fi in favor or Zigbee, and a dash of BLE as well for some functions.

But you can still do a lot with the lowly ESP-12x for a couple of bucks, a MOSFET, and a relay. You can create compatible hardware that interfaces with a variety of existing systems. But it will have no growth potential.

If someone really wants to learn this area, I would suggest starting with Home Assistant and cheap, compatible devices like Sonoff or (somewhat less cheap but excellent) Shelly. And, for voice assistant, you can use Alexa, or Siri and HomeKit, or Google Home—or Home Assistant’s built in option.

But, you need to survey the landscape and choose the direction with some foresight because you could do an awful lot of “learning” that won’t be portable, and won’t age well—and doesn’t teach the key principles, just particular implementations.

Good luck.

 

“Smart Bulbs” can mean a lot of things. The Home Automation and IoT landscape is far from simple.

First, I don’t see any benefit in making a smart bulb. I would start with a smart outlet or switch. The space constraints and high temperatures in a light bulb are needless complications for a learing exercise, and there is not practical advantage to be had.

Second, there is a great deal to learn about the variety of IoT ecosystems before you can select the one you want to participate in. Things definitely seem to be converging on Thread and Matter, and also moving away from Wi-Fi in favor or Zigbee, and a dash of BLE as well for some functions.

But you can still do a lot with the lowly ESP-12x for a couple of bucks, a MOSFET, and a relay. You can create compatible hardware that interfaces with a variety of existing systems. But it will have no growth potential.

If someone really wants to learn this area, I would suggest starting with Home Assistant and cheap, compatible devices like Sonoff or (somewhat less cheap but excellent) Shelly. And, for voice assistant, you can use Alexa, or Siri and HomeKit, or Google Home—or Home Assistant’s built in option.

But, you need to survey the landscape and choose the direction with some foresight because you could do an awful lot of “learning” that won’t be portable, and won’t age well—and doesn’t teach the key principles, just particular implementations.

Good luck.

Solved on this crosspost:

https://forums.tomshardware.com/thr...nal-to-make-bulbs-smart.3842524/post-23244571

 

Solved on this crosspost:

https://forums.tomshardware.com/thr...nal-to-make-bulbs-smart.3842524/post-23244571

That‘s not any more of a solution than what was discussed here, nor is it the only or necessarily the best development board for the purpose. I like the Seeed Xiao, I use it—but there is nothing special about it.

You are missing the point, but good luck anyway and do I hope you learn something and have fun.