Overview
The intelligent flashlight project is an ongoing microcontroller-based flashlight. Unlike more basic flashlights; the intelligent flashlight has multiple colors and modes. One key feature is it's ability to preserve the batteries by turning itself off after a while. The goal of the flashlight project was to provide a simple, easy-to-use flashlight for around the house at night.
While the original flashlights were quite clunky; one of the most recent versions created in 2021 (pictured right), is small and compact. It uses a rechargeable battery system which has largely negated the need to for an automatic shutoff.
While the original flashlights were quite clunky; one of the most recent versions created in 2021 (pictured right), is small and compact. It uses a rechargeable battery system which has largely negated the need to for an automatic shutoff.
Origination
Flashlights have always been a mild interest (obsession) of mine. My first introduction into electronic circuits involved using batteries and aluminum foil to power a small incandescent light bulb. Later I used that technique to build some of my very first electronic devices (a Phaser Rifle replica from Star Trek First Contact with a working light) .
Primitive Development
During my college years I wanted to play around with some of the concepts I learned so I created a multi-color flashlight; I called it the "police light" because it could flash red and blue in various patterns. This early flashlight used JK flip-flops to store control data about which lights flashed and when. It used a 555 timer IC to time the LED flash. You could input your pattern using DIP switches and a special clocking system.
Microcontroller Integration
After college I played around extensively with the MSP430 microcontroller; it became my bread and butter for a number of years. My first prototypes (PVC lights pictured above); used an MSP430 and simple BJT circuits to drive LED. This was the first time I used a microcontroller for something like this. I developed the key feature; automatic shutoff because my kids would leave the flashlights on and forget to turn them off. This would kill the batteries, requiring me to open the case (a delicate operation) to change the batteries.
Primitive Development
During my college years I wanted to play around with some of the concepts I learned so I created a multi-color flashlight; I called it the "police light" because it could flash red and blue in various patterns. This early flashlight used JK flip-flops to store control data about which lights flashed and when. It used a 555 timer IC to time the LED flash. You could input your pattern using DIP switches and a special clocking system.
Microcontroller Integration
After college I played around extensively with the MSP430 microcontroller; it became my bread and butter for a number of years. My first prototypes (PVC lights pictured above); used an MSP430 and simple BJT circuits to drive LED. This was the first time I used a microcontroller for something like this. I developed the key feature; automatic shutoff because my kids would leave the flashlights on and forget to turn them off. This would kill the batteries, requiring me to open the case (a delicate operation) to change the batteries.
A New Design
PVC pipes proved to be...cumbersome. Replacing the batteries was also problematic as it required the rear cap on the PVC tube to be pried off and the entire circuit pulled out so the batteries could be accessed. The solution was to use a structure of wood and carboard covered by paper. These new flashlights had replaceable batteries on the outside allowing them to be easily changed.
To the left you can see what the enclosures looked like before they were finished. Wooden sticks were used to reinforce the structure. What you can't see here are the 4.5V 3x AAA battery holders underneath. The flashlights would sit on their batteries.
On the outside of the flashlight was cardstock. I used plain paper for the first few models but cardstock proved to be more durable.
This wasn't a water proof design but the flashlights' primary environment was indoors so that wasn't a problem. What I found though is having that much weight on the outside of the box did made the flashlights vulnerable to damage from being dropped (i.e. thrown by my kids)
On the outside of the flashlight was cardstock. I used plain paper for the first few models but cardstock proved to be more durable.
This wasn't a water proof design but the flashlights' primary environment was indoors so that wasn't a problem. What I found though is having that much weight on the outside of the box did made the flashlights vulnerable to damage from being dropped (i.e. thrown by my kids)
Rechargeable
Originally I thought that a rechargeable battery system would be too complex to implement. After all I am just a software engineer. After some research I found that a rechargeable solution required just a few inexpensive components.
What I found is that one 3.7V lithium cell (or battery) could be "bucked" up to 5V with a simple buck converter/booster. This device also acted as a charging circuit and fuel gauge.
Moving to a rechargeable system meant that the flashlights would last longer than their AAA driven predecessors. It also meant the battery could be enclosed inside the box; making a sleek, well protected flashlight.
What I found is that one 3.7V lithium cell (or battery) could be "bucked" up to 5V with a simple buck converter/booster. This device also acted as a charging circuit and fuel gauge.
Moving to a rechargeable system meant that the flashlights would last longer than their AAA driven predecessors. It also meant the battery could be enclosed inside the box; making a sleek, well protected flashlight.
One Step Too Far
As of 2022 my latest flashlights (pictured right) have been less than successful. I wanted to eliminate the slide switch from the system. I attempted to incorporate a push-on, push-off switch using a BJT-based OR gate. Unfortunately this circuit didn't work well with the charging board which required a regular pulse to stay active.
In the end I had to get DPDT momentary contact switches (which are rare with not many good options) and use them to control the power. The lights required the button to be clicked twice for on and twice for off which I found to be problematic.
In the end I had to get DPDT momentary contact switches (which are rare with not many good options) and use them to control the power. The lights required the button to be clicked twice for on and twice for off which I found to be problematic.
For these lights I enclosed them using the same wooden frame inside of carboard. This time I used brightly colored cardstock to help them show up better at night. Although I found the previous model of flashlight to be more user-friendly.
Future
I don't know how many flashlights I have in my house right now. Its probably not 100....probably. I will continue to improve upon my flashlights. Each iteration has new features and allow me to explore new ideas. Its amazing how something so simple can be so complex and go in so many different directions.
I am 100% going to create a new batch of flashlights in 2022. Probably during the summer. The new design will incorporate a new charger/booster that doesn't have a fuel gauge. It will also likely include some new buttons and switches.
As far as 2023 goes; who knows. The phrase "running out of flashlights" is a joke in my household so not building flashlights for a year or two won't hurt me especially since I have many other projects on my plate.
I am 100% going to create a new batch of flashlights in 2022. Probably during the summer. The new design will incorporate a new charger/booster that doesn't have a fuel gauge. It will also likely include some new buttons and switches.
As far as 2023 goes; who knows. The phrase "running out of flashlights" is a joke in my household so not building flashlights for a year or two won't hurt me especially since I have many other projects on my plate.