Group Actions
Engagement
  • Author Author: shabaz
  • Date Created: Date Created
  • Views 4549 views
  • Likes 17 likes
  • Comments 28 comments
Related
Recommended

Smart Doorbell System - Part 10 - Complete Project Video

shabaz

For all the parts to this project, click here! Smart Doorbell System

 

Introduction

The Smart Doorbell System is a Bluetooth Low Energy (BLE) project to build a better doorbell! However, it could also have other uses - to open and close gates, to act as an emergency attendant button or to control or monitor all sorts of things. It can also be connected up to a Pi or BeagleBone, for extending reach and capabilities even further. For example, the Pi could log and timestamp activity, or activate a camera, or tunnel the information to another Pi or BeagleBone, and then pass the information onward over BLE to the far end, providing all the benefits of BLE combined with the compute horsepower and cloud connecting capability of the Pi.

 

All software and schematics and PCB files and bill-of-materials (BoM) are available.

 

For more information, click on the link above to explore all the blog posts, or (easier) watch the twenty-minute video below.

 

 

What Was Built?

Two modules were constructed; the bell push module, and the chime module. The can communicate directly to each other, or they can connect to a Linux system for extra features such as tunnelling alerts over the Internet, or for logging events or taking photos of who is at the door. With some effort the same system could be used for patient care, or for alerting people in a warehouse / shop floor.

image

 

The Bell Push Module runs from an AA sized battery and should run for four years (I intend to make code changes to improve that further - currently it consumes 45uA but around 20uA should be feasible). It uses a Texas Instruments chip, the CC2640R2FCC2640R2F. When connecting to the wireless Chime Module, it consumes around 12-13mA average, briefly until the communication is complete. More accurate tests still need to be done, but these values were promising.

image

 

The Chime Module was constructed from a custom PCB plugged on top of a ready-made Texas Instruments microcontroller board called the CC2640R2 LaunchPadCC2640R2 LaunchPad. The custom PCB contains digital audio capability and dual supply capability (external DC, or the attached rechargeable battery which gets charged up automatically). The Chime Module can be used plugged-in to the DC supply, or it can be detached and taken outdoors (e.g. into the garden). When used on battery power, the current consumption is about 8mA, which allows for several days of outdoor use with the 600mA battery that was used. The battery will charge itself so there should be no maintenance required.

image

 

Next Steps

Although the system functions, the code is very proof-of-concept and I'd like to take the time to improve it and add more features. I've already tested communications with the Pi (so that the Bell Push module connects to the Pi, and another Pi connects to the Chime Module) and the code functions nicely (see blog post 9 for the source code). So, with little effort, it should be possible to use this system over any distance.

Thanks for following the project!

  • For a revision 2 PCB for the bell push, these are the enclosures I'm considering. Any alternate options welcome!

    A 3D printed custom version would provide a lot of flexibility, but that can be a revision 3. For now, I wanted to use an off-the-shelf enclosure.

     

    The problem is that most small bell or alarm push boxes are designed for the small old-school A23 sized 12V battery. But this project requires around 3V, and I'd like to use large non-rechargeable cells for longer life, and they would be ideally be a single non-rechargeable Lithium AA sized 3V battery, or two 1.5V AA or two 1.5V AAA batteries. So, space is needed for a holder or clips for the single or double AA, or double AAA.

    Around 7mm depth is needed for the PCB (less is possible, but the microswitch needs a few more millimeters for the lever that will need to be implemented. The current PCB is about 77x28mm, and only a few millimetres thickness apart from the microswitch. The PCB could be reduced from 77mm to maybe 60mm if required, by expanding to 35mm width maybe.

     

    These are the current two candidates:

    Option 1:

    I think it looks quite ugly (I'd rather a more classical shape)  but it may be good enough to fit a couple of AAA batteries.

    image

    This is the other side of it, and I think it could be possible to fashion a plate to connect to the rear to attach to a wall, even if that was with foam tape to the plate screwed on the wall. The battery will last long enough that foam tape wouldn't need replacing often:

    image

    The enclosure would need spray-painting to blend in, i.e. spray it white or something.

     

    Option 2:

    This too looks ugly, but maybe less ugly. It is near-similar size but 10mm wider than Option 1. The cutout ellipse is great for an LED.

    image

    This enclosure would need spray-painting too, although it does come in a more acceptable color than black:

    image

    Option B enclosure is slightly unusual, in that it is 3-part, and so I believe has less room inside than Option 1, despite overall being 10mm wider:

    image

     

     

    Both options are not small, they are 120x45x22 and 120x55x22 outside dimensions, and I still need to order them to examine closely. I think the size is still reasonable for a smart bell push though. And leaves some flexibility for adding more features like a display or buzzer.

    Both options cost about $5 each.

  • in reply to shabaz

    I'm liking those Vandle proof buttons you can get illuminated ones for a couple of quid all stainless steel and everything.

    I used one for a similar project inspired by  your Giant Red Eye project!

  • in reply to stevesmythe

    Hi Steve,

     

    Thanks!

    A beep or sound is a good idea, it shouldn't need a lot of energy. Actually I'm hoping to re-design the PCB for the bell-push (so that it can fit in an enclosure, and clean up the design slightly). I can route a couple of pins to solder a piezo sounder. I'm still deciding on the enclosure.. need to order some candidates to see what they are like!

  • A very interesting and useful project for others to learn from. I am always impressed at your circuit board design skills.

     

    Would it use too much energy to have a small piezo sounder on the bell-push end? I'm thinking that the person pressing the bell needs some audible feedback that it has worked (I know there is an LED but the visitor might not appreciate what that means). In some houses (e.g. those with an enclosed porch), the visitor wouldn't be able to hear the sounder from outside the house. It would also act as a quick debugging check that the bell-push end battery has enough power left for a bluetooth transmission. My own wireless bell-push blew a capacitor and wasn't working and that wasn't obvious for a while, until people started hammering on the door!

  • in reply to aspork42

    Hi James,

     

    I think BoM could be reduced to $10 ballpark, especially if using (say) lower cost AAA cells, and eliminating the microswitch of course.

    One thing I'd suggest is examining the antenna closely - I thought I had got all measurements spot-on, but I've made a small error, there is a portion of the PCB antenna that looks like it has the incorrect dimension, I must have mistyped when I copied co-ordinates from the gerber file. So, although the system works, the range is just adequate until I get that sorted. Quite annoying that I missed it, but it is like a millimetre difference that was near-impossible to visually spot until I had the revision 1 board in my hand. I hope to have newer CAD files in the next few weeks. Also, it would be worth spending time seeking cost-efficient parts, because some of the inductors and capacitors are expensive, but cheaper alternatives will exist. There seems to be an industry-wide capacitor shortage recently which isn't helping : (

    There are also a few other tweaks I'd like to make to the board, although it functions today. One is that I really shouldn't have ground plane directly under the positive battery contact terminal! in case the solder resist ever accidentally scratches off : ( This was an oversight (or maybe I'm being over-paranoid about the likelihood), but ok for a rev 1 prototype where I can be extra careful with handling it all.

    Also, the tiny ceramic caps have a tendency to short when damaged, so I'd want to re-examine their positions, or perhaps specify flexible-termination caps, because an outdoor unit may end up with a flexed board when installing or over time with impacts. So it's an important thing to consider. A small resistor in series with the supply, or a polyswitch will also help to protect things in that circumstance.. there is space for such a resistor, currently a zero-ohm resistor is specified for that position. So in summary a few things are needed to improve long-term reliability.

    It would also be cool to see if there is a convenient sized enclosure out there (or 3D print something) so that the board is a good fit.