Smart Home Carbon Monoxide (CO) Detector
This project is similar to our smart home smoke sensor, where we have taken a standard device and connected to our smart home, to enable it to become much more intelligent.
For this project we are using a Kidde 7CO Carbon Monoxide Alarm. We have chosen this device because we like its looks, it has a 10-year warranty and is reasonably priced at around £15. It is also nice and compact (115mm × 70mm × 35mm) and comes with a nice wall mounting plate. It also comes with batteries.
There are four clips holding the front and back of the case together. To open it you first need to cut through the sticker wiht the manufacturing date on the side. This will invalidate the warranty but we are not worried about warranties on such cheap devices. There is a pin that holds three battery springs in place and we have removed the pin and these spring clips as we no longer need them.
The back battery cover is also the mounting plate.
Front view of the circuit board, clearly showing the battery connection points. The LED on the left is the red alarm LED and we are taking our signal feed from the anode of this LED.
The rear view of the circuit board. We are going to solder a very fine wire to the LED anode on the back of the circuit board. We are also soldering fine wires to the battery clip points.
So we have three fine wires leading off of the circuit board: red = +5V, black = ground, blue = signal. These will then feed into a tiny circuit board with the regulator and transistor to convert the LED signal to a 12V dc signal. This circuit board sits in the battery compartment, so that the device can be mounted flush onto a surface and the wires will be out of sight.
This device is powered by three AA (1.5V) batteries and three new alkaline AA batteries supply up to 4.9V. Our testing showed that this device is happy with any voltage over 4.0V dc. Below this voltage it will start to signal that the 'battery' is low. The actual device will keep working with voltages as low as 2.25V though. Below this voltage it is not happy and sounds continuously.
This is our circuit board with a LM317H adjustable voltage regulator (we happened to have one to hand). It is much smaller than an AA battery so that it easily fits inside the batttery compartment. It is designed to provide a 4.65V dc supply voltage from the the 12V dc feed from our input board. This will ensure that the device never thinks the 'batteries' are running flat.
Our optically isolated input board is designed to accept a 12V dc signal back into the opto-isolator input stage, we use the feed to the LED on the front of the device to drive the input stage of a 4N26 opto-isolator. This means we have two opto-isolators in use for this device but they are very cheap and it keeps the interfaces simple and consistent between numerous devices.
This is the circuit board with the four wires soldered to it:
The blue wire goes to the anode of the red LED. Our testing shows that this is fixed at the supply voltage, which tells us that the LED is lit by the circuit pulling the cathode down to ground (via a resistor). The green wire is connected to the cathode of the red LED. These are used to wire the input stage of an opto-isolator in parallel with the existing red LED. A resistor is also used to limit the current into the opto-isolator. The red wire is the +4.5V dc feed and an the black wire is ground.
The whole point about this alarm is that once installed, we don't have to worry about it. It the alarm goes off, the sounder built in to the alarm will make some noise and our smart home will get notified and take appropriate action.
The operation of the LEDs on this alarm also mean that we will know when it has failed or reached the end of its life. If it malfunctions, the red LED will light (sending a signal to our smart home) and the alarm emits a continuous tone. When it reaches the end of its life, the alarm chirps twice every 30 seconds and the red LED flashes twice (sending a signal to our smart home).
As you can probably tell, we like to wire things up to our smart home. This gives us the ability to use cheap 'off the shelf' sensors and devices that can quickly and easily be replaced if need be.
By using a wired solution we also eliminate the need to change batteries, whilst enabling the alarm to work through mains power outages.
A wired interface also ensures that the interface to our smart home is extremely reliable when compared to wireless technologies. This is important when we are talking about safety devices like this. It is also not subject to interference by other devices and wireless technologies.
A simple wired interface like this is also not going to be made redundant any time soon and superceded by new variants and updates (unlike Wi-Fi, Z-Wave, ZigBee, etc.). It is very much future-proof when compared to wireless technologies.
- Our feature: Why we won't be buying the Nest Protect (October 2013).