Energy harvesting to build a Carbon Monoxide Detector with no battery


There are a few devices that work tirelessly to protect our lives. We’re talking about smoke detectors and carbon monoxide detectors. Increasingly these either need to be hardwired into the home, or have a sealed battery which is good for ten years (in the case of smoke detectors). [Gelmi] recently had to change the battery in his Carbon Monoxide detector — which happens very rarely — and he it got him to thinking. If the batteries need to be changed so rarely, how hard would it be to harvest energy to power the device?

Our first thought was that he’d use inductance like those spy birds which perch on power lines. But instead he went for the heat lost from using the hot water spigot. Above you can see his test rig which attached a Peltier device to the faucet in his bathroom. Whenever you turn on the hot water the faucet also heats up. The differential between faucet temperature and ambient room temperature generates a small amount of power. This is a suitable source, but only if he could also cut the amount of power needed by the detector. This adventure takes him down the rabbit hole, learning about how the sensors work and designing for reliability at the lowest consumption level possible.

The faucet application might seem peculiar. But if you use a natural gas water heater you want a carbon monoxide detector near it. Attach the Peltier to the outflow and every time any hot water tap in the house is opened your system will get a bit of a recharge.


  1. chango says:

    This is yak shaving at its finest.

  2. Sovereign says:

    Why not just use the heat from the water heater? especially since its a gas heater, the exhaust vent should get pretty hot

  3. hboy007 says:

    when you can have shower heads with generators that light up LEDs, why not use a turbine for this kind of harvesting? Maxwell Supercaps will also help with the job (yes, those 310F ones ;-) )

  4. Hirudinea says:

    Makes me wonder why gas appliances aren’t mandated to have CO detectors built into them by law?

    • steven-x says:

      When I upgraded my power (from 60 amp service to 200A), I had to install one AC-powered smoke detector per floor to pass code. And all three had to be tied together. Each does have a 9v battery, but they are essentially shelf-life. I added one smoke detector in the living-dining room area, as it was some distance from the AC-powered unit.

      I think it is an intersting hack, I just wanted to point out the “need to be hardwired” was literal in my case.

  5. echodelta says:

    One splash and crash! Better mounted on the hot water pipe under the sink out of sight, not the warm mixing part of the faucet. More “juice”!
    I have to assume that this strictly a lab demo. Ultimate negative WAF. I require nothing be left on the back of the sink, kitchen or bath. This zone is filthy and needs cleaning very often, no antibiotic soap allowed, just stop being so probiotic.
    This zone may be the nest of the next killer germ, resistant to all antibiotics. Forget sneezing and tissues, most germs are handed around!

    • Victor says:

      This was part of the “Energy Harvesting Solutions” design challenge on element14. With an energy harvesting kit (multiple sources) and a EnergyMicro development board participants made an energy harvesting application and blogged about it.
      In the same challenge I made a device that calculates the temperature of the egg yolk in an egg (second order function) from the measured water temperature, using a second order thermal model. Both mine and gelmi’s application were meant to show the possibilities AND difficulties in designing an energy harvesting application.

  6. Kevin says:

    Yay, more energy harvesting demonstrating ignorance of the basic laws of physics!

    The heat sink attached to the faucet slightly increases its heat loss, which slightly decreases the temperature of the water coming out of it. To achieve the same water temperature, the user will then use slightly more hot water. The energy that powers the Peltier junction ultimately comes from the hot water heater, with losses in transit.

    And if the user isn’t going to compensate for the slightly cooler water, it would be better to just use cooler water in the first place. It would be better to insulate the faucet. The fact that no energy conversion can ever be 100% efficient means that it’s almost always better to not waste energy in the first place than to try to reclaim it.

    The only time parasitic energy reclamation makes any sense is when the expenditure of energy cannot realistically be made more efficient. For example, those piezoelectric floor tiles that capture energy from the human gait. I don’t anticipate human bipedal locomotion becoming more efficient any time soon… unless we all get Segways.

    • Tinkerer says:

      So do it yourself in a better way, you nitpicker!
      * This is using microwatts, with high efficiency, so your water temperature will be tens of degrees cooler, if any
      * If you had read the post, you could have read that he’s also using the energy that is wasted when the faucet is still hot, AFTER you’ve let it run. Or would you compensate that by turning up the airco this summer?
      * In your way: those piezoelectric floor tiles impede my walking a bit. I’ll get hungry sooner. I’ll eat more. The world will starve.

      Be fair to energy harvesting. It’s all physics and it’s not free energy, but a lot of energy is wasted that could be used for something very energy-efficient, like a CO-sensor

      • Kevin says:

        “so your water temperature will be tens of degrees cooler, if any” – It doesn’t matter how much cooler it is. The point is that the energy gained is *guaranteed* to be less than the energy lost. If cooler water is acceptable, it would be better not to heat it in the first place.

        “he’s also using the energy that is wasted when the faucet is still hot, AFTER you’ve let it run” – But as I said, it would still be better to insulate the faucet so it doesn’t get hot in the first place.

        “those piezoelectric floor tiles impede my walking a bit” – True of rolling wheels, but not of walking. Every time your foot impacts the floor, some energy is wasted as heat. Piezo tiles just convert it into something else. We can’t do much to change the efficiency of human locomotion, and that’s what makes it valid as a target for energy reclamation. As others have suggested, it might make more sense to try to reclaim energy on the drain side… but only after exhausting every practical means of avoiding using that energy in the first place, and every practical means of reusing the energy in the form in which it already exists.

        • Greenaum says:

          Who the hell insulates taps (or “faucets” in American)? Seriously, how much rockwool is wrapped around your bathroom taps? Or do you use expanding foam?

          You can’t insulate taps. This device, yes, will steal microwatts of heat while the tap is running, but mostly it’s using heat left behind in the tap afterwards. This heat will always be there, can’t be helped. Even if the tap were perfectly insulated, at best, it’d heat up the first splash of water out of the tap the next time you turn it on.

          Though since hot taps give cold water when they’re first turned on, til the hot arrives through the pipes, it’d be heating those first drops that nobody uses.

          I can’t think of any practical way to insulate a tap, and moreso, to use the stored heat afterward, without Dr Seuss levels of plumbing. Which all uses pipe which is melted from the Earth in furnaces etc etc.

          The heat from a cooling hot tap is lost, always will be. It’s not worth keeping as a heat source, there’s too little to be useful. But if you need microwatts for something, why not?

          Probably bathroom plumbing isn’t the future of energy harvesting, but it’s a growing field, and you must know electronics use less and less power each year. Many useful circuits use less power than a battery’s self-discharge, they don’t noticably make the battery life shorter!

    • Mojoe says:

      “The heat sink attached to the faucet slightly increases its heat loss, which slightly decreases the temperature of the water coming out of it. To achieve the same water temperature, the user will then use slightly more hot water. The energy that powers the Peltier junction ultimately comes from the hot water heater, with losses in transit.”

      He can just attach the device to the drain side instead of the faucet side. No compensating or extra losses. Perhaps a custom heatpipe setup coiled around the drain pipe for better heat transfer to the peltier device and thus more energy harvested.

      • Kevin says:

        True. But the water is much cooler on the drain side, having been mixed with cold water and possibly cooled by evaporation. With a smaller temperature difference, it’ll be harder to recover any usable energy from it. But if he can make that work, it might actually be practical.

        • Mojoe says:

          It was already mixed with the cold water in the picture above. It was placed after the valves. Some losses would be incurred in the sink, but as I said previously he could impliment a more efficient heat transfer setup for the final product.

          Transferring the waste heat from the drains back to the hot water heater INPUT would be the best way to recover energy in my opinion, but that was not his goal in the first place.

      • Tinkerer says:

        Drain side is indeed very good. Also for showering, and kitchen sinks. Newest generation of harvesting chips need only a little input to get started…

    • Eirinn says:

      “The only time parasitic energy reclamation makes any sense is when the expenditure of energy cannot realistically be made more efficient.” – this would be correct mr. smart arse if the only reason you’d use parasitic energy reclamation would be to reduce energy dissipation. However using parasitic energy reclamation to cut out a battery of an equation may very well by enough to be worth it. Energy loss and dissipation aside.

  7. dALE says:

    ” and he it got him to thinking. ” Hey, Give us $500k and maybe we will proof read our articles.

    • Greenaum says:

      I don’t care about a few typos. I don’t even notice them unless I look. Human brains are good like that. I’d rather read this site than one of those horrible Wikipedia-news-Outbrain-aggregator sort of things. The horrible ones with endless lists of distracting trivia, big pictures, and popups. They’ve taken over half the web, and are basically cynical money-machines.

      I’d rather read a website with a purpose, written by people who are interested, for readers. Not “generated” by some machine that makes fodder to pump to people it sees as brains attached to credit cards, like the horrible kafkaesque dairy-farm the WWW’s turning into.

      So yeah, typos. Could be worse! It takes humans to make typos!

  8. Dracula says:

    Hey guy’s, this is JUST an example to recoverv wasted energy. If you open your eyes you will see that we are still “operate” lot of things where we are wasting energy and have less then 20% Eff. If we could improve that, using energy harvesting to operate some ouf our devices, our energy balance and life would be much more comfortable, cleaner and easier.

  9. Gelmi says:

    Hi guys,
    Thanks for the feedback about my project. I value both positive and negative comments. This project is for the RoadTest and its goal was to show that some of the customer products can be supplied from renewable energy. I understand that if you use faucet, the water would be slightly colder. As some of you suggested it would be better to use it under the sink (safer place and all this heat is also wasted), but the mounting gear I got for the Peltier module only fit the faucet, not the drain pipe. The efficiency is rather poor (max 40%), but still in my opinion it is worth to get the most from all the heat that would be lost. I hope you enjoyed my post and maybe one day the technology would make possible to supply more energy from these types of harvesters.

    • Greenaum says:

      I don’t think waste-energy harvesting will ever be used to save energy, at least not domestically. Probably in factories and power plants there’s enough energy about that you can sweep kilowatts up off the floor with the dust. The heat lost off pipes on a sink is piffling, and low-temperature heat is the least useful form of energy. Much more to be gained in using less energy to start with.

      But on a small scale like this, the aim is surely a solution to powering small gadgets without having a mains connection, or needing to replace batteries. The US military are very into stuff like this, they want a way of generating a few watts off a soldier to run the electronics they’re all going to end up mounted with. Also stuff like wireless sensors in the middle of nowhere which make the odd low-bandwidth radio transmission. Or the same sensors in other places.

      The main advantage is, no maintenance. No batteries or power wires to maintain. No need to be accessible, no need to hire the staff and vehicles to go change the battery, no need to even keep track of when you fitted the battery. The thing does it’s job, intrinsically, forever. It’ll be a godsend for lots of things.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s