I have been working on a project over the last month for my kitchen - a smart range hood. The plan is to use a RaspberryPi, four MQ air quality sensors, DHT22, and a GridEye infrared 8x8 pixel temperature sensor. I have the blog lost here: (full disclosure this is part of a contest; I just wanted to share it here since this is a group of like minded people. If this doesn’t belong PM me and I’ll take it down)
I plan to use the humidity sensor to turn the fan ON at something like 80% and rising and back off around 70% and falling (for histeresis). The air sensors are Methane, Hydrogen, Carbon Monoxide, and Propane. It will turn the fan on if any of these gasses get above safe levels.
I am using the PiCamera to take images of the Stovetop so my wife can text me when dinner is ready with a picture
The GridEye sensor will be used as an on/off sensor to see if any burners are on when we leave the house (using Owntracks geofence). I think I can detect the oven itself from the GridEye as well since it would make the entire field of view look warm.
What do you guys think? Any ideas I should include?
I think it’s a great idea. I am curious how you plan to keep the gear clean or at least make it cleanable. The underside of a range hood is a tough environment with the humidity, condensation and oils that build up relatively quickly. Most DIY kit isn’t robust enough to handle the brutality. I will go check it out when i get a few minutes.
I read your blog post about your smart hood. Ambitious project. One very important thing to keep in mind when designing the hood. The hood’s primary function is to exhaust smoke and particulates out of the kitchen air space. Detecting humidity and heat should give you the primary trigger to start the blower. Be careful about all of the extra sensors or you might find out why no one puts smoke or carbon monoxide detectors in or near a kitchen. Constant false alarms will drive you to suicide long before you die from Co2 or other gases.
Another important consideration is sensor contamination from all of the grease that goes up into a hood. When I installed my hood I didn’t use the blower in my Micro hood. I used a really big squirrel cage blower from a pop-up counter vent I salvaged. I installed the blower in the attic to cut down the noise. I hate a noisy kitchen blower. The blower is controlled by a relay so it would be easy to install sensors and a controller (pi). I also did the same thing with my bathroom vents. I put a big blower in the attic to keep the noise down so I could hear the music piped into the shower marine ceiling speakers. Also to save money(and extra hassle) since my bathrooms are stacked on top of each other, I only needed one blower for both showers. If I decide that I need one, I have some vent servos that I could install to close off an unused vent. The ductwork has a sound baffle to reduce sound transmission between bathrooms. This vent will definitely get hooked up to the house controller with a humidity sensor to activate the blower and servos.
Thanks for the comments. I like the idea of “outsourcing” the blower to the attic.
The visible light camera and the IR camera will be the most “in the line of fire” - I will experiment with different windows to put them behind. The IR camera will be a problem though since I need a special material which is transparent in the IR wavelengths. I have looked at Zinc sulphide which is used in commercial IR sensors as a window but a 1 cm square costs $250 on Edmunds Optics.
The air quality sensors will be tucked away a little more and I may use a little 3D printed air tunnel to ensure that no solids like grease splatter get on them. The rest will be packed away in a small box.
Nice about using a 3D printer to make an air baffle case. I just bought a Tronxy x5s. I’m still in the process of putting it together but I can’t wait to start printing something. Be sure to use PETG or some other higher temp filament for your box. If for no other reason than you might need to clean it in hot soapy water occasionally.
Just posted an update on this project. I had a local fab shop create the sheet metal shell for me and I have established basic controls for the system.
If @jon says “the best UI is no UI” then I win because my touchscreen blew up last week. The sensors are good enough to know when the stove is on and turn itself on right away. Check out the video at the end of the blog post.
This is a fantastic project. Thanks for sharing all the details!
I’m really interested to know how well the positive pressure method keeps the camera clean. Our rangehood sat for several years as just the cover without the insert installed, so I put an Arduino up inside the hood with a DHT22 sensor and found that it ended up with a coating of oil condensed onto it. The underside of a rangehood is one of the nastiest places you could put electronics.
Now that you’ve had it running for a little while, have you found that the sensors stay clean?
So far, so good. After a month and still clean. My wife cooks a lot; almost every night. I think that the physical barrier - the 90 degree turn with the baffle - probably will do most the filtering for the air sensors (DHT22 and MQ sensors). The mesh screen will also help a little. I wanted to do more with the cameras to get them behind a window or use a mirror like @MisterFixit1952 suggested to get them out of the ‘line of fire’ but a) ran out of time and b) couldn’t have even if I wanted to since the opening in the metal insert was too small to back the cameras away any further. In fact, The last modifications I did had to do with lowering both cameras in the mount as much as possible so the field of view wasn’t cut off.
