There are a lot of things we don't need. That's not a good reason not to try something. The author learned a new skill and built something with his hands, which is the reward, forgetting that his kid also had a stable light switch now. Seems pretty fine to me.
I'm currently building a cat feeder using a Raspberry Pi and a servo. Is it way overkill? Well, yeah, probably. It's not that difficult to pour a cup of cat food into the dish.
But I'm learning a whole lot of neat stuff in the process. I've spent my entire professional career in software, dealing with hardware only at a high level. So this kind of low level stuff - wiring, soldering, etc - is new to me and something I've wanted to really dive into for awhile.
The point of the exercise is not because the cats need feeding, or that I couldn't buy a programmable feeder already on the market. It's a pretext to learn a new skill and actually create something physical that I can interact with. So much of what I do every day is ephemeral - it exists and bits stored on various computers. This will be something that I can touch and see and interact with. I can look at it and say, "I made that, hardware and software."
How are you handling dosing/delivery method? I've been (over)thinking this for a while now and my prototype "lego hopper with RC servo controlled bottom flap" didn't work particularly well, and would get jammed with kibbles, and wasn't particularly consistent in delivering a fixed portion of food.
I'd like to find a nice big & cheap auger that could pull or push to dispense, and have a lot more control.
The other suggestion someone made was a hopper with a close-toleranced vibratory table/spout to bounce the appropriate amount out.
All this assumes you want arbitrary duration/portion dispensing, otherwise you just need N bowls, and a rotary cover that opens each at a set time (which is most commercial ones I've seen).
> All this assumes you want arbitrary duration/portion dispensing, otherwise you just need N bowls, and a rotary cover that opens each at a set time (which is most commercial ones I've seen).
Mine current design will allow some level of arbitrary feeds and duration.
My current plans call for a sealed rotating drum with an opening that contains ~1/2 cup of kibbles, so each spin of the drum will dispense that much. Using door brush to seal around the bottom of the hopper and the drum to keep kibbles from escaping. The servo will rotate it from 0 to 180 and back, allowing the kibbles to fall down a small ramp and into the bowl. My concern with this design is that I may not be able to rotate the drum fast enough, but I'll see if that's an issue when I get there.
Just spin it X times for X/2 cups of kibbles. I figure 1/2 a cup is good resolution for a home project.
