Which servo (attachment?) to use for automatically dispensing dry food?

[Deadgye]

I'm trying to figure out which servo, or type of servo, would be most appropriate for dispensing dry pet food. I'd really appre

[Mojo]

Are you designing some kind of pet operated pet food dispenser? I don't think you've provided enough information for anyone to be of help for you, start by better describing what you want it to do (as specifically as you can.)

[Deadgye]

Ah, my bad. The background information is probably quite important. I definitely didn't explain my requirements enough too. The overall project is an automated outdoor cat feeder of sorts. There will be a microcontroller present along with other subsystems, most of which will be chosen/defined, programmed, and prototyped by us. The subsystem we're aiming to have fully defined first is the food dispenser mechanism.

Only dry food is being dispensed. (Water too, but that's a different subsystem.) There will be a storage compartment containing enough dry food for a decent number of days, manually filled. We need to separate a dynamic amount of food, static might be acceptable, from this pool of food -- thus we need to use a servo (more specifically a servomotor I think?). That separated portion will then need to find its way to a bowl, but I'm not really concerned about that specific part right now since it can most likely be handled independently.

The following requirements aren't completely set in stone, they may have some leeway or they might be dependent on other things, but I'll list them regardless: We need to handle possible problems, so an absolute encoder is preferred over an incremental one. We need to be able to tell if the servo is stuck, so we can do something to fix it. (Solution is likely handled one level above.) If we need to utilize something additional, like an H-bridge, we can probably include it.

The current method we're favoring is a turning cereal type thingy. But that might not be the best way to go about it and a gravity drop method or something else might be better, so I wanted to reach out for some second opinions.

How to actually use a servo to do a turning cereal dispenser thingy is what's confusing me the most. I understand generic servo stuff and basic robotics, but I don't actually know exactly how a simple plastic manual cereal turning mechanism thingy works; attempts at googling just gives me crappy product pages that explain nothing. When I try to imagine a setup I picture a rotational servomotor connected to a small plastic gerbil ball with its cover off and then I feel stupid lol.

And lastly whatever servo setup we decide to go with needs to be something we can actually purchase in the near future; we also need to create our own small documentation concerning it.

Hopefully that additional information makes stuff clearer, if it just made stuff worse let me know. Or if anybody has any questions feel free to ask. Any help at all will be greatly appreciated.

[Mojo]

A lot of the servos you'll find can't actually rotate completely, so keep that in mind. You could try finding an old gumball machine and taking it apart, I think they work similarly to what you're thinking about. But really, the first thought that comes to my mind is something like this.



Drive the top of it with a gear servo attachment. This will require a servo that can continue actually rotate completely to work how I've designed it. I think those ones actually use potentiometers to keep track of where they are. Either way, the discs would rotate through a chute that the pet food would fall through (which I didn't draw here.) When at rest, the empty region of the top disc and the solid region of the bottom disc are positioned over the chute. This allows pet food to accumulate in the space between them during this time. When the shaft is driven, the top part of the chute is closed, then the bottom chute is opened. The pet food that accumulated will fall down. As the shaft continues to rotate, the bottom chute closes, and eventually the top one re-opens. That probably will not occur with the angles I've drawn, but you can make it work if you think about it more. Hope this helps.

[Deadgye]

Rather late response, but oh well. Also it seems my original post was somehow eaten.

The picture above was actually very helpful. We're most likely going to go with that instead of our original idea. We're looking at possibly using something like this or this.

Now we need to decide which microcontroller we're going to use. Probably going to aim for something Atmel.