Dual Pet Food Dispenser Is Doubly Convenient

Does your dog or cat wake you up every morning, demanding to be fed? Maybe you feed Sparky in the evenings instead. But doesn’t that limit your spontaneity? It sure limited [Jorge]’s after-work plans. He has two dogs that eat the same type of food, but in different quantities. This was a big factor in the design and execution of his dual pet food dispenser.

[Jorge] started by modeling his requirements in 3D. Dispensing takes place in two stages as food moves from the storage hopper to the bowls. A 12V printer motor turns the 3D-printed auger, which transports the nuggets to the staging area. Here, a servo controls a ramp in a see-saw motion, sending the food sliding sideways into one bowl or the other.

The dispenser is designed around a PIC18F2420. Although this micro was [Jorge] ‘s second choice, it ticks all the boxes in the design. His acrylic enclosure features four push buttons for navigation and selection through the 16×2 LCD. [Jorge] has an issue with the food getting stuck in the first stage. A friend suggested that he use vibration to agitate the food, but that didn’t work. [Jorge] ultimately added a stirring shaft with spokes that helps keep the morsels moving. Take the tour after the break.

If you want to dispense single doses of food on a timer, check out this automatic cat feeder made from scavenged parts.

21 thoughts on “Dual Pet Food Dispenser Is Doubly Convenient

  1. Couldn’t this be simplified by using a rotating paddle rather than an auger and see-saw? If arranged correctly, you’d just have to turn it one way to fil one dog’s bowl, and the other for the other. Kind of along the lines of a cereal dispenser, but with a directional element. You wouldn’t get as smooth a delivery, but it’d be just as functional.

    1. Couldn’t this be simplified by using a rotating paddle rather than an auger and see-saw?
      no because that would jam very easily. unfortunately i think this design will also jam. non-jamming dispensers are extremely difficult to design so your best bet is to look at professional grade candy machine dispensers designs.

  2. I have two automated dispensers but one cat steals food from the other cats bowl and the victim cat decides to file a vocal complaint with me about at around 4 am. Is there any inexpensive DIY tools for detecting the under-the-skin-chip in a cat when the cat head is close to a food bowl? Ideally some sensor that can be connected to a raspi or arduino and that can be hidden in a flat box below the bowl.

    1. There was an automatic per door lock on the Internet some time ago that used a separate collar tag rather than the under-the-skin tag. That would likely be easier to read just-in-time. My vet needs to hold the reader close to the back shoulder blade area to read the embedded one.

      1. Thanks. I googled “rfid cat collar” and found several products. That will be my fallback solution. However the stealing cat is not only a thief, it also dislikes wearing a collar which is why detecting the under the skin tag – if feasible – would be so good.

    2. I’m pretty sure that nobody has a way to read it from under the bowl. I think though that a lot of cats would be very able to learn to brush up against a reader just so if it reliably resulted in food – after all, that’s a classic Skinner-ian training situation. Another thought that occurred to me is that using a cat sized platform with some load cells might do the trick if the two cats have substantially different weights.

      I’m even wondering if you could use a cheap color camera and some equally cheap image processing to distinguish cats by average coat color.

      I saw that somebody made one with a servo and a cereal dispenser – I’ll probably start from there before too long and throw some things at the wall to see what sticks.

      1. I did find the commercial product SureFlap and their FAQ ( https://www.sureflap.com/en-gb/support/frequently-asked-questions ) state “Do SureFlap microchip products work with my pet’s existing microchip? Yes. SureFlap microchip products are compatible with the following types of identification microchip commonly used for pets: 15-digit microchips (also known as FDXB, standard in Europe) 10-digit microchips* (also known as FDXA, most commonly used in the USA) 9-digit microchips (also known as Avid Secure/Encrypted Chip)”. I don’t want their product however, I’d rather make something myself. But their claims indicate that it can be done. I could construct an arch over the bowl, like their product does, if that would make a DIY reading system more feasible.

        I’ve actually tried using a a camera and computer vision software but it is hard to reliably separate both two cats and also human motion in different lighting conditions (shadows etc).

          1. My thought was to detect the thief cat and trigger a thief cat scaring device (noise and/or water). But yeah I could collar and detect the innocent cat and have a motorized lid on the bowl.

    3. A tiny Neo magnet on the victim cat’s collar and a large coil where the cat’s bowl is. When the magnet passes the coil, it can generate a small electrical signal… An analog comparator can probably trigger on that.

      1. That is a good idea as an alternative to a rfid tag on a collar. But I’d prefer to not put a collar on the cat if possible. Also the thief cat tends to sleep on all the stuff in the house. Isn’t there a risk to have a strong magnet walking around freely when there are laptops and things like that laying around? Maybe there’s no risk nowadays but I’ve believed that strong magnets should be kept away from harddrives.

        1. It is the *victim* cat, not the thief cat to put the collar on.

          HDD have very strong magnet inside their voice coil and also their BLDC motors. A tiny magnet (2 – 3 mm diameter 1 mm thick) magnetic field drops off very very rapidly so it should not be an issue.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.