A feeder does one job well: it puts food down when nobody is home. It also adds two failure modes that a bowl never had, a portion that drifts from the number on the screen and a machine that stops without telling anyone. Both are testable in about 20 minutes with a kitchen scale.
Portion accuracy is a range, not a number
Most rotor and auger feeders dispense by volume, with a nominal portion of 8 to 12 g. Real output typically runs 8 to 14 g depending on kibble density, and the same setting can vary by 1 to 2 g between consecutive drops. That is acceptable. What is not acceptable is a spread wider than 4 g across ten portions, which usually means the kibble is the wrong size, the rotor is worn, or the hopper is running low.
The test takes ten minutes. Dispense ten portions onto a scale that reads to 1 g, weigh each one, and write down the mean and the range. Then do the arithmetic for your cat: a 4 kg cat needs about 200 kcal a day, kibble at 3.8 kcal/g is 53 g, split into five meals of 10 to 11 g. If the feeder is really dropping 13 g, the cat is eating 250 kcal a day, a 25 percent surplus, and roughly 1.8 kg of body weight over a year if nothing else changes. The starting numbers come from the food calculator.
Kibble size decides whether it jams
Round kibble between 8 and 12 mm across feeds reliably through most rotors. Below 6 mm the pieces double-drop and the fines pack into the outlet. Above 14 mm, or with flat disc and cross shapes, pieces bridge over the opening and the next two meals arrive at once, which is the failure that actually harms a cat.
Mixed textures are the other trap. Kibble plus freeze-dried pieces, or a prescription diet with a coated surface, will separate in the hopper: the small pieces fall first and the large ones clog. Keep the hopper one-third to two-thirds full, since a nearly empty hopper drops uneven portions, and check for dust at the bottom every refill.
Dry food only, for reasons that are not snobbery
Wet food bridges in every auger design, and the residue that stays in the chute is a bacterial culture at room temperature. A bowl of wet food left out for more than 2 hours above 4°C should be discarded, which rules out the whole point of a timed feeder. The exception is the carousel type with an ice pack, which holds two 85 g portions cold for 8 to 12 hours and is worth considering for one wet meal a day.
Dry food has its own limit. Kibble left in the bowl absorbs humidity, and a cat eating the last of a portion 10 hours later is eating stale food. Clean the bowl daily and the hopper every 2 weeks.
Power cuts, Wi-Fi drops and the double dump
Check the backup before you need it. Unplug the feeder with a full hopper and watch what happens: a unit with 4 AA batteries usually keeps its schedule for 24 to 72 hours, a unit without batteries stops and then dumps a missed meal when power returns. That dump is the single worst behavior in the category, so set a maximum single dispense of 20 g and, if the app allows it, a missed-meal limit of one.
Wi-Fi is the second weak point. Most feeders only join a 2.4 GHz network, so a router that hides the 2.4 GHz band or runs a single SSID with band steering will fail at setup. Test a scheduled meal with the router's 5 GHz band off, then test again with the phone off the network. A feeder whose schedule lives in the cloud will skip meals during an outage; one that stores the schedule on the device will not.
Multi-cat homes: who gets which bowl
A single-outlet feeder in a two-cat home is a resource competition, not a feeding plan. The bolder cat eats first and the other cat learns to eat fast or not at all, which is the pattern behind food guarding. Splitting the day into 5 or 6 small drops helps only if there are two feeding stations far enough apart that neither cat can watch the other.
Microchip and RFID feeders solve the sharing problem properly: the lid opens for one registered chip or collar tag and stays shut for everyone else. They cost two to three times a plain feeder, hold 300 to 500 g per bowl, and are the only practical way to feed two cats different prescription diets. Expect a 3 to 7 day training period with the lid taped open.
Placement and the cleaning schedule
Put the feeder on a hard floor against a wall, with the cord run behind furniture, at least 1 m from the water bowl so kibble dust does not end up in the water. Cats fed on a schedule that matches their natural pattern of several small meals eat more calmly; the timing options are in how often to feed a cat.
Cleaning is the part that gets skipped. Bowl daily, hopper and chute every 2 weeks with hot water and a bottle brush, desiccant pack changed monthly. A hopper that smells sour is growing biofilm, and no amount of fresh kibble on top fixes that.
Types and what they are actually for
| Type | Capacity | Best for |
|---|---|---|
| Gravity feeder | 3 to 5 L | Free-choice dry food; no portion control, no schedule |
| Programmable rotor | 2 to 6 L | 1 to 6 meals a day, workdays, no app needed |
| App-connected with camera | 3 to 6 L | Travel, remote top-ups, checking intake |
| Microchip or RFID | 2 bowls, 300 to 500 g each | Two cats on different diets, food guarding |
| Wet-food carousel with ice pack | 2 × 85 g | One wet meal, 8 to 12 hours cold |
Five checks before you rely on one
- Weigh ten portions and record the mean and the range; reject anything wider than 4 g.
- Confirm kibble size is 8 to 12 mm and the diet is not a mix of shapes and textures.
- Unplug it with a full hopper and see whether the schedule survives and whether it double-dumps.
- Set a 20 g maximum per dispense and test the app with the phone off the network.
- Wash the bowl daily, the hopper every 2 weeks, and keep a manual bowl and a measured bag of food as the fallback.
Frequently asked
How accurate are automatic feeders in grams?
Most land within 1 to 2 g of the setting from one drop to the next, because they dispense by volume rather than weight. The spread is what matters: if ten weighed portions vary by more than 4 g, the kibble size, the rotor or a low hopper is at fault, and the fix is usually a different kibble diet.
Can I put wet food in an automatic feeder?
Only in the carousel models built with an ice pack, which keep two 85 g portions cold for 8 to 12 hours. In an auger or rotor feeder, wet food bridges in the chute and the residue grows bacteria at room temperature, and food above 4°C for more than 2 hours should be thrown out.
What happens to my cat if the power goes out?
A feeder with 4 AA batteries usually holds its schedule for 24 to 72 hours. Without batteries it stops, and many units then dispense the missed meals at once when power returns. Set a 20 g cap per dispense, keep a manual bowl, and test the outage before you travel.