Kessil

DIY Fresh Food Auto-Feeder

FreshFoodFeeder_001_squist.png
Background

I learned of DIY fresh food feeders first through Scott Sweet’s fresh food auto-feeder design. Where the Scott Sweet Feeder uses 1/2” rigid plumbing and similarly scaled components for a continuously flowing system, I went minimalist in construction and with low water flow.

I started sourcing parts summer ‘20, sketching, prototyping and testing in October and November, and began running the system Thanksgiving weekend 2020.

The system cost me less than $250 to build using all new parts. Food not included.

Design
The fresh food auto-feeder uses 1/4” flexible waterline, push fittings, Neptune PMUP, and peristaltic dosing meter to transfer fresh food from refrigerated bottles to the tank. While Neptune Apex and PMUP are used in this version, any low flow timer-activated pump will work just as effectively.

Goals
- Easily sourced from inexpensive and common parts;
- Easily built using simple tools;
- Reliable low-risk continuous performance;
- Easy to maintain with self-flushing manifold.

How it works
In a small refrigerator
FreshFoodFeeder_002_squist.png
connected together: dosing pump, simple push-fitting manifold, and liquid fresh food. The refrigerated manifold and food source is connected to the tank through 1/4” waterline tubing loop. Water circulates through the loop using a small powerhead on a program timer.

With the dosing pump’s clock and circulation pump’s timer in-synch,1.5 minutes after the dosing pump deposits fresh food into the manifold, a Neptune Apex PMUP in the return chamber of the sump (any similarly sized timer-activated pump for non-Apex systems) turns on for 30 seconds; the time needed to circulate tank water through the 40 ft loop. The pump, placed in the return chamber of the sump and attached to a 1/4” waterline, cycles water from the sump, through the refrigerator and manifold, and back to the mouth of the return pump in sump’s return chamber.

A dosed portion of food, as a result, moves from the refrigerator, is injected into the mouth of the return pump, then arrives in the display tank.


FreshFoodFeeder_005_squist.png


FreshFoodFeeder_003_squist.png

On the next hour, and every hour 24-times throughout the day regardless of feeding or not, the water cycle repeats for 30 seconds clearing the line of any food that might not have made it to the tank on the feed cycle.

In addition to the dosing pump depositing food, there is a manual port for injecting larger food using a syringe. I use this port for feeding thawed frozen food that the dosing pump would smash to mush. I keep 1-2 days of thawed mysis shrimp at hand in a container in the fridge, squeeze 2ml of shrimp into the port using a syringe, and follow-up with 10ml of fresh salt water to clear the port and push food into the manifold.

Outcomes
Since Thanksgiving 2020, the system has delivered more than 500 mL of Reef Nutrition food (approximately 8 mL/day total) including: Oyster Feast, Roti Feast, and Photo Feast. In addition to the dosed bottled foods, I manually inject into a feeder line frozen mysis shrimp (after thawing, 2 mL) and baby brine shrimp (0.5 mL). For the last two months, to Feb 6, 2021, I have used three of the four dosing heads and food sources.

There have been zero leaks. No indication of blockages or residue. And there is zero reverse water flow back into the food bottles. I credit this with a low-pressure pump that’s just powerful enough to push the food to the tank.

In terms of nutrients (NO3 and PO4), there has been -20% downswing of nitrate (15 to 12 ppm) and a 66% uptick in phosphate (0.03 to 0.05) since starting the system. I’m not drawing any conclusions of the cause of change; there are many other factors at play. It’s relatively stable with no visible adverse changes to the tank.

FreshFoodFeeder_007_squist.png


FreshFoodFeeder_004_squist.png

If I could do it again
If there were anything I’d do differently, I’d go with a bigger fridge. Although everything perfectly fits in this refrigerator and with room for large bottles, there is not a whole lot of space for moving things or reprogramming the dosing pump. For my use though, it is set-it and forget-it. If you think you might want to swap things out or change things up you might want a larger refrigerator.

Upgrades coming next
I am adding an unrefrigerated dosing unit for four more additives. I am going to create a second manifold like the one above and connect it to the water line. And I added RedSea Reef Energy Coral Food to the lineup.


Materials and Cost: $250-$400 total

FRIGIDAIRE EFMIS164 70 Can, Glass Door Beverage Center Refrigerator (a "used" returned purchase from Amazon) $98 (I scored a “used” one for $98)
Whynter BR-062WS Refrigerator $180
Kuppet 62-Can Refrigerator $150
Frigidaire 70 Can Beverage Refrigerator $125
FreshFoodFeeder_006_squist.png

Auto dosing pump, bottled fresh food, and food supply manifold are connected, chilled, and contained inside the refrigerator. The supply line enters and exits the fridge through drilled holes and connects to the food supply manifold using a 1/4” ’T’ press fitting. Any refrigerator will work.

Jebao Programmable Auto Dosing Pump 4-Head $60
Sits in the refrigerator. Each source in-flow line is inserted into refrigerated fresh food container. Each dosing out-flow line is connected to a manifold T push fitting. Any dosing pump that fits in fridge will work.

Manifold $10 (purchased in bulk from Amazon)
Series of T push fittings, one per container of live food. Each T-fitting connects to one of the dosing pump head’s out-flow lines. The manifold is attached to the closed-loop from and to the sump.

Neptune PMUP $40
FreshFoodFeeder_008_squist.png
Sits in return chamber of sump. Feeds a closed-loop 1/4” water line that enters the fridge, passes through food delivery manifold, and returns to the sump into the mouth of the return pump. Any pump capable of connecting to 1/4” flexible waterline and a timing system will work. 3/8” to 1/4” adapter needed.

Waterline and push fittings $20
40 ft 1/4” waterline
manifold parts: 1 T-fitting per food container ending at a T-fitting that the closed-loop passes though; 1/4” press-end cap
 

Attachments

  • FreshFoodFeeder_009_squist.png
    FreshFoodFeeder_009_squist.png
    453.7 KB · Views: 463
Last edited:
Great write up, thanks for taking the time to put this together.
Fridges are usually a dry environment right? So condensation isn’t a concern?
 
Great post! Glad to hear it's working well!
Not sure if you're using one of the other products, but I believe reef energy AB+ is supposed to be refrigerated.
 
Great write up, thanks for taking the time to put this together.
Fridges are usually a dry environment right? So condensation isn’t a concern?

Thanks! Condensation ... haven't had any issues. There is some ice that has formed on the chilling plate in the back of the fridge, but nothing out of the ordinary. I live in a pretty dry environment too. The fridge has a glass door and it is always clear.
 
Great post! Glad to hear it's working well!
Not sure if you're using one of the other products, but I believe reef energy AB+ is supposed to be refrigerated.

Thank you! Yes--I have the reef energy AB+ in the fridge. And on a stirring plate that does a gentle stir for 2 minutes before dosing. I'm pretty excited about that product -- I just started dosing it over the weekend. For AB+ I'm dosing at night 1/2 the qty and the other half during the day.
 
Very nice write-up Squist...looks like you did a great job on the project. This is on my to-do list when the time comes. Quick question and please correct me if I'm wrong. It appears you drilled straight through the inside of the fridge and through the outside casing to run the in/out lines connected to a "L" push connector. Is there a locking nut that holds that "L" in place outside of the fridge? I also see a valve on the in and out lines. Is there a reason where a reduced flow is required?
 
Thanks!

Drilling—yes—a simple 1/4” hole straight thru the sides.

As far as the connector goes, an L fitting isn’t necessary but I used them to clean things up. Otherwise there’s a semi loop of line coming out that might crimp and eventually, potentially, break.

There’s no locking nut. The line from the manifold inside extends outside the fridge 5/8” (the internal length of line a fitting accommodates). The L fitting rests flush against the exterior.

Valves close the line off in one or both directions so I can remove the manifold to clean if necessary or flush in one direction or the other. There is a siphon if the loop is opened. Valves let me disconnect things for maintenance.

66B5A4BE-AC50-4C57-986F-90232DAB003F.jpeg


On left side, L fittings. High traffic side.


CC946284-8D47-48D2-8D89-9398DAF1F52E.jpeg


On the right side of the fridge, no L fittings. After adding the 2nd dosing pump, there’s one line on the right. This earlier photo shows both.
 
Last edited:
Thanks for the info...didn't think about the maintenance part. I wonder if condensation builds up inside the doser while it's in the fridge. Maybe the heat generated by the unit keeps it dry...just pulling that out of my rear as I have no idea lol.
 
No condensation for me. I am in a dry climate with no fog. My fridge-feeder is near a water heater (gas hot water tank for house). If there were condensation I’d put the dosing pump in a plastic bag or Tupperware container most likely. That’s the concern, I suspect, right? Electronics?
 
Rereading this, you say: “inside the doser” not “inside the fridge” as I read it. That’s to be determined. No issues so far after 2.5 mos.

One more point: this method has been used for years by others too. This isn’t anything new. Maybe the use of flex waterline and press fittings is, but housing a dosing pump in a fridge isn’t.
 
Back
Top