Vending Machine for Birds

A vending machine for birds that rewards them with food when they drop something in a hole. Electronics and enclosure materials cost <100 USD/EUR. I've designed a PCB and PVC pipe enclosure and dispenser (described below). The circuit is not complex, but I think a custom PCB makes it easier for people who are not electronics hobbyists to build this. I've made a few different enclosures out of scrap for this, but will share the PVC pipe enclosure details since it's sturdy, waterproof, and the materials can all be found at the hardware store.

There were two main inspirations for me to start and document this project

Hans Forsberg - designed his own machine and was the first person I learned of that had birds (magpies) bringing items to deposit on thier own
Josh Klein - makes his Crowbox designs freely available and supports people who want to build it

I would like to thank PCBWay for helping me get the PCB made.

MAIN FEATURES

portable
saves power until a warm body is present
battery or mains powered
one moving part
accessible components and materials
rodents can't break in and get the food
cheap to build
customizable

ELECTRONICS, SENSORS, AND POWER

The circuit is all analog, using four NE555 timers and one LM358 dual op amp. The sensors are all made using IR phototransistors and IR LEDs. The PCB has headers broken out so you can monitor and control it with a microcontroller or SBC. It runs on 5-6VDC, but be sure to use 5V if you are planning to power a USB device (like a camera) with the vending machine's USB ports. If you use a USB cable to power the vending machine, be sure it can deliver enough current. I found that some 10-year-old 2-meter USB 2.0 cables were resulting in a voltage drop from 5 to 4.5-4.8 volts on the vending machine power rail. Cables that didn't work so well had a resistance >1Ω. Ones that did had a resistance of <0.5Ω. NE555's should get at least 4.5 and Raspberry Pis need at least 4.7V. If things are not working as expected, use a multimeter to check the rail voltage - it should be around 4.9V. Shorter cables are less lossy and take up less space inside the enclosure.

Power consumption at 5V is as follows:

Without RPi Zero W

Idle with Powerbank Keep-Alive and TImer enabled: 18mA
Timer uses 8mA
Powerbank Keep-Alive uses 4mA
^With both of these disabed it uses 6 mA
IR LEDs on: 320mA
IR LEDs and motor on: 510mA

With RPi Zero W and wide-angle camera module

Idle with Powerbank Keep-Alive and TImer enabled: 220mA
IR LEDs on: 520mA
IR LEDs and motor on: 820mA

So with the Rpi and camera you could get about 2 days out of a 20,000 mAh power bank. With just the vending machine you could get about 10 days.

Modules

Powerbank Keep-Alive: 555 sinks current every 7-22 seconds for 0.7-2.2 seconds to keep USB power banks on while sensors are in standby - the powerbank keep-alive can be disabled.
IR LED ON/OFF: 555 keeps the IR LEDs used by the sensors on for 22 seconds after the last PIR trigger
Op Amps: LM358 used as comparators to adjust the threshold of the sensors and send signals when they are triggered
Motor Controller: 555 turns on the dispenser vibration motor when the deposit sensors are triggered, tuns it off when the dispense sensor is triggered
Timed Dispense: 555 dispenses food every 10-100min - timed dispense can be disabled
Daytimer: Phototransistor that puts the timer in reset while it's dark outside - brightness threshold can be adjusted or disabled (Rev. B)
PIR Sensor: Sends a signal to turn on the sensor IR LEDs
Deposit/Dispense Sensors: IR phototransistors and IR LEDs used as proximity sensors to monitor the deposit and dispense chutes
Hopper Level Sensor: IR phototransistor and IR LED to detect when the food level gets low (Rev. B)
Monitoring/Control: lines for monitoring and triggering events with an external controller (Rev. B)

PCB, BOM AND GERBERS

Kicad Files (Github): https://github.com/src1138/VMFB
Gerbers, PCB and BOM (PCBWay): https://www.pcbway.com/project/shareproject/Vending_Machine_for_Birds_11f2a4da.html...

Dispensing Unshelled Peanuts
Stephen Chasey • 16 hours ago • 0 comments
I've made many adjustments to the dispenser assembly in the past two weeks of testing. Unshelled peanuts are more dificult to reliably dispense than shelled peanuts - they vary more in size and shape, they are less dense with a rougher surface. They jam very easily, and the size/shape variance makes it difficult to dispense just one. I usually get 2 or 3 peanuts at a time with this dispenser.

First I had occasional jams in the hopper itself, leading me to
- increase the size of the pathway from the hopper to the dispense chute
- add a debriging arm
Here is the underside of the vibration plate showing where the motor and sensors are mounted.

These changes solved the hopper jams. Then I had jamming in the dispense chute. To solve this I removed the red cup that caught peanuts on their way out. This helped reduce the frequency of jams, but the real issue was using several segments of pipe in the chute. Each joint was an opportunity for a jam. Replacing this with a single piece of pipe seems to have solved this for now. If there is a jam, it is at the opening at the bottom and the birds can pick it out and remove the jam themselves.

I'm getting a pretty steady stream of magpies and jackdaws now. Along with some pigeons that remember when this dispensed shelled peanuts :)

https://cdn.hackaday.io/files/1847547904515136/VMFB-MC_05-49-22.mp4
VMFB-MC for unshelled peanuts deployed
Stephen Chasey • 03/27/2024 at 12:50 • 0 comments

It passed some basic tests and is on my back balcony now.

I think I may have put too much strain on the camera cable as blue and green are faded. I will fix that later.

UPDATE - I was using an IR camera... I replaced it with a normal one.

Now hoping to get some jackdaws and magpies interested. I got my first visitors already.
https://cdn.hackaday.io/files/1847547904515136/VMFB-MC_16-56-28.mp4
Here's what it looks liike on the balcony.
New enclosure ready for testing
Stephen Chasey • 03/25/2024 at 15:31 • 0 comments

To avoid problems associated with feeding pigeons in my neighborhood I revised the design to accomodate peanuts in the shell, which corvids love. After some trial and error (and some poor designs) I finally have my test candidate assembled. It's not as small or pretty as I want it to be, but I think it will work. This is my 7th enclosure design and the 5th one based on 110mm PVC. I also made a base out of a discarded wooden pallet so I can move it around easier.

It does have enough room for batteries and an opening below the platform to get a mains line or solar charging cable in. The bottom comes off by loosening a hose clamp to replace batteries. The electronics are accessible by removing the top (everything above the camera arm).

A wide-angle camera is mounted on a rotatable mount so you can position it as you like. The deposit hole and dispense chute are on either side of the platform to reduce the chance that a bird drops food into the deposit hole and the dispense chute is 50mm instead of 32mm to accomodate the larger unshelled peanuts.
Now to put the electronics inside and do some more tests before setting it outside. It should be out back by this weekend.
PCBWay 6th Project Design Contest
Stephen Chasey • 03/14/2024 at 13:57 • 0 comments

A while back I submitted the analog version of this project to the PCBWay 6th Project Design Contest, and I actually got a prize!

PCBWay has been very supportive of this project since the beginning and I am excited that the first PCB I made with Kicad got some recognition.

Now to finish assembling and testing the peanuts-in-the-shell enclosure for the microcontroller version.
Code is ready (functional anyways)
Stephen Chasey • 02/23/2024 at 15:13 • 0 comments
After a month of occasionally working on moving all the main logic to one python script and debugging I think it's ready. It's all interrupt- and timer-driven, just like the analog version. It took a little work to make sure it worked well with both op amps and comparators, especially since I did not really design the circuit to work with comparators (no feedback to reduce hysteresis) and did not include capacitors to absorb the initial blip you sometimes get with phototransistors.

In the analog version, I had the camera recording while the sensor IR LEDs were on instead of using motion detection. For this version I'm still just enabling and disabling motion detection to trigger recording.

I also optimized the status overlay for motioneye so it's more informative.

https://github.com/src1138/VMFB-MC

The code works, but is as a friend of mine described it, "janky". There is still room for improvement, but for now it's usable and appears to work consistently.
- You can enable and disable the powerbank keep-alive and timed dispense from the web interface.
- You can trigger a manual dispense event from the web interface.
- The video overlay indicates the state of PIR, deposit, dispense, feed level, sensor IR LEDs and the dispense motor.
- The video overlay also shows the count of how many times PIR, deposit, dispense, timed dispense and manual dispense were triggered that day.
- You can easily use other types of sensors or switches in place of the PIR and phototransistors if your situation calls for it.
Here is a screenshot. The numbers are the result of testing and debugging. The sensor IR LEDs are on (SIR is red). The peanut level is low and the timer is enabled.

I will share some better screenshots and a short video soon.

I found a better way to mount a camera in the enclosure so it gets a good shot of the platform without being in the way or using a side-mounted arm.

Finally I found a better solution for rodent-proofing that should allow me to make the enclosure a little smaller without making it easy for rodents to get in there and have a feast. I have not had any rodent issues since I started this project, but this is a pervasive problem with bird feeders and I want to mitigate it as much as possible.

I expect to have another complete prototype based on the VMFB-MC with the above improvements and dispensing peanuts in the shell ready in the next 3 weeks. Since the peanuts will be in the shell, pigeons will not be able to eat them, but the jackdaws, magpies and crows that come by my balcony love them. Hopefully this will keep me from running afoul of the HOA for feeding pigeons.

I think these changes, along with the event logging and dashboard, will make this a more practical solution for students and researchers wanting to experiment with behavioral conditioning and collect data automatically. The VMFB can be adapted to use with other animals like rats or mice and apply to different use cases.
VMFB-MC PCB Populated
Stephen Chasey • 01/07/2024 at 02:59 • 0 comments

I found time to populate the board and am working on the code to make it work with an RPi Zero W and motioneyeos. Hope to be able to demonstrate it soon.

From top left to bottom:

Dispense Sensor, Empty Sensor (for peanut level)

Deposit Sensor

Main Board
VMFB-MC PCB Received
Stephen Chasey • 01/03/2024 at 18:56 • 0 comments

Fabricated, shipped and received in less than a week - and over the New Year holiday!

Quality looks great as usual (PCBWay) - looking forward to populating and testing it.

For anyone insterested, here's the PCBWay share project link: https://www.pcbway.com/project/shareproject/W586910ASV7_VMFB_MC_RA_V1_0_20231227_ecb23676.html

The all-analog VMFB and the VMFB-MC PCBs side-by-side to show how much smaller it is without all the 555's :) The main board is now about the same size as a Raspberry Pi Zero.

Update and smaller PCB footprint/BOM

Stephen Chasey • 12/15/2023 at 18:08 • 0 comments

It's been a little over a month since the student in Warsaw deployed one of these and despite freezing temps and lots of snow the thing has been working fine. It does get clogged with snow build-up though, so a small resistive heater on the plarform is a good idea if you are doing this when there is a lot of snow.

I experimented with dispenser designs that could reliably dispense unshelled peanuts. The one I landed on is just a simplified version of the shelled peanut dispenser. One baffle at a 10-15 degree angle, one vibrating plate with an opening on one side. It works pretty well, and would allow me to deploy it where feeding pigeons is prohibited since pigeons can't eat unshelled peanuts and can't figure put how to open them. The larger openings makes keeping it rodent-proof a challenge. The dispense tube went from 32mm to 50mm. A rat or small squirrel could get in there and cause problems. I'll have to think of a solution for that.

In an effort to make this a more useful research tool I designed a new PCB that relies on external GPIO to work. Sadly, no 555's in this one, but the BOM is smaller (44 components including headers) and so is the footprint (6.5x6.8cm including sensors). This should make it easier to build and install, and you can use prety much any microcontroller, development board, ot SBC to control and monitor it.

It's not fully baked yet - I will create a version tag after I get one made and teseted.

https://github.com/src1138/VMFB-MC

Current Schematic

Current PCB

Board Dimensions: 65x68mm

Front

Back

BOM

Component Count: 44

Ref	Qnty	Value
D1, D2, D3, D4, D5, D6, D8	7	IR LED
D7	1	1N4001
J1, J2, J3, J6, J7, J8, J10	7	Conn_01x04_Pin
J4	1	Conn_01x02_Pin
J5	1	Conn_01x03_Pin
J9	1	Conn_01x05_Pin
Q1, Q2, Q3, Q4, Q5, Q6, Q9	7	SFH313FA-2/3
Q7, Q8	2	TIP120
R1, R2, R4, R5, R6, R7	6	150R
R3	1	1K
R8, R9, R10, R11, R12, R13, R14	7	10K
RV1, RV2	2	100K
U1	1	LM393

Total BOM cost for all electronic components including a Raspberry Pi Zero W, 8GB micro-SD card, a wide-angle camera module, PIR sensor, a motor and a terminal strip (to make the vibration motor) cost €59.62 at Mouser (not the cheapest place, but while I'm in the Dallas area for the holidays delivery is next day). Jumper wires (36x F/F) are not included in this BOM, but cost around €6 in most cases. The PCB will cost about €7 per board at PCBWay.

Here is the Mouser cart (minus some non-informative columns).

Mfr. No	Manufacturer	Description	Order Qty.	Price (EUR)	Ext.: (EUR)
TIP120	STMicroelectronics	Darlington Transistors NPN Power Darlington	2	0.744	1.49
SFH 313 FA-2/3	ams OSRAM	Phototransistors PHOTOTRANSISTOR	7	0.642	4.49
LM393P	Texas Instruments	Analog Comparators Dual Differential	1	0.428	0.43
1N4001	Diotec Semiconductor	Rectifiers Diode, DO-41, 50V, 1A	1	0.214	0.21
15400594F3590	Wurth Elektronik	Infrared Emitters WL-TIRC IR THT Blue 5mm 940nm	7	0.4	2.8
61304011121	Wurth Elektronik	Headers & Wire Housings WR-PHD 2.54mm Hdr 40 P Single Str Gold	1	1.42	1.42
MF1/4LCT52R151J	KOA Speer	Metal Film Resistors - Through Hole 1/4 WATT 150 OHM 5>#/td###	6	0.101	0.61
MF1/4LCT52R102J	KOA Speer	Metal Film Resistors - Through Hole 1/4 WATT 1K OHM 5>#/td###	1	0.101	0.1
MF1/4LCT52R103J	KOA Speer	Metal Film Resistors - Through Hole 1/4 WATT 10K OHM 5>#/td###	7	0.101	0.71
PT10WV10-104A2020-1CRPM	Amphenol	Trimmer Resistors - Through Hole 100K 20% 10mm POT CREAM COLOR SHAFT	2	0.886	1.77
114992065	Seeed Studio	Camera Accessories Camera Cable for Raspberry Pi Zero	1	2.3	2.3
SC0065	Raspberry Pi	Single Board Computers Raspberry Pi Zero WH (With Pre-soldered GPIO Header)	1	14.88	14.88
U054	M5Stack	Infrared Detectors An M5StickC compatible human body induction sensor	1	3.26	3.26
PPN7PA12C1	MinebeaMitsumi	AC, DC & Servo Motors DC Brush Motor, Flat, 11605RPM, 5VDC	1	2.7	2.7
39100-1802	Molex	Barrier Terminal Blocks TERM STRIP 8MM HIGH PROF 2 CIR WHT	1	0.735	0.74
114110130	Seeed Studio	Cameras & Camera Modules OV5647-160 FOV IR Camera module for Raspberry Pi 3B+4B, suitable for large or night...

Shipped one
Stephen Chasey • 11/08/2023 at 13:47 • 0 comments

I received my packaging materials today, so I packed up and shipped one to Warsaw for the student I mentioned a couple of updates ago.

This is what they will get:

I had to break it down a little for shipping (I included mounting hardware):

So it would fit in a 100x25x35cm box (weight 7kg):

I used soooo much bubble wrap - I hope it arrives in good condition.

UPDATE

They received it in good condition. Hopefully setup will go ok. I really want this to be easy to setup and use, so this is a good test.
UPDATE
After fixing a few connections that came loose in shipping, it seems to work and is mounted. Now it's up to the birds. Kudos to the student for some excellent troubleshooting and fixing.
More efficient remote monitoring
Stephen Chasey • 11/07/2023 at 16:01 • 0 comments

I have been experimenting with ESP32 and RPi camera modules to record video and pictures as well as log events in log files. While cameras typically demand 100-150 mA,, the boards themselves can make due with ~80mA and can be put to sleep to consume less power when nothing is happening.

Since the ESP32 is less power-hungry than an RPi I will use one to make a simple event logger that is accessible over wifi and bluetooth so experimenters will have an easy way to monitor activity without video. The parts to do this with an ESP32 cost about 10EUR.

This would also make it more practical to run the vending machine off of batteries/solar and allow remote log collection via the internet, internal network, or when in proximity of bluetooth.
UPDATE
After looking more into it I could use an ESP32 and provide the ability to monitor, get logs, do some simple graphing, configure and control the vending machine via wifi or bluetooth using just a few additional parts. I could even integrate one of the ESP32 CAM modules without more wires. Even with an ESP32 dev board and an ESP32 CAM board, all the electronic components (there would be a lot less of them) would cost ~25 EUR.
This would be a big shift away for the current all-analog approach, but since ESP32 boards are pretty inexpensive, have radios, have lots of GPIO and are not too difficult to program I think this would make this project much more functional and more attractive to people that just want to use the thing without spending an hour or so soldering.

View all 104 project logs

VMFB-MC_05-49-22.mp4 Getting a steady stream of magpies and jackdaws. Still in Phase 1 at this point. MPEG-4 Video - 1.49 MB - 04/17/2024 at 13:21		Download
VMFB-MC_16-56-28.mp4 First visitors to the unshelled peanuts version with VMFB-MC MPEG-4 Video - 781.75 kB - 03/27/2024 at 17:04		Download
20231029_175433.mp4 Brief functional demo with the Rev B V 1.1 board from PCBWay and improved dispenser. MPEG-4 Video - 2.79 MB - 10/29/2023 at 19:47		Download
20230601_233103~2.mp4 Brief functional demo with the new board from PCBWay MPEG-4 Video - 27.06 MB - 06/04/2023 at 12:53		Download
Vending Machine For Birds_16-29-47.mp4 Jackdaws are starting to use the bottlecap magazine. MPEG-4 Video - 922.07 kB - 05/25/2023 at 14:31		Download

Vending Machine for Birds

Description

Details

Files

VMFB-MC_05-49-22.mp4

VMFB-MC_16-56-28.mp4

20231029_175433.mp4

20230601_233103~2.mp4