How to neatly transport a bunch of small objects, such as the hex nuts, one-by-one from point-A to point-B? A linear vibratory feeder is an excellent yet simple approach to do this, which offers stable, orientation-specified movement for transporting small objects with high speed. A conveyor belt system can do the work too, but when it comes to the need for specific feeding orientations or equipment’s space limitations, a linear vibratory feeder would be a better option.
Here, I built a simple linear vibratory feeder with 3D printed parts and a frequency control circuit. The function of smoothly transporting M6 hex nuts was achieved as well as speed adjstments. The purpose of this work was to setup a testing platform for developing high-throughput products selection system, from a emerging project that I involved in while working in ITRI. Before this, I had never really understood how does a vibratory feeder work that can transport things without a "moving part”. Let’s break it down.
Overview of the linear vibratory feeder.
Mechanical Design
I referred to this DIY work and a wide variety of commercially avaliable models of the linear vibratory feeders for the design. They all share a basic framework: two parallel shelfs being conneted by two oblique spring plates as a parallelogram shape, and a vibration source being fixed on the top shelf to create a back-and-forth vibrating motion. The vibration source can be a vibration motor (a motor + an eccentric wheel) or a set of electromagnet. I choose using a vibration motor for its simplicity and smaller size, and it worked pretty well in the testing runs. I designed the device’s structure using Solidworks and printed all the parts including the spring plates (thickness 1.5 mm) by a Prusa MK4 3D printer with PLA material. The motor was a small 130 type DC motor and a 3D printed eccentric wheel was tightly fixed on the motor shaft with a maximum diameter of 13 mm. The base was fixed on an aluminum extrusion to ensure a solid foundation that won’t make my device “walking” on my desk instead.
Working Principle
The rotating eccentric wheel causes the motor body to vibrate due to the inertial force generated by an oscillating mass. As the upper tray vibrates, it produces a slanted back-and-forth trajectory instead of a purely horizontal one because of the tilted angle of the spring plates, which is 70° in this case. Here's a brief explanation of how the object moves on the feeder:
When the tray is lifted and swings forward, the object remains in contact with a spot on the surface due to static friction between them.
As the tray drops, the inertia of the object keeps it in the same spatial position while the friction is reduced.
The object then lands on a point further forward on the tray’s surface because the tray is now swinging backward. This creates a linear displacement of the object along the tray.
Promising as it might sound, the successful transportation of the targeted objects is not guarenteed with an arbitrary setup. The frequency and amplitude of the vibrations should be carefully adjusted according to the properties of different objects, as variations in size, shape, and mass can significantly affect their transportation.
Solidworks 3D modeling for the linear vibratory feeder design.
Animation of how the feeder works
Control & Circuit Implementation
Since the vibratory movement comes from the rotation of the eccentric wheel, it’s straightforward that controlling the motor’s rotational speed will directly affect the vibrational frequency and amplitude. In terms of mechanics, this system can be simplified as a mass-spring system undergoing forced vibration, where the external force, the rotating eccentric wheel, is a simple harmonic oscillator. The steady-state motion of the mass can be described as a simple harmonic motion as well, with the same vibrational frequency as the external force plus a phase shift. The vibrational amplitude is characterized by how “hard” the eccentric wheel is rotating, i.e. its rotational kinetic energy, which is proportional to the object’s mass and the square of the rotational frequency. I actually added a set of screw on the eccentric wheel eventually to increase the vibrational amplitude since the 3D printed part itself wasn’t heavy enough. Good, now I can control the vibration with only one factor–the motor’s rotational frequency, but how exactly?
There are so many solutions to control a DC motor, such as the PWM control, a potentiameter directly in series with the motor, and a potentiameter with a MOSFET. Here, I used a 10k potentiometer combined with an IRFZ44N MOSFET to control the motor's rotational speed for the vibratory feeder. While PWM offers precise and automated control, it adds complexity and cost in the circuit setup, which may not be necessary for prototyping. On the other hand, using only a potentiameter in series with the motor is a basic speed control method but is highly inefficient, as it dissipates excess power as heat and offers limited speed control. In contrast, the potentiometer-MOSFET combination provides a simple, efficient setup for motor control, making it an ideal choice in this application.
I referred the circuit setup from this video, which was the simplest I found. This allows for manual adjustment of the motor speed by varying the gate voltage of the MOSFET, which regulates the current supplied to the motor. The power supply was 9V 2.5A DC, and the circuit was assembled on a small breadboard attached on the device’s base. I also used the LTSpice for designing and simulating the circuit, which I’ve learned in the university but completely forgot how to use it at first. It felt great to pick up using a powerful tool!
Schematic diagram and simulation result of the vibration control circuit by LTSpice.
Picture of the vibration control circuit.
Performance Testing
I tested the device by manually put the M6 hex nuts onto the top tray and adjusted the potentiameter until seeing them moving smoothly. The transportations were achieved at about 0.5 wiper on the potentiameter, which was outputting about 4.5 V, and it took approximately 8-10 seconds on average for a hex nut to traveling from one end to the other. The vibrational frequency was not really stable so I actually had to adjust it after a while during the operation. I also made a small plate as a loading station and placed it 1 mm away from the end of the feeder’s tray. As shown in the first testing video, the hex nuts were successfully transported and stood on the loading station, and they were then pushed by the following ones continuously. In industrial applications, the feeder system can be coupled with optical or image-based sensors to measure the critical parameters of products one-by-one, and expelled the NG products using mechanisms such as air-blowing with high-speed solenoid actuation.
Resonance
During testing, the resonance occurred at about 0.3 wiper on the potentiameter, and the amplitude became so large that all the objects on the feeder would madly fly away. While recording this resonance phenomenon, I found that the video looked interesting as the feeder’s movement becoming relatively slow and blurry–the stroboscopic effect! Since the framerate of my iPhone camera was set as 30 hz, the device’s resonance frequency could be deduced at somewhere around the integer multiples of 30 hz. The resonance should be avoided in practice to ensure a stable transportation. Therefore, the spring plate’s properties including the material, length, and thickness should be carefully designed such that the resonance frequency was set far away from the working frequencies of the device.
Transporting M6 hex nuts with the linear vibratory feeder
Resonance phenomenon of the device (recorded by iphone SE2 at 30 fps)
Conclusion
While I didn’t have quantified test or analyses for this feeder’s performance, I think it’s a great prototype that effectively simulates the linear feeding of small products as well as offering the flexibility to integrate with other modules. The working principle of the linear vibratory feeder is really impressive, since it provided a simple, energy-efficient approach to transport objects using vibrations instead of an actual moving part. I think this concept can be further utilized into micro-nano engineering that precisely manipulates the movements of irregularly shaped objects, such as arranging bio-particles on a microstructure.