ScienceMadness: Construction of an inexpensive adjustable speed magnetic stirrer
This is an old piece that I wrote and posted to the ScienceMadness
discussion forums back in July of 2013. If you prefer, the original
thread is here. I rather like this writing style. I was attempting to copy the feel of a proper scientific paper.
Construction of an inexpensive adjustable speed magnetic stirrer
Abstract:
This paper describes the construction of an inexpensive adjustable speed magnetic stirrer for use with a non-ferrous hotplate or oil bath. Design decisions, Pitfalls, Schematics of the speed controller, and a bill of materials are included.
Action Shot:
Description:
A rectangular magnet is attached to the top of an inexpensive double ball-bearing CPU fan. The fan is driven by a PWM motor speed controller and powered by a 500 mW wall-wart power supply. JB-Weld adhesive is used to secure the magnet, and great care was taken to balance the assembly. A square of non-skid padding is attached to the bottom of the stirrer to prevent it from vibrating off the workbench.
Design Decisions and Pitfalls:
This fan was specifically selected for its dual ball-bearing construction. The unit was originally constructed with a junk-drawer sleeve-bearing fan. This unit failed very quickly.
The chosen fan is nominally rated for 3,000 rotations per minute. This speed would make the finished device unusable, sending both the stir bar and solutions flying across the lab. A simple potentiometer was attempted for speed control, but the stirrer would not rotate at any speed with the potentiometer connected.
Failing the simple speed control option, the second option was to use an Arduino to output a PWM signal to a simple motor driver circuit. The following circuit was built and tested on a breadboard.
To drive the circuit, the Arduino AnalogWrite function was used. See http://arduino.cc/en/Reference/AnalogWrite. Permanently dedicating an Arduino microcontroller to this project violates the “Low-Cost” objective of this project. Additionally, the movers broke my Arduino touch screen shield. :( A third option, a dedicated PWM generator, was designed, tested, and built.
An Altoids tin was selected as a project enclosure. This was lined with duct-tape to prevent shorting out the controller. Any container of suitable size would suffice, but the likelihood of exposure to heat and chemical spills suggests the use of a metal enclosure.
Cyanoacrylate “Super-Glue” adhesive was tried as the adhesive for the fan-to-drive-magnet junction. This failed repeatedly during testing. The first failure was at approximately 75% speed, impacting the wall behind the workbench. After reattaching, the magnet snapped off when the stirrer was accidentally dropped onto a wooden workbench. Hot-Glue was discarded as an option due to heat and chemical compatibility. Roughing the surface of the drive magnet and fan with 80-grit sandpaper and securing the magnet with JB-Weld adhesive provided a satisfactory solution.
The magnet must be centered very carefully on the fan to prevent excessive vibration. This is a painstaking process, and the JB-Weld allows plenty of time to make final adjustments. Before applying the adhesive, use a stir-bar to verify the magnet’s polarity. Improper orientation will cause the stir-bar to stand on end.
An attempt to use two magnets for greater magnetic flux was made, but this was too much load on the fan bearings. It would not rotate at low speeds.
Self-Criticism:
The author has assumed there is a fuse built-in to the wall-wart power supply. This is a foolish and unsafe assumption, and a discrete fuse will be added.
The author’s electronics skills are very limited, and skill in drawing schematics even more so. Those attempting construction of this project are strongly recommended to breadboard the circuits before construction.
Bill of Materials:
Note: Radio-Shack part numbers are included in parenthesis below, but other vendors may be significantly less expensive. Excluding the Fan, Magnet, and 12V power brick, all of the electronic components to breadboard this project are included in the Radio Shack “Electronics Learning Lab” Kit.
To permanently construct the motor controller off the breadboard, the following additional components are required.
Suggestions for Future Work:
A motor is a series of electromagnets energized in such a way to make a magnetic rotor rotate. It seems Rube-Goldberg-ian to physically attach this apparatus to a magnet for the purposes of rotating a third magnet. A more efficient design would use a number of electromagnet coils turned on and off in the correct sequence cause the stir bar to rotate. This apparatus could be physically constructed in a relatively inert substance like glass or ceramic to make a silent and vibration free stir-plate or stir-hockey-puck. It may be possible to adapt a stepper-motor controller IC for the difficult work of generating the correct polarity, sequence, and timing for this apparatus. To the author’s dismay, a patent search found an example of this idea in patent 3554497 and a number of similar products on the market.
A larger drive current could be used safely by limiting the PWM duty cycle. This would allow significantly higher starting and low-speed torque.
A humming sound can be heard when the device is in use, particularly at low speeds. Increasing the frequency of the PWM drive to > 20kHz should shift this humming sound out of the range of human hearing. With the addition of another drive transistor, a second fan/magnet could be added to agitate large solutions or an oil bath.
References:
Simple PWM DC Motor Driver Circuit
NE555 PWM Motor Controller Circuit
Homemade Show: 2$ Magnetic Stirrer Tutorial
Bonus Material (Not on SM):
Construction of an inexpensive adjustable speed magnetic stirrer
Abstract:
This paper describes the construction of an inexpensive adjustable speed magnetic stirrer for use with a non-ferrous hotplate or oil bath. Design decisions, Pitfalls, Schematics of the speed controller, and a bill of materials are included.
Action Shot:
Description:
A rectangular magnet is attached to the top of an inexpensive double ball-bearing CPU fan. The fan is driven by a PWM motor speed controller and powered by a 500 mW wall-wart power supply. JB-Weld adhesive is used to secure the magnet, and great care was taken to balance the assembly. A square of non-skid padding is attached to the bottom of the stirrer to prevent it from vibrating off the workbench.
Design Decisions and Pitfalls:
This fan was specifically selected for its dual ball-bearing construction. The unit was originally constructed with a junk-drawer sleeve-bearing fan. This unit failed very quickly.
The chosen fan is nominally rated for 3,000 rotations per minute. This speed would make the finished device unusable, sending both the stir bar and solutions flying across the lab. A simple potentiometer was attempted for speed control, but the stirrer would not rotate at any speed with the potentiometer connected.
Failing the simple speed control option, the second option was to use an Arduino to output a PWM signal to a simple motor driver circuit. The following circuit was built and tested on a breadboard.
 |
| Figure: Schematic of the dead-simple motor driver circuit. |
To drive the circuit, the Arduino AnalogWrite function was used. See http://arduino.cc/en/Reference/AnalogWrite. Permanently dedicating an Arduino microcontroller to this project violates the “Low-Cost” objective of this project. Additionally, the movers broke my Arduino touch screen shield. :( A third option, a dedicated PWM generator, was designed, tested, and built.
 |
| Figure: A low-cost PWM DM motor speed control circuit |
An Altoids tin was selected as a project enclosure. This was lined with duct-tape to prevent shorting out the controller. Any container of suitable size would suffice, but the likelihood of exposure to heat and chemical spills suggests the use of a metal enclosure.
Cyanoacrylate “Super-Glue” adhesive was tried as the adhesive for the fan-to-drive-magnet junction. This failed repeatedly during testing. The first failure was at approximately 75% speed, impacting the wall behind the workbench. After reattaching, the magnet snapped off when the stirrer was accidentally dropped onto a wooden workbench. Hot-Glue was discarded as an option due to heat and chemical compatibility. Roughing the surface of the drive magnet and fan with 80-grit sandpaper and securing the magnet with JB-Weld adhesive provided a satisfactory solution.
The magnet must be centered very carefully on the fan to prevent excessive vibration. This is a painstaking process, and the JB-Weld allows plenty of time to make final adjustments. Before applying the adhesive, use a stir-bar to verify the magnet’s polarity. Improper orientation will cause the stir-bar to stand on end.
 |
| Figure: Magnet Placement Polarity Failure Indication |
An attempt to use two magnets for greater magnetic flux was made, but this was too much load on the fan bearings. It would not rotate at low speeds.
Self-Criticism:
The author has assumed there is a fuse built-in to the wall-wart power supply. This is a foolish and unsafe assumption, and a discrete fuse will be added.
The author’s electronics skills are very limited, and skill in drawing schematics even more so. Those attempting construction of this project are strongly recommended to breadboard the circuits before construction.
Bill of Materials:
Note: Radio-Shack part numbers are included in parenthesis below, but other vendors may be significantly less expensive. Excluding the Fan, Magnet, and 12V power brick, all of the electronic components to breadboard this project are included in the Radio Shack “Electronics Learning Lab” Kit.
- Inland brand “80mm Performance Fan”. (Micro-Center $2.99)
- Magnet Source 3/8” x 7/8” x 1 7/8“ Permanent Ceramic Magnet (MagnetSource.com CB60N)
- JB-Weld Adhesive (Home Depot)
- 12 volt 500 mA wall-wart power supply (#273-357)
- NE555 timer IC (#276-1723)
- 1 NPN Transistor (#276-1617)
- 3 1N4001 Diodes (# 276-1653)
- 2 0.1uF Ceramic capacitors (#55047557)
- 1 0.01uF Ceramic capacitor (#55047551)
- 1 10K Ohm resistor (#271-306 assortment Pack)
- 1 1K Ohm resistor (Included in above assortment)
To permanently construct the motor controller off the breadboard, the following additional components are required.
- Altoids tin or other project enclosure
- Printed circuit board or perfboard (#276-148)
- 100K ohm potentiometer (#271-092)
- SPST toggle switch (#275-324)
- Solder, soldering iron, & hookup wire.
- Non-Slip drawer liner for “feet” (Source unknown)
Suggestions for Future Work:
A motor is a series of electromagnets energized in such a way to make a magnetic rotor rotate. It seems Rube-Goldberg-ian to physically attach this apparatus to a magnet for the purposes of rotating a third magnet. A more efficient design would use a number of electromagnet coils turned on and off in the correct sequence cause the stir bar to rotate. This apparatus could be physically constructed in a relatively inert substance like glass or ceramic to make a silent and vibration free stir-plate or stir-hockey-puck. It may be possible to adapt a stepper-motor controller IC for the difficult work of generating the correct polarity, sequence, and timing for this apparatus. To the author’s dismay, a patent search found an example of this idea in patent 3554497 and a number of similar products on the market.
A larger drive current could be used safely by limiting the PWM duty cycle. This would allow significantly higher starting and low-speed torque.
A humming sound can be heard when the device is in use, particularly at low speeds. Increasing the frequency of the PWM drive to > 20kHz should shift this humming sound out of the range of human hearing. With the addition of another drive transistor, a second fan/magnet could be added to agitate large solutions or an oil bath.
References:
Simple PWM DC Motor Driver Circuit
NE555 PWM Motor Controller Circuit
Homemade Show: 2$ Magnetic Stirrer Tutorial
Bonus Material (Not on SM):
 |
| Completed Magnetic Stirrer External Shot |
 | ||
| Completed Magnetic Stirrer Internal Shot |
 |
| How not to breadboard (USE SHORTER WIRES!) |
Comments