1
\$\begingroup\$

What's the simplest and cheapest way to use one or more electromagnet to to stir a stir bar ? That is, how to control the speed at which the polarities are reversed.
I don't want any mechanical part (no motor) or software (no PIC microcontroller) and the speed must be controllable by hand.

I could use a fan with a magnet but I'm curious to see an e-magnet at work.
I can figure out the rest of the circuit.

\$\endgroup\$
  • 1
    \$\begingroup\$ When you say "I can figure out the rest of the circuit", which part of the circuit do you need help with? Or do you need help with the mechanical design of the motor? Do you mean this type of Stir Bar? (It helps to provide links with more obscure stuff which many people won't know about) \$\endgroup\$ – Oli Glaser Mar 7 '13 at 6:17
  • \$\begingroup\$ Would it be the same structure as a three-phase variable frequency electric motor? \$\endgroup\$ – pjc50 Mar 7 '13 at 9:34
4
\$\begingroup\$

This is no different than controlling a brushless DC motor open loop. Actually, this is a brushless DC motor.

I would not use a single electromagnet. With a single pair of poles, you can only make the field flip, not actually spin at some controlled rate. Or put another way, you can make the field rotate at any frequency you want, but it won't rotate smoothly. It will always jump 180° at a time and the torque goes to 0 twice per cycle.

The minimum to make a rotating field requires 3 poles, which is why so many motors have 3 wires going to the field windings. The three windings would be arranged in a even "Y" layout. You can connect the inside ends of each of the coils together and control everything from the 3 remaining outside connections. Each of these 3 lines is then connected to a half bridge, meaning a high side and a low side switch that can be controlled independently.

Since the rotation speed is going to be very slow compared to the speed of a microcontroller, you have the micro step thru all the phases of a rotation sequentially. The easiest is a 12 step process. Each line is driven with a OPEN - HIGH - OPEN - LOW repeating sequence. Draw it out on paper and you can see how you make one change to one line each step, which eventually walks the magnetic field thru steps of about 30° each. One advantage of this 12 step scheme is that it naturally does break before make. There are other scheme that can deliver more power thru the same set of windings, but the 12 step scheme is very easy to implement and more forgiving of screwups, making it all around better for beginners.

\$\endgroup\$
  • \$\begingroup\$ Ok and how would I know what I need exactly ? cause I am told that coils are difficult to simulate and should be built by hand for experimentation \$\endgroup\$ – WaterBearer Mar 9 '13 at 1:33
  • \$\begingroup\$ @user: For starters, you can experiment by winding magnet wire around three ferrite sticks, then arranging them in a "Y" layout mechanially. Doing this well and efficiently will take some careful mechanical design. Electrically, it's just three coils with one end connected to a common node. \$\endgroup\$ – Olin Lathrop Mar 9 '13 at 14:01
0
\$\begingroup\$

It seems that the basic blocks would be: variable frequency oscillator, power amplifier, electromagnet. It helps that the oscillator doesn't have to tune to exact frequencies or generate a perfect waveform, and the amplifier doesn't have to meet any stringent distortion and noise figures.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.