Electro magnet project

Thread Starter

JCOX

Joined Nov 29, 2011
136
Hello AAC, with this being my first post, allow me to introduce myself, My name is J. Cox and I am active duty military, I teach automotive mechanics to new military member starting their career. Out of all the automotive objectives we teach, I particularly enjoy teaching basic electrical we basically scratch the surface with theory, laws, follow current flow, a few semiconductors (transistors diodes etc..)

With that said I would love to start a personal project to incorporate into my lesson, however it is beyond my scope of expertise, and hopefully I will be able to have AAC help and guide me along the way.

Because automobiles use electro magnets to control essentially everything on a vehicle these days, I want to create a series of electro magnets that pulse on and off spinning a permanent magnet on its axis, and to complicate things, I don’t want it to be something simple as a DC motor.

Here’s what I propose:



Each colored circle represents an electro magnet… and each color has 2 electro magnets on a circuit (one spun clockwise and the other counter clockwise) to create a north and south pole, and when the direction of current is changed they will flip.

This part is done, here are my 6 electro magnets:




Note: that I bumped up to a thicker gauge wire than what is in the picture.




What I need help with is to design a circuit to have the electro magnets energize and collapse in succession so that north and south poles are constantly rotating counter clockwise in order to rotate a permanent magnet.

I want this to be able to speed up, slow down with the help of a potentiometer, and hopefully randomly run at varying speeds on its own.

Is this something that you guys can help me with? If there are circuits already out there that I can modify slightly to fit my needs that would be great. But if we have to go from the ground up, I’m game. Just please keep in mind that I’m enlisted and I’m not serving my country for the pay check ;), so I don’t have money to throw around. “On a budget” is key. The good news is that I already have a DC power source that will be perfect for this application. It is 12vdc with 7 amps constant. I have already used it to simulate the electro magnets and it works great.

I am eager to get this project started...
 

gerty

Joined Aug 30, 2007
1,305
Do a search here for Bill Marsden's articles on 555/4017 circuits.
He's written several articles for sequencing leds, and that's kinda what you need. We'll need to know how much current the coils draw so tha power transistors can be added to the output of the 4017. The 555 acts as an oscillilator/clock and the 4017 is a Johnson counter. It has 10 outputs but you'll only need 3.
 
Hello,
to make it go cw & ccw and vary speed can be done by logic circuits like 555 and 4029 up/down counter and decoder 4514-4515 and mosfets to supply current
random action needs micro controller to generate the random behavior
 

Thread Starter

JCOX

Joined Nov 29, 2011
136
Thanks guys, to start out with Gerty, if I understand correctly This is the the circuit you're talking about, If so, can I simply replace the 1M ohm resistor with a potentiometer to vary the sequencing speed? and another question is: If I were to only use 3 of the 10 outputs of the 4017 is it going to just sequence those 3 or will there be a delay between 0 and 2 while it runs through the other 7?

Another question would be: how would I switch the polarity of the 3 circuits so that north pole on the electromagnet changes to south and vise versa?


Thanks for your help!!!
 

Thread Starter

JCOX

Joined Nov 29, 2011
136
Hello,
to make it go cw & ccw and vary speed can be done by logic circuits like 555 and 4029 up/down counter and decoder 4514-4515 and mosfets to supply current
random action needs micro controller to generate the random behavior
Thank you so much Maged, are you suggesting that I can incorporate this to the link I posted above, or is this simply another way to skin the cat?

I don't need it to switch directions, however I would almost make it a necessity to have it spin at random varying speeds.

With the link that I posted, do I simply replace the LED's with transistors, with the signal coming from the 4017 going to the base of each to allow for the transistor to act as a switch for the higher 12 volts?

Thank You so much for your help as well.

J. C.
 

gerty

Joined Aug 30, 2007
1,305
That is the basis of what I was referring to. Yes changing that resistor value changes the speed. Back to the coils, we'll need to know the resistance of the coils to calculate what other changes will be needed. The output of the 4017 is only capable of driving a few(10?) milliamps, and I fairly sure your coils will need more than that. As for counting, the reset pin is tied to the last output, and that starts the counting sequence all over again.
Here's the data sheet for the 4017
http://www.ti.com/product/cd4017b?CMP=AFC-conv_SF_SEP&partnerName=DSA
 

Thread Starter

JCOX

Joined Nov 29, 2011
136
Thank You Gerty,
I meant to post the resistance but forgot to. each coil has 0.7 ohms of resistance.

Here is what I drew up, let me know if I'm at least on the right track.

 

praondevou

Joined Jul 9, 2011
2,942
The circuit will apply voltage to two opposite coils which have to be connected in opposite polarity to create a north and a south pole.

If the coils resistance is 0.7 Ohm that would be 17A if 12VDC was applied! Since you are applying pulses it would be less, this however depends on the coils inductance which is unknown.

You mentioned that you tested the electromagnets with your power supply. Question is at what voltage/current? How much current do you need to create the desired magnetic field strength?

I put 3V for the coil power supply because of the diodes current rating. In order to choose the proper diode we will need to know the current through the coil.

The switch is a DPDT switch. It allows to invert the switching sequence.

I'm still thinking about the random part.

EDIT: I just noticed that the FETs gate signals are to low. They would better be driven by at least 12V. I.e. you will need to use other FETs or a level shifter.



gate signals
 

Attachments

Last edited:

praondevou

Joined Jul 9, 2011
2,942
Here comes a modified schematic. As I mentioned, 5V VGS is not enough.

15V would be ideal, 12V should do it.

missing information:
- desired current through coils to determine what components to use for the power part
- specify the random part



I have to call it a day. 0:40
 

Attachments

Thread Starter

JCOX

Joined Nov 29, 2011
136
Praondevou,
Thank you so much for drawing that out for me.
The circuit will apply voltage to two opposite coils which have to be connected in opposite polarity to create a north and a south pole.
They are, out of the 3 sets of coils, each set has 2 coils, one would clockwise and the other counter, making them the opposite when current flows through either direction.
You mentioned that you tested the electromagnets with your power supply. Question is at what voltage/current? How much current do you need to create the desired magnetic field strength?
I just went down and tested the current and voltage, the current flowing through the coils are 1.09A and the voltage is 13.85VDC. The 1.09A seems to be plenty to make the permanent magnet oppose and move to attraction.
I put 3V for the coil power supply because of the diodes current rating. In order to choose the proper diode we will need to know the current through the coil.
I'm assuming that because the current and the voltage is higher, that the diodes and IRF540's can't be used???

I honestly thought that I wasn't going to get the help that I so desperately needed, and this project was never going to come to fruition, but I have a feeling that it will get done!!!
Thanks to all of you.
 

praondevou

Joined Jul 9, 2011
2,942
through the coils are 1.09A and the voltage is 13.85VDC. The 1.09A seems to be plenty to make the permanent magnet oppose and move to attraction.
I'm assuming that because the current and the voltage is higher, that the diodes and IRF540's can't be used???
Well that's perfect then. Instead of the 3V in my drawing use 12V then.

I still don't get why 0.7Ohm (from your other post) gives 1A @ 13.85V...

With this current the IRF540 should be ok. As a diode a schottky like the 1N5820 would be better.

What about the random feature?
 

THE_RB

Joined Feb 11, 2008
5,438
One thing that jumped out is that your 3-step driver always has the north and south coils in the same places, when that pair is energised. So a permanent magnet will not spin (as you originally wanted) it will step 1-2-3 then jump back to 1.

To make the rotor spin, it needs to go 1-2-3-4-5-6 etc.

There are two ways to do it, the easiest way is not to use a bar magnet as the rotor but a plain bar of steel, so it won't care about the N and S poles, it will just step to the nearest powered coils. That will do 1-2-3-4-5-6 fine.

The hard way is to use a bar magnet rotor, then you need to reverse the polarity of the coils, which requires a full-bridge for each coil pair, so that is 3 full-bridges. That is a heap more complexity.

In the interest of keeping it very simple I would use a plain steel rotor, and the simple 1-2-3 step driver as you were already discussing (to give 1-2-3-4-5-6 motion). That is assuming it is still going to demonstrate the principles you need to the students.
 

praondevou

Joined Jul 9, 2011
2,942
THE RB is right.

I see 3 possibilities if a bar magnet is being used as rotor.

Use a fullbridge as was mentioned.
Use a halfbridge (requires two power supplies)
Wind another coil on the same rod in the same sense and connect the middle point to 12VDC.

Unless I got totally confused with all the coils this should create a rotating magnetic field. I only drew it for one coil pair otherwise it gets to cluttered.
(Diodes are also missing for the other legs. It's one diode for each MOSFET)
The only thing is that the way I drew it the coils are no longer in series so current will increase with same DC voltage.

 

Attachments

shortbus

Joined Sep 30, 2009
10,045
Wouldn't be simpler to add another set of coils (8 total) and use stepper motor driver chip? Then you would just need a 555 timer to supply pulses to the step driver.
 

praondevou

Joined Jul 9, 2011
2,942
Wouldn't be simpler to add another set of coils (8 total) and use stepper motor driver chip? Then you would just need a 555 timer to supply pulses to the step driver.
Why 1 more set of coils? Can you elaborate?

The stepper motor driver idea crossed my mind too. Do you have a particular chip in mind? The advantage would be that the driver bridge is already built-in. No external MOSFETs necessary.
However, are you referring to the chip you recently purchased for your project? That has 2 fullbridges, right?

Here we need 3 fullbridges, is that 2 ICs? I wonder if it's possible to configure the right sequence for the OPs' circuit.

I preferred the 555/4017/FET combination because it's easy to understand how it works (unlike some stepper motor driver/controller datasheets I saw)
 

Thread Starter

JCOX

Joined Nov 29, 2011
136
One thing that jumped out is that your 3-step driver always has the north and south coils in the same places, when that pair is energised. So a permanent magnet will not spin (as you originally wanted) it will step 1-2-3 then jump back to 1.

To make the rotor spin, it needs to go 1-2-3-4-5-6 etc.

There are two ways to do it, the easiest way is not to use a bar magnet as the rotor but a plain bar of steel, so it won't care about the N and S poles, it will just step to the nearest powered coils. That will do 1-2-3-4-5-6 fine.

The hard way is to use a bar magnet rotor, then you need to reverse the polarity of the coils, which requires a full-bridge for each coil pair, so that is 3 full-bridges. That is a heap more complexity.

In the interest of keeping it very simple I would use a plain steel rotor, and the simple 1-2-3 step driver as you were already discussing (to give 1-2-3-4-5-6 motion). That is assuming it is still going to demonstrate the principles you need to the students.
You are correct, I want the current in the coils to switch directions in order to change magnetic poles so that it does what you explained.

That is assuming it is still going to demonstrate the principles you need to the students.
the principle is using repulsion to rotate a magnet on it's access. So I would really want to stay with the permanent magnet if at all possible.
 

Thread Starter

JCOX

Joined Nov 29, 2011
136
Well that's perfect then. Instead of the 3V in my drawing use 12V then.

I still don't get why 0.7Ohm (from your other post) gives 1A @ 13.85V...

With this current the IRF540 should be ok. As a diode a schottky like the 1N5820 would be better.

What about the random feature?
The power supply that I'm using is an Astron RS-7A. We have an abundance of them at work, the students use them to build simple series and parallel circuits to prove the laws in the 2 different circuits.

The sticker on the bottom of the power supply says:
OUTPUT: 13.8 VDC
5 AMPS CONTINUOUS
7 AMPS ICS @50% DUTY CYCLE

As far as the randomness, just for the sake of pulling a number out, If 1000 RPM was at 100%, I would like to have it slowly ramp up to 100% slowly drop to 23% ramp up to 55% then down to 40% then up to 80% and so on, completely random though. Is that at all possible?


Your last schematic will work if I were to re-wrap the electromagnets to have 2 sets of coils (each wrapped in different directions) and set them up to output 0 and 3 that will change the magnetic poles. The only problem I see is that I would have to double the amount windings around each iron core to maintian the 0.7 ohms of resistance.
 
Last edited:
Top