Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Electromagnets and switching polarity?

  1. Apr 5, 2008 #1
    [SOLVED] Electromagnets and switching polarity?

    Hi people,

    I am relatively new to the forums and am experimenting with a new toy I want to try to make. I am stuck on the subject of changing polarity quickly in an electromagnetic coil and automatically. Let me explain.

    I have two rubber balls, hand made and with an electromagnetic coil in the centre of the ball. They are approximately the size of a golf ball and weigh just under 200g each so should roll easily. My aim is to use a small powering circuit that will allow me to set one ball at a constant polarity at the same time as being able to charge the other ball will the same voltage but as a swappable voltage, so that it changes the coil from polarity to polarity so that it will attract and then repel the other ball that has the constant polarity. I want to make the balls bounce toward each other and then away from each other.

    I plan to limit the voltage so as to make the impact of the balls safe to be around adults fingers, so that they cannot be bruised but that they will feel the balls impact. The balls will only be able to travel a maximum of 4 cm apart due to walls and so I am hoping I can limit the voltage to acheive the minimum impact.

    The part I am having trouble on is this! What is the easiest way to swap voltages to the ball with the varying polarity? Is there an IC that will do this or a basic circuit I should study to understand how to do what I need? Would it be easier to create two coils inside the second rubber ball that are wired in reverse and simply run 4 leads into the ball and swap which leads get power to them or was I thinking right the first time, is there an easier way.

    I would also like to control the speed at which they attract and repel and so I am envisioning my solution as being a circuit as opposed to a component?

    Any help, advice or ideas would be a great help.
    Thanks in Advance
  2. jcsd
  3. Apr 5, 2008 #2


    User Avatar
    Science Advisor
    Homework Helper

    The mention of wires says you don't expect these balls to rotate.
    So the circuit you want is an H-bridge.
    You can buy one made up.
    You would also need a clock circuit(or manual switch) to control the H-bridge polarity and a power supply.
    Do a search on H-bridge for more info.
  4. Apr 5, 2008 #3
    Thank you very much for the quick reply NoTime. I will go have a study of H-Bridges and see if this is what I need. And yes you are correct in assuming the are not made to spin, they may of course do so in use but only to a certain degree, they are not freely rotating on the wires. I will post back again as soon as I have had a study and let you know how I am progressing. Thank you again.

    Ok back again.... I just had a look at a shematic for a H-Bridge Push / Pull Converter and I beleive from the shematics that this is what I need. I found the schematics for this converter at the following address: http://en.wikipedia.org/wiki/Image:Push-pull_converter_schematic.svg

    Will this however change the voltage of the output as it says it is an amplifier as well or is that simply due to the windings of the secondary coil being larger than the first. If so if my primary and secondary coils are the same as in an isolator transformer, will the output voltage remain the same?

    Also, using an H-Bridge seems to swap my polarity and so my first issue seems to be solved now but my second issue on this design is this:

    I want to leave the output voltage as say 9v IN to 9v Out but I also want to be able to adjust the speed at which my circuit will alternate between polarities. Without adjusting the input voltage, am I able to make this circuit do this? Am I thinking correctly in beleiving that this would be done by amperage control?

    I must also clear up that my input is DC 12v or DC 9v, I am electing for the 9v supply, so that the end toy can be powered from a 9v battery. Drainage time on the battery will not be so much of an issue. If it runs solidly for one day on a single 9v battery then it has worked exceedingly well for what it needs to do.

    Hope this helps, and thank you very much for your help so far. I am not looking for the answers to be handed to me, just a helpful hand in pointing me in the right direction, so thanx again NoTime, I appreciate your quick reply.

    Thanx again!!!!!!
    Last edited: Apr 5, 2008
  5. Apr 6, 2008 #4
    Um, I don't think that circuit is the one you want. You just need the two switches on the right-hand side. The switches would be inside a relay, but the switch contacts need to be protected against the reactive current of the coil using diodes.

    You might find a url that addresses transistor full wave bridges. But on second thought you might be better off sticking to a double-pull double-throw relay, and driving the relay with a 555 timer. Otherwise you're in way over your head, i'd guess.
    If I dont miss my guess, you're hanging these two balls on wires and their frequency of oscillation will depend on how they behave as pendulems rather than the frequency of your circuit. You'll have some oscillator driving this bridge. The frequency should be adjusted with a potentiometer on the 555 timer IC. Or you could sense contact between balls and change the polarity at that time.
  6. Apr 8, 2008 #5
    Thanx Phrak,

    You gave me a good angle to think from here. I also thought about a relay but with the relay being mounted at the circuit end as opposed to the coil end, which would be about a metre away from the circuit.

    There will definately be a lot of research needed yet on how to sense contact between the balls and I am looking at making the circuit recognise contact between the balls and swap polarity on impact, then swap polarity again at a particular timed interval set by a 555 chip. My new angle of approach on this is to soften the walls of my playing feild so that impact between the ball and the wall is lesser than impact between the two balls. Then set inside the rubber would be a switch that is basicaly a hollowed out rectangle with a contact top and bottom, so that when the ball impacts with a hard enough surface, that the rubber will compress and push the contacts together and this will be the base for a NPN transistor with a large voltage collector to emitter, that will then feed a rverse polarity coil.

    This is as far as I have gotten with this idea of switching polarities. However you did raise a pont which I had not consider being that I really need to keep the controller circuit away from the coils to avoid magnetic interference. So I am redesigning it a little.

    Thank you both for the great leads on this concept, you have given me a lot to think about and I am sure have given at least some of us less technical minded people a direction to look in to learn this stuff.

    I appreciate the help and will keep posting as I get results or find a new area that makes me stumble. Also thank you for the inspiration, I am always open to new ideas and ways of approaching issues.
  7. Apr 8, 2008 #6
    A glancing blow between the two balls would make them stick if you didn't sense the impact. You need to change polarity at the slightest excuse.

    You don't need to worry about magnetic interference. Heck, you could glue your components to the side of the relay body, if it would help compactify things.

    Both the magnet coil and the relay coil are inductive loads. They need to be decoupled with capacitors to get ride of switching noise over the supply wires.

    Put a diode across the relay coil to keep the inductive reactance from burning whatever semiconducor that drives it. You'll want 2 diodes at both ends of the coil. One going to the positive supply, and the other to ground.

    You'll want some sort of inertial sensors to tell you that you have an impact. I double you could buy one that's cost effective for a consumer product. But if you're just doing the prototype you can leave that to the design engineers. I envision something like an old light bulb filament that would go into motion and hit the sides of a conductive container from a tiny impact from nearly any direction. You could clean up the signal if necessary with a one-shot (like a 555) or a flip-flop.

    You need the 555 you already mentioned to set a delay period which prevents the field from switching for a short time interval after a field switch, I think. Also, I see these balls locking-up, stuck together, after bouncing around a while. After perhaps 4 seconds, if nothing happens, always change the coil polarity.

    I really don't know the details of what you're making so I'm just guessing are what's required.

    With all this going on, you might want a tiny microcontroller instead of the LSI. These things come in packages about 1/4 inch square with onboad clocks and all, and programmable through your PC with a cheap interface board. Microchip sells the simplest pogrammable things around. Motorola has tint stuff too. Somebody else would be better at telling you the best selection. You don't want to go and learn motorola's codewarrior stuff for a one-time thing.
    Last edited: Apr 8, 2008
  8. Apr 9, 2008 #7
    Thanx again Phrak

    Your knowledge is invaluable to me as I stumble through this vast field of electronics and magnetism. Yes it is very hard to see what I want done as I am deliberately holding back on a lot of the information about what this device will be used for, which makes it even harder to explain what it should do.

    Phrak if you email me at al-street@hotmail.com, I will explain my design more fully to you and once I do you will understand a lot more clearly what I need. The reason I am being so secretive, is not to hide my idea from the everyday user but because these devices are not yet being produced as far as I can tell and I dont want to hand my idea out to any big companies scouring the net for inspiration on products to create, at least not until I have a crack at it.

    Otherwise, for now I am letting this project rest for a while until I have learnt some more about components like IC's and what can do the tasks I need done. I am also learning about magnetics and fields so I can now see that I have a lot to learn yet to make this idea a success.

    Thanx again guys, I appreciate the help and advice.
  9. Apr 10, 2008 #8
    Sorry if I gave too much away. I've been too many years designing to what the customer really wants instead of what he asks for. You can send private messages within the physics forum. It's under Quick Links in the menu bar.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook