Science Buddies: "Ask an Expert"

Posted: **Mon Mar 05, 2007 7:57 am**

I had some good results in creating three shake-a-lights using the Science Buddies outline, but I came up with some unusual resluts in some cases, and I was hoping someone might help me explain the.

First, I created three shake a lights, but I used 4-inch-long PVC tubes with PVC caps at both end (in lieu of 35mm film tubes), whiuch added another 2 inches total to the tube length. I used three, six, and nine 1-inch-diameter D250H magnets from Amazing magnets, which fit inside the tubes' 1.25-inch opening.
Tube 1 was wound with 560 windings of 30AWG. Tube 2 was wound with 280 windings of 30 AWG. Tube 3 was wound with 280 windings of thicker 26 AWG. The wire was enameld copper as specified on the web site, and the insulation was scraped away at the ends of the wire to attach an LED from RadioShack, rated at about 3 milliamps or so.
Into each tube, I placed three magnets, then six, and then nine to determine the effect of the number of magnets on the voltage generated. We shook the tubes to determine brightness levels, and we used a small voltmeter to quantify the results.
So here's what we found (in terms of voltage levels):

30 AWG (560 windings) 30 AWG (280) 26AWG (280)

3 magnets 1-volt peaks 0.1-volt peak 0.3-volt peak
6 magnets 1.9-volt peaks 0.2-volt peak 2.9-volt peak
9 magnets 1-volt peaks 0.1-volt peak 1.5-volt peak

With 6 magnets, I got more voltage than three, but voltage went down with nine magnets. Any reason why? Perhaps because the pack of 9 magnets could not completely go outside the coil, reducing the amount of change in the magentic field inside the coil?

Also, why would the voltage produced be lower when using 280 windings of 30AWG versus 560 windings? Thank you for your help.

Posted: **Mon Mar 05, 2007 8:20 am**

Joseph Palenchar wrote:I had some good results in creating three shake-a-lights using the Science Buddies outline, but I came up with some unusual results in some cases, and I was hoping someone might help me explain them.

First, I created three shake a lights, but I used 4-inch-long PVC tubes with PVC caps at both end (in lieu of 35mm film tubes), whiuch added another 2 inches total to the tube length. I used three, six, and nine 1-inch-diameter D250H magnets from Amazing magnets, which fit inside the tubes' 1.25-inch opening.
Tube 1 was wound with 560 windings of 30AWG. Tube 2 was wound with 280 windings of 30 AWG. Tube 3 was wound with 280 windings of thicker 26 AWG. The wire was enameld copper as specified on the web site, and the insulation was scraped away at the ends of the wire to attach an LED from RadioShack, rated at about 3 milliamps or so.
Into each tube, I placed three magnets, then six, and then nine to determine the effect of the number of magnets on the voltage generated. We shook the tubes to determine brightness levels, and we used a small voltmeter to quantify the results.
So here's what we found (in terms of voltage levels):

30 AWG (560 windings) 30 AWG (280) 26AWG (280)

3 magnets 1-volt peaks 0.1-volt peak 0.3-volt peak
6 magnets 1.9-volt peaks 0.2-volt peak 2.9-volt peak
9 magnets 1-volt peaks 0.1-volt peak 1.5-volt peak

With 6 magnets, I got more voltage than three, but voltage went down with nine magnets. Any reason why? Perhaps because the pack of 9 magnets could not completely go outside the coil, reducing the amount of change in the magentic field inside the coil?

Also, why would the voltage produced be lower when using 280 windings of 30AWG versus 560 windings? Thank you for your help.

Posted: **Mon Mar 05, 2007 9:37 pm**

Dear Joseph,

I believe that the amount of voltage is most likely directly related to the number of turns. That is the more turns the greated the voltage.

Here's a document that might help to explain this concept:

http://www.abdn.ac.uk/physics/s6/labeandm.pdf

Check out the section that says: If the loop is a flat coil of N turns, then the total flux through the coil is N times the flux for a single turn.

Hope this helps,
Janelle

Posted: **Tue Mar 06, 2007 10:12 am**

Dear Janelle:
Thank you for the link. It provided some insights and good illustration.
However, I am looking for guidance on a plain-English explanation of the magentic induction effect and why the voltage would rise when 6 magents are used instead of three and then fall when 9 magnets are used.
Can anyone help?

Posted: **Tue Mar 06, 2007 1:32 pm**

Joseph,

That does seem counterintuitive, to say the least!

What data did you collect?

Posted: **Tue Mar 06, 2007 1:54 pm**

Thanks for your interest.
I've pasted my original query below.
I think that because the 9 magnets were so long that many of them stayed farther within the coil when shaken, the resulting change in the magnetic field was not so great, and therefore the voltage produced would be lower.
Re question 2, I think I got more voltage with the thicker wire versus the thinner wire because there was less resistance.
But, I'll let you look the data over (below). Thanks again

First, I created three shake a lights, but I used 4-inch-long PVC tubes with PVC caps at both end (in lieu of 35mm film tubes), whiuch added another 2 inches total to the tube length. I used three, six, and nine 1-inch-diameter D250H magnets from Amazing magnets, which fit inside the tubes' 1.25-inch opening.
Tube 1 was wound with 560 windings of 30AWG. Tube 2 was wound with 280 windings of 30 AWG. Tube 3 was wound with 280 windings of thicker 26 AWG. The wire was enameld copper as specified on the web site, and the insulation was scraped away at the ends of the wire to attach an LED from RadioShack, rated at about 3 milliamps or so.
Into each tube, I placed three magnets, then six, and then nine to determine the effect of the number of magnets on the voltage generated. We shook the tubes to determine brightness levels, and we used a small voltmeter to quantify the results.
So here's what we found (in terms of voltage levels):

30 AWG (560 windings) 30 AWG (280) 26AWG (280)

3 magnets 1-volt peaks 0.1-volt peak 0.3-volt peak
6 magnets 1.9-volt peaks 0.2-volt peak 2.9-volt peak
9 magnets 1-volt peaks 0.1-volt peak 1.5-volt peak

With 6 magnets, I got more voltage than three, but voltage went down with nine magnets. Any reason why? Perhaps because the pack of 9 magnets could not completely go outside the coil, reducing the amount of change in the magentic field inside the coil?

Also, why would the voltage produced be lower when using 280 windings of 30AWG versus 560 windings? Thank you for your help.

Posted: **Wed Mar 07, 2007 9:01 am**

Also, why would the voltage produced be lower when using 280 windings of 30AWG versus 560 windings?

You haven't descibed the complete "circuit" involved for an engineer to do any calculations to answer your question accurately. The coil with the moving magnets is a current source. The LED and/or the meter is a load. The impedance of both the load and the source will affect the voltage. Maximum power transfer will occur when the source and load impedances match. The impedance of the two sources are different, so one would expect different results.

I'm sure your eyes are glazed over as you got lost in the above technical jargon....

Lets start over. If you wrap a single turn coil and a two turn coil out of the same wire next to each other on the same insulating core (same diameter). Now if you move a magnet through the core, the changing magnetic field will be the same for both coils. However, the two turn coil will have twice the induced effect compared to the single turn coil.

The above assumed that the magnetic field changes were consistent for the area of the two coils. This does not necessarily hold true for longer coils.

With 6 magnets, I got more voltage than three, but voltage went down with nine magnets. Any reason why?

Magents interact with other magnets which can effect the orientation and strength of the "magnetic flux" lines. Stick a bunch of identical magnets in a chain (N-S)(N-S)....(N-S) and the external magnetic flux is actually weaker because it is just like one long magnet. For the most part, only the two outside ends, which are farther appart, contribute to the external magnetic flux pattern.

Posted: **Wed Mar 07, 2007 9:30 am**

Thanks, Craig. This is helpful.
Given that I haven't had to delve into basic science in recent DECADES

, you're helping me frame some insights.

So if I can summarize a point or three, thinner wire creates more resistance to the electron flow, thus (all other things being equal), I would get less voltage through a coil with thinner wire. Correct? (This is basic, but I want to doublecheck so my son gets a good grade.)

Next, the strength of the magnetic field is greatest at the north-south poles of a magnet, so in going to nine magnets from six, I may have actually reduced the strength of the north and south magnetic fields to a level lower than the north and south fields of a string of 3 magnets? An insight on how so?

And, is it the change in the strength of the magnetic field, not the strength of the magentic field itself, that creates voltage in the coil? By change, are we talking about going from a zero value to a value above zero (like an on/off situation) as the north magnetic field passes through a coil, then the south magnetic field passes through the coil as I shake the tube in alternate directions? Hence, alternating current?

Finally, doesn't a magnetic field actually contain an electric charge, which in my experiment would be conducted by the coil to the LED? So would the change (or back and forth movement) of the magnet's electric field "excite" the electrons in the coil and moves them forward, then backward, when I shake the tube?

I am trying to frame the questions in the simplest way possible so the information can be shared in the project's report.

Thanks for your time.
Sincerely,
Joe

Posted: **Wed Mar 07, 2007 11:58 am**

thinner wire creates more resistance to the electron flow, thus (all other things being equal), I would get less voltage through a coil with thinner wire. Correct?

The difference in the pure resistive component of the wire usually doesn't make as much difference as the inductive component. Thinner wire probably meant you have a greater turns per inch which will increase the inductive component. The impedance of the load and the mismatch between the impedance of the coil are the more likely cause based on my experience.

I may have actually reduced the strength of the north and south magnetic fields to a level lower than the north and south fields of a string of 3 magnets? An insight on how so?

Correct.

And, is it the change in the strength of the magnetic field, not the strength of the magentic field itself, that creates voltage in the coil?

Yes. Differential form of Faraday's law -dB/dt, the change in the magnetic flux density per unit time what is involved. Given the earth has a magnetic field, the only place and time you have a zero density is when the magnetic flux density from the magnet is precisely equal and opposite polality of the earth's field. This is hard to visualize. A messy experiment is to spread some iron filings evenly on a piece of cardboard and orient a bar magnet so the N-S poles are parrallel to the plane of the cardboard and slowly approach the cardboard from the bottom to reveal lines of flux. I spent a couple minutes looking for a picture of this on the web but didn't find one quickly.

doesn't a magnetic field actually contain an electric charge?

No! Maybe you are thinking about the complex relationship between H (magnetic field) and E (electric field). Look up Maxwell's equations if you want the complete picture.

So would the change (or back and forth movement) of the magnet's electric field "excite" the electrons in the coil and moves them forward, then backward, when I shake the tube?

Yes. That is a simple statement of Michael Faraday's induction principle.

The area you are having the most trouble describing is the magnetic flux density. Something that many EE's have trouble with because it is hard to demonstrate and visualize.

Posted: **Wed Mar 07, 2007 12:27 pm**

Thanks Craig. There was a lot in your last message that I could only "deduce" from the many sources that I've read. No source has come out and stated some of these things as clearly as you.

Regarding thinner wire having higher resistance, that was the only reason I could figure out for getting a higher voltage out of 280 windings of 26AWG compared to 280 windings of thinner 30AWG. Is resistance the likely cause in my experiment in this case?

Another thought: Is it accurate to say that when the magnet is running in one direction through the coil, it is repelling the coil's electrons in the opposite direction. Then, when the magnet moves in the other direction, the coil's electronocs flow in the direction opposite. Hence, AC.

Thanks again.

Posted: **Wed Mar 07, 2007 1:59 pm**

Is resistance the likely cause in my experiment in this case?

No. It is the inductance of the coil. Impedance of a coil with this many turns has a small resistive component and a large inductive component. See equation (2) in http://www.midnightscience.com/coilinfo.html for one formula that relates the number of turns, the radius, and the length to the inductance. This "reactive" or frequency dependent component of impedance will be much larger than the resistance.

Is it accurate to say that when the magnet is running in one direction through the coil, it is repelling the coil's electrons in the opposite direction.

No. See http://hyperphysics.phy-astr.gsu.edu/hb ... agfor.html and http://hyperphysics.phy-astr.gsu.edu/hb ... or.html#c3. Also remember that electrons are negatively charged so current is flowing in the opposite direction of the electrons. Search for "left hand rule fields" if you want things reversed for what happens to an electron.

Posted: **Thu Mar 08, 2007 5:52 am**

Thanks.
Re "Thinner wire probably meant you have a greater turns per inch which will increase the inductive component." In this case, can you define inductive component? And, if in using my thinner wire, I get more turns per inch(despite counting off visually the same number of turns around my PVC tube as I did with thicker wire) , then with more turns per inch of the thinner wire, I should get more or less voltage to my LED?
Thanks again.

Posted: **Thu Mar 08, 2007 6:40 am**

can you define inductive component?

Impedance (commonly denoted Z) has two components.
1) Resistance - the direct current pure resistive component measured with an Ohm meter or looked up in a wire size table (ohms per foot or meter). This pure resistive component is the same at all frequencies making it the simplest to understand.
2) Reactance (capacitance and inductive component) - is a frequency or rate of change dependent component of impedance. Typically Xc or Xl (or Zc or Zl) for reactance of a capacitor or inductor again in ohms. For inductors, the reactive component Xl increases perportionately to frequency or rate of change. See http://tpub.com/neets/book2/4a.htm

then with more turns per inch of the thinner wire, I should get more or less voltage to my LED?

Unfortunately, it is far more compicated than that.

I intruduced turns per inch into this discussion because it affects the inductive component of the source. See the coil reference in an earlier reply for more specifics on this relationship. This source impedance difference affects the power or energy transfer to the load.

Until now, I have avoided bringing up another complicating factor. The geometry and uniformity of the changing magnetic flux field interacting with the turn density and geometry of the windings (single layer, multi-layer, etc.) can make a difference.

If you thought EE's heads hurt dealing with simple uniform field assumptions, here is where even the field theory nerds cringe and walk away or start writing computer programs to do complex contour path specific integral calculus programs and spending months analyzing them.

Don't go there, it hurts too much unless you are after a doctorial thesis or have a practical problem that is worth a lot to solve.

Posted: **Thu Mar 08, 2007 5:10 pm**

Craig, I think I can summarize a few things.
When I went to 9 magnets from six, I actually decreased the strength (flux) of the magnetic field, and thus the amount of change in the flux also declined when i shaked the tubes. That reduced the voltage of the induced electric current. Safe to say?

When I went to 560 windings of 30AWG from 280 windings of 30AWG, I increased the inductance of the coil and thus the voltage produced. Correct?

When I went to 280 windings of thicker 26AWG from 280 windings of thinner 30AWG, I also increased inductance and thus the voltage produced. But I'm still not sure why? Is it because the thicker wire has more electrons that can flow through the "pipe" at a gioven interval of time, or the electrons can flow with greater force through a thicker wire, all other things being equal?

Thanks.

Posted: **Thu Mar 08, 2007 7:20 pm**

When I went to 9 magnets from six, I actually decreased the strength (flux) of the magnetic field, and thus the amount of change in the flux also declined when i shaked the tubes. That reduced the voltage of the induced electric current. Safe to say?

That is my best guess without knowing more about your magnet geometry.

When I went to 280 windings of thicker 26AWG from 280 windings of thinner 30AWG, I also increased inductance and thus the voltage produced. But I'm still not sure why? Is it because the thicker wire has more electrons that can flow through the "pipe" at a gioven interval of time, or the electrons can flow with greater force through a thicker wire, all other things being equal?

No and No. Vl(t) {voltage across an inductor at time t} = L di(t)/dt {Inductance times change in current at time t with respect to time} or said another way {Inductance multiplied by the rate of change of current with respect to time at time t}. Assuming the same magnetic coupling (probably good if the coil diameter, magnets and geometric relationship, and shake were the same), the change in the source impedance was the likely cause. Your data (0.1 v to 0.3 v) tends to indicate the inductance went up by a factor of 3. Use the coil inductance formula in the previous link and calculate an estimate of the inductance of the two different coils and see if you get agreement! If yes, then you have your explaination. If not, then parasitic (unintended circuit elements) measurement effects are probably involved.

When I went to 560 windings of 30AWG from 280 windings of 30AWG, I increased the inductance of the coil and thus the voltage produced. Correct?

Partially. By increasing the number of turns you increased the magnetic coupling and changed the source impedance. If the magnetic coupling effect were uniform over the entire coil (and I discussed this might not be the case earlier), the coupling effect would double the induced current. Using Kirkoff's current law for circuits {the sum of the current entering and leaving any point in a circuit is zero}, V(t) = I(t) * R {Voltage across a pure resistive load at time t = current at time t multiplied by the load resistance}. This pure resistive load assumption is probably pretty good for LEDs and meters. Since your data (0.1 v to 1.0 v) indicates a factor of 10 change, something else is involved. Again use the coil inductance formula in the reference previously cited to predict the ratio of the coil inductances. If it comes out 5, then you know 2x was caused by magnetic coupling and 5x was caused by a source impedance change. If it is less than 5x then you need to look at the turns per inch and how coupling might have increased by more than 2. If it is more than 5x then the coupling zone is probably less than the whole coil.

The wild card in all of this (even if you understand all of the physics) is the meter. Most meters are "low pass" by design. They don't measure high frequency voltages or currents accurately.

And you just thought this was a simple science project...
This one is like peeling an onion. The more you dive into, the more you learn, and the more layers you uncover. And you and/or your student cry when you try to explain it. I'm glad I don't have your job as parent of science fair project student. You've been asking questions that involve more than two semesters of physics plus two semesters of circuit analysis for EE majors.

Science Buddies: "Ask an Expert"

SHake-a-light experiment

SHake-a-light experiment

Re: SHake-a-light experiment

Shake a light

Re: Shake a light

SHake-a-light experiment

Craig,s response to shake-a-light

Craig's second response

For Craig:

getting much closer