Understanding Reverse Current in Linear Generators

[BilPrestonEsq]

Hello, so another question I can't find the answer to anywhere else: at what point in a magnets travels does it reverse the current in a linear generator? I am imagining a solenoid with a cylindrical magnet N on top S on bottom and as its moving up to the top of the solenoid its 'pushing' current one way and when it reaches the top and starts moving back down the direction of current changes, so that its peak voltage is in the middle of the coil. Is this right?

 

[BilPrestonEsq]

also in my description it is a cylindrical magnet axially magnetized( I think i made that clear but just in case). That doesn't seem right. From what I know ( which obviously isn't much Ha!) the poles of the magnets in a 'regular' generator are perpendicular to the wire so... man I need a teacher.

 

[BilPrestonEsq]

and one more thing the magnet in my description would be 1/4 the length of the solenoid and just pass the coil before heading back the other way

 

[berkeman]

Hello, so another question I can't find the answer to anywhere else: at what point in a magnets travels does it reverse the current in a linear generator? I am imagining a solenoid with a cylindrical magnet N on top S on bottom and as its moving up to the top of the solenoid its 'pushing' current one way and when it reaches the top and starts moving back down the direction of current changes, so that its peak voltage is in the middle of the coil. Is this right?

Do you know the equation that determines the voltage induced in a coil when the magnetic flux through the coil is changing?

 

[vk6kro]

Hello, so another question I can't find the answer to anywhere else: at what point in a magnets travels does it reverse the current in a linear generator? I am imagining a solenoid with a cylindrical magnet N on top S on bottom and as its moving up to the top of the solenoid its 'pushing' current one way and when it reaches the top and starts moving back down the direction of current changes, so that its peak voltage is in the middle of the coil. Is this right?

[PLAIN]http://dl.dropbox.com/u/4222062/magnet%20in%20coil.PNG

I used an old diagram so the directions are different.

It is the field that goes to the sides of the magnet that produces current in a solenoid.

The problem is that the South pole generates a voltage in the coil that is opposite to the one generated by the North pole.
So when both of them are in the coil, they tend to cancel each other out.
It is mostly when they approach the coil that you get an output, because one pole then has more effect than the other.

 

[BilPrestonEsq]

Do you know the equation that determines the voltage induced in a coil when the magnetic flux through the coil is changing?

I bet I wrote it down somewhere, but I do know that voltage depends on the magnetic flux density, the coil turn density and the speed at which the the magnetic flux is changing.Oh and thanks for moving this over here

 

[BilPrestonEsq]

So when both of them are in the coil, they tend to cancel each other out.
It is mostly when they approach the coil that you get an output, because one pole then has more effect than the other.

First of all thanks for the reply and I really appreciate the illustration as well(I am more of a visual learner).So how is that problem solved currently? I was thinking about a square shaped housing with a square magnet, with coils, say 4 on 2 sides of the square housing(8 coils in total). The magnet would be polarized to point in the direction of the coils.The coils would be connected to each other in a crisscross pattern so that the two coils on the bottom would be connected to the next two coils up on the opposite side. https://www.physicsforums.com/attachments/31173 as you might notice I didnt add the last connection because i didnt know where to put it, maybe two different circuits? I have a feeling I am painfully wrong with this one but your comments and corrections would be really appreciated.

 

[vk6kro]

With a magnet like that, the fields coming out of the ends of the magnet would be generating voltages in the coils.

However, you would have to space the sides of the coils so that the North pole (say) was not overlapping the two sides of the coil at the same time. Otherwise it would generate opposite voltages in the coil and these would tend to cancel each other out.

That arrangement would need some soft iron at the top and bottom of the picture to increase the magnetic field of the magnet.

You might like to see this diagram (below) which shows conventional generator operation. You can see that the windings are arranged so that the voltages add up even though they come from magnetic fields of opposite polarity.[PLAIN]http://dl.dropbox.com/u/4222062/generator.PNG

 

[Bob S]

Hi Bill-
Your original description of the cylindrical magnet inside s solenoidal coil sounds more like the flashlight in the thumbnail. A cylindrical permanent magnet moves up and down inside a copper solenoid coil seen in the center of the flashlight.

Bob S

 

[BilPrestonEsq]

Thanks guys both images are very helpful. So in that last diagram from vk6kro when the coil is in the middle of the two magnets that is when the voltage output peaks, right? Because both fields are pushing the current in the same direction. So when does the voltage peak in the flashlight? Right when the magnet first enters the coil right? That seems pretty inefficient. Problem is I really need a linear generator for what I am trying to design.

 

[BilPrestonEsq]

https://www.physicsforums.com/attachments/31177


This would be back to the solenoid design, cylindrical shape housing and magnet, except the coils would be separated and wired like this (the lines on the outside).Would this solve the problem?

 

[vk6kro]

As always, the detail is important. The diagram is pretty vague about where the wires actually go. Don't forget you need a lot of turns of wire to get a decent output.

As long as the two poles of the magnet are working together and not cancelling each other out, then you will get output.

So, you need to design your coils so this is what happens. You can learn a lot just with a small coil of wire and a strong magnet.

 

[Bob S]

If a solenoid coil is the same length as a cylindrical magnet, then the N pole exiting one end of the coil at the same time the S pole is entering the other end, the two generated voltages will add. If the N pole were simultaneously entering another adjacent identical coil, the voltage generated in the second coil would cancel the voltage produced by the S pole entering the first coil. But what if the second coil were wound in the opposite helicity (e.g., left-handed corkscrew rather than right handed)? Would the voltage generated in the second coil then add rather than cancel? Suppose there were many equal length coils with alternating helicity wired in series. Would this work?

Bob S

 

[BilPrestonEsq]

https://www.physicsforums.com/attachments/31179Arrows pointing in direction that the coil is wrapped. This time all coils are right in series.

Like this bob right?

 

[BilPrestonEsq]

As always, the detail is important. The diagram is pretty vague about where the wires actually go. Don't forget you need a lot of turns of wire to get a decent output.

As long as the two poles of the magnet are working together and not cancelling each other out, then you will get output.

So, you need to design your coils so this is what happens. You can learn a lot just with a small coil of wire and a strong magnet.

HA! I know, sorry I'm just using paint. I am just trying to illustrate the way the coils connected and the magnets size and polarity, other design factors not considered

 

[Bob S]

Yes. If the coils are all the same length as the permanent magnet, and with alternating helicity. I visualize a 3/4 inch diameter PM about 1 inch long (similar to the PM and coil in the flashlight in post #9).

On second thought, I am not so sure. Any opinions?

Bob S

 

[BilPrestonEsq]

well i was thinking the coils would be half the size of the magnet that way the magnet would be between two coils. I am just going to go back to the drawing board its easier to illustratehttps://www.physicsforums.com/attachments/31181yea the drawing is getting worse. But you see what I am saying right?

 

[Bob S]

If you had identical coils the same length as the PM, and spaced apart by the length of the PM, then the N pole of the PM would be entering the second coil when the S pole is exiting the first coil. these coils would be wired in series. This would work.

So what should we do with all the empty gaps. Put an identical set of coils in them, also wired in series. You could take the voltage output of the two series of coils, and connect them in series. In one polarity, the two outputs will cancel. In the other polarity, the two outputs will add.

Bob S

 

[vk6kro]

Yes. If the coils are all the same length as the permanent magnet, and with alternating helicity. I visualize a 3/4 inch diameter PM about 1 inch long (similar to the PM and coil in the flashlight in post #9).

On second thought, I am not so sure. Any opinions?

Bob S

It is a clever idea, but I think you could get the same result just by reversing the leads.

I was wondering if you could have it so the magnet was longer than the coils and only one pole could be adjacent to one coil end at a time.
[PLAIN]http://dl.dropbox.com/u/4222062/magnet%20in%202%20coils.PNG
Then you would get more pulses per transit of the magnet

 

[BilPrestonEsq]

Yea my drawings only really pertain to my design requirements that's why the magnet and cylinder are that size in relation to each other

 

[BilPrestonEsq]

Wow this thread has really been SO helpful. I guess its time to build one now.

 

[Bob S]

Although I cannot find software to calculate off-axis radial fields around cylindrical neodymium permanent magnets, images imply that the effective magnetic length for maximum radial magnetic fields is about half to 3/4 a pole-tip diameter longer than the physical magnet. So a 3/4" dia by 1" long cylindrical magnet may have an effective length of about 1.5".

Bob S

 

[BilPrestonEsq]

https://www.physicsforums.com/attachments/31207 that's the magnet you desrcibed but I drew the coils an inch thick instead of 1.5". I think if I build one I might make it two coils longer. Thats a good size little prototype to mess around with. I was thinking about wrapping the wire around a thin acrylic cylinder and encasing with an iron pipe. Also I would want to have dividers between the coil sections. Those things aren't really described in the drawing. I wonder what the wave would look like? I think it would deliver peak output in the middle of the stroke.

 

[BilPrestonEsq]

I wonder if iron dividers would be better than plastic or if that would interfere with the field and have a negative effect, actually they probably would probably just heat up. With an arcylic tube on the inside and the iron casing on the outside, the magnetic field's 'circuit' should try to 'flow' through the iron right? I was wondering if you had copper or aluminum pipe wrapping around the whole casing, just 1/2" flexible pipe going to a little radiator, would that pipe become energized too? I was also wondering if the power needed to run the pump would be wasted if you did something like that. I would imagine it wouldn't be wasted energy, it isn't wasted energy that powers the fan and water pump in a car.

 

[BilPrestonEsq]

https://www.physicsforums.com/attachments/31210 I found this on the K+J magnetics website this is there N42 grade neo. Its a cylinder, 1" long with 3/4" diameter. The image is blown up slightly. On the website there's a color key with the gauss rating.

 

[Bob S]

I wonder what the wave would look like? I think it would deliver peak output in the middle of the stroke.

Have you looked up the equation Berkeman suggested in his post #4? What is the name of the equation? Do you understand why a voltage is induced? Do you understand why no voltage is induced when a magnet is completely inside a coil. Do you understand why no voltage is induced when the magnet is not moving?

You do get a voltage out like shown in the thumbnail. Explain the plot in terms of the equation Berkeman asked you to look up, and what you know about the magnetic field around the moving magnet.

Bob S

 

[BilPrestonEsq]

e=N*dMF/dt e=instantaneous voltage in volts N=number of turns in wire coil MF= magnetic flux in webers t=time in seconds d=rate of change of flux over time. Is this the one? I thought I understood it after this thread. I thought that's what your design suggestions were trying to solve. I was thinking that because of the way the coils are wrapped in opposite directions, and sized in relation to the magnet and its two poles. That except for small changes moving between the separated coils the current would flow in one direction until the magnet got to the end and moved back the other way. Thats why I asked if the wave would peak in the middle of that generator in the drawing, not just one coil wound in one direction, because that's when the magnet is moving the fastest in that one direction. When it stops at the end, no voltage is produced and when it moves back the other way the current moves in the opposite direction also. So I'm wrong about this?

 

[Bob S]

e=N*d(MF)/dt e=instantaneous voltage in volts N=number of turns in wire coil MF= magnetic flux in webers t=time in seconds d=rate of change of flux over time. Is this the one? I thought I understood it after this thread. I thought that's what your design suggestions were trying to solve. I was thinking that because of the way the coils are wrapped in opposite directions, and sized in relation to the magnet and its two poles. That except for small changes moving between the separated coils the current would flow in one direction until the magnet got to the end and moved back the other way. Thats why I asked if the wave would peak in the middle of that generator in the drawing, not just one coil wound in one direction, because that's when the magnet is moving the fastest in that one direction. When it stops at the end, no voltage is produced and when it moves back the other way the current moves in the opposite direction also. So I'm wrong about this?

Not quite. Here is one change:
"e=N*dMF/dt e=instantaneous voltage in volts N=number of turns in wire coil MF= magnetic flux in webers t=time in seconds d=rate of change of flux linking coil over time. Is this the one? "

Here are two thumbnails showing the magnet in the center and at the end of a solenoid coil. Is the flux linking the coil in either thumbnail? If the magnet is moving in the center of a long solenoid, is there any d(MF)/dt in the solenoid?

Thanks for the link to the K & J neodymium magnet field calculator: http://www.kjmagnetics.com/fieldcalculator.asp

Bob S

 

[BilPrestonEsq]

If the magnet is moving in the center of a long solenoid, is there any d(MF)/dt in the solenoid?

No the magnet in the center of the coil is cancelling out voltage output in either direction.

I thought that's why that last design would work. Because it's constantly entering the coil as it moves through the generator as a whole

 

[BilPrestonEsq]

https://www.physicsforums.com/attachments/31246

All the coils are wound in the same direction. There are 3 coils in series and another 3 coils in series. Those sets of coils are in parallel to each other. Wouldn't that design create the same effect as a magnet entering a new coil as it passes through?

I thought the way the coils are wrapped in opposite directions as Bob suggested solve the same problem but in a different way.

Am I confusing this?