Flux Gate Interrupter, BEMF Redirector

[JimBoot]

Yeah I did see that.using VCR tape head bearings. I think my next one will be the same except have the coils perpendicular.

[TinMan]

Here you go http://youtu.be/v_Kf5RsTGZ4 feedback welcome

Lol-Jim had a blow out
It's always good to see smoke and exploding parts--done it many times my self.
Dont forget your safty glasses Jim

[T-1000]

Here you go http://youtu.be/v_Kf5RsTGZ4 feedback welcome

Nice one!

The 90 degrees angle is due 2 magnets facing directly each other over rotor. You can mod another cooler and instead of having magnets on outside casing put coils there. Then see what will happen

[T-1000]

Hi all,

To speed process with CAD drawings I spend my day with doing them in FeeeCAD.
The drawing is based on my concept explained earler and someone has to do a physical assembly for verification..

Please see screenshots and FreeCAD drawing attached which is calculated for 20mm diameter magnets and 20mm diameter coil cores(remove .sch in .FCSTD file extension):

[JimBoot]

Thanks mate. I'll have to learn cad now I guess

[JimBoot]

Nice one!

The 90 degrees angle is due 2 magnets facing directly each other over rotor. You can mod another cooler and instead of having magnets on outside casing put coils there. Then see what will happen

I think the flux would be redirected and I would lose a significant amount of output.

[JimBoot]

I think the flux would be redirected and I would lose a significant amount of output.

Yep pretty much as I thought. Tried it tonight. Did a rebuild of the rotor and stator. At least now I know.

[Matt Watts]

So Jim, do you think it's the cogging that is the secret here?  The strong and rapid pole flips are the source of the electrical energy you are able to extract?

For the "RamaGen",

Syairchairun used 40 poles and 162 permanent magnets, then wired all the stators in parallel.  So I'm wondering with that many poles, the cogging points would be so tight, so close together and the flipping would be extremely fast even at very low RPM.  With 40 poles, that's a flip every 9 degrees of rotation, I'm not sure you could even really feel exactly where those would be by hand.

[Centraflow]

From what I have found the cogging reduces with speed of rotation, the first movements have huge cogging effect, this leads me to the following conclusion:-

I think it is the speed of magnetisation and demagnetisation of the iron, the pickup coils need to be between the magnets, the iron picks up the magnetism then gives it up over the coil, then depleted it picks up the reverse from the coil, BEMF. The iron then passes to the next magnet and the process repeats, basically the iron acts as a magnet of very short life or one that is switched off by the BEMF as opposed to causing load on the rotor. The important main thing is the iron has to be fast switching on and off, so silicon iron as in transformers is a must.

This would lead to a setup of magnet coil next to one another, then small space to allow the BEMF to be lost, that part needs to be tested, then magnet and coil again and so on. The effect is the BEMF is thrown to waste and so no load is created on rotor, and that is the secret as I see it.

Lets face it, the BEMF is the load that stops the possibility of a loop, without it the only load is friction in turning the rotor, so why not throw it out via a metal that looses it fast to the ambient or adds to the next magnetic impulse, " that last part is a very strong possibility and so the spacing would not be needed", to be tested soon.

Regards

Mike

[T-1000]

Yep pretty much as I thought. Tried it tonight. Did a rebuild of the rotor and stator. At least now I know.

Syairchairun had magnet -> iron in rotor -> coil in his CAD and this arrangement does not compensate magnetic attraction between magnet and iron and the forces are fighting there against movement.
The coil also has magnetic field which obviously is changing polarity in movement so we have to overcome this block by balancing magnetic forces. This is why the coil is between 2 magnetic poles in my design...
You can try to see it with 2 magnets, iron coil's core between them and moving iron bar between coil and magnets with going over one magnet under coil then reaching second magnet for Bloch wall then moving away over second magnet. The results will talk for themselves..

[wattsup]

Sorry!

[T-1000]

About the BEMF, there should not be any BEMF if the coils are only being induced by the magnets, there is no pulse so there is no open condition of the coils.

We are talking about situation when coils have maximum load or are shorted in other words So you have Lenz force there.
If you have it shorted in conventional generator, the coil will try to push away approaching magnet and attract when it goes away.
When you have "magnet" which is switching polarity to opposite in sync with coil polarity change to be same polarity again when it goes away, the coil always will push away it. And there iron bar can be used for solution as we can use it in same was as in solenoid transformer with separate coils on same core. Just North and South magnets instead of primary transformer coil and the core movement across like in conventional generator... The rest can be seen when experimenting

Cheers!

[JimBoot]

So Jim, do you think it's the cogging that is the secret here?  The strong and rapid pole flips are the source of the electrical energy you are able to extract?

For the "RamaGen",

Syairchairun used 40 poles and 162 permanent magnets, then wired all the stators in parallel.  So I'm wondering with that many poles, the cogging points would be so tight, so close together and the flipping would be extremely fast even at very low RPM.  With 40 poles, that's a flip every 9 degrees of rotation, I'm not sure you could even really feel exactly where those would be by hand.

Yep the stronger the cogging the better the output. I'm tempted to do attracting poles facing but hot glue I fear will not hold the mags in place in that config. Any theory if that is worth the test? I'm only looking for more output

[Smudge]

What you see there as drag is because it is pure magnet to coil drag, where as the coil becomes energized it proportionally becomes an electro-magnet thus you have the magnet passage to electro-magnet drag which is very normal regardless if you are using a magnetic diversion method.

And to be clear, the coil becoming "energized" is the coil carrying current.  Of course that current is determined by the voltage induced into the coil (by rate-of-change of flux) and the load impedance, but see later comment.

The only way I can think of to beat drag is to not let the coils increase in their energized state. If the coil is connected to a capacitor via a diode or a bridge rectifier, the capacitor grows in potential and proportionally so does the energized state of the coil and so does the drag.

That's not quite true.  The capacitor can grow in potential (volts) while the current is reducing.  For example normal pulse charging of a capacitor does this where initially the current is a high value but them reduces exponentially while the capacitor voltage increases exponentially.  The energized state of the coil (current) does not follow the energized state of the capacitor (volts).

What I think a lot of people miss is that the flux that is changing (and creating the coil voltage) is actually the sum of the flux from the moving magnet and the flux from the coil current, it is not just the magnet flux itself.  IOW the coil inductance comes into the equation.  So it is necessary to delve deeper into this situation to find out what actually happens when you put a capacitor across the coil.  And it is interesting to consider a capacitor without your diode.

If you have an array of magnets whizzing past the coil the induced open circuit voltage can look like a sine wave, and that can help in understanding what really goes on.  You can consider the coil circuit as a closed series LCR tank circuit with a (zero impedance) voltage generator in series with the L, and the value of that voltage is the open circuit voltage of the coil.  Note that this includes the coil inductance L which for an air coil or one where the core does not saturate is constant.  It is then easy to deduce the phase between the current and that induced voltage where it is found that under the right conditions the current can be almost 90 degrees shifted from the voltage.  That puts the "energized coil" in a different position relative to the magnets, indeed it can now not create a drag at all but actually a boost.  Unfortunately at typical rotation speeds this situation demands that (a) the coils be shunted by a low ESR capacitor and (b) the coils have impossibly low resistance.  IOW there is a practical limitation that is the reason this effect has not been discovered or utilized.  But maybe the type of motor here that has many magnets (thus creating much higher frequency) can resolve this dilemma.

Smudge

[JimBoot]

Yep the stronger the cogging the better the output. I'm tempted to do attracting poles facing but hot glue I fear will not hold the mags in place in that config. Any theory if that is worth the test? I'm only looking for more output

New rotor built , cogging is so strong now I can't turn it, doh! I have epoxied the rotor to the new axle and am waiting for it to dry. I'll spin it up with the drill tomorrow and doo some measurements. I'm thinking of putting the stator on rails and moving it into place once the rotor has momentum.

[gotoluc]

Hi everyone,

I was wondering what the Wave Form of one of these devices may look like.
So I put together a simple test device using a shaded motor coil with two N - S ceramic magnets between it and placed a thin board over it so I can manually slide different cores over it to observe the wave forms and see if there's anything unusual we can see.

I find the waveform to be different then the standard and would like the opinion of others as to why you would think this would make a difference in a generator.

I also noticed the wave form is quite different depending on how wide the core I manually move over magnets and coil.

First picture is the coil and magnet test setup and cores to be used. Second picture is with the thin board over the test setup with a core ready to be moved by hand.
The below scope shots are the results of U-core, then thin core, then wide core.

Please share your thoughts

Luc

[T-1000]

So I put together a simple test device using a shaded motor coil with two N - S ceramic magnets between it and placed a thin board over it so I can manually slide different cores over it to observe the wave forms and see if there's anything unusual we can see.

The waveform is different due magnetic field discharge in middle of cycle.
Also if you move coil away and keep its core in center then move iron bar between coil and magnetic poles, this will give even better results on coil...

[JimBoot]

Thanks luc and arunas. I built another rotor and having some major issues with the cogging. Not getting anywhere near the output of my first design. The main difference is I'm  using a drive shaft so I'm not getting the benefit of the pulley magnification. Even with good Rpms though my alloy rotor does perform anywhere near that of the plastic rotor

[Matt Watts]

The waveform is different due magnetic field discharge in middle of cycle.
Also if you move coil away and keep its core in center then move iron bar between coil and magnetic poles, this will give even better results on coil...

T,

This pendulum example you posted, do you have any idea how fundamental this is?

If someone gets this working, placing the concept into a M/G should be almost child's play.

All this design needs is some geometric tuning to find the proper magnets, dimensions and spacing; "The rest is history" as they say.



The only thing I may add, is unlike a pendulum, in a M/G the iron bar will continually flip, since it does not change its direction of motion.

[T-1000]

T,

This pendulum example you posted, do you have any idea how fundamental this is?

If someone gets this working, placing the concept into a M/G should be almost child's play.

All this design needs is some geometric tuning to find the proper magnets, dimensions and spacing; "The rest is history" as they say.



The only thing I may add, is unlike a pendulum, in a motor the iron bar will continually flip, since it does not change its direction of motion.

It is fundamental and this is why I repeat myself over and over again to get it communicated right

Also here is first scope shot of open coil from actual assembly from my CAD drawing. Shorting coil does not affect rotor at all as I predicted!
And 16 stator magnets with 9 blades on rotor still have not very balanced magnetic forces when running free but that is geometry issue to solve later on...

[Matt Watts]

It is fundamental and this is why I repeat myself over and over again to get it communicated right

Maybe it's that accent of yours.  hehe  

Also here is first scope shot of open coil from actual assembly from my CAD drawing. Shorting coil does not affect rotor at all as I predicted!

Okay, good.

Having done a lot of work with Bill Alek's Split Flux Transformer, I would expect no load and full short to behave that way.  Now what happens in the middle?  Can you find a particular resistance where the rotor speed is most effected?

And 16 stator magnets with 9 blades on rotor still have not very balanced magnetic forces when running free but that is geometry issue to solve later on...

Again, I figured this ratio would be better, but not optimal.  I have a hunch an arrangement using the Golden Angle will prove to be much smoother for rotor motion.  Unfortunately we can't take this series out to infinity, so it still won't be perfect.  We just add as many irons as will fit and then mechanically balance the assembly by weight.

[T-1000]

Now what happens in the middle?

That bump in bottom of scope shot is the middle
The BEMF gets support from opposite magnetic polarity and shoots back up to complete sine wave...

[JimBoot]

Looking good arunas.

[gyula]

...
I find the waveform to be different then the standard and would like the opinion of others as to why you would think this would make a difference in a generator.

I also noticed the wave form is quite different depending on how wide the core I manually move over magnets and coil.
...

Hi Luc,

Have been thinking on the induced waveforms you have shown. I assume you tried to apply the same speed for all the 3 cores by your hand, right?

When I first saw your scope shots, I recalled Naudin's test with two orientations for a coil, where he moved a magnet in front of the coil, then tangentially to the coil, see here: http://jnaudin.free.fr/images/magconfig.gif  from his site http://jnaudin.free.fr/html/mromexp.htm 
I think the right hand side setup shown in the picture gives a similarly shaped induced voltage you received, even though there are certain differences, I think the tendency is similar.
What I mean on similarity is that all 3 induced voltage waves of yours start with a small (negative) peak which suddenly goes up and peaks in one or two (positive) peaks, then suddenly goes down and change sign again and ends in a small (negative) peak. The order of the peaks (negative) and (positive) could be upside down, depending on the NS or SN placement for you parallel magnets, just like in case of Naudin if he had used a South pole up instead of the North up, his waveform would have started with a small positive polarity first as the S pole moved from left to right.

For your U core and the wide core the 'split into two peaks' of the single positive peaks comes about, I think, because these two cores are able to bridge directly the two magnets with overlaps and the flux of the perm magnets (mainly in the top part) would prefer closing via the appearing U or wide cores when they are just above the magnets symmetrically and most flux on top is "sucked" away from the coil core and goes through the moving cores. This is not so for the moving thin core that cannot form a shunting bridge fot the facing magnets.

Which waveform would make a difference in a generator, you ask. I would prefer the U shaped core but with its thickness matching the length of your magnets (now the U core thickness is less than the magnet length if I see it correctly). The U shape gives the biggest flux change out of the three I think and perhaps you wish to consider moving an U core under your magnet-coil setup too, now you show the U core moves only above the setup of course. This way the flux change could be even higher in the coil core. I mean two U cores moving alltogether as a rotor, sandwiching the stator magnets + the coil whenever they pass but then the total setup would become 'clumsy'...     surely a mechanically better setup could be devised.

Gyula

Gyula

[Farmhand]

Maybe these clips show something of interest, 2nd Clip shows regular generator action load (rpm slows), neutral action with load (rpm no change), acceleration action with load, and also a more dramatic acceleration under short circuit, Waveforms show the "tune".

Fair warning the setup is loud on a (wooden frame), turn volume down to begin with.

Clip #1 shows the setup.
https://www.youtube.com/watch?v=W2NfmyyhbZs

Clip #2 shows the entire run.
https://www.youtube.com/watch?v=TV_dm8COKBY

Short clip of actions.
https://www.youtube.com/watch?v=iFWin-crxQY
..

I can show a speed up under load or short circuit using almost any gen coil with a core, some inductance, some capacitance and a prime mover.
..

Theoretically cogging should be almost "net drag neutral" because the cog consists of a pull and a push of almost equal magnitude/strength.

..