Energy gain by induction change.

[broli]

Since I didn't want to distract the on going Steorn thread I decided to make a new thread about this subject. Since I don't build much I can spend more time thinking, and this is what Steorn led me to think...

http://ziosproject.com/NJ/magPres/index.htm

It's almost self explanatory but here's a summary. Basically it's a system where an inductor changes inductance during a certain time period. But the energy gain that comes with this is unexplained as the mechanical energy loss is in comparison very small. This system is based on two facts, core saturation and seemingly small mechanical forces. When you combine these two energy might get created in big quantities. In the presentation every rotation gives about 800 Joules. This is considering ideal conditions like giving the current enough time to rise during the small period where the magnet is at TDC.

As a side note I should add that hysteresis losses plays a big role too. The smaller the remnant magnetic field, after collapsing it, the better.

I hope this will make people think.

[teslaalset]

Broli, that is a very good analysis indeed. Thanks for sharing this insight.
I had a look at the effect of the changing permeability of magnetic material  under the influence of magnetic fields. Below a graph of this effect which you described.

Zooming into Naudin's experiment to determine the current lag and what may be the reason for his findings that may seem a contradiction to this theory, I think Naudins findings may be explained as follows:
If the magnets Naudin is using are not strong enough to saturate the core of the toroid, he is measuring at the left hand area, before u_max in the graph, so too weak magnetic field strenght.

He should place his magnets even more close to his toroid to achief the required effect.
This may then lead to too weak induction values of the toroid to get the motor going with the maximum current he has available.

In my view, it's very important to have a good balance in the magnet strenght, coil value and the permeability value of the toroid core. So, it's not so simple to obtain the perfect situation.

[edit] additional remarks:
Instead of decreasing the distance of the magnet to the toroid there are two alternatives to obtain a saturated toroid coil while having too weak rotor magnets:
1) mount a strong enough magnet to the (back)side of the toroid with the proper polarity facing the rotor magnets
2) use bifilar windings and use one set of the winding to add DC current to the coil. This one is very useful to do measurements on the core saturation

The measurement method of Naudin is very useful to optimize the setup anyway.

[gyulasun]

Since I didn't want to distract the on going Steorn thread I decided to make a new thread about this subject. Since I don't build much I can spend more time thinking, and this is what Steorn led me to think...

http://ziosproject.com/NJ/magPres/index.htm

It's almost self explanatory but here's a summary. Basically it's a system where an inductor changes inductance during a certain time period. But the energy gain that comes with this is unexplained as the mechanical energy loss is in comparison very small. This system is based on two facts, core saturation and seemingly small mechanical forces. When you combine these two energy might get created in big quantities. In the presentation every rotation gives about 800 Joules. This is considering ideal conditions like giving the current enough time to rise during the small period where the magnet is at TDC.

As a side note I should add that hysteresis losses plays a big role too. The smaller the remnant magnetic field, after collapsing it, the better.

I hope this will make people think.

Hi Broli,

Last night I saw your drawings in your link but now they disappeared... 

Anyway I understood and agreed on your first two drawings but the third puzzled me much and I did not get it. It would be best to see it again for my  question.

So what's happened I wonder.

Thanks, Gyula

[broli]

Hi Broli,

Last night I saw your drawings in your link but now they disappeared... 

Anyway I understood and agreed on your first two drawings but the third puzzled me much and I did not get it. It would be best to see it again for my  question.

So what's happened I wonder.

Thanks, Gyula

The link should still be operating, try again.

Basically in the third diagram the current in the coil is keeping the core saturated as the magnet moves away. But the magnet is being pulled back as it moves away, this is an energy loss. I estimated this energy loss by saying that on average the magnet wil be pulled back by a force of 5N during the entire 90°. This is not really correct since we know the magnet will have a big force close up and very fast get down due to the 1/r^2 nature of the force, this is why I just took the average to keep things simple.

While that is happening though there is no flux change in the coil since the current is holding the saturation. I will include a very ideal hysteresis to explain why flux doesn't change.

[gyulasun]

Hi,

I am still in office where I do  not see the pictures but a fully empty page (I meant the link works but with no pictures seen). Last night I saw them from home, there was no problem with seeing the pictures and hopefully from home I will be able to see them again.

Ok on your explanation for the third picture I understand it now but it needs a practical test and it is difficult to measure. 
As the flux of the permanent magnet moves out of the core the question is how much input power is needed to keep the core in saturation (or in near saturation) so that a minimal loss should occur from the attraction momentum. Present science says that the same flux should be needed to keep it in  saturation like the permanent magnet established in it at the facing position moment and I hope that hysteresis and / or viscosity of the core may modify it to the ou area...
Cores with rectangular hysteresis do exist but I wonder if the B-H curve needs the rectangular shape with a narrow area in between, I tend to believe such feature is needed for the curve, what do you think?

Thanks,  Gyula