Self running coil?

[IotaYodi]

Your not the first person who says the link doesn't work

My MSN explorer doesn't bring it up. Firefox,Netscape and Opera do. I havent put Googles chrome on. I havent used the MS browser since its inception for a few reasons. Id stay far away from it as possible.   

[gyulasun]

...
But you say more inductance may cost more energy. Is this because it usually takes more windings to get more inductance which = more wire resistance which = loss of energy?
...

Hi Luc,

No, because normally if you double the number of turns, the R copper resistance also doubles (linear relationship) but the coil's L inductance quadruples  (quadratic relationship in the number of turns in the L formula).

This means the L/R time constant eventually increases too (because its nominator grows faster than its denominator).

I think this is the answer to your findings in your latest video with the big air core coil of the same inductance like the toroidal one.
Consider the followings:
Battery voltage  V=12V
toroidal coil     R_t=7 Ohm     Air core coil R_a=5 Ohm
                       L=.1135H                   L=.1135H

See the formula for calculating the current at any moment in an inductance here:
http://www.ac.wwu.edu/~vawter/PhysicsNet/Topics/Inductance/LRCircuits.html

I_max is the maximum current that can flow in a series R-L circuit connected to a voltage source, comes from Ohms law when at least 5 times the L/R time already passed from the switch ON time.

This max current is different for the above two coils (12/7=1.714A and 12/5=2.4A).

The L/R time constant is 16.21 ms for the toroidal and 22.7 ms for the air core coil.

Now if you use the formula from the above link for getting the instantenous current first in the toroidal coil then in the air core coil, say, 10 millisecond after you switch the 12V onto them, you get the followings:

current in the toroidal coil i_t=1.714(1-e^-10/16.21)= .789A 
and current in the air core coil i_a=2.4(1-e^-10/22.7)= .855A

So we got higher current in the air core coil under the same circumstances, the only difference is the 2 Ohm copper resistance between the two coils. 
In fact there is another difference between the two coils: the iron core of the toroidal coil is surely biased by a magnet so it may behave nonlinearly, due to the nonlinear behavior of its permeabilty in the function of the coil current, the air core coil is 'linear' in this respect

So the need for the higher input drive to the air core is due to its less copper resistance, when you compare it to the more lossy toroidal core, sounds as a strange result indeed.

rgds,  Gyula

EDIT:  It would come from the above that if you reduce the supply voltage from 12.7 to about 8.57V (1.714A*5 Ohm)then the current in the air core coil would be about the same 1.714A like in the toroidal coil with the 12.7V supply (this Amper value is meant after the 5*L/R time of course).  BUT the MOSFET would have a higher input and output interelectrode capacitances at the reduced drain source voltage you may have to compensate with a slight retuning of the input frequency, and try using the same drive level as in the 12.7V toroidal core case.

[mscoffman]

The reason that these two coils is different is exactly the reason one uses
a ferrite to increase the inductance of a coil. Just so we understand each
other the brookes coil is a standard coil - is that correct? That is it doesn't have
any attributes that would cancel part of it's inductance by backward winding.
Some OU type coils have partial backward winding. A wire wound resistor cancels
it's parasitic inductance by winding half in one direction flipping in the middle
and the other half in the other direction.

The large gauge wire coil has tremendous interwinding self capacitance.
This capacitance is called distributed rather then the lump capacitance
of a capacitor. Each winding has a small value of capacitance relative
to the next turn and the first winding has some capacitance to the last.
How the coil is wound will determine how the capacitance is distributed.

The whole point of the inductive core is to increases the inductance without
increasing the interwinding capacitance so the toroid has much less
interwinding capacitance. Because of the way this works each segment of
the large coil is filtering RF. And so a ferrite is used to increase the inductance
of an RF coil. 10Khz is apparently RF for the big coil. These things effect the
Q of the reasonace at a certain frequencies. The Q of the coil for our purposes
is the ratio of the pp signal across the toroid to the bulk driving power...
You will note that the toroid we are using is picking up some good Q at a ratio of
about 2.5/(1).; at 30Khz...This is good. The big coil is displaying a Q of very much
less than 1. Now put a large cap on that big coil and excite it at hundreds of times
less frequency and I suspect it will show a reasonable Q also.   f=1/(2pi * sqrt(LC))

Your inductance meter operates at probably one frequency and this may make it
susceptible to misreading the inductance. You may want to carefully attach your
scope on the inductance meter to see what it is doing. A frequency agile series
of signals would indicate a more sophisticated measurement technique. I'll bet it's
using one sinewave frequency and the low Q of the big coil may be causing it grief.

The following link contains a document that shows how to measure the
distributed capacitance of a toroid and would work for the big coil as well.
Several pf's of distributed capacitance of the toroid are not going to hold
a candle to 1100pf of the transistor gate. The capacitance of the toroid
is also low because the intermediate windings splay out at all angles relative
to one another. That is not optimal for forming self capacitance.

http://g3ynh.info/zdocs/magnetics/appendix/Toroid_selfC.html

:S:MarkSCoffman

[HarryV]

Great !
To me, it makes perfect sense.
And it opens a good field for optimizations, i think, regarding the spacer's material.
In my imagination, It would require a material with a good shape memory and superelasticity.
As an orthodontist, i instantly thought of nickel-titanium (Ni-Ti), or nitinol alloys.
http://herkules.oulu.fi/isbn9514252217/html/x317.html
As far as i know, but not entirely sure, this alloy is non ferro magnetic.
It would not be expensive or hard to make spacers with a small acrylic disc holding 4 or more small 0.012 wires as "legs",  providing an spring effect.
The choice of material could provide better amplitude on magnets micro movements, optimizing it's work.

Just my 2 cents.
Hope it sounds not too much stupid

Men, please keep this great work.

Thanks.
Look at this new alloy.

http://blogs.physicstoday.org/update/2010/03/stretchy-metals-recoil.html

In materials, as the axiom goes, structure follows function: A metal’s tightly bonded atomic crystal lattice gives it strength, and a polymer’s mesh of macromolecular chains makes it elastic. Medical implants, electronic components, and other similar devices call for multifunctional materials that are both strong and stretchy. One such material is the shape-memory alloy (SMA), a polycrystalline arrangement of assorted metals that, when stressed, undergoes a structural phase transition from high to low symmetry. The transition is reversible, and above a critical temperature SMAs are superelasticâ€"they fully recover after being stretched well beyond the reversible-deformation strain values of pure metals. Now, materials scientists at Tohoku University in Sendai, Japan, have presented evidence for an iron-based SMA that is 35 times as elastic as pure metals. The new alloy, which also features nickel, cobalt, aluminum, tantalum, and boron, has an elastic strain of 13%, as shown in the figure, almost double the value of the more expensive commercial-standard nickelâ€"titanium alloy. Furthermore, the material’s yield strength, 800 MPa, is 1.5 times that of the nickelâ€"titanium SMA. The researchers say that microstructured precipitates similar in composition to the bulk matrix and interspersed through it are a key to the improved mechanical strength. The greater elastic strain and strength could be exploited for mechanical damping in building materials. Also, the ferrous SMA’s magnetism is phase dependent, which makes it potentially useful for electromechanical sensing applications. (Y. Tanaka et al., Science 327, 1488, 2010.)â€"Jermey N. A. Matthews

[gyulasun]

Hi Luc,

One more thing for you:  if you have a 2 Ohm resistor, you could connect it in series with the air core coil, this way its loss would 'seem' to be the same as that of the toroidal one, so the current would also be similar from working the 12V battery in the same way. IT would be worth for a quick test too, then no need to use a 8.5-9V supply voltage.

Gyula