Self running coil?

[gotoluc]

Hi Luc,

The so called A_L value for the Ferroxcube core is 11400nH/N² (taken from the Farnell link). If you wound 24 turns on it then the formula for L is L=N²*A_L=24*24*11.4=6.566mH
You measured 9mH instead, this difference can only be explained by a high manufacturing tolerance. 
On you other 'usual'  core you got 1.22mH for the 30 turns from the same length of wire. If I recall the A_L value I calculated for your core is A_L=1298.6nH/N²   This gives 30*30*1.2986=1.168mH, very close to your measured 1.22mH.

The big difference comes from the big A_L value differences. Putting it otherwise: the Ferroxcube core has a much higher permeability than the other core.  (7730 vs 1196)

rgds, Gyula

Thanks once again Gyula for your time in explaining this!  much appreciated

I have more questions regarding the BH curves and permeability and was wondering if we could just talk on Skype, it would be much faster than writing and save time. Are you set up and able to do this?

Thanks

Luc

[void109]

I'm wondering if the magnets effect on the toroid may be interfering with whatever method the LCR meter uses to determine the inductance.  I've paired this toroid with several capacitors, and using the inductance it appears to have while positioned on the magnet, and I have not yet been able to get a resonant sine wave out of it.  So I'm wondering if the inductance reading isn't just incorrect.

[gotoluc]

I'm wondering if the magnets effect on the toroid may be interfering with whatever method the LCR meter uses to determine the inductance.  I've paired this toroid with several capacitors, and using the inductance it appears to have while positioned on the magnet, and I have not yet been able to get a resonant sine wave out of it.  So I'm wondering if the inductance reading isn't just incorrect.

Hi void,

at what frequency are you trying to get resonance?

Luc

[gyulasun]

...
Yes it is a shaded pole motor inductor that I use as pulse coil because it's easy to adjust the exact inductance needed by just sliding in or out a long ferrite rod that I made by super gluing 2 AM radio loop stick antenna ferrite rods together. With it I can vary the inductance from 35mH to 350mH. Anyways, it's DC resistance is 67 Ohms. I know many will say that it's a waste of energy but I tried it with my single wound 6.9 Ohm toroid coil and I see no difference. Perhaps because at resonance the coils resistance is not seen? Let me know what you think of this.   

Yes I tried the pulse coil with the toroid as stated above but found no gain.

I don't know much about coil Q so I don't know what to answer

Hi Luc,

It is interesting you did not find any advantage in using a toroidal core+coil instead of the shaded pole motor coil.  I wonder if the two coils had the same inductance then? (it would mean the oscillator had run nearly on the same frequency with the two different kind of coils.

Yes, the motor coil's 67 Ohm DC resistance is very high, unusual to use it in an oscillator where generally everything is done to reduce losses in the resonant LC circuit that also determines the oscillating frequency 

Well, normally the higher a coil Q, the less loss it has, this means the less input energy is needed for maintaining oscillation. (coil loss=DC resistance + core hysteresis and eddy current losses)  At resonance the coil's DC resistance 'disappears' indeed, a resonant impedance appears across the coil instead, this can be from several tens to a few hundreds of kOhm range at your frequencies involved, just because the Q 'magnifies' the resonant impedance hence voltage.
Maybe it would be wise to check again at one frequency in your oscillator, what advantage if any a high Q coil has versus the low Q motor coil. This would involve picking a frequency where you can insure the same inductance for both coils, especially if you have already a ready toroidal coil in the 100-240mH range to which you can easily match the motor coil's inductance with the ferrite rods.
In a high Q LC circuit the resonant AC voltage can be much higher than in a low Q one because of the less losses involved. Normally this is beneficial for an oscillator but in your MOSFET circuit the peak to peak voltage should not be let to be higher than about 40V_pp because the allowable maximum gate-source voltage must be +/-20V (FET type dependent though but valid for most).
The higher resonant AC voltage means that the LC tank can store higher reactive peak power  i.e. the circuit would need even less input current to make up for the less loss.  Could you recall what the gate-source voltage was when you used a toroidal coil for the pulse coil?
IF you found no significant difference, maybe the first step would be to check what the Q of that toroidal core actually is. Whenever you have time you could measure the Q of these coils as I outlined in an earlier post:
http://www.overunity.com/index.php?topic=8892.510
This way you can also gain some further insite what the Q means for coils.

rgds,  Gyula

PS 1)  Thank you for asking on the Skype, I do not use it (not yet installed). By the way I am not an expert on the B-H curves... for instance I wish I knew the real explanation why the L increases on a toroidal coil when you parallel facing its core with a big magnet...

2) I have seen the data sheets of your newer FETs, perhaps IRLML2502 seems the most promising with its low gate threshold voltage and gate charge needs.

[gyulasun]

Oh I wasnt saying anyone proposed that example, that was just a thought experiment on my part - given assertions that were made, which were:

1 - Higher inductance means it takes more current to saturate the core (?)

2 - Saturation means that the core material cannot contain any further flux (?)

3 - The inductance of the coil increases in a certain position and arrangement relative to the permanent magnet (?)

4 - If the saturation increases - that implies that the coil now has increased capacity to contain further flux (?)

Those are my assumptions based on my current understanding.

So what I said before was a thought experiment based on the above postulates - if any of those assertions are wrong - please tell me    And number 3 is just from my own experiment, the first two are factoids I've gleaned from folks here more knowledgeable than myself like you fine gentlemen.  The thought experiment just outlined my confusion as to the *how* the inductance can rise given its proximity to the magnet.  I appreciate the feedback.

1) All I can say is that it is a complex problem, depends on several factors.   (What I wrote to Luc on it was in connection of his ORBO question, see it in my earlier posts.)

2) Here is explanation from this link http://en.wikipedia.org/wiki/Saturation_%28magnetic%29 :
"Ferromagnetic materials like iron that show saturation are composed of microscopic regions called magnetic domains that act like tiny permanent magnets. Before an external magnetic field is applied to the material, the domains are oriented in random directions. Their tiny magnetic fields point in random directions and cancel each other out, so the material has no overall net magnetic field. When an external magnetizing field H is applied to the material, it penetrates the material and aligns the domains, causing their tiny magnetic fields to turn and align parallel to the external field, adding together to create a large magnetic field which extends out from the material. This is called magnetization. The stronger the external magnetic field, the more the domains align. Saturation occurs when practically all the domains are lined up, so further increases in applied field can't cause further alignment of the domains. This is a simplified account; a more complete explanation can be found in Ferromagnetism."

3) You have made nice experiments on your question, carry on and learn.

4) See the link at 2) above.