Resonance Effects For Everyone To Share

[gotoluc]

Yes.99 that makes perfect sense and that was how I understood it.

All I was saying is if we have 100 feet of thick wire and we coil it in a single layer we will be able to get many more turns = Inductance then if we coil it in a multi layer geometry. Agree?

Luc

[poynt99]

Yes.99 that makes perfect sense and that was how I understood it.

All I was saying is if we have 100 feet of thick wire and we coil it in a single layer we will be able to get many more turns = Inductance then if we coil it in a multi layer geometry. Agree?

Luc

No, I think you are misunderstanding one factor. If we use my hypothetical coil as an example, that coil has 100 turns, not 10 turns as I think you are thinking.

10 turns x 10 layers = 100 turns.

If you wind a single layer coil with the same wire and it is 100 turns, a) it will be a very long and skinny coil, b) it will not conform to the Brooks geometry, and c) it will have less inductance than the same 100 turns wound in 10 layers as per the Brooks geometry.

The multi-layer coil will require a little more wire length than the single layer coil (assuming the same wire diameter), but the increased inductance benefits of the Brooks configuration will be there.

I hope that is clear now.

.99

[ren]

but would a single layer coil have a higher Q?

I think I understand what you are saying Luc, you are pointing out that everytime you overlap a wound layer your wire has to stretch that little bit futher because your former gets larger with every layer. I dont think this will prove to be too great, in our 100ft example it might cost you another 10feet.

I think there are arguments for each type, depending on the coils purpose. If one is after a super high inductance then it is usually for two reasons.

1. Choking current/establishing large magnetic field.

2. Inductive discharge.

For these purposes I would guess a solenoid style coil is the best bet.

My 0.02 cents

[gotoluc]

No, I think you are misunderstanding one factor. If we use my hypothetical coil as an example, that coil has 100 turns, not 10 turns as I think you are thinking.

Sorry to say .99 but I do and did understanding the coil has 100 turns.

If you wind a single layer coil with the same wire and it is 100 turns, a) it will be a very long and skinny coil,

Correct

b) it will not conform to the Brooks geometry,

Correct

and c) it will have less inductance than the same 100 turns wound in 10 layers as per the Brooks geometry.

Okay now this is what I wanted to get too! you are now saying something different. Before you mentioned the Brook coil you said the amount of turns a length of copper wire has is what will determine its inductance. The more turns the more inductance. Now you are saying that a long and skinny coil having 100 turns it will not have as high of inductance then a 100 turn Brooks geometry coil. Is this correct?

If this is fact, then we need to make it clear and maybe explain it another way, example:

It is not necessarily the amount of turns that will give a coil the highest inductance but rather utilizing the length of wire available by carefully calculating it to obtain a coil geometry known as Brooks since a single layer 100 turn coil will have less inductance then a multi layer 100 turn Brooks coil geometry.

Is this correct and factual?

The multi-layer coil will require a little more wire length than the single layer coil (assuming the same wire diameter),

YES!... this was what I was saying all along. If it is only about turns a single layer coil will give you more turns then a multi layer if you use the same length of wire, mostly if the wire is thick like 10 gauge.

but the increased inductance benefits of the Brooks configuration will be there.

I hope that is clear now.

.99

ya that clears it, as long as you know this for fact because I will trust your teaching me this

Thanks for helping and sharing.

Luc

[poynt99]

Hi Luc,

It's been a long day for me. 27, now 28 posts today.

Anyway, the Brooks method of making a coil is supposed to give you more inductance for the same length of wire.

So I guess compared to a long solenoid, you will in fact have less turns on a Brooks coil for the same length of wire, but the inductance will be higher. This is due to geometry. There comes a point where too long a coil (compared to its cross-section) causes a fall-off of inductance and how well the calculations hold true.

Sorry, sometimes it's easy to overlook something when two people are looking at it from different perspectives.

The point was and still is that you should build a brooks coil configuration if you want to maximize the coil inductance for the length of wire you have.

Building a long solenoid precludes you from the Brooks configuration straight away, so that was confusing me why you wanted to go that way when clearly that would not be a Brooks coil.

It still stands that the coil inductance is proportional to the number of turns squared, but that is within similar geometries. Each geometry has a slightly different equation for determining inductance, but they all vary as the square of turns. The equation for a solenoid is not the same equation as a Brooks coil, so even though they both might have 100 turns, their inductance will be different.

Sorry for the confusion.

So in summary:
Brooks coil configuration with the max number of turns you can muster. It is good to try to subtract one layer from your calculation of number of layers you think you will be able to get so that you can ensure that your last layer will be full. This will keep the coil symmetrical. You want to avoid having only 5 turns on your final top layer when each layer below has 10 turns. I guess you could always space those last 5 turns over the entire width of the 10 turns if you really wanted to use as much wire as possible though.

It wasn't my intention to get bogged down in details. It was just a simple suggestion; of course you can build what you want . All coils will resonate. btw, I think the coil you made (the big one for the big neo magnet) for the BEMF recirc. thread is pretty close to a Brooks geometry.

Regards,
.99

[poynt99]

btw Luc,

Do you have a schematic version of Groudloop's build diagram he gave you? Could you ask him for one if you don't have it?

What is it about his circuit that is not giving you what you want again?

Thanks,
.99

[poynt99]

OK Luc.

I went back and looked at the deficiencies you guys would like corrected:

1) Duty Cycle adjustment
2) Higher voltage (are you sure you want to use 1000V?)
3) Oscillator built in
4) Wider frequency range (min 3 MHz) (could be a tough one )
5) any I missed?

So far I have determined that you do not need a full H-Bridge. A Half bridge is all you need.

.99

[gotoluc]

btw Luc,

Do you have a schematic version of Groudloop's build diagram he gave you? Could you ask him for one if you don't have it?

What is it about his circuit that is not giving you what you want again?

Thanks,
.99

Hi .99,

I just looked through my stuff, topic at OU and could not find a real schematic other then the illustrative circuit. Also Groundloop is away on vacation so no luck there.

I've attached a circuit that was made by Groundloop some time later but it's not the same as the first illustrative one he made for me.

FYI, Gyula mentioned that the the CMOS version of the 555 family LMC555C or TLC555C is capable of going up to 3Mhz.

I also have 4 of IRFPG50 and 4 of 4N35 in my stock.

1000vdc handling capability is not necessary just a wish

Thanks for your help

Luc

[poynt99]

Thanks Luc,

It's not a problem re. the schematic.

I've already got a good notion of where I'm heading with my design.

It may or may not use the high speed 555, but it will give you the duty cycle and frequency range you're after. At least I'm hopeful of the frequency, that will depend on the MOSFET and a couple other things.

As TK has shown, the IRFPG50 is a slow MOSFET, so I may not use it. I will certainly try it if you wish. It's probably more rare and more expensive than the good old IRF820 or IRF840's though. One advantage of the G50, is it's a 1000V FET, so I'll definitely try it.

Doubtful I'll need opto-isolation for my design, but I will keep it in mind.

So far so good

.99

PS. You realize that 1000VDC input would likely result in 10kV or more resonant voltage right? The highest voltage MOSFET out there is about 1500V, so it would blow quite readily with 10kV across it.

[gotoluc]

Thanks Luc,

It's not a problem re. the schematic.

I've already got a good notion of where I'm heading with my design.

It may or may not use the high speed 555, but it will give you the duty cycle and frequency range you're after. At least I'm hopeful of the frequency, that will depend on the MOSFET and a couple other things.

As TK has shown, the IRFPG50 is a slow MOSFET, so I may not use it. I will certainly try it if you wish. It's probably more rare and more expensive than the good old IRF820 or IRF840's though. One advantage of the G50, is it's a 1000V FET, so I'll definitely try it.

Doubtful I'll need opto-isolation for my design, but I will keep it in mind.

So far so good

.99

PS. You realize that 1000VDC input would likely result in 10kV or more resonant voltage right? The highest voltage MOSFET out there is about 1500V, so it would blow quite readily with 10kV across it.

Thanks .99,

I've got a bag full of IRF840's so if they work better then the PG50's that is fine with me and I think they are rated at 600v

Is there not a way we can prevent (isolate) the Resonating coil's voltage rise to come back and fry the mosfet? Even considering coil construction!

Thanks for your help.

Luc

[Armagdn03]

YouTube - fun with lcs


I would like to share a few "thoughts" on the subject of this video (device no longer exists).


First the obvious,

For energy to get from the exciter coil to the pickup coil it must pass through a region of space in which the characteristic permittivity and permeability appear as translucent to the frequency used as possible.

We accomplish this by creating a secondary which has the correct capacitance and inductance. Energy of the correct frequency will be guided along this secondary to the pickup coil. Anything even slightly off of a resonant mode goes absolutely nowhere, and the pickup coil sees nothing.

Now,

For every action there is an equal and opposite reaction. So if we create a disturbance, and it travels along the secondary, picked up by the pickup coil, and powering a dc motor, the motor will have to provide a counter electromotive force in reaction.

Now the question is....Is this counter electromotive force from the motor of the proper character to travel back down the secondary back to the primary?

Second question,

Increasing harmonics create multiple magnetically dominant, and electrically dominant nodes and peaks along the secondary. Each magnetic peak, can be used to pick up power.

If I were to place several loops of stout copper wire over a magnetic peak (as large a load with as little resistance and general impedance as possible) does this change the input power required to resonate the secondary in any way.

something to think about.

[poynt99]

YouTube - fun with lcs

Now the question is....Is this counter electromotive force from the motor of the proper character to travel back down the secondary back to the primary?

I don't think it needs to travel back down to the other primary. The CEMF is created locally, and it's effects should be "felt" locally, therefore the actual load seen by the secondary will be higher due to this CEMF.

Second question,

Increasing harmonics create multiple magnetically dominant, and electrically dominant nodes and peaks along the secondary. Each magnetic peak, can be used to pick up power.

If I were to place several loops of stout copper wire over a magnetic peak (as large a load with as little resistance and general impedance as possible) does this change the input power required to resonate the secondary in any way.

something to think about.

The secondary may still resonate, but the power drawn off any one peak node will or should subtract power from the other nodes. The input power would need to be increased to make up for the extra loads (if present) along the secondary.

Well, that's my take on it anyway, without actually try this.

.99

[Guest]

Hi

I think you are looking for something like this...


The duty cycle can be adjusted by increasing the resistance of R1 or R2 resistor's. Period is calculated from the equation T1=R1*C1 and T2=R2*C2, where T1 is on period and T2 is off period, and frequency can be adjusted by "Increase frequency" Pot.

Here is a data sheet:
http://www.nxp.com/acrobat_download/...otes/AN171.pdf
It is possible to achieve 100MHz frequency...

Regards,
Nenad

[gyula]

I think you are looking for something like this...


The duty cycle can be adjusted by increasing the resistance of R1 or R2 resistor's. Period is calculated from the equation T1=R1*C1 and T2=R2*C2, where T1 is on period and T2 is off period, and frequency can be adjusted by "Increase frequency" Pot.

Here is a data sheet:
http://www.nxp.com/acrobat_download/...otes/AN171.pdf
It is possible to achieve 100MHz frequency...

Regards,
Nenad

Hi Nenad,

The NE558 was basically a quad version of the normal monolitic bipolar process NE555 timer IC and as such it works up to about 100kHz or so.
And the NE558 is obsolote now, and very hard to buy.
So I do not think the NE558 is helpful in this application, sorry.
The CMOS version of the NE556 however still exists: like TLC556 Texas or TL556 from ST, if you can think of a dual timer solution, then it may help. They work up to about 2MHz or so.

rgds, Gyula

[Guest]

I used it for my speed control of my window type motor...
it can go up to 100MHz (not my motor, just the NE558)
YouTube - Speed control of the Window type motor

Regards,
Nenad