magnetic flux / faraday disc idea . / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

magnetic flux / faraday disc idea .

first previous 2 3 4 5 next

Move Thread

LAN_403

Dr. Slack

Mon Sept 21 2015, 08:54AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Any practical air-spaced movable capacitor will have such a low capacitance to be useless for power transfer. That's power transfer at any level. not just inadequate for 1kW windmills, but inadequate for bike-light dynamos, or even solar-cell level nodding toys.

Excited generator as a power amplifier. This is something with a long and illustrious history. See the Ward-Leonard system for DC output. Have a look in the engine of the nearest car and check out the alternator, multi-kW AC output controlled by a few 10s to 100s of watts of DC field excitation. But, the output frequency depends on the rotational speed.

So, how do we rebuild a car alternator to uncouple the rotation speed of the field from the rotation speed of the rotor? Simples, well, sort of. Have a 3 phase (or n-phase for that matter) armature, driven by an n-phase inverter. The static output coils will only see the movement of the field, as it progresses around the armature.

Now, this is where I thought my intuition would fail me, as it's not initially clear to me how much power input into the spinning the field electrically will end up in the power output of the generator. However, as I type, a picture is forming. I suspect the following. When the armature is stationary, obviously all the power output comes from the electrical input. When the armature field excitation frequency is at DC, obviously the armature power supply is only driving losses like in the car alternator, and all the power output comes from the mechanical input. So, what I suspect is, that the power output comes from the electrical field input and the mechanical shaft input proportional to their rotation speeds. This is appealing conceptually, voltage is proportional to speed, so if each energy source contributes to the same output current (this is where the intuition is weakest, help invited to corroborate or improve), power is also proportional to speed.

This would have a number of consequences.
a) For small shaft/output frequency differences, the power going into the field would be small.
b) For big differences, an uncomfortable amount of power needs to go into the field.
c) For high input speed, low output frequency, so negative field speed, you are actually extracting power *from* the armature field. This is a generator after all, some of the power is coming from this winding, some from that winding. This means your field inverter would need to support bi-directional power transfer.

So, in summary. I think it can be done. Is it worth going to all that trouble, other than to demonstrate it can be done?

Are there any other simpler ways to achieve a generator with output frequency independent of input shaft speed?

What about have a n-phase PM generator, followed by a MOSFET rectifier/inverter producing the frequency you want? Off the shelf components, available commercially with the price benefits and spares availability of mass production, rather than having to create your own multi-slip-ringed one-off electrical machine.

Now the field inverter in the hybrid machine I've described would only have to support up to (for example) 33% of the total output of the generator, if we were to restrict the output frequency variation to +/- 33% of the input shaft speed (well, that, and losses). In the off the shelf PM machine, the output inverter must run 100% of the output power. But, with silicon cheap and well-scalable, and machines big, heavy, and one-off versus commodity, is it worth shifting some of the power handling from an output inverter to a more complicated machine. It restricts frequency range (any frequency is possible with PM+inverter), and if you're suddenly asked to supply a system 3x as powerful, you'd need to build a new machine, rather than buy a larger size of the peg.

Sorry to apply a strong magnetic field

to the speeding copper plate of your ideas

<edit> It occurs to me that with this definition of power transfers at each terminal of the machine, if we revisit your invention that I said was a multiplier, then for the DC output condition, all of the mechanical shaft input is absorbed as electrical output in the field excitation, and the power output from the terminals is zero, which is consistent with Uspring's observation that it cannot generate DC. </edit/>

Salvador

Mon Sept 21 2015, 10:11PM

Registered Member #54402 Joined: Mon Feb 02 2015, 11:09PM
Location:
Posts: 86

In the middle of your reply i believe you are sort of describing an AC induction generator of which the simplest form is to take an AC induction squirrel cage motor and exvcite it with some low frequency and then just drive the rotor above the field frequency and it will push power back into the grid or whatever excites the field , if I remember correctly. I guess the only two types of generators that have solid rotors (no coils on them just metal) are the induction like you are speaking of and magnetic reluctance ones.

In this sense the faraday disc or any homopolar generator for that matter is the perfect and quite frankly only choice , for a given frequecy and waveform field it would respond with a similar, just amplified output as you confirmed.

as for the rotating capacitor, well I'm not saying this would solve the very low capacitance problem but it would consist of many discs half of them on the rotor hald stator , basically a mounted assembly , I dunno maybe if I use a high dielectric constant material inbetween , definately using just air would result in low capacitance, but maybe having a very tight clearance multiple disc compact assemly with some specific good dielectric sprayed on the discs surface or something maybe it can have some merit ? atleast something that I could try out.

Also you definately put some eddy current brake on my ideas but I will try to use all my brainpower to turn those brakes into the more sophisticated and useful regenerative brake. :D:D

The last part of your post were you said about the mutiplier and the one you write edit at the beginning , what was that about as I can't recall of what was it about.?

Uspring

Tue Sept 22 2015, 10:05AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

@Ash:

Wikipedia says the 'Faraday Disc' was also known as 'Faraday's Paradox' because it couldn't be explained by 'Faraday's Law', and it was another 20 years or so before Lorentz arrived to explain it.

Any chance of a bit more 'clarification'?

The Lorentz force and Faradays law have the same origin. Think of a magnetic field (i.e. field lines) pointing upward. If you then move a charge horizontally through this field, it will experience a sideways force, the Lorentz force. If you move a horizontal wire horizontally perpendicular to its length through the field, the charges in it will also experience a force, causing a voltage along the wire. The general principle is, that you'll always get a voltage as long as field lines are crossing the wire.

Faradays law states, that an increase of flux through a wire loop will cause a voltage. In terms of field lines, this means, that there will be more field lines crossing into the loop then crossing out of it, since the total flux is proportional to the number of lines going through the loop. The lines crossing into will cause a voltage and the lines crossing out of will also cause a voltage, but of opposite polarity. So as long as there are more lines going into the loop than out of, there will be a non zero total voltage, just as Faradays law states.

I can't resist to tell about one of the for me most exciting lessons I learned during my physics student days on what causes the magnetic field:

Special relativity tells us the weird story, that an observer moving relative to an object will see it contracted along the direction of its motion, i.e. the Lorentz contraction. The contraction will be larger for higher speeds.
Now think of a wire horizontally in front of you, carrying a current, e.g. electrons moving in it from the left to the right. Now place an electron above the wire and move it parallel to the wire to the right. Now something amazing happens. The electron will see the wire contracted. It will also see the electrons, which carry the current, contracted but to a lesser degree, since they move at a lower relative speed. The net effect is, that the density of electrons in the wire will be less than those of the positive charges in the wire. That amounts to a positive net charge of the wire, which causes an attraction of the electron to the wire. This implies, that currents going the same direction in parallel wires feel an attraction. Note, that for this effect I didn't use the notion of a magnetic field.

Basically magnetic fields are a consequence of relativistic contraction. Historically it's interesting, that Maxwell came up with his equations, which comply with special relativity long before Einstein formulated it. It's a beautiful example, how these seemingly disparate theories work together.

Salvador

Mon Sept 28 2015, 07:02PM

Registered Member #54402 Joined: Mon Feb 02 2015, 11:09PM
Location:
Posts: 86

so I'm back after a little while.
I few questions.First of all , lets say I wanted to make the oridnary faraday disc with two brushes attached , we know well that the outer brush wears out faster due to higher speed at the rim than at the center.Now could I take attached one brushes to the start of the disc at the center then make a wire connection at the outer rim part of the disc and attach some low reluctance material to the side of the disc and use that current carrying wire to loop around those materials and then take that wire back to the shaft to attach it to another brush.
with this I could use the generated current in the disc to also power the magnetic field neccessary , since the faraday disc produced current both when te magnet was spiinning and also when it was stationary i thought one could just use the rim to route the current back to the fhast via some coils that could also excite the b field for the disc, ofcourse when first started it would need some permanent magnet to induce the first current.

Now the second idea is this, one of the most prominent problems of the homopolar generator is it's single current loop low voltage thing.Now it's complicated to make many discs one after the other or any other configuration beacuse each opf the discs or wire loops needs a b field going in one direction across the whole loop.
I wonder if I were to have many discs were the current goes from center to rim and then from rim back to center in the other disc and so again in the next set of discs, could it be possible to for example have the first disc made out of ordinary copper for example but the second (return current) disc made out of a material in which the charge carriers are positive rather than negative , now I think you know what I mean if one disc has negative charge carriers and the other has positive charge carriers as the main current carriers then one could use a simple N-S magnetic field since in each of the disc the charges would be pushed into opposite directions and they would add instead of cancelling as would happen if both discs had the same polarity charge carriers as in ordinary copper for example.

Ok this is what I had thought recently , please be gentle but anyhow I like to hear your thoughts,
thanks.

Ash Small

Mon Sept 28 2015, 09:11PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

What happens if you keep the magnetic field the same for all discs, but spin alternate discs in opposite directions?

Salvador

Mon Sept 28 2015, 09:32PM

Registered Member #54402 Joined: Mon Feb 02 2015, 11:09PM
Location:
Posts: 86

well one can change current flow direction by either changing direction of motion, polarity of magnetic field or as i stated have a material whose charge carriers are of opposite charge (even though im not 100% sure about the last one but i hope someone will jump in on my previous post about this)

well if I understand what you say Ash then that would also work but it would be technically very problematic , I mean one already requires brushes , spinning each disc in separate directions, not to mention how one could do that on a single shaft , would also require brushes not only on the shaft but between discs , if I understood your question correctly?

I wonder what about the idea of having a disc and the return current path also serves as the field coil ?

Salvador

Sat Oct 03 2015, 05:13PM

Registered Member #54402 Joined: Mon Feb 02 2015, 11:09PM
Location:
Posts: 86

hey folks, isn't there anything someone would like to add?

radiotech

Sun Oct 04 2015, 05:11PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

The concept that shaft speed is proportional to voltage holds where the field, (static or rotating)
has a polarity that is the reference for the instantaneous output. This could give rise to machines
rotating within machines. Self excited generators. What ever happens, at some point, a magnetic
pole needs to be created. A mass of rotating copper, can't do this unless it 'fakes' ferromagnetism.

Also to consider, AC machines an be connected for constant speed, constant horsepower, or
constant torque. In the constant horsepower connection, output power does not depend on speed.

The Tesla motor, is a generator too. A damnable result is that it can simultaneously act as both
at two or more frequencies.

This whole thread of a generator that produces its own flux works only if the magnetizing vars
come from external capacitors.

Salvador

Mon Oct 05 2015, 07:53PM

Registered Member #54402 Joined: Mon Feb 02 2015, 11:09PM
Location:
Posts: 86

thanks radiotech for kicking in.
now I could make a set of thin discs and eac one would have N-S field passing through and also at the stator each one would have two plates facing the disc one at the center one at the rim , this would then act as a capacitor and by connecting the plates in series that from ones rim to the next ones center we could have a capacitive coupled series homopolar generator but ofcourse the problem is the very very low capacitance achieved by the low surface area.

I didin't quite understood your last sentence radiotech , but here is what I assume you meant in terms of my idea.
I was talking about how to make the disc generator more efficient , and to make it series one would have to take the current path from the rim and somehow attach it to the next discs center.
What do you think of the case were there is a disc and the current from its rim then leads through a coil attached to the very same disc back to the center to go into the next discs center connection, in such a way each discs return path to the next discs forms also a field for the very disc. and then the outputs from both sides are connected to the rotating capacitor plate were half of them rotate on the shaft and the other are stationary , as a means of transferring power ?
the whole situation is ofcourse high frequency AC not DC.

radiotech

Mon Oct 05 2015, 11:56PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

There are perhaps two approaches to your first paragraph. The first is a mathematical approach in
which the magnitude of the capacitive current (I) is calculated numerically.

The second approach, following the first, is to calculate the magnitude of the current (I) through
the copper or (whatever) metallic path, and from that, the magnetic fields(B). With the fields,
the torque can be calculated with B*A*I*N* (cos theta). From the torque and the speed,
the power can be calculated, and from that, and the current, the voltage.

The generator needs to be connected to an external load. Some geometric spoiler
is needed to cyclically vary a determinant of capacitance.

A= area of coil
N= number of turns

Capacitive generators predated magnetic generators. Faraday proved every form
of electricity could causes electrolysis, and declared all to be of the same nature.

An 'EXISTOLOGICAL' approach would require the plates be separated,
(assumption) by a distance boosting the capacitance in Farads to the
range of the disk inductance in Henrys.

The term 'somehow' only applies to a path out of confusion, if you are successful.

first previous 2 3 4 5 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob