View this page "Eldarion's TinyMEG"
eldarion
Click on
http://groups-beta.google.com/group/energymover/web/eldarions-tinymeg?hl=en
- or copy & paste it into your browser's address bar if that doesn't
work.
Trump
I hope you can keep us posted on your progress. Who knows if the core
will be the right one of not? I hope you document everything. With
people like you trying things I am sure it will not be long before we
have a working model. I think the best way to do things is in a group
like this with other people who may have tried things like this in the
past or may have wanted to try it. There are people in the groups that
have a lot of knowledge and will share there thoughts on what you are
trying to do.
Trump
eldarion
everything I do.
Does anyone know about what gauge magnet wire should be used on a small
core like this? Would #30 AWG work?
Thanks,
Eldarion
softwa...@yahoo.com
Eldarion,
I agree with Trump. Keep trying different things; i.e., different
magnets of various strengths and widths, various wire thickness, total
turns, etc. Since Naudin used a different core with different
dimensions we cannot say for certain what magnet strength or type to
use or what wire gauge, etc. So as far as I know you would be the
first person to try the MEG with an AMCC 4 core. Unfortunately there's
no equation or way of telling what size wire or PM type or width will
work best.
There are fine details to MCE theory that I have not really touched
upon such as potential magnetic energy. Here are more details -->
All unsaturated magnetic materials contain a certain amount of inherent
magnetic energy, mostly in the domain walls, but I theorize there are
ways of vastly increasing this PE by means of a PM. It's too difficult
to accurately simulate in the mind and hence the need for computer
simulations. In a nutshell, the PM in the MEG rotates part of the cores
moments' perpendicular. Imagine you are holding two 1" round PM's that
are separated by say 1" that are magnetically aligned. Now it requires
a force to rotate the PM's so they are not in alignment. So when the
PM's are forcibly rotated out of alignment that is what I called
potential magnetic energy (PME). Going back to the atomic world in the
MEG things are more complex due to thermal vibrations, etc. Consider
the MEG at near absolute zero Kelvin. We know that when the atoms and
such are not violently vibrating that the magnetic material (Metglas in
this case) saturates like a switch. IOW, as we increase the applied
field there reaches a threshold where the core will suddenly flip.
That's a really square Hysteresis loop. Because there are no
appreciable thermal vibrations the atomic moments (small magnets) are
tightly bonded making the core tightly saturated. Now we insert the PM
in the middle of the core to make our MEG. Under normally temperatures
the PM would easily influence the core, but at such low temperatures
the core resists any change due to lack of vibrating atoms
(temperature). Of course if the PM is strong enough it will cause the
core to flip to align with the PM. My point is --> the idea is to pick
a PM that would be close to appreciably influencing the Metglas under
such low temperatures, but at room temperature the PM would appreciably
influence the Metglas. Therefore, as you can see, if this is true then
the magnetic strength of the PM in the MEG is *VITALLY* important! Too
strong a PM and we have no potential magnetic energy. Too weak and we
again have no potential magnetic energy.
In further detail, we have a PM that would change the moments at room
temperature, but not near zero Kelvin. So the ambient temperature is
helping break the tight magnetic moment bonds, which will actually cool
the magnetic material because it takes energy to help break such tight
bonds. If the PM is the precise strength then this results in a very
sensitive "free energy" balance where a slight applied field can cause
the core to go from being aligned with the PM to being saturated in a
closed loop around the core. During every MEG cycle there would be two
flips-- one flip on the left side of the MEG and then another flip on
the right side of the MEG. During each flip the ambient temperature is
assisting in this flip, if everything in your MEG is correct. Such
ambient temperature assists in magnetic avalanches. So the magnetic
moments in such an avalanche will rotate from one atom to the next like
a burning dynamite fuse. Thankfully ambient temperature is not biased
to any atomic rotational direction. IOW, as the magnetic moments flip
(thereby gaining kinetic energy) the ambient temperature (vibrating
atoms) will *not* apply a counter rotational force on the rotating
atoms in the avalanche. Thankfully on average the ambient temperature
applies an equal rotational force on all magnetic moments, which means
the magnetic avalanche kinetic energy is up for grabs. Normally micro
eddy currents absorb nearly 100% of such energy. In the case of
materials such as magnetite that have no appreciable eddy currents the
magnetic moment flips hundreds to thousands times faster and occur in
the nano & pico second range, which is in the hundreds of MHz to GHz
region. At such high frequencies the magnetic material efficiently
absorbs such radiation. Metglas is electrically conductive so there
will be micro eddy currents, which will slow down the magnetic moment
flip rate. Metglas cores are tape wound, meaning there will be no
macro eddy currents, but this will by no means prevent the micro eddy
currents. So we need another trick to rob the avalanche energy from the
micro eddy currents. One method is to pulse the core quickly enough so
the avalanches tend to occur simultaneously, for the most part. We need
the avalanches to be in phase. Under normal core usage the avalanches
are out of phase for the most part. So normally the macro size coil
cannot absorb the energy from the micro avalanches.
In a nutshell, if the pulse is too quick then the process is
inefficient. If the pulse is too long then the avalanches will mostly
be out of phase. So depending upon your core material type, core size
and shape we need the precise pulse duration. What works for the AMCC
320 will not necessary be correct for the AMCC 4 core. Furthermore,
the PM shape and strength is vitally important. In fact, the PM that
works for the AMCC 4 may be an entirely different PM material type than
what works for an AMCC 320. Perhaps the best PM for the AMCC 320 is a
ceramic 8, but that might be too strong for an AMCC 4 core.
I believe your small core has a potential of being more efficient than
a AMCC 320 core. Understandably people think bigger is better, but I
strongly disagree in this case. Given the same weight, a dozen small
MEG's should potentially generate more "free energy" than one big MEG.
Regards,
Paul Lowrance
MeggerMan
Phew, that was a long reply, I sort of understood some of it.
The number of turns on the input is 100 and because of the core size
then perhaps 0.3mm to 0.4mm wire would be ok, not sure what that is in
SWG, AWG.
I don't think the thickness of the wire is too critical, but the
thicker the wire the lower the losses through resistance. Thicker wire
will give a higher coil capacitance too, this may or may not be an
advantage.
The secondary windings, perhaps 0.2mm as you have 1500 turns to pack in
a small space.
I plan to make up three lots of 500 for each of the secondary windings,
purely so I have more options and to get around the high voltage issue
that JLN mentioned.
JLN was concerned that the high COP (4 watts input and 29 watts output)
may have been a measurement artifact.
I may get time to build a second mosfet drive circuit this evening and
update my MEG replication page.
Regards
Rob
softwa...@yahoo.com
> JLN was concerned that the high COP (4 watts input and 29 watts output)
> may have been a measurement artifact.
Hi Rob,
I just can't see that as a possibility. The power source is DC. Naudin
was measuring the DC current from the battery. So regardless how much
AC or spikes is on the battery, so long as the scope is able to measure
the DC current and given the DC voltage we know the consumed power.
That's the input. The output measurements should also be reliable.
I've never seen any appreciable reactance in Carbon resistors. So at
any given moment if you know the current and voltage on the load
resistor then you know the dissipated power. Although there could be
high frequencies on the resistor that Naudin's scope did not measure,
but that would only mean the device is generating more power than we
realize. One reason I think the MEG v2.1 is the best choice is because
Naudin shows scope shots of both current and voltage on the load
resistor. So if anything, the unit is more efficient then Naudin
thought. :-)
One other possibility is Naudin's scope was faulty, but it's just hard
to believe he could have used such a faulty scope for so long and not
know it.
Am I missing something?
Regards,
Paul
Eldarion Telcontar
Thank you for all of the information! It took a while to read, but it was
well worth it.
It takes very little electrical energy to create a small, steady magnetic
field. I wonder if a computer could keep track of how close the core is to
maximum potential magnetic energy (based on power output or something) and
automatically adjust a small electromagnet that is creating the bias field?
Obviously, this is for a much later time, after the basic effect is fully
mastered. But it could temperature stabilize the device...
Just my $0.02, and thanks again for all of the info!
Eldarion
> --
> Internal Virus Database is out-of-date.
> Checked by AVG Free Edition.
> Version: 7.1.409 / Virus Database: 268.15.26/594 - Release Date:
> 12/20/2006
>
>
g trump
|
Very good info Paul, Eldarion may still be in a mind spin, that was a lot of info. Pardon my thoughts, but if I must get some of my thoughts out.
1) One of the topics was the PM size and strength, it appears very crucial to obtain the right PM, What would the chances be to have a electromagnet to obtain a strength and then when the proper strength is obtained through testing then the PM of that gauss strength could be obtained?
Working with a core ( AMCC 320 core ) seems to have so much variable situations it seems almost impossible to know what to do. Wire size, windings, PM size and strength and so on. How close to the core is the PM and what dangers would be present in case of voltage and at what point?
If I was Eldarion, I may not have any idea on which foot to start off with, just a few windings more or less in a case like this or a PM to strong or weak could mess up the whole test, or am I wrong?
Trump
-------Original Message------- |
|
From: softwa...@yahoo.com
Date: 12/28/06 09:53:51
To: Energy Mover
Subject: Re: Discussion on eldarions-tinymeg |
eldarion
No, I think I am OK and understood most of what Paul had to say. :-)
I am thinking that it maight be a good idea (for initial testing
purposes) to substitute an electromagnet coil for the magnet. That
way, I could easily vary the field strength until I find out what
strength works. Once I know the desired strength (and if the
electromagnet is simply drawing too much power to close the loop), I
could find a magnet with the correct characteristics to replace the
coil. Or am I thinking wrong?
As far as dangers with respect to voltages generated, etc. I think I
know how to deal with that. I have built several solid state Tesla
coils (that have died spectacular deaths :-)) that ultilized resonance
to build up extremely high voltage.
The bit about several smaller MEGs creating more power than one big one
makes sense knowing that the core tends to absorb the MCE radiation.
Eldarion
g trump
|
Eldarion,
No, I think with your experience working with Tesla Coils, you should have a lot more understanding on what you are looking for and also on how to obtain it. Even though the thought of using an electromagnet was just a brain storm I thought it may have some advantages, but there again it may also just add a little more confusion to the whole thing.
I am sure I am wrong on this part, but my question would be "could the PM field strength be affected by the number of turns of wire, wire size and temp?"
Trump
-------Original Message------- |
|
From: eldarion
Date: 12/28/2006 9:18:35 PM
To: Energy Mover
Subject: Re: Discussion on eldarions-tinymeg |
softwa...@yahoo.com
least see what happens. Although this will have very noticeable side
effects because iron has high permeability where as most PM's have very
low permeability. Also iron does not have the coercivity of a PM.
Essentially the iron will act as a partial short for the magnetic lines
on the toroid, but who knows, may be Eldarion will discover a new
effect. :-)
Perhaps a good choice is Alnico 5. According to FEMM it's 4.6 times
weaker than Ceramic 8, but it still has very low permeability. On the
other hand, Alnico 8 could be a bad choice because its permeability is
4.5 times greater than Alnico 5.
One great thing about NdFeB is it's extremely low permeability and high
coercivity. Perhaps you could use NdFeB, but place dozens of thin
paper spacers between the PM and core. See the image "
variable-PM-MEG.gif" in the files in beta google groups -->
http://groups-beta.google.com/group/energymover/files
Direct link to image -->
http://energymover.googlegroups.com/web/variable%20PM%20MEG.gif?gda=c73EpUkAAADFHwvm2Z-cCLf7kGThbonc1aLPmE3L4jqp5oUQgt5ccBVJVT3hrgKrn_IySc8la3LMbVQ0zVLszEdAETfVITh2PgvYu07UMpClZNRuJp1kEA&hl=en
Black = Metglas core
Red = NdFeB PM
Green = lots of thin spacers, perhaps paper
The green is dozens of thin slices of paper. So you could gradually
move one piece of paper at a time from the outer green to the center
green. This would gradually bring both PM's closer to the Metglas core.
Regards,
Paul Lowrance
softwa...@yahoo.com
I created 3 FEMM files for you in the google groups beta files area -->
http://groups-beta.google.com/group/energymover/files
MEG AMCC 4 test 1.FEM
MEG AMCC 4 test 2a.FEM
MEG AMCC 4 test 2b.FEM
MEG AMCC 4 test 1.FEM uses an electro iron PM with a gap in the center.
The gap is important.
MEG AMCC 4 test 2a.FEM and MEG AMCC 4 test 2b.FEM are PM versions, but
there's very little field variance on the metglas core. Actually the
biggest variance is when both PM's are evenly spaced between top and
bottom. You can play around with it. At one time I had a version that
varied between 1.2 T and 0.75 T, but I overwrote it.
MEG AMCC 4 test 1.FEM has the best potential as far PM variance, but it
could have appreciable side effects because the iron electromagnetic
still has some effective permeability, but the air gap greatly
decreases that.
The FEMM core dimensions are not exactly an AMCC 4. I simply took an
AMCC 320 core and scaled it down 1/3.
Rob,
I created a FEMM materials properties for your core, 2605SA1. It seems
close. You can find it inside the above FEM files. Also there's an
updated 2714AF FEMM materials properties.
Regards,
Paul
Eldarion Telcontar
Paul,
Thank you, that will be a great help! I am currently constructing the driver board; I should have it done in a day or so.
Eldarion
-----Original Message-----
From: energ...@googlegroups.com on behalf of softwa...@yahoo.com
Sent: Fri 12/29/2006 9:16 AM
To: Energy Mover
Subject: Re: Discussion on eldarions-tinymeg
eldarion
MeggerMan
Nice work, at this rate you will be finished before me.
Word of warning about the metglas core, do NOT do like I did and wrap
pvc insulation tape direct to the core, because when you come to remove
it, it will take the metglass layers with it.
Lots of sharp slivers of metal, and sharp as glass, I have already got
some metal splinters.
I wrapped the waxed paper over the core first, then pvc tape then the
copper coils.
You really have to treat this core like its a very delicate ornament.
It even worries me to clamp the two halves too tight.
Regards
Rob
Paul
much. Wow, tape was enough to remove some Metglas?
Eldarion,
Nice detailed pics! I wish I had a 1536 x 986 pixel camera! I see
there's a small motor in picture. Could that be a load? That's an
interesting idea. I think that motor probably has a very high
breakdown voltage. Some motors could perhaps have high enough
resistance to work as a MEG load with a little help getting the motor
spinning.
Regards,
Paul
eldarion
Thanks for the words of wisdom! I had no idea the core was that
brittle.
Don't worry, I'm going to slow down a bit here :-) I still have to
wind the core, and have not found a good way to do so yet. Having
wound several-thousand-turns cores by hand in the past, I am reluctant
to do so again.
Paul,
I had not thought of using the motor as a load; it was just sitting
there from a different project. Good idea! I guess sometimes it pays
off to have a messy workbench... :-)
Happy new year to you all, and good luck in 2007!
Eldarion
MeggerMan
Well I did a quick test this afternoon with an input coil of about 32
turns, input voltage was about 4V and the current virtually zero (less
than 10mA). The output coil, 42 turns of 0.56mm, voltage was around 60V
at certain frequencies unloaded, the voltage drops to about 2 to 3
volts when the lamp is connected.
The core made some very odd screeching sounds at the lower frequencies.
I managed to get my 5W 12V lamp to fully light with 12V input.
Only problem is I have now cooked two mosfets in the process.
I tried to apply a capacitor (150nF) and resistor (1K) to reduce the
back EMF but to no avail, it still cooked the mosfets.
May need a lower resistance for the resistor.
Sorry I have no frequencies as it cooked before I could start measuring
frequencies.
The wave forms are very strange and far from a sine wave, more like a
Swiss mountain range.
The odd thing is that the voltage across the mosfet climbs to just over
30V as I increase the input voltage from 0 to 4V and then stays there
all the way to 12V, I think at about 13V it cooked.
It is almost as if the mosfet has some kind of 30V zener diode built
in.
I have not updated my page as I did not document anything before things
started to heat up.
Its funny really to see a mosfet suddenly max out the current (2.5 to
3A) at only 2.5V, then you know something is up.
Luckily the mosfet is in a socket and there are 23 more mosfets left.
Your messy workbench looks just like mine, wires, components
everywhere, you're in good company. ;o)
Regards
Rob
Eldarion Telcontar
The MOSFET itself is a type of diode, but I think you may have been
triggering some kind of ESD protection mechanism. You could try putting a
fast-acting Schottky (sp?) diode across the load to snub an back-EMF.
I just had another thought. Would it be a good idea to run the core at its
resonant frequency? If so, it would eliminate the back-EMF problem. If
this wouldn't hinder energy extraction, I could elaborate on it later.
Eldarion
----- Original Message -----
From: "MeggerMan" <k1n...@ntlworld.com>
To: "Energy Mover" <energ...@googlegroups.com>
Sent: Saturday, December 30, 2006 3:55 PM
Subject: Re: Discussion on eldarions-tinymeg
>
> --
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.1.409 / Virus Database: 268.16.0/610 - Release Date: 12/30/2006
>
>
PL
> Paul,
>
> I had not thought of using the motor as a load; it was just sitting
> there from a different project. Good idea! I guess sometimes it pays
> off to have a messy workbench... :-)
>
> Happy new year to you all, and good luck in 2007!
Thanks. Happy New Year to you, Rob, Gerry, and all!
Paul
Paul
My vote is to try any type of load at least once. Although for the
MEG I think something like an avalanche diode or high voltage zener
could work well. You could pick a diode with the appropriate
breakdown voltage.
Regarding the windings. I'm sure you and Eldarion know, but it's
important to wind the core the best way to minimize capacitance and
possible voltage breakdown. One no-no is to not place some type of
electrical insulating tape (with high breakdown voltage) between
windings. The problem is you have say 100 windings of wire and then
all of a sudden the two wires are very close to each other. Without a
nice thick layer of high voltage electrically insulated tape that's
the same thing as placing hundreds of substantial capacitors between
winding #1 <--> #100, #2 <--> #101, #3 <--> #102, #4 <--> #103, #5
<--> #104, #6 <--> #105, #7 <--> #106, #8 <--> #107, etc. etc.
Regards,
Paul Lowrance
On 12/31/06, MeggerMan <k1n...@ntlworld.com> wrote:
>
> Hi Paul and Gerry(Eldarion?),
> Happy new year to you all.
> Hopefully in 2007 we can prove that the MEG works and close the loop.
>
> Motor is one idea, how about a full size 40watt fluorescent tube, as I
> understand it, virtually no resistance until you reach a certain
> voltage, then the gas is ionised and current can flow.
> Possibly the actual output windings will work as a current limiting
> ballast anyway, if you load up the output coil with a lower resistance,
> the voltage drops, so it should self regulate.
>
> Regards
> Rob
>
>
> >
>
MeggerMan
multi-layers probably easier.
I think I will need to make up a jig to wind the output coils on. This
needs to clamp the core and have a counter balance weight. Also needs
to tension the wire and have a one way ratchet.
I bought a step counter/walking meter yesterday to use to count the
turns (for the winding jig).
I took it apart and it has a reed switch and magnet that increments the
display by one for each switch of the reed.
It would be useful to know what the capacitance of JLNs coils were.
I have a LCR meter that can measure coil capacitance.
Regards
Rob
Paul
As far as measuring the capacitance between windings, remember this is
not so easy. For example, if you take a cap and then place a resistor
in parallel then your meter cannot tell you what that cap is. Same
with a wound core, as it's a complex matrix composed of countless C's
and R's.
Regards,
Paul
MeggerMan
Yes, it turns out you are right, the Tenma 72-960 measures ordinary
capacitance not coil capacitance.
I just tested a 150uH pre-wound inductor and it gave a capacitance of
159mF, wow that is some capacitor for just a few turns of copper wire,
if only it were true. ;o)
I just analyzed my test core setup and realized that the coils were
creating the following:
N---S || S---N
S
|
N
Where N---S is one coil and
S
|
N
is the central magnet.
Which as you can see the central magnet is upside down, there was a 50%
chance I would get the polarity right first time and I got it wrong.
Perhaps I should have done the test I have just this minute carried out
(passing 2.5A through the coil and holding a marked up magnet on a
piece of string near to the core face.
Regards
Rob
Eldarion Telcontar
What frequency were you at? I tried a test of my core with 30 turns on the
input coils and 50 on the output coils. Even at 5 volts, the input coils
just sucked down a lot of current (several AMPS!) and the output coils
didn't even twitch.
Hmmm....
Any ideas?
Eldarion
----- Original Message -----
From: "MeggerMan" <k1n...@ntlworld.com>
To: "Energy Mover" <energ...@googlegroups.com>
Sent: Saturday, December 30, 2006 3:55 PM
Subject: Re: Discussion on eldarions-tinymeg
>
MeggerMan
Have you blown your mosfets already?
Not sure of the frequency but it was audiable. Could be the magnet that
was vibrating against the core.
I used a 2.2nF capacitor ( I think) and the magnet was the wrong way
round.
Unloaded you should see plenty of voltage on the output coil.
I would need to replace my mosfets and try again to give you the
frequency.
Try a bigger capacitor to start with.
Regards
Rob
Eldarion Telcontar
I found the problem after poking around a bit with an ohmmeter. Apparently,
the core is so sharp that the coils that I wound are all shorted out through
the core! :-(
I will be putting some paper or something in between my coil and the core to
avoid this problem next time.
Eldarion
----- Original Message -----
From: "MeggerMan" <k1n...@ntlworld.com>
To: "Energy Mover" <energ...@googlegroups.com>
Sent: Monday, January 01, 2007 5:04 AM
Subject: Re: Discussion on eldarions-tinymeg
>
MeggerMan
Yes, never wind direct to the core.
For one, when you take the windings off I bet some of the corners may
have crumbled very slightly.
I recommend a couple of layers of paper then a couple of turns of pcv
tape to hold it in place.
The windings do not need to be tight against the core material.
I suspect a circular former or in this case an oval former would not be
too in-efficient.
When they build high voltage transformers then weight several hundred
tonnes, the core/widing gap is big enough for a small cat to crawl
through.
Regards
Rob
Eldarion Telcontar
I rewound the core with 30 turn on the input and 100 turns on the output, as
well as a layer of paper in between the core and the coils. When I
energized the drive coil with a 9v battery, the magnetic field was
sufficient to slam the two core halves together from about 1/2-inch apart!
Fortunately, no damage was done to the core. The oscilloscope registered a
large voltage from the output coil at the same time; I guess it is starting
to work now!
Eldarion
----- Original Message -----
From: "MeggerMan" <k1n...@ntlworld.com>
To: "Energy Mover" <energ...@googlegroups.com>
Sent: Tuesday, January 02, 2007 4:41 AM
Subject: Re: Discussion on eldarions-tinymeg
>
MeggerMan
> large voltage from the output coil at the same time; I guess it is starting
> to work now!
Thats good to hear.
> energized the drive coil with a 9v battery, the magnetic field was
> sufficient to slam the two core halves together from about 1/2-inch apart!
Ohh, and don't do that either, I smashed a perfectly good ferrite E
core into about 4 pieces when I tried that trick, looks good, but I
won't be doing it again.
More experiments this evening, I'll let you know how it goes.
Regards
Rob
On Jan 2, 5:38 pm, "Eldarion Telcontar" <eldar...@pearsoncomputing.net>
wrote:
Eldarion Telcontar
No, I won't be doing that again! :-) Which brings me to another
question--what's a good way to clamp the two core halves together so that
they *won't* smash into each other on each drive coil pulse?
Thanks,
MeggerMan
together like I did.
They pull the halves together without being brutal, and you can adjust
the tension using finger force.
When you are ready to try a different coil you just carefully snip off
the ties at the clip, bit wasteful, but who cares.
eldarion
g trump
|
Eldarion,
Appears you are gaining leaps and bounds on the MEG project. Do you have any close up pictures of your set up showing the PM and the wound coils? I would like to see that part if you have the pictures.
What is your next step now?
Trump
-------Original Message------- |
eldarion
I have updated this page with the info you requested.
g trump
|
Seems kind of quiet the last two days, I hope I did not offend anyone on the links I found. Hope the project is still going well on the MEG.
Trump
-------Original Message------- |
Eldarion Telcontar
Internal Virus Database is out-of-date.