Metglas, MEG, Ambient Energy, Simulations
Paul
purchase the Metglas AMCC 320 cores for $110 from http://elnamagnetics.com
There's some very important information MEG researchers should know
about. I spoke with the engineer at Metglas and he said it's difficult
to couple the two U-shaped cut cores together tight enough to achieve
the materials exceptionally high permeability and coercivity. In fact,
the Metglas engineer said his test, which consisted of taping two
pressed cores together, resulted in a completely flat BH-curve with
relatively low permeability. I'm not certain how much pressure one
needs to apply on the core halves, but one should be very cautious as
such nanocrystalline and amorphous material is brittle. My suggestion
is to rub the two cores together thereby creating some micro powder to
fill in any gaps, and then press the two cores together. To verify the
two cores are correctly coupled you could apply two static AC signals.
One signal would be strong low frequency, no higher than 1 Hz. The
other will be a weak higher frequency signal of a few KHz's. It's
important the KHz signals peak to peak current remain relatively
constant. To achieve this you could place a high resistance resistor
in series with the KHz signal so it's peak to peak current does not
change much as the 1 Hz signal changes. The 1 Hz signal peak must be
high enough to saturate the core. Next, view the induced voltage on a
secondary coil. If the two cores are correctly coupled then the KHz
signal should mostly be low in amplitude along with a blip high in
amplitude. On the other hand, an improperly coupled core should
result in a flat KHz signal with no blip.
Also you'll want to store such cores stored in a desiccator-- low
humidity environment. Believe it or not, cheap cat litter that
contains silica gel works great. You'll want to microwave the cat
litter to expel the absorbed moisture. Once it cools then place it in
a seal tight bag or container along with the nanocrystalline core.
This will prevent the core from oxidizing.
I learned some very concerning information from the Metglas engineer.
The AMCC cores are not longitudinally annealed, but no-field annealed.
I don't know if they were ever longitudinally annealed, but this was a
shocker since my computer simulations (based on conventional physics)
shows the "free energy" comes from a coil robbing Magnetic entropy
away from Lattice entropy, which occurs when the core goes to
saturation. Present simulations indicate longitudinally annealed cores
have appreciable magnetic entropy, and non-annealed cores have
significantly less magnetic entropy, and transversely annealed cores
have negative entropy when such material is saturated. Actually
simulations show transversely annealed cores behave erratically
depending on various situations. Note that two U-shaped cores not
properly coupled perform just like a transversely annealed core. This
gap effect is seen in simulations due to the micro gaps between the
cut core.
In a nutshell, it's very important to verify your cut AMCC core is
properly coupled and exhibits the materials natural exceptionally high
permeability and coercivity. According to simulations the high
coercivity is very important in achieving magnetic entropy.
Here's an outline of what the simulation software reveals. With no
applied field the magnetic dipole moments on such material on average
are appreciably unaligned because the longitudinally annealing
discourages domain structure-- lower order, higher magnetic entropy.
When the core is saturated the magnetic moments are aligned-- high
order, low magnetic entropy. Normally when magnetic entropy decreases
there's an increase in lattice entropy. Meaning, magnetic entropy is
converted to heat. This is a well-understood process known as MCE
(Magnetocaloric effect). MCE is difficult to notice in typical
magnetic materials at room temp because the domains are mostly
saturated with no applied field; i.e., low magnetic entropy. When the
material is above Curie temp the magnetic moments are in disorder, the
domains are destroyed; i.e., high magnetic entropy. The problem is
such materials have hardly any permeability above Curie. Although,
Superparamagnetic materials have high magnetic entropy well below
Curie. According to simulations, such material with exceptionally
high permeability require extraordinarily small amount of energy from
the coil used as a catalyst to convert magnetic entropy to lattice
entropy (heat). Last year I witnessed MCE in a Metglas transversely
annealed core, but at the time I couldn't figure out why the effect
was so erratic, and even reverse some times. Actually I asked Metglas
for a longitudinally annealed core, but oddly enough they sent me the
worst type of core for "free energy," a transversely annealed core.
Recently, after discovering this, I bought some Metglas MAGAMP cores,
which are uncut longitudinally annealed 2714A core material. I'm still
in the process of measuring MCE at room temperature in such cores, but
so far it appears they definitely exhibit appreciable MCE at room
temperature. :-)
How to capture such magnetic entropy away from lattice entropy is
another story. What occurs is electrons flip, which generates a pulse,
emits photons. The flip rate depends on the type of atoms, lattice,
and material electrical resistance. Such Metglas cores have low
electrical resistance, which can significantly slow down the flip rate
due to eddy currents. Note that 18 micro tape wound cores reduce macro
eddy currents, but not nano scale eddy currents around the atomic
flip. Ideally the coil would collect a percentage of the pulse,
magnetic entropy. So now the core is saturated-- align magnetic
moments, low magnetic entropy. According to standard physics it
requires energy to break such magnetic alignments. Temperature
(vibrating atoms) breaks such magnetic bonds, which is why such
magnetic materials cool down when the applied field is removed-- the
later half of MCE. Therefore, the coil captures energy that is
normally converted to lattice entropy. When the field is removed the
material cools slightly more than it stated.
The end results are a device that moves ambient energy at the output,
which cools the core. Recent simulations are showing the core domain
structure can abruptly change when this occurs. If true, then it would
require a machine that adapts and re-tunes itself continuously and
dynamically. That could explain why Naudin had trouble closing the
loop. There seems to be controversy what Naudin failed. All I know is
that a person who lives in France, was in direct contact with Naudin,
and claimed that Naudin did indeed closed the loop, but the MEG would
run for a very brief time. He said Naudin could not figure it out. If
Naudin was capturing ambient energy from the core, then it's my
opinion that such a device is finely tuned and highly balanced, and a
slight change in core temperature would cause significant domain
structure. I think the Metglas MEG is legitimate, but someone needs to
spend the time to figure out how to make the circuit adapt to the
cores changes and/or somehow keep the inner core at a stable
temperature.
My final comment on the MEG is that presently I see no way of capture
such Magnetic entropy from ferrite cores. In a nutshell, such ferrite
cores are made of powdered iron. A bonding material separates the iron
particles. Therefore, with no applied field, each iron particle will
be appreciably saturated due to strong domains. Furthermore, a
saturated core would actually have *less* B-field magnetic entropy due
to the iron particle separations as caused by the bonding material.
Perhaps an intensely longitudinally annealed ferrite core could work.
Regardless, it's difficult to beat nanocrystalline and amorphous cores
that have permeability over 500000, especially if they are
longitudinally annealed! :-)
Regards,
Paul Lowrance
Paul
> Someone was asking about Metglas cores. At this very moment you can
> purchase the Metglas AMCC 320 cores for $110 fromhttp://elnamagnetics.com
I should provide instructions how to find the correct Metglas core as
used by Naudin. Go to -->
and click on the yellow button, "Search Inventory," which should take
you to -->
http://www.elnamagnetics.com/inventory/
Then type -->
AMCC-320
At this moment it shows -->
Item Number Description Alternate Item Number Qty In Stock Available
Soon
AMCC-320 00009109 7 12 SET
They have 7 in stock and 5 on order (12 available soon). Last month a
set (2 U-cores) cost $110.
Note that the AMCC cores are no longer longitudinally annealed, but no-
field annealed. I have no idea if Metglas was longitudinally annealing
the AMCC cores when Naudin purchased his. For the moment, it requires
a custom order to buy a longitudinally annealed AMCC core, but you
would probably have to sell your house to pay for it. Although Metglas
does sell a nanocrystalline and amorphous longitudinally annealed core
in a different material. Bad part is this material made of 2714A is
roughly 1/3 the saturation. That makes me highly suspicious what's
really occurring in the lattice to cause such low saturation. For
those who would like to research this material you can buy an uncut
toroid core about the size of a U.S. silver dollar for $6.10 each. The
part number is MP2510P4AS. elnamagnetics.com presently has 148 in
stock.
A perhaps even better material type is Finemet circumferentially/
longitudinally annealed or their no-field annealed. It has an equal
saturation as the AMCC cores. elnamagnetics.com presently has 882 of
FT-3KM K1208A in stock, but this particular core is only no-field
annealed. Also it's a tiny core, smaller than a penny in diameter.
This core is also ***NOT*** longitudinally annealed, but no-field
annealed. Even so, it's an uncut toroid, so you don't have to worry
about properly coupling the two cut cores. Therefore, this no-field
annealed core will have relatively high coercivity. Another problem is
it's a round, a toroid, so the permanent magnets will not flush
against the core, but I don't know if that's important.
You can get longitudinally/circumferentially annealed Finemet cores
directly from Metglas. That would be their FT-3AH core material, but
again they are very small. Presently elnamagnetics.com does not carry
these cores. Perhaps if enough people requested these cores they would
carry them. I was able to get one from Metglas, part number
MP1603VF3T.
For those not interested in researching and experimenting, who would
rather stick to the core as used by Naudin, but who do not want to
spend a lot of money there's always a small AMCC-4 that
elnamagnetics.com presently sells for $8.50 a set. Of course you'll
have to start from scratch to find the exact number of primary and
secondary turns, the correct resonance, etc. since this core is
considerably smaller than Naudin's AMCC-320 core. Also we don't know
if Naudin has a longitudinally annealed core. Present standard AMCC
cores are no-field annealed. Just remember you'll need to verify such
a cut core is properly coupled to achieve the materials
extraordinarily high permeability and coercivity. Please see the first
post in the thread "Metglas, MEG, Ambient Energy, Simulations" at this
google group forum on how to verify your core is properly coupled.
Regards,
Paul Lowrance
MeggerMan
I did some tests a while back using an inductance meter coupled to the
AMCC-320 core and did find that I needed to press the two halves very
firmly together to get a good inductance.
I think that provided you follow the white marked manufacture lines
(some kind of white paint marked down one side) and compress the core
halves with some nylon bolts I think it will be as good as you can
get.
Certainly rubbing them together may help but at the same time you may
end up separating the laminations or introducing gritty particles that
may separate two flush faces.
Your best bet may be to use applied pressure from a bolt/studding with
some elasticity like a M5 or M6 nylon studding. You could use some
10mm acrylic end-plates joined by the studding to compress the two
halves.
I am currently exploring the John Bedini/Mark Window motor before I
flip back to the MEG, I have not forgotten it, just side tracked.
I can perform some tests on the core if you want?
Rob
Paul Lowrance
> Hi Paul,
> I did some tests a while back using an inductance meter coupled to the
> AMCC-320 core and did find that I needed to press the two halves very
> firmly together to get a good inductance.
> I think that provided you follow the white marked manufacture lines
> (some kind of white paint marked down one side) and compress the core
> halves with some nylon bolts I think it will be as good as you can
> get.
> Certainly rubbing them together may help but at the same time you may
> end up separating the laminations or introducing gritty particles that
> may separate two flush faces.
> Your best bet may be to use applied pressure from a bolt/studding with
> some elasticity like a M5 or M6 nylon studding. You could use some
> 10mm acrylic end-plates joined by the studding to compress the two
> halves.
Hi Rob,
Sounds like good advice. Measuring the inductance could help. Although that's
not a guarantee. The only way to know for certain if the cores are properly
coupled is to view the hysteresis loop. A cheap an easy way is to apply a very
low frequency AC signal in addition to a KHz signal. The low frequency current
(say 1 Hz) should be strong enough to saturate the core. The KHz signal should
be weak, just enough to measure on the secondary coil. The KHz induced voltage
magnitude on the secondary will show the relative permeability. What you want to
see is a sudden rise in permeability that lasts a short duration relative to the
entire cycle.
> I can perform some tests on the core if you want?
I'm looking forward to your MEG replication. Maybe you could work on it part
time. I think it's time as many people as possible purchase a Metglas core and
replicate Naudin's MEGv2.1. I just bought an AMCC-320 for $110, and an AMCC-4
for $8.50 plus a few dozen other Metglas cores including some Finemet. :-) Also
I purchased some 2W carbon composite resistors. When time permits I'll create
Naudin's conditioned resistors and replicate the MEG. It appears Naudin spent a
lot of time on getting the MEG just right. Perhaps he just got luck, I don't
know, but it seems everything needs to be just correct on the MEG.
Therefore, anyone who is interested in "free energy" I would now strongly
encourage them to replicate the MEGv2.1. The more people working on it will
increase the odds of someone getting the correct combinations. Someone just sent
me some quotes from Naudin. Take special notice on Naudin's strong emphasis on
the correct permanent magnet and how the Metglas core must absorb all the field
from the magnet. Also Naudin claims the load must be non-linear -->
---
From: jnaud...@aol.com
Subject: For the MEG explorers....
Dear MEG explorers,
For a successfull replication of the MEG, I recommend you to use a non linear
load ( but a non inductive and a non capacitive load ) :
- a "conditionned" RLoad (100 Kohms, non inductive carbon, 5Watts)
- or a MOV (Metal Oxide Varistor) is REQUIRED for getting the output datas
that I have measured with my MEG.
You may put this non linear component in serie with a linear component ( a
pure resistive linear load ) for measuring the I/O.
I give you a very important tip about the MEG design :
ALL the B-field lines of the magnet MUST BE fully absorbed by the core, so,
if you use a Teslameter and you try to measure the magnetic field around the
core, you will not be able to detect any magnetic field, this is the key for a
successfull replication. In all other cases, you build a common step up
transformer ( with about 70% of efficiency ).....
You will find all the data with the wires size and components references
about my Meg device at
:http://jnaudin.free.fr/html/megv21.htmhttp://jnaudin.free.fr/html/megv21.htm">http://jnaudin.free.fr/html/megv21.htm</A>
I have used stacked ferrite magnets :
Magnets : Ferrite Barium 40 x 25 x 10 mm ( anisotrop )
Energy produced ( BxH ) max : 29.5 kJ/m3 - 3.7 MGOe
Remanence ( Br) : 400 mT - 4000 G
Coercitivity ( T=20°C ) :
bHc : 160 kA/m - 2000 Oe
jHc : 165 kA/m - 2050 Oe
Permeability : 1.1 mT/(kA/m)
Temp. coef. -0.20%
Max operating temp. : 200 °C
Density : 4.9 g/cm3
Curie point : 450 °C
PLEASE, read carreully the main MEG paper from Tom Bearden :
The Motionless Electromagnetic Generator:Extracting Energy from a Permanent
Magnet with Energy-Replenishingfrom the Active Vacuum.
Don't forget that the results published in my web site are the result of an
attempt of a private and a fully independant replication of the Bearden's
MEG.
The diagrams published in my web site are NOT the original MEG diagrams
being tested by the Bearden's teamwork or by some accredited labs.
On January 2001, I have decided to stop the publishing of the results on my
web site about my MEG because the MEG is in a final patent pending phase by
the Bearden Teamwork and it will be soon granted.
Good experiments and good luck during your replication,
Best Regards
Jean-Louis Naudin
---
Regards,
Paul Lowrance
sandpiper
Looks like I need to get off my butt and start to read about Naudin's
MEGv2.1, and then
when I understand more about it I will purchase a Metglas core and
then start working on
the project.
Sandpiper
> From: jnaudin...@aol.com
> Subject: For the MEG explorers....
>
> Dear MEG explorers,
> For a successfull replication of the MEG, I recommend you to use a non linear
> load ( but a non inductive and a non capacitive load ) :
>
> - a "conditionned" RLoad (100 Kohms, non inductive carbon, 5Watts)
> - or a MOV (Metal Oxide Varistor) is REQUIRED for getting the output datas
> that I have measured with my MEG.
>
> You may put this non linear component in serie with a linear component ( a
> pure resistive linear load ) for measuring the I/O.
>
> I give you a very important tip about the MEG design :
> ALL the B-field lines of the magnet MUST BE fully absorbed by the core, so,
> if you use a Teslameter and you try to measure the magnetic field around the
> core, you will not be able to detect any magnetic field, this is the key for a
> successfull replication. In all other cases, you build a common step up
> transformer ( with about 70% of efficiency ).....
>
> You will find all the data with the wires size and components references
> about my Meg device at
> :http://jnaudin.free.fr/html/megv21.htmhttp://jnaudin.free.fr/html/meg...">http://jnaudin.free.fr/html/megv21.htm</A>
Paul Lowrance
how you can see light right between both top and bottom gaps. I think it's
extremely difficult to achieve a properly coupled nanocrystalline core. Ron
Martis, the Metglas engineer was unable to properly couple two AMCC cores. When
asked how Metglas achieved the square looking hysteresis curves on their sites
he said they used uncut cores. This does not mean it's impossible with cut
cores, and actually I did not ask Ron.
http://www.conspirovniscience.com/meg/expe2005.php
http://www.skif.biz/index.php?name=Forums&file=viewtopic&t=982&pagenum=ALL
Not sure what language the sites are in, but both sites look like great MEG
research sites. Although the MEG in the first site is terribly coupled, LOL.
Really the only way to test how well such a core is coupled is to perform the
simple test as previously described, but IMHO a first step is to close the gaps
so no light shines through. If you can see light through any part of the gap
then that's equal to at least 250000 atoms in thickness, which will absolutely
destroy this materials exceptional properties.
Another potential problem is not storing these cores in a desiccator-- low
humidity environment. Again, cheap cat litter containing silica gel works great.
One final note, the MEG in the first link uses tiny magnets, probably NdFeB
magnets. Although it's worth trying, IMHO it's not the way to achieve "free
energy" from the MEG. I think the magnets should cover the entire meg, and IMHO
NdFeB are too strong. Refer to my previous post for Naudin's ferrite magnet
datasheet.
Regards,
Paul Lowrance
MeggerMan
Could you ask the people at Metglas if it is possible to buy a
AMCC-320 uncut.
It will be difficult to wind the coils but not impossible, especially
if we use multi-stranded wire.
The first site I think is French and the second may be Russian, and
you are right about the first core setup, it does look like a half
hearted effort to secure the windings, but there is no attempt to
compress the core halves together, what were they thinking???
When we build large transformer cores at the company where I work,
they have to clamp the whole lot together using first hydrolic rams
then tie bolts and large clamping frames so that nothing can move or
fall apart.
Even the windings are glued together.
I think their circuit is lacking the pre-driver for the mosfets that I
have shown in my circuit.
I will try and get the artwork completed this weekend so I can build
the pulse generator.
Regards
Rob
sandpiper
I see the light between the cores in that link you posted. The cores
are shaped like a U, as you mentioned earlier, and have the windings
around them. I will look into getting a pair of cores in the near
future. I will verify the list from Jean-Louis Naudin link for the
wire size and so fourth. I may be able to convert both of those links
into English if you feel it would do any good. From what Ron posted
earlier, "I think the first one is French and the second may be
Russian".
Sandpiper.
On May 3, 12:49 am, Paul Lowrance <energymo...@gmail.com> wrote:
> Here's an example of a MEG core that may as well be a cheap ferrite core. Notice
> how you can see light right between both top and bottom gaps. I think it's
> extremely difficult to achieve a properly coupled nanocrystalline core. Ron
> Martis, the Metglas engineer was unable to properly couple two AMCC cores. When
> asked how Metglas achieved the square looking hysteresis curves on their sites
> he said they used uncut cores. This does not mean it's impossible with cut
> cores, and actually I did not ask Ron.
>
> http://www.conspirovniscience.com/meg/expe2005.php
>
> http://www.skif.biz/index.php?name=Forums&file=viewtopic&t=982&pagenu...
Paul Lowrance
MeggerMan wrote:
> Hi Paul,
> Could you ask the people at Metglas if it is possible to buy a
> AMCC-320 uncut.
I haven't asked Metglas, but an uncut AMCC core probably falls under a custom
order-- $$$. Although for $6.10 each you can buy a Metglas MP2510P4AS core,
which is uncut longitudinally annealed 2714A material. It's a little small,
1.09" outer diameter. Also it's a round toroid.
> I will try and get the artwork completed this weekend so I can build
> the pulse generator.
That would be great. Hopefully other people also will begin replicating Naudin's
Metglas MEGv2.1. I'm also looking forward to replicating Naudin's MEG since I
just bought two AMCC cores, 320 & 4. Presently I'm redoing my MCE experiments
with a MP2510P4AS core. The core Metglas sent me last year was transversely
annealed!!! I asked for a longitudinally annealed core, which explains the MCE
instability that was driving me crazy. That pretty much wasted 4 to 6 months of
research.
IMHO it's very important to at least perform a cheap hysteresis test to see what
Metglas really sent you. What's interesting is that Ron, the Metglas engineer,
was unable to get a cut AMCC core to properly couple. Notice Metglas website
does not specify how the AMCC cores are annealed. So they could send people any
type of annealing processed core. Wouldn't it be crazy if upper management was
recently forced to transversely anneal all AMCC cores? That would pretty much
destroy the MEG design. Ron Martis sent me the URL of his hysteresis graph of a
taped cut AMCC core -->
http://www.metglas.com/downloads/powerlite.pdf
Now take a look at what the standard stock AMCC core should look like -->
http://www.metglas.com/downloads/2605sa1.pdf
It should look like the "no-field annealed" square BH curve, but it does not. It
looks just like a transversely annealed core. Two things can cause that:
1. The 2 cut core halves were not properly coupled.
2. The core was truly a transversely annealed core and not a no-field annealed.
Regards,
Paul Lowrance
Paul Lowrance
> Ron Martis sent me the URL of his hysteresis graph of a taped cut AMCC
> core -->
>
> http://www.metglas.com/downloads/powerlite.pdf
>
> Now take a look at what the standard stock AMCC core should look like -->
>
> http://www.metglas.com/downloads/2605sa1.pdf
>
> It should look like the "no-field annealed" square BH curve, but it does
> not. It looks just like a transversely annealed core. Two things can
> cause that:
>
> 1. The 2 cut core halves were not properly coupled.
> 2. The core was truly a transversely annealed core and not a no-field
> annealed.
>
Actually I think it'2 #2 because Ron's BH graph shows less permeability than a
transversely annealed AMCC core. I think Ron is correct in that the core is
no-field annealed core, but just not fully coupled. I wonder which is worse, a
coupled transversely annealed core or an improperly coupled no-field annealed
core. :-)
Regards,
Paul Lowrance