Source Impedance II

Thu Jan 19, 2012 1:53 am   

Hi,

Schematic Link:

http://img269.imageshack.us/img269/8108/seriesw.jpg


Power Supply: +12V, 8.5 A.

The two coils are 8 inches apart parallel to each other (Helm
Holtz) . Both coils Both coils ( primary coils )are rectangle coils.
Length = 18 inches, width = 9 inches. The wire guage is 38. Turn /
coil = 7. Number of strands per coil = 1050. ( Litz Wire)

Problems:

1. How can I effectively measure the flux produce by each coil plus
coefficient of coupling? I am using magnetic sensor but it is showing
2.3 volts DC all the time and shows 2.3V plus ripples (sine wave with
DC offset of 2.3 V) when I put it in the field. The ripple amplitude
is 500mV peak to peak with 100KHz frequency when it is closer to the
coil and dropped to 50mV when put it in the middle of the field.

Can anyone suggest a better and cheaper way to do this flux
measurement? Plus how can make sure that the field stays uniform
throughout the enclosure. I pumped up the current to 8 A but no luck
so far.

link is as below

http://www51.honeywell.com/aero/common/documents/myaerospacecatalog-documents/Missiles-Munitions3/HMC1043.pdf



Thanks

jess

Thu Jan 19, 2012 2:11 am   

On Jan 18, 3:53 pm, Jessica Shaw <jsscsha...@gmail.com> wrote:

For coefficient of coupling:

k = M/sqrt(L1*L2)

to determine M, use an inductance meter plus measurements for the
series-aiding and series-opposing configurations.

series-aiding:
Lta = L1 + L2 + 2M

series-opposing:
Lto = L1 + L2 - 2M

therefore,
M = (Lta - Lto)/4

Thu Jan 19, 2012 2:51 am   

Why do consider opposing? How to use inductance meter to measure M? I
am using meter 875B.

jess

Thu Jan 19, 2012 3:10 am   

Jessica Shaw wrote:

"moisture sensitivity level" of three? Three _what_?? Furlongs per
fortnight? Raccoons per garbage can? How about milligauss per %RH?
2)
Treat the pair like a transformer and measure inductance of L1, L2,
L1+L2, L1-L2, and since it looks symmetrical short L1 and measure
inductance of L2.
With that info, you can calculate the coupling coefficients M and K.
3)
Then with a little math you can calculate the field inside center of
them (series aiding is what i assume you want).
A bit more and one can make a plot of field strength off-center.
4)
That info can be used as calibration information for that fancy
whatza doozie sensor.

Thu Jan 19, 2012 3:23 am   

Jessica Shaw wrote:

http://www.allaboutcircuits.com/vol_2/chpt_9/1.html
And:
http://hyperphysics.phy-astr.gsu.edu/hbase/magnetic/indmut.html
There are scads of other references; "jess" use the baby bird, Goo-Gull.

Thu Jan 19, 2012 3:29 am   

Jessica Shaw wrote:


I used to have the "A" version of that, it was a nice tool, it worked
and it was accurate enough for quick checks however, I lend it out to
some one and you know the rest of the story.

The unit was replaced with the B version, in comparison, it is a
big peace of junk! It is nothing like the original model. I am glade
I don't rely on it, but, it works ok for repairing and doing basic
stuff.

If you opening the limited manual that comes with it, there are some
conversion formulas in there. But I am sure you have already looked there.

If two coils of the same value are tightly bound (Physically)
together, phases the same, and connected in parallel. You'll get the
same L as one coil when using that meter. Or any L meter for that matter.

That's because they are sharing the same B field and equalizing, to
put it simply.

That meter will not tell you the (M12)/(M21)(M) directly, How ever,
this all depends on the positioning of the coils from each other and how
they merge the B fields. Try pulling them apart using the Radius as a
measuring point and watch the L change on the meter.

With no change of L between 1/2 coils together normally depicts
a M or 1 how ever, you also have to factor in the case where the 2 coils
may not be exact to start with.


Jamie

Thu Jan 19, 2012 3:32 am   

On Jan 19, 1:51 am, Jessica Shaw <jsscsha...@gmail.com> wrote:

You use your inductance meter to measure the inductance of both coils
connected in series, twice, once with the winding hooked up so that
that the coils cooperate to create a magnetic flux (Lta), and once
with leads to one coil swapped over so that the coils oppose one
another (Lto).

As Simon Aysdie points out above, you can then calculate M from M (Lta - Lto)/4

--
Bill Sloman, Nijmegen

Thu Jan 19, 2012 12:33 pm   

On Jan 19, 3:23 am, Robert Baer <robertb...@localnet.com> wrote:

I don't think much of that exposition - or most textbook expositions
on transformers. I'd read everything that I could on transformers for
years before I came across the transformer equation

V1 = L1. dI1/dt + M. dI2/dt

V2 = M. dI1/dt + L2. dI2/dt

Where V1 and V2 are the voltages developed across coil 1 and coil 2 by
currents I1 and I2 respectively.

M = K. Square root of L1.L2 and is always less than 1 (though not much
less than one for good transformers).

I found it to be a revelation.

--
Bill Sloman, Nijmegen

Thu Jan 19, 2012 2:31 pm   

Bill Sloman wrote:

That's just the Fourier transform of the usual method.

Cheers

Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

Thu Jan 19, 2012 4:30 pm   

Bill Sloman wrote:

Whichever way I read that last bit, it always parses wrong.
It's the dimensionless coupling constant K that's always less
than 1, not the inductance M.

Jeroen Belleman

Thu Jan 19, 2012 7:45 pm   

Can you give any advice on the math?


jess

Thu Jan 19, 2012 8:53 pm   

On Wed, 18 Jan 2012 16:51:37 -0800, Jessica Shaw wrote:



You've been told all this before.

Connect the two coils so that the finish of the first goes to the start of
the second. That is series-aiding, the field of one coil reinforces that
of the other, which is normal Helmholz connection.

Measure the inductance of the pair between the free ends you have left.
That will be L1+L2+2M, where M is the mutual inductance, one coil to the
other.

Reverse the connections of one coil. The fields from each coil now oppose,
hence M is negative for each coil.

Measure the inductance of the pair again, That will be L1+L2-2M

Subtract the two readings: (L1+L2+2m)-(L1+L2-2M)=4M

Simple (??) algebra.



--
"For a successful technology, reality must take precedence
over public relations, for nature cannot be fooled."
(Richard Feynman)

Thu Jan 19, 2012 9:07 pm   

On Thu, 19 Jan 2012 02:33:20 -0800, Bill Sloman wrote:


Considerably less than 1 for coupled tuned circuits, such as IFTs.

K=1/sqrt(Q1.Q2) for critical coupling.

--
"For a successful technology, reality must take precedence
over public relations, for nature cannot be fooled."
(Richard Feynman)

Thu Jan 19, 2012 10:21 pm   

On Jan 18, 4:51 pm, Jessica Shaw <jsscsha...@gmail.com> wrote:

So you can derive M.


You don't "measure M" via the method I and others explained.

You need to *derive* M via the two actual measurements of inductance,
one in the series-aiding configuration, and one in the series-opposing
configuration. That is, you obtain M indirectly.

Thu Jan 19, 2012 10:51 pm   

On Jan 19, 12:45 pm, Jessica Shaw <jsscsha...@gmail.com> wrote:

Are you an EE? No E&M course? Freshman Physics?
Circular coils are easy, 'cause you can get an algebraic expression
for the field along the center line. (Using Ampere’s law) For a
rectangular coil you’ll have to do the integrals. (With messy
limits)

George H.