RE: Current Limiting and Impedence

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "Mark Dunn" <mdunn@xxxxxxxxxxxx>


Paul:

I will probably elicit howls from the experts, but here goes:

Ve, Ae, & Le are the Volume, Cross-Sectional Area, and Magnetic Path
Length respectively for a toroid equivelent to a core.  They are the
properties of that core.  They normally must be obtained from the mfg
specs of the core.

For salvaged cores, I measure the total volume of the core(Ve) and the
cross-sectional area(Ae).  I then compute Le = Ve/Ae.  I work in mm,
mm^2, mm^3 because my spreadsheets sre set up to work in those units.

Note that with an E-core Ae = the area of the center leg.  The outer
legs should be 1/2 the area of the center leg.

C is the core factor. I take C = Le/Ae.  This is an approximation that I
leavce for future discussion.

AL is the inductance factor for the core. Units typically
NanoHenries(nH).  It is the inductance of one turn on an ungapped core
of that geometry and that material.

L = AL * N^2  N=# turns

When salvaging cores I measure L(ungapped) with a known # of turns and
work backwards to get AL.  Normally, AL is provided in the technical
specs for the core.

B(magnetic Flux Density) can be computed in Tesla or milliTesla or Gauss
for that matter.  The core material has a peak limit and saturation
limit for B.  For example my ferrite material that has been discussed
has a Bpk = 320 mT & Bsat = 500 mT.  I think we learned the other day
that the steel in our MOT's is approximately Bpk = 1.6 Tesla and Bsat =
1.9 Tesla.

H is the magnetic field strength, typically expressed in amps/meter.

The permeability is then U = B/H (I am confused here on units, but they
I think they must be H/M)

The realtive permeability is Ur = U/Uo, Ur is unitless
Uo is the magnetic constant 4*3.14159*10^-7 H/M.

Ui is the initial permeability.  I think it would be the permeability of
an ungapped core.  It is a specified in the core technical
documentation, but I am having a hard time computing it from a salvaged
core.

Ue is the effective permeability.
Both Ui and Ue are unitless(They are relative to Uo).

Ue = Ui/(1+G*Ui/Le) where G = the total gap(2Xspacer)

Note Syd's post of May 6.  He uses the same formula but puts Ur in place
of Ui.  I think the semantics are different.

I am hoping to pursue discussion of the Ux's with the experts and
clarify which Ux is which.

Mark


-----Original Message-----
From: Tesla list [mailto:tesla@xxxxxxxxxx]
Sent: Tuesday, May 10, 2005 5:51 PM
To: tesla@xxxxxxxxxx
Subject: Re: Current Limiting and Impedence


Original poster: "Paul B. Brodie" <pbbrodie@xxxxxxxxxxxxx>

Mark,
Please excuse my ignorance. Will you please define the following: Ve &
Ae, Le & C (capacitance?), AL, 10555 nH (nano henries?), mT
(miliTesla's?), and finally, Ui, Ue, and Ur. I want to make sure I'm on
the same page with you. I really appreciate what
you are doing and your sharing it with me. You have been way more than
helpful. Regards.
Paul
Think Positive

----- Original Message -----
From: "Tesla list" <<mailto:tesla@xxxxxxxxxx>tesla@xxxxxxxxxx>
To: <<mailto:tesla@xxxxxxxxxx>tesla@xxxxxxxxxx>
Sent: Tuesday, May 10, 2005 12:44 PM
Subject: Re: Current Limiting and Impedence

 > Original poster: "Mark Dunn"
<<mailto:mdunn@xxxxxxxxxxxx>mdunn@xxxxxxxxxxxx>
 >
 >
 > Steve, Gerry, Jim, Paul, et all:
 >
 > When I started this thread it was to discuss issues with current  >
limiting and has now progressed into Charging Reactor Design.  That's  >
OK, but we need to keep straight what we are talking about.  >  > Right
now I am returning to the original current limiter issue and will  >
come back to Jim's charging reactor in later post.  >  > As you know, my
current limiter was saturating.  It was very interesting  > to follow
the discussion between Gerry and Steve, because what they  > discussed
was exactly what I was running into.  I needed many turns to  > get the
B(Core Flux Density) low enough, but the high turns drove the  >
inductance through the roof and demanded a huge gap to get it back down.
>  > Last week I began analyzing a "bad" MOT(coincidently a destroyed  >
charging reactor)  I have to see if I could use it for a current  >
limiter.  A similar "good" MOT had a primary inductance of ~300 mH(sec
> open).  I measured Ve & Ae and calculated Le & C.  Then back figured
> ~AL.  Preliminary analysis suggested B would be around 500 mT which  >
should be OK for steel.  >  > I cut the "bad" MOT unit apart and was
surprised to find that all the
 > E-core laminations were in the same section.   All the I-core
 > laminations were in the other section.  This meant I could easily gap
> this device.  (Note: I sawed it apart right at the weld on either
side.  > I ran a weld bead parallel, and above and below, each original
weld so  > that the laminations would not fall apart before I sawed
through it.)  >  > I made a 60 turn coil(R=0.2 Ohms) and placed it on
the E-core, clamped  > the I-core on top, measured V,I and computed Z &
L with no gap.  L = 38  > mH.  This gave me an AL = 10555 nH.  Does this
make sense for steel?  > I then measured L for a number of gap spacers
up to about .070"(note  > this means total gap of .140") when I reached
~4 mH.  My theoretical  > math suggests that I should be around .070"
total gap or .035" spacers.  > Are gapping calculations highly accurate
or do they just give an  > approximation?  B will be around 620 mT which
is good for steel.  > (Another data point of interest was .009" spacer
for .018" total gap and  > L = ~15 mH.  This would meet the criteria of
my original current  > limiting requirements.)  >  > Pls comment about
this design.  Like to delve into the math issues with  > you guys if
your willing to work with me.  I think my issues are in the  > Ui, Ue,
Ur area.  >  > Thanks.  > Mark  >  > Mark  >  >  >