[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: xfrmr question, was: Re: That other HV List



Original poster: "Ed Phillips by way of Terry Fritz <teslalist-at-qwest-dot-net>" <evp-at-pacbell-dot-net>

Tesla list wrote:
 >
 > Original poster: "Stephen Conner by way of Terry Fritz 
<teslalist-at-qwest-dot-net>" <steve-at-scopeboy-dot-com>
 >
 > I think you can use this formula for ferrite cores. It won't be totally
 > accurate because it assumes a sinusoidal voltage and most switched-mode
 > power supplies run with a square wave. So:
 >
 > For E use the peak value of the fundamental component of whatever waveform
 > you're feeding the transformer. If you can't be bothered with Fourier
 > analysis then just multiply the amplitude of the square wave by 1.5 :)
 >
 > Also the saturation flux of ferrite is not the same as transformer iron. I
 > found this explanation of the equation on
 > http://www.tortech-dot-com.au/toranalysis.html
 >
 > "In terms of real-world transformer design, Faraday's low becomes:
 > e = 4.44NBfA x 10^8
 > where the voltage in a winding is proportional to the number of turns (N),
 > the flux density 'fl gauss (B), the Ac
 > -frequency-/ (I), and the core cross-sectional area in square 
centimeters (A)."
 >
 > So B is in gauss. (There are 10,000 gauss in one Tesla) You want to aim for
 > a value that's about 75% of the saturation flux density for the material.
 > So for your ferrite core, B would be about 0.75*0.51*10000= 3825.
 >
 > I think the 60,000 figure for transformer iron might be wrong, given that
 > it has a Bsat of between 12000 and 17000 gauss depending on the type. Or
 > 'B' might be in different units than gauss, in which case the 3825 figure
 > for ferrite is obviously B.S.
 >
 > To give you a ballpark idea: Ordinary sized ferrite cores (0.5 in2 cross
 > section) running at about 30kHz usually generate about 5 volts per turn.
 >
 > Steve C.
 >
 > >the "roll your own transformer"
 > >web page <http://members.tripod-dot-com/~schematics/xform/xformer1.htm> has
 > >the formula for Primary Turns:
 > >
 > >N = (E X 10^8) / (4.44 X F X A X B)  Where F = frequency   A = area in sq.
 > >inches.   B = magnetic flux assume 60,000
 > >
 > >This is clearly for silicon-steel transformer core material. If I am using
 > >a ferrite core with these parameters what is the formula?
 > >         type 77 ferrite
 > >         permiability    u = 1800
 > >         saturation flux Bs = 510 mT (at H=1194 A/m)
 > >         residual flux   Br = 170 mT
 > >         coercive force  Hc = 16 A/m

	The value of 60,000 sounds like like the units are in "lines", long
practice.  There are several sites with good introductions to design of
magnetic circuits and conversion factors for the units. In the early
1930's edition of the ARRL handbook a working value of 12,000 GAUSS was
used, although modern steels are a lot better.  Use of 15,000 would be
pretty conservative unless the core is 50 years old or more.  The rest
of the values look good "in gauss", which I find more convenient than
Tesla.

	Anyone contemplating the use of ferrite cores for RF choses should read
and reread the formula above and then remember the "E" involved is of
the order of 15,000 volts.  Results will be discouraging!!!

Ed

P.S. Magnetic units are "fun" because there are so many different
systems of them.  Most engineers I know still use gauss.