Z0 = (60/sqrt(er)) * ln((4*b)/(0.67 * pi * w * (0.8 + t/w)))
where
er is the dielectric constant of the dielectric
b is spacing between the upper and lower ground planes
w is the width of the conductor
t is the thickness of the conductor
t, w, and b can be in any units but must all be in the same units.
It's assumed that the conductor is centered between the ground planes.
The equation is claimed to be accurate when w/(b-t)<0.35 and t/b<0.25.
This is from the Motorola MECL System Design Handbook. I've found the
microstrip equation in that book to be good in the vicinity of 50 ohms
and moderate er, but have never used the strip line equation quoted
above. The book shows measured results of a number of strip lines which
they say are very close to the equation's results, but all the
measuements are for er = 4.7, so I don't know the accuracy for air (er =
1).
Roy Lewallen, W7EL
----------- de Woody, WB4QXE -----------
Ham Homebrew/ATV Circuits -*- Scanning Electron
Microscope Images
The actual calculations are pretty complex and generally approximations
are used. I have some papers on this subject that I could photocopy and
mail them, or perhaps scan. (I'll make a followup post if I end up
scanning them.)
There was some stripline stuff in QST awhile back, and I ported some BASIC
code into C; you can get the ref's from my web page below...
best regards,
--
Dave * N3WTK (DM04xf) * http://isi9.mtwilson.edu * n3...@arrl.net