# RE: [SI-LIST] : Trace Impedance Selection

From: Mary ([email protected])
Date: Wed May 10 2000 - 05:50:06 PDT

Mike,

In your last message, you said:

>"The actual equation that computes the radiation for a given line uses the
>log of the ratio of the trace characteristic impedance to the impedance of
>free space (120*Pi=377 ohms)."
>
>Therefore, when Zline = Zfreespace, the log(base 10) of one is zero and
this
>term (one of many in the equation) contributes nothing.

This is the way I interpreted your message at first, but if this is true,
the
magnitude of this term decreases as the characteristic impedance increases.
Since this was contrary to the point you were making, I assumed that this
term
appeared in the denominator of an expression for the radiated field.

If this term is added to other terms, I guess we'd have to know what the
other terms were before drawing conclusions about the effect of the
characteristic impedance. However, since characteristic impedance is not
inherently related to radiation, I believe that that equation can only be
applied to a fairly specific trace configuration.

>Although I do not know (nor does the EMCAD1 literature indicate) the
>boundaries within which their software/equations apply, I agree with you.
My
>tests on a demonstration board with 50 ohm and 100 ohm microstrip traces
>(verified by HP TDR equipment) indicated field increases of 8-9 dB for the
>100 ohm line relative to the 50 ohm line. This result is between the 6 dB
>and 12 dB boundaries of our discussion. Isn't that always the way it is?

This is an interesting experiment. As others have pointed out, lower
impedance
lines will have a higher current if the voltage is the same. Did you keep
the
voltage or the current constant when you switched impedances?

>Nevertheless, in case any of the above has misled anyone, the fact remains
>that surface traces DO emit electromagnetic fields and therefore DO
>The measured levels in the far field, even for simple topologies, will be
>influenced by any and all near by materials (conductors, dielectrics,
>permeabilities > 1, etc.) that can concentrate or otherwise distort the