antenna-theory.com

[Maxxpower]

Hi,
When designing a planar (PCB trace) antenna the wavelength should be with regards to the Er of the material correct? Or do I use Er=1 for the wavelength?
What does the width of my PCB trace do to the antenna properties?
PCB antennas are not routed over ground planes so I guess I'm a little confused as to how the trace width and Er affect it.
Thanks
Ken

[helix]

There is not a straightforward relationship between the effective wavelength (i.e., twice length at which a half-wave printed dipole becomes resonant) and the relative permittivity and thickness of the PCB.

If you check out the NEC (old Fortran MoM code) theory manual, you can find some info on analytical corrections for thin/cylindrical wires insulated with a cylindrical material with relative permittivity greater than 1. How that quantitatively relates to a printed/PCB problem is beyond me.

Sounds like you're kinda thinking about PCB trace width in the context of microstrip characteristic impedance. The thing is that antennas have antenna mode currents (also known as unbalanced currents) on them, and microstrip lines have transmisison line mode (also known as balanced currents) currents on them. These two types of currents serve very different purposes and behave very differently. On microstrip, increasing the trace width decreases the characteristic impedance. For a dipole, increasing the trace width lowers inductance, and tends to increase bandwidth. These two effects are related, but the way they're typically thought about is different.

Search for "fat dipole" for an explanation.