PCB Antennas April 25, 2008
From time to time, I get questions about antennas (particularly PCB antennas), so I figured I would at least make a small “stub” of a post to serve as a place to discuss the topic.
For example, Sia writes “How can we find the impedance?”
For antenna impedance (and other antenna properties), there is no one simple answer. The impedance of an antenna varies with the geometry of the antenna, frequency used, and proximity to ground planes and other nearby conductive objects. That is what makes them fun to learn about for some people and evil black magic to others.
The best thing I can recommend is to do a Google search on “antenna impedance” and become familiar with the various standard geometry antennas that you will find (e.g. quarter wave, half wave dipole, folded dipole, yagi, loop, etc.). Each type has specific impedance, gain, directionality, and other properties, so you can make some design trade-offs. Since you asked about impedance, you may also want to Google “RF impedance matching”.
The actual math behind antennas is pretty esoteric (3-D partial differential equations) and is usually taught in the 3rd or 4th year of electrical engineering programs, and even then, only if RF is selected as the area of concentration. It is worth noting that specialized software is often used to solve these kinds of problems.
Since this is a stub article, at this point I will open the floor to reader comments!
- Posted in : Electrical Engineering, PCB
- Author : Brad
Comments»
Try EZNEC at http://www.eznec.com/
Your article caught my attention because I actually needed to calculate the power being driven into a PCB loop antenna to see whether I needed to qualify the product under FCC regulations.
I modeled the antenna with an inductance and series resistance, the latter being the actual resistance of the trace on the PCB. Then I did a simple I2R calculation knowing that I am driving the antenna with essentially a 200 ma square wave (100ma RMS current then).
The antenna is 6 concentic loops, each loop about 17mm long (total length of 4″) using 10 mil traces in 1 ounce copper. I came up with 3.9 milliwatts.
Don’t know if I’m being too simplistic. The antenna has a range of less then 1 inch and is used in a RFID application, so the low power did not surprise me.
Just wanted to give you some feedback and if you ever determine a definitive way to calculate antenna power, then I’m interested.
Hi George,
It sounds to me like you are calculating the Joule heating in the trace due to I^2*R losses instead of the RF power output of the antenna. The RF output will depend on the characteristic impedance and efficiency of the antenna at the frequency of interest as well as the matching to the source. It might be advisable to just get a measurement of the RF output.
Brad