January 2017 13
connectors. If an electromagnetically shielded switch is
available, that would be a good choice.
Extending Wi-Fi Range on the Cheap
QMy (step) granddaughter wants to make a Wi-Fi signal extender for a science fair. I was wondering if a purely passive circuit might work. Could we put a dipole at one end of some coax
and then at the far end just have the thing loop back? If the
loop was big and part of it passed near a laptop, would it
rebroadcast? Would it help if the circuit was tuned to 2. 4
MHz or 5.whateveritis MHz? Am I being crazy?
Palo Alto, CA
ANo Randy, you’re not being crazy at all! What you propose is close to what you want. It’s called a passive repeater. I’ve done it myself on the two-meter amateur band.
Instead of a loop at the far end, you’d want another
dipole. The idea is to construct a dipole that is resonant
in the middle of the Wi-Fi channel spectrum (or on a
particular channel that she’s using), pass it through some
coax, and then build a dipole of the same dimensions at
the other end of the coax.
If the dipole is resonant, in other words 1/2l (or
wavelength), then the feed point impedance would be 72
ohms resistive, which is very close to the characteristic
impedance of RG- 6 75 ohm low loss coax — what’s used
in cable television everywhere. One dipole would couple
energy from the propagating electromagnetic energy from
the Wi-Fi base station into the coax. The energy would flow
down the coax, shielded from radiating externally. Then,
the dipole at the far end would reradiate that energy into
the destination area.
Calculating the dipole length requires a slight
adjustment to the speed of light in a vacuum. This is a
somewhat complicated calculation that is dependent on
the thickness of the wire relative to the wavelength. We
call this unit-less number the velocity factor, and for this we
will just guess that it will be around 0.95. A simple rule-of-thumb calculation to determine the length is shown in
Figure 3, where A is this velocity factor. When I run the
numbers with a frequency of 5. 5 GHz, I get 0.026 meters,
or 26 millimeters. That’s the entire length of the dipole, so
each leg is half of that.
Cable TV coax is usually terminated with F connectors.
She could get some solderable F connectors and then
construct the dipole from copper wire. The skin effect
is the tendency for electromagnetic energy to flow on
the surface of a conductor at AC. This effect is more
pronounced with increased frequency. As less depth of the
conductor is used, the effective resistance increases.
The skin effect is quite big at the frequencies used for
Wi-Fi, so low resistivity material is important, and copper
is one of the metals with the lowest resistivity (though not
the lowest). RG- 6 coax is usually made with aluminum
which is also quite good, but not as good as copper. It’s
significantly less expensive, though. So, building the dipole
out of copper wire is recommended.
Supporting structures should be made with something
that exhibits a low relative dielectric constant so that the
pattern and impedance aren’t distorted much. While the
dimensions are important, they aren’t critical. A dipole
has a fairly low Q (quality factor — a measure of how
narrow its operational bandwidth is) given its high radiation
resistance, so it works without too much impedance shift
over a fairly wide range. Even if it’s off, it’s not fatal. There
is just a slightly higher loss in the coax and changes to
the antenna’s field strength pattern. Careful construction
is warranted because of the skin effect, though. Figure 4
gives you a basic idea of what the passive repeater looks
With this arrangement, place one dipole near the Wi-Fi
base station’s antenna — preferably polarized the same
way as its antenna for maximum coupling — then place
the other one in another room and take signal strength
measurements both with and without the apparatus.
Almost all Wi-Fi transceivers have a way to measure relative
field strength, often called RSSI. It is usually displayed in
dBm (decibels against 1 m W).
A similar measurement of noise is usually available
from the transceiver. Measurements of the RSSI and
noise can be combined to produce a signal-to-noise ratio.
Another study for the science fair might be how critical
the dipole dimensions are to performance; both the dipole
length and the separation of the wires at the feed point.
To refine the passive repeater a bit, you’d want to
put around two or three ferrite beads (little toroids) on
the coax near the feed point, or even just loop the coax
tightly two or three turns near the feed point to decouple
the outer side of the coax shield from the dipole and
QUESTIONS and ANSWERS
Post comments on this article and find any associated files and/or downloads at
n FIGURE 2. Example Cough Button Circuit.
n FIGURE 3. Dipole length rule of thumb.