antenna-theory.com

[SM.Physics]

Under the section explaining voltage standing wave ratio (VSWR), Figure 3 (VSWR specs and the VSWR Curves) shows the specs needed to build a proper antenna. Specs 2 and 3, situated just above Figure 3, state that the VSWR needs to have a ratio higher than 4 and 3, respectively. Firstly, why would any antenna be purposely engineered to have a VSWR greater than 1? (I know having a VSWR of 1 is asymptotic, but why are we choosing to travel away from this number?) Wouldn't the ideal graph be y=1?

Secondly, why does spec 3 call for a linearly decreasing VSWR? Why isn't the line straight like the others? How are the data for these specs chosen?

[bigSteve]

I assume you are referring to this page on VSWR:
http://www.antenna-theory.com/definitions/vswr.php

I'll try to shed some light on your questions.

First, do you remember the old linksys routers, that had two 3" antennas hanging away from the device? These were 2.4 GHz antennas. Nowadays, if you look at your cell phone, it has a WIFI antenna, but no exposed stub. As a result, the antennas are significantly shrunk down from an optimal size in order to fit the product.

The main result of this is that the bandwidth of the antennas significantly decreases. This means that we can match the antenna to 50 Ohms (VSWR=1) at any single frequency, but the antenna impedance will vary wildly around this point. As a result, instead of going for a perfect match the antenna impedance can be looped around a circle on the Smith Chart, such that the VSWR may be between 2 and 3 across the frequency range of interest.

The same story goes for cell phones - we used to have these huge device with an external antenna on there, now we have more bands and more frequency coverage. So the point of the last two paragraphs is that no, we can't engineer the antenna to have a VSWR of 1 everywhere, particularly if the bandwidth required is high or the volume given is very small.

Now, back to the mentioned Figure 3. Note that this does not define the specs needed for any proper antenna, just an example of a real-world VSWR curve and some example specs set around them.

And what is the reason you may choose to define a lower-bound VSWR spec [for instance VSWR>4 from 1200MHz<f<1600MHz]? First, once your antenna has acceptable performance in the bands of interest, you may measure a large quantity of acceptable antennas to get a feel for where the acceptable VSWR is. In this case, 1200-1600 MHz would not be the band where the antenna is radiating, but possibly you know that the good antennas all have VSWR's above 4 in this region. So this spec really has two goals:

1) Consistency
2) Let's say the antenna has a broken connector or some undesirable resistance in parallel with the antenna feed. Then this would destroy your antenna efficiency, while dropping the VSWR

On the second point, note that if you solder a 50 Ohm resistor to a cable, the VSWR would be 1.0 across the entire band. And of course you would have no radiation. So I find it is good practice to use at least 1-2 lower bound VSWR specs to prevent against this sort of thing.

Finally, for the linearly decreasing VSWR spec, sometimes that is used because a resonance is needed at a certain frequency. [The resonance in Figure 3 I'm referring to may be the dip in VSWR around 1900 MHz]. To capture this decrease in VSWR, a linearly decreasing spec may be used.

Anyway, that's a lot of information for now. I suspect a lot of people may have had the same questions so I wrote a longer than normal response. The specs I give in the Figure might look like BS to the uninformed, but I assure you there was no funny stuff going on there.

Peace

[Bobby Digital]

Great response. Hats off to you, bigSteve.