Handheld Antenna Performance Shootout!

Status
Not open for further replies.

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,409
Location
VA
There's always a lot of discussion about the relative merits of various antennas, so I decided it was time to do a real-world comparison of various antennas, looking at actual performance, rather than just SWR. Here's the test setup:

20210218_080854_HDR.jpg

I'm taking two identical antennas, attaching them to ports 1 and 2 of my Nano VNA, then logging the loss between port 1 and 2, which is dictated by the coupling between the antennas at a given frequency. Then I average 13 iterations of a 136-512MHz sweep, while holding the VNA like a walkie-talkie away from nearby metal objects. The data of interest is the red trace, and higher is better (indicates less loss).

Baofeng Standard.png
Baofeng Standard Rubber Duck

Baofeng High Gain.png
Baofeng "High Gain"

Nagoya NA-701.png
Nagoya NA-701

Nagoya NA-771.png
Nagoya NA-771

Nagoya NA-810 Micro.png
Nagoya NA-810 Stubby

Fake NA-771.png
Fake NA-771 Clone

ABBREE 48cm.png
ABBREE 48cm "tactical" folding whip

Uniden Rubber Duck.png
Uniden Rubber Duck

Diamond RH77CA.png
Diamond RH77CA​
 

Ubbe

Member
Joined
Sep 8, 2006
Messages
10,217
Location
Stockholm, Sweden
It looks as if some antennas would be really good for VHF air. Is it possible for you ta make a 108-136MHz sweep to see if any antenna continues to be good over the whole air band?

Some antennas have best performance where they resonate but not others like the 701 that must have some design and impedance issues. Uniden seems to be a solid antenna and the best one for 150Mhz 390MHz and the 460-512MHz band. Many european systems are in the 425MHz band and that antenna are among the worst ones for particular frequency band.

/Ubbe
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,409
Location
VA
Not to mention the two antennas causing the VSWR to go all over the map.
Getting a "real world" SWR plot is sort of like nailing jelly to a waterfall. It shifts around considerably, depending on how you grip the device the antenna is attached to. Having 2 antennas in close proximity changes the SWR plot somewhat, but not as much as changing your grip on the VNA, or moving the antenna close to your body. The test as performed is reasonably analogous to putting the radio in a belt pouch on your hip.
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,409
Location
VA
Exactly "test" antennas way too close, and no ground plane or counterpoise.
The body of the VNA is about the same size as the typical handheld. So the undersized ground plane is a fairly realistic test.

And it will depend on what 'radio' the antenna is attached. Antenna Range 101.
Hence my "nailing jelly to a waterfall" comment. Moving the antenna closer or farther from your body can shift the frequency with the lowest SWR up or down by 5% or more.
 

W5lz

Active Member
Joined
Feb 28, 2019
Messages
617
Just from the looks of your graphs I'd have to say results aren't terribly optimistic. But, as also pointed out the testing procedure isn't very good either. Just in very broad terms, it appears as the longer the 'shortened' antenna, the better. Also not surprising, huh? All 'duck' antennas are for convenience, -not- profeciency. Don't expect performance near that of a longer antenna (at least close to a 1/4 wave, for example). Just for grins, substitute a 50 ohm resistor for that 'duck' and see what happens!
(Height still makes a difference too.)
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,409
Location
VA
If I put a dummy load on either port, the trace drops down below -40dB across the board. That's the difference between this test and measuring SWR. It's measuring how much of the energy going into one antenna comes out the other.
 

rumcajs_tr

Member
Joined
Feb 16, 2014
Messages
259
Location
Europe, Czech Republic
Interesting measurements(y)
Just thinking how to improve the conditions - what about connecting the S11 port to a coax line, routing it some 10 meters away from the NanoVNA and connect it to a reference antenna with good ground plane?
Then the antenna under test would be connected to the S21 port as in your setup and holding the NanoVNA normally as a walkie-talkie exactly as you did. In this way, you cour solve the "close proximity" issue while keeping the rest of the conditions the same.
Still thinking what the "reference antenna" should be - the same as the measure one? Or maybe a discone, which is know for it's equal characteristics over wide spectrum?
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,409
Location
VA
Using two identical antennas is best, because you can take the port 1 ---> port 2 loss figure and divide it by two to determine the performance of one antenna. That eliminates the problem of figuring out a calibrated reference antenna, and averages out sample variation a little bit.
 

prcguy

Member
Joined
Jun 30, 2006
Messages
17,464
Location
So Cal - Richardson, TX - Tewksbury, MA
Under the test condition you show with two antennas 2 inches apart, the coupling values are meaningless. You must get the two antennas far enough apart to be in the far field. They must be far enough apart so they don't skew each others VSWR.

A better way to run the test in a confined area is with a log periodic that covers all the frequencies to be tested as one of the antennas, make it the receive antenna in this case and the transmit antenna under test should be on a metal box to approximate the size of a radio with feedlines to both antennas run through very effective common mode chokes so the feedline interaction is minimized. Calculate the far field and 10ft is probably fine at these frequencies since one of the antennas has no gain.

Then elevate both antennas so the likelihood of a ground bounce between antennas is minimized. The log periodic will help reject reflections from several directions and give reasonably consistent gain across a wide range. Cal the instrument with an antenna you want to consider the generic sample, then you can compare any other similar antenna for gain, VSWR, etc.

Using two identical antennas is best, because you can take the port 1 ---> port 2 loss figure and divide it by two to determine the performance of one antenna. That eliminates the problem of figuring out a calibrated reference antenna, and averages out sample variation a little bit.
 

jonwienke

More Info Coming Soon!
Joined
Jul 18, 2014
Messages
13,409
Location
VA
I totally disagree. The skew in SWR from having 2 antennas near each other is far less than the skew you get from placing the antenna next to your body or other objects--something that is absolutely going to happen in real-world handheld usage. It's not a lab-perfect methodology, but it's far more indicative of real-world performance under less-than-ideal conditions than the process you're suggesting.
 

n3obl

ØAES-1024
Database Admin
Joined
Dec 19, 2002
Messages
2,180
Location
PA
Jon,

do you have a vhf/uhf sweep of the rem820s?
 
Status
Not open for further replies.
Top