Antenna testing is hard (II)

This article follows up on this one. The title could also be: doing measurements on hand held radios is hard. Actually, some measurements I did in the past might not be so accurate after all.

Measuring in dBm
If you measure in dBm, what I always did, you assume that the device at the other end is 50 Ohms, just like the manufacturer promises you in the specs. Slowly but surely we are finding out that this is not always the case. As a result I might have to switch to microvolt (µV), a method which doesn’t require a perfect 50 Ohms at the other end.

Can’t we just convert it? 0.5 µV @ 50 ohms = -113 dBm, a piece of cake, right? No. You must know the actual impedance of the receiver if you want to convert from dBm to µV. Erik PE1RQF did some measurements on a few hand held radios. The outcome was, well, a bit scary.


If a device is 50 Ohms, the SWR should be 1:1. Only a perfect dummy load and the Marconi RF generator are.

What this means for us
What this outcome basically means is that generating cold numbers on sensitivity and antenna performance are nice, but don’t tell the whole story, or could be misleading.

  • Antennas which prove to be the best performers (RX/TX, VSWR), tested under optimal conditions, are not necessarily the best performers on a specific brand/model of hand held radio.
  • There are antennas which aren’t 50 Ohms at all, but could very well outperform everything else on the market because your radio isn’t 50 Ohms either.

Do the test yourself
The best example I can give you is this one: take a Baofeng BF-666S, 777S or 888S, use the stock antenna, and make notes of the performance in the field. Try to hit repeaters which are barely in range. If you have a field strength meter, measure field strength when transmitting.

Remove the stock antenna and replace it by a Nagoya NA-701, NA-771, the $3.79 antenna, or the Baofeng UV-B5 antenna. I tried all of these; just pick whatever 3rd party antenna you have. Repeat the tests.

What happened here is that the short stock antenna was the best of the bunch, all 3rd party antennas had a negative influence on the performance of this specific Baofeng model. The same 3rd party antennas mentioned above did improve the performance of the Baofeng UV-5R, often by a wide margin.

Exception to the rules
The funny thing is that reception suffered greatly too. This is quite uncommon and a sign that hand held radios don’t follow the rules. For example, if you just want to receive on 20 meters and your dipole is 2×6 meters instead of 2×5 meters, you would never notice the difference. With your hand held radio however you will.

9 comments on “Antenna testing is hard (II)

  1. Can you elaborate on the test setup that Erik used to measure the source impedance of the radios? The chart shows SWR which is a bit limiting for interpreting the results.

    • The SWR is an indication that the impedance is not 50 ohms – it could be higher or lower. He’s working out more details now, which will be posted later.

  2. Using time to test HT antennas is, well, an activity that someone who has plenty of time to waste performs. (grin) When my wife was in Tibet for six weeks, YES, I tested a few HT antennas with calibrated Motorola test equipment. BOTTOM LINE: Longer is better. Almost all stock ducks are -6bd or worse. Want better performance from an HT? It will not be from any aftermarket 7″ antenna. My favorite improvements for many years have been the Smiley 270A and the Comet SMA24. The Smiley is a collapsible metal antenna, BUT with a spring in the base to protect our delicate SMA connectors. Collapsed, it performs much like any stock duck. But extend it to achieve ¼ wave on 2 meters and 5/8 wave 440 MHz performance. I like the “flexibility of the 17” Comet SMA24 – I keep one coiled up and in a ZipLoc(tm) sandwich bag in the glove compartment of my car. Great for outdoor enthusiasts who might walk through brush – VERY flexible.

    • What you say is generally true (longer in better), until you run into something like a Baofeng BF-666S. This was what started all this, because the odd behavior didn’t make sense at all. But yes, trying to get as close to an antenna which doesn’t require much ‘coiling up’ to achieve an a quarter or half wave length is, under most circumstances, the best.

  3. Ah ha! Radio is radio! I’m new to ham, but I was on CB for years and installed many of radios on mobile and homes. I trained 6 employees to install. I always used an swr and pi matcher. Here I am in ham (2m and 70 cm) and I’m lost! I’ve been told no matcher is needed, but an swr is. You do this, so what’s the truth?

  4. HT antenna’s are actually a linear dipole,
    the bit you plug in is generally the signal end, but the plug, socket and metal internals of the radio are the oposing ground section.

    i have found that extending the connector a few cm on the uv82 for example, improves the signal on the original antenna.
    the element length is the same, but the grounding is closer to it in length now.

    i intend to do some experiments with this concept of matching the element and overal ground length soon.

  5. Here is a different way of doing things , haven’t tried it as yet ..
    HT antenna testing , very interesting ..
    Currently running a Nagoya NL-R2 on my GT3 , very interesting ( Good )

  6. What I meant is that showing complex impedance provides more useful information than simple SWR.

    Measuring receiver input impedance is normally troublesome so it would also be nice to know the test fixturing Erik used for future reference and comparative purposes.

Comments are closed.