3
\$\begingroup\$

edit: In view of the helpful comments below, I just wanted to show an example of a cheap plastic antenna that does have an inductor inside. In this case, the antenna is shorter and the wavelength longer than the 900 MHz antenna discussed in my question, so the inductor is strictly for loading the short vertical to get it to resonate down at 433MHz. Take home message: not all cheap plastic antennas are just a piece of (straight) wire, some are coiled!

enter image description here

above: annotated screenshot from this video.


On most 2.45GHz WiFi routers the the black vertical antenna is thick at the base (10-12mm) and narrows (4-6mm) near the top. I've been poking around various catalogs and sites and it seems to me that the lower frequency 900 MHz antennas are simply cylindrical in shape with a uniform diameter of about 10-12mm.

Of course what matters is the electrical resonant frequency (center and width) of the antenna, but the shape reveals differences in things like inductive loading schemes for example, and so can reflect differences wavelength with respect to the physical length.

I noticed on the PyCom website they show antennas for the LoPy (LoRa) and SiPy (SigFox) boards as having the tapered shape (that I normally see on 2.45GHz antennas), while the antenna sold separately even though they operate in roughy the same frequency range. They also sell the antenna separately and it is labeled for both.

enter image description here

above: LoPY module from here. Two identical coaxial antenna connectors for external (software-switchable) WiFi antenna, and LoRa antenna are in the top corners.

Since the LoPy has TWO coaxial antenna connectors (top right and top left in image above), one for an optional external 2.45GHz WiFi and one for 800 - 900 MHz LoRa, it is conceivable that one could have both kins of antennas laying around at the same time.

Can the two types (usually) be distinguished by appearance? I left my grid-dip meter at home.

enter image description here

Antenna Kit for use with the LoPy LoRa, WiFi and BLE board or the SiPy Sigfox, WiFi and BLE board.

Includes:

  • External antenna.
  • RF Cable Assemblies RP-SMA (Female) JK-IPEX MHF U.FL 1.13 100MM
  • RP-SMA (Male) Tilt Swivel 1/2 Wave Whip antenna

NOTE: Using the LoRa/Sigfox radio without the external antenna can lead to damage of the device and is therefore not recommended.

enter image description here

LoPy Twin Pack

The bundle includes:

  • 2 LoPy modules
  • 1 Expansion Board
  • 2 LoRa Antenna kits
\$\endgroup\$
12
  • 1
    \$\begingroup\$ On most 2.45GHz WiFi routers the the black vertical antenna is thick at the base (10-12mm) and narrows (4-6mm) near the top. That is NOT the shape of the antenna but the shape of the plastic covering the antenna. The antenna itself is usually just a thin strip of metal. This is mostly done by removing a certain distance (1/4 lambda) of the coax cable shield leaving the inner conductor free, that is then used as antenna. \$\endgroup\$ Dec 27, 2016 at 15:26
  • \$\begingroup\$ @FakeMoustache you know this for a fact? Have a link? I'm pretty sure there are some passive elements in there as well. To increase the useable bandwidth perhaps. A naked half-wave dipole will only have gain over a few percent range in frequency. \$\endgroup\$
    – uhoh
    Dec 27, 2016 at 15:29
  • 2
    \$\begingroup\$ Yes I am 100% sure, I have some of these router antennas lying around here and I have opened them and found what I just described. There are no passive elements (how would that work anyway). These antennas are extremely cheap, there's no money for anything fancy. \$\endgroup\$ Dec 27, 2016 at 15:30
  • 1
    \$\begingroup\$ I'm only talking about 2.4 GHz antennas. I do not have any separate 900 MHz antennas. You do not show any picture or link to the 900 MHz antennas you're talking about. \$\endgroup\$ Dec 27, 2016 at 16:13
  • 1
    \$\begingroup\$ I've done this for 900MHz many years ago. It's the same story (just coax with shielding removed after feed point). I've never seen passive elements in the antenna itself for anything linear. You'd build your matching network on the feed-side. Doing it in the antenna would be inefficient since there could be any range of devices feeding it and you'd have to match their output to the new value of antenna input, which is now matched internal to the antenna to the antenna's real impedance. That's just extra components, cost, and wasted energy (lower efficiency). \$\endgroup\$ Dec 27, 2016 at 16:32

1 Answer 1

0
\$\begingroup\$

It's not entirely clear what you're asking...

The only real question in there I can find is:

Can the two types (usually) be distinguished by appearance?

To which, the answer is no. The overmolding is not indicative of the shape of the wire inside and is decorated for design/branding reasons. If this were not the case, it would look like a wire (obviously).

Here's the classic "rubber duck" antenna. It's a normal-mode helix, in this case targeting ~850MHz (US cellular band)

enter image description here

Some other questions are implied like:

Does the shape/length tell you the target frequency band?

Also, generally no, but the there are very real limits to size -> frequency band with any practical efficiency so, with some caveats, you can make some (very) educated guesses as what was intended. The biggest giveaway is the dielectric medium. In free-air, the practical lengths are in a certain length-to-band range, in ceramic media, the practical lengths are different but also in a limited range.

I'm pretty sure there are some passive elements in there as well. To increase the useable bandwidth perhaps.

To which the community responded that this is not the case. The reason is that the passive element to which you were referring (based on your later edits) is a loading coil, not a matching element. It doesn't increase the usable bandwidth (in most cases it actually decreases it). Instead, it moves the center frequency lower and lowers the radiating efficiency in the process.

Loading coils for short antennas feature high reactance values and extremely low capacitance at one end relative to the other. This makes the coil vulnerable to environmental capacitances as they will be differential to each turn of the coil increasing circulating currents.

Turn-to-turn capacitance increases effective inductance, effective resistance, yet reduces system bandwidth and ironically quality factor!

Charles Rauch summarizes this nicely:

In tank circuits, or with loading coils used with longer antennas or with capacitance hats, the optimum inductor shape becomes shorter and larger in diameter.

In a short mobile antenna or very high impedance tank system or circuit, optimum inductor form factor becomes longer when compared to diameter. Long high inductance coils are generally good, compact large diameter inductors often work better in low inductance ranges.

So most of the engineers didn't consider your claim of "passive components" in the overmolding to include loading coils as "inductors" since long helice behave more like a radiating transmission line than a conventional inductor.

\$\endgroup\$
2
  • \$\begingroup\$ Excellent, thank you very much for the thorough answer! If you read again you will see I was just speculating - not claiming - that there could plausably be something in there besides empty space, and so I asked here. Opportunistic condescention asside, it's a great answer! \$\endgroup\$
    – uhoh
    Jan 4, 2017 at 8:16
  • \$\begingroup\$ ...and I think "Can 900MHz and 2.45GHz be distinguished from each other by their shape?" is pretty clear, the only other question mark appears here in this clear reiteration of the one question: "Can the two types (usually) be distinguished by appearance? I left my grid-dip meter at home." \$\endgroup\$
    – uhoh
    Jan 4, 2017 at 10:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.