1
\$\begingroup\$

The transmitters for hobby projects working on 433/868/915 MHz (RFM69) require an antenna connected to the board. The board is battery powered, without metallic chassis and overall much smaller than the wavelength (i.e. 20x20 mm board vs 700 mm or 345 mm wavelength).

I can choose between full, half- and quarter-wave. I also read I can use a straight wire or a coiled antenna.

Would a coiled half-wave perform better than a quarter-wave straight? A coiled one with large coils could also allow placement of the micro inside the coil itself, resulting in a compact device.

What are advantages and disadvantages of longer, coiled vs shorter, straight?

\$\endgroup\$
2
\$\begingroup\$

Quarter vs. half-wave

A quarter-wave antenna, in the sense of "a wire or rod 1/4 wavelength long", is not a complete antenna; it is half of one. The other half is whatever is connected to the other terminal of your transmitter output — probably its ground, i.e. the ground plane or chassis (if any) of your board. This works well as long as there is a quarter-wave-or-longer surface adjacent to the antenna feed point (the place where the two connections spatially diverge from each other, also the place you start measuring the length from).

A half-wave antenna is a good choice when your entire device is much smaller than a quarter-wavelength, or when you are building an antenna that is physically separated from the transmitter.

Given that you mentioned placing components inside of the coil (this may not be actually a good idea depending on the specifics) I presume your device is small and therefore a half-wave antenna would be wanted.

Helical antennas

Coiling an antenna element — I assume you mean a helix, not a flat coil (loop antenna), which has yet different properties — is one way of shortening it. A shortened antenna is always less efficient than a full-length one.

Summary

  • With a small transmitter (board/chassis ≤ 1/4 wave), if you have a choice between a helical half-wave and a straight quarter-wave occupying the same space, you want the half-wave because otherwise the ground for your quarter-wave is too small so you have an even worse shortened antenna.

  • With a large transmitter, a quarter-wave antenna is a better choice because the overall space taken up is less.

(Note that if the device has any cables of any sort running to it, they may also act as part of the antenna system unless you specifically design so they won't. This can be an advantage in saving space or it can be an interference problem, depending.)

\$\endgroup\$
  • \$\begingroup\$ I added info about the device: 1) battery operated (no earth) 2) no metallic chassis 3) overall size. I understand that all goes towards half-wave, as you explained. \$\endgroup\$ – FarO Sep 13 '17 at 15:58
  • \$\begingroup\$ For info: the only long cables I have are the ones going to a rugged temperature sensor. Will foil shielding be enough to avoid it behaving as antenna? or are capacitors/chokes needed? \$\endgroup\$ – FarO Sep 13 '17 at 16:05
  • 1
    \$\begingroup\$ No, shielding by itself does not help. Chokes do. The full answer is much more complex and not one I'm really able to write. \$\endgroup\$ – Kevin Reid Sep 13 '17 at 16:30
  • \$\begingroup\$ Soldering the antenna to the PCB at one end or in the middle is equivalent? \$\endgroup\$ – FarO Sep 14 '17 at 8:54
  • \$\begingroup\$ No, it should be quarter wave on GND and quarter on the second connector... to create a dipole. \$\endgroup\$ – FarO Sep 14 '17 at 9:12
0
\$\begingroup\$

Coiled vs straight does not matter that much for this kind of application. What is a lot more important is the correct feed point impedance. A 1/4 antenna has about 30R impedance, which means you will need an impedance converter to get good coupling into the antenna. I would suggest to use a 5/8 antenna, which has approximately 50R feed point impedance (plus some capacitance which you can easily compensate by a small inductor if you want). This way you get a much better coupling and thus better radiation. In an 2.4GHz application I once build, we only had a "ground plane" of 2x2cm. The 5/8 antenna performed so well, that in the limited space we had (50m in a building), we couldn't get to the limit of the antenna reception, even when we coiled up most of the antenna.

I never had any good practical experience with half or full wave antennas. They were way to finicky to get reliable performance out of them.

As for straight or coiled: go for the straight if you have the space. If not, coil up the antenna, starting from the tip, and as loosely as possible, until you are down to the maximum size.

\$\endgroup\$
  • \$\begingroup\$ I'm using ready-made modules. I cannot change the impedance and I'm only required to choose the antenna length. I would assume there is some compensation already in the chip. \$\endgroup\$ – FarO Sep 14 '17 at 7:28
  • \$\begingroup\$ You can always add an impedance conversion circuit (which is basically just an LC filter) to the antenna output. All modules I have seen have an 50R impedance at their antenna port. But better check the datasheet. \$\endgroup\$ – Attila Kinali Sep 14 '17 at 8:12
  • \$\begingroup\$ yes, it has a selectable impedance of 50R or 200R. \$\endgroup\$ – FarO Sep 14 '17 at 8:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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