1
$\begingroup$

I’m designing an antenna that I’ll implement on a printed circuit board. The antenna will be working on both sides of 900 MHz, so it’s not exactly for ham use. I’m planning to use a bandstop filter in the antenna element, but because the two bands are close to each other I’ll need more narrow bandstop filters that are realistic with normal lumped components. The idea for using such filters came from the tribander yagi, but it is not an actual Yagi-Uda antenna and is not even that directive. A wide band antenna just isn't going to work for the case I have in mind.

The SAW and BAW filters that I found look promising, but unfortunately require connection to the ground that my antenna doesn’t have. Or at least there are ground pads in the packages.

Are there narrow band SMD packaged bandstop filters that don’t need ground to function around 900 MHz?

$\endgroup$
6
  • 1
    $\begingroup$ What's wrong with just using discrete components, as you suggest? 900 MHz is realistic with a small hand wound or off-the-shelf coil, and an smd capacitor. $\endgroup$
    – tomnexus
    Nov 1 '21 at 11:36
  • 2
    $\begingroup$ And, trapped antennas work best when the bands are well separated. You won't easily make a trapped dipole (or yagi) for (say) 850 and 950 MHz. But you could make a yagi to cover that whole band. And wider, if it's pcb, what about an LPDA? Far fewer pcb revisions too. $\endgroup$
    – tomnexus
    Nov 1 '21 at 11:39
  • $\begingroup$ Unfortunately the two bands I’m interested in are only 70 MHz away from each other. This means that a normal C and L in parallel filter has relatively high impedance at the next band. I’m simulating the idea in CST and there I can compensate the impedance with a third L or C in series with the filter, but I doubt I can make it work in real life. $\endgroup$
    – OH2FXN
    Nov 1 '21 at 12:54
  • 1
    $\begingroup$ 70 MHz is quite realistic for a single yagi. We made them for 880-960. If you need 698-960 then you need an LPDA - which can have quite high gain over such a narrow range. It looks like a trapezium rather than a triangle. I can't see the value in a trapped yagi. Traps would be for 900/1800, for example, and only if you're desperate to keep it small (as Hams are, when the wavelength is 20 metres...) $\endgroup$
    – tomnexus
    Nov 1 '21 at 23:31
  • 2
    $\begingroup$ Ok - more like what we call a Trap Dipole then. But on PCB there are so many ways to make multiband antennas, why must you resort to a discrete component? The most common is 900/1800 fan dipole, basically a 900 dipole (or monopole) with short legs next to it that make it work at 1800. And traps on PBC are made with transmission lines and slots, not discrete components. Still sounds like an XY problem. What are your real antenna requirements? Why do you say "A wide band antenna just isn't going to work for the case I have in mind."? $\endgroup$
    – tomnexus
    Nov 3 '21 at 16:51

Your Answer

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

Browse other questions tagged or ask your own question.