0
\$\begingroup\$

I have to put a GPS antenna on a PCB layout, but not sure how to calculate a matching network, and how to check that it will work correctly. I've chosen components with desired features (radiation diagram, amplification, and cost).

Actually, there are two or three networks, antenna - LNA - SAW - output.

But how do you choose the impedance matching components? For example antenna's and LNAs datasheets suggest different inductances. Do I have to put both inductances or choose one that fits?

SAW filter datasheet says that its input and output is 50 Ohm and nothing more. But here in this article Adding A GPS Chipset To Your Next Design Is Easy they use a capacitor after SAW.

This device in general is a small wearable GPS tracker - gsm+gps.

Components

Is there some software for network calculation, or maybe some open-source ready schematics?

GPS antenna & amplifier schematic

\$\endgroup\$
4
  • \$\begingroup\$ Be clear and unambiguous about the matching network you are talking about. You appear to start talking about matching the antenna but then drift onto the SAW filter. Be clear. \$\endgroup\$ – Andy aka Oct 30 '20 at 12:55
  • \$\begingroup\$ I’m a newbie in this topic. I’m talking about two networks ( or maybe three!) antenna - lna, lna - saw, saw - output \$\endgroup\$ – Denis Jelesniak Oct 30 '20 at 13:20
  • \$\begingroup\$ What needs matching that you don't understand? Pick one not three. And put inductance values on your schematic - I for one don't want to go looking up some obscure inductor part number. Be specific. \$\endgroup\$ – Andy aka Oct 30 '20 at 13:36
  • \$\begingroup\$ Try and copy a working reference design or commercial product. How many have you opened up to see what they did right and wrong? I suggest 5-20. \$\endgroup\$ – KalleMP Oct 31 '20 at 9:49
1
\$\begingroup\$

All of these components are nominally 50 Ohms. For a receiver, improving the impedance match doesn't have a big payoff, so if you're not experienced, you are probably better off putting a DC blocking capacitor between the each stage and moving on.

I'd suggest this, because the LNA may be biased internally in such a way as to have a DC voltage at either the input, the output or both. If you antenna has a DC short, then you'll effectively short out the LNA on the input. On the Output, SAW filters can be sensitive to a DC Bias, so a DC Block could help that last longer. The same on the Output of the SAW/Input to the GNSS chip. The GNSS chip may be able to feed +3V out of the RF port to power a remote antenna, which could cause problems with your SAW.

\$\endgroup\$
4
  • \$\begingroup\$ Thanks, but here is an unclear thing, does for example antenna’s 50 Ohms Impedance Include the matching network from data sheet? Do I have to use both inductances from antenna’s design as well as from lna’s reference? Or only one of them? \$\endgroup\$ – Denis Jelesniak Oct 31 '20 at 14:26
  • \$\begingroup\$ Looks like I was wrong, neither of those parts are 50Ohm input. That's really annoying. \$\endgroup\$ – rfdave Oct 31 '20 at 15:43
  • \$\begingroup\$ So, the datasheet says that the antenna has 50 Ohm Impedance. So maybe inductance is needed for some other reasons? \$\endgroup\$ – Denis Jelesniak Nov 2 '20 at 10:24
  • \$\begingroup\$ The application note you linked seems to imply that the antenna requires the two inductors in a L configuration at C2 and C3 to provide a 50ohm point, so you do have to use those two inductors to provide a 50ohm point looking into the antenna. Looking at the LNA, it requires a single input matching network consisting of that series inductor, L1, which provides a 50ohm load looking into that series inductor. Based on that, I'd suggest you layout a Tee matching configuration, with all three inductors. Do you understand why that is? \$\endgroup\$ – rfdave Nov 3 '20 at 2:48

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.