Author Topic: New to impedance matching | 868MHz | Ceramic Chip Antenna  (Read 156 times)

0 Members and 1 Guest are viewing this topic.

Offline mrburnzie

  • Contributor
  • Posts: 25
  • Country: cs
New to impedance matching | 868MHz | Ceramic Chip Antenna
« on: October 19, 2019, 06:44:09 pm »
Hi!

I'm new to building RF stuff and I need a bit of help for matching the 50 \$\Omega\$ rule.

I have chosen the CMWX1ZZABZ-091 module that will run LoRa at 868MHz.
For the antenna I have chosen this one: http://www.farnell.com/datasheets/2280483.pdf

I have copied the antenna footprint with all the dimensions required.


1. In the datasheet it says that it is highly recommended to add a pi-network, which I have added. Do I have to use the values that are given in the datasheet? (by have to use I mean do I have to solder them)
2. I want to a use a FR4, 2 layer board, made by aisler.net. I have calculated the trace width and clearance via kicads calculator: . I'm not sure if I did it right, can I get confirmation on it?
3. The calculator says that I need a 1.05mm width trace, but for the module used (CMWX) I can't directly connect the 1.05mm trace to the pin, because the pins width is 0.7mm... what should I do?
4. Should I use it without the bottom ground plane? The calculator (if I use it right, not sure) gives me this:
5. I don't have the right instruments at hand to properly match the impedance. What is the best recommendation for newbies to match the impedance?

Cheers!
« Last Edit: October 19, 2019, 06:47:38 pm by mrburnzie »
 

Offline mayor

  • Regular Contributor
  • *
  • Posts: 142
  • Country: ca
Re: New to impedance matching | 868MHz | Ceramic Chip Antenna
« Reply #1 on: October 23, 2019, 10:30:25 am »
Hi,

1. Pi Network: you can't just use values from the datasheet. Your match will be influenced by a number of things like ground plane size, layout, enclosure, etc. The best way to do this is to use a VNA (there are cheap ones which might do the trick). I've used the xaVNA, happy with it.

For impedance matching of the trace from source to antenna, you can get away without matching it at all if it is short enough (say less than 1/10 wavelength). Can't speak to the tool you're using or your board, but in all cases, it was easier for me to use CPWG than without ground.

As far as going to a smaller pin, I don't think that's an issue. Get as close to the device as you can, then match em up. If TI does it in their reference designs, that's good enough for me! They do it for a good number of antennas here:
http://www.ti.com/lit/an/swra351a/swra351a.pdf
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf