CC1310: CC1310 - Wurth Chip Antenna - Matching Circuit

If the RX and TX units are close you will get a link even if the antenna impedance matching on your board is not optimum. The reason why you don't get a link is most likely not linked to the matching components (although proper matching is needed to get maximum range).

Setting up an RF link should be the very last test. If you don't, you have no way of knowing why the range is poor. In order to debug and test properly you need to do the testing in steps.

First you need to test conducted RX and TX performance using proper lab equipment (signal generator and spectrum analyzer respectively). Disconnect the antenna and solder a semi-rigid coax cable to the antenna feedpoint. Use a good known SW for this; e.g. SmartRF Studio or CC13xx SW examples from TI web. Next you need to test the antenna separately using a network analyzer. Tune the impedance matching network if needed. When the above is done you can set up an RF link.

Thanks,

Wondering if you can comment on the Antenna matching circuit. Its just a chip antenna, should be quite "plug and play". I don't have access to any test equipment unfortunately.

Our firmware works well on CC1310 LaunchXL, but not on our boards with the chip antenna. The distance between units is only 1 meter at the moment and we are using the LaunchXL for transmission and our board with the chip antenna for reception.

We have three possible cases:

1. Using the matching filter as shown in the schematic above - Does not work.
2. Using a direct connection from capacitor C2 to antenna - Does not work.
3. Some other solution?

Can you advise how to match the antenna or even if that matching system I've shown will even work.

Thanks!

The matching network depends on the environment. The Wurth matching network is valid for the Wurth test board; not necessarily yours. For your design you might need a new set of matching components for optimum range. That said, you should get a link at 1 m distance even if the matching is not optimum so both ) and 2) above should work IF the RX path on your PCB is designed correctly and the SW is OK.

Connect to SmartRF Studio and check the RSSI reading in the continuous RX tab. Does the reading indicate that there is a transmitter? If not, then you need to debug your PCB design and not only the antenna.

Regarding 3). For testing you can use a different antenna approach. Instead of the Wurth antenna you can simply use a piece of copper wire as antenna. At 868 MHz the length is roughly 8 cm.

Thanks, that was a good idea.

I can confirm that I'm not getting any Tx or Rx in either condition with the Wurth matching circuit or not.

All I have at my disposal is a 100MHz scope. Primitive measurements a the antenna's on a LaunchXL and my hardware show similar things during transmit.

For reasons unknown I can't get my PCB to work in either Tx or Rx.

I will try the 80mm of wire and see how that goes.

Thanks.

No luck with the 8cm of wire.  

Going back to another board with the Wurth antenna and matching filter. I get this from the SmartRF tool:

However, cannot receive or transmit any other data.

So I'm wondering what I am seeing?

1. The regular interval RSSI spikes are most likely my test signal.  868MHz, 50Kbaud, 25KHZ Deviation, 98KHz Rx Filter, 14dbm TX, No Whitening.

Any further thoughts?

Thanks

Please confirm the following:
1) Communication between CC1310LP in RX and CC1310LP in TX works
2) Communication between CC1310LP in RX and your board in TX does not work
3) Communication between your board in RX and CC1310LP board in TX does not work

What boards are used in the plot you sent? CC1310LP in RX and your board in TX? If this is correct, it seems like your board is transmitting. If you don't get a link you need to check
1) Frequency offset between RX and TX units. Do you get link if you increase the rxBW to 118 or 155 kHz?
2) SW used on your board. Is this custom SW or are you using SmartRF Studio or PacketTX SW example from TI web?

Just a quick note: CC1310 crystal oscillator can be used with external capacitors, but then you need to disable the internal ones. Otherwise the crystal capacitive loading will be too high and the crystal will be oscillating at a lower frequency than specified.

Recommendation is to use the internal capacitors as this will save 2 external components. Performance-wise there is no difference between external vs internal capacitors.