I noticed that the RFM69 library seems to use the default of 200 ohms for the LNA input impedance setting (LnaZin bit in RegLna at 0x18). Given the 1/4 wave monopole antennas that come with the Moteinos, might the 50 ohm setting be a better choice for impedance matching?
Right, so LNA is on the receiver side, not really sure if that's the optimal value, but certainly good enough for very decent operation.
I guess the 2 possibilities are either the RF experts in the forum to confirm that it needs to be 50ohm or try it out side by side and compare results...? Even though I don't have an analyzer so I don't know for sure if the RSSI will be the best way to tell if incoming signal has an advantage.
Maybe johnk2ox can contribute his expertise...? :)
I did check that a long time ago. I found 50 gave me an edge, about 1 db better sensitivity.
I've often wondered why they default to 200 ohms. I should look at it again now that I understand the differences in the radios. It might be different between the 'w' and 'hw' radio's.
The biggest problem is knowing the impedance (Z) of your antenna. If it is 50ohms then the LNA Z should be set to 50. If the antenna Z is 200, then 200. What if it is 100 ohms? 100/50 is a 2:1 mismatch and 200/100 is a 2:1 mismatch, in this case either should yield the same results.
If you use 50 ohm coaxial cable between the Mote and the antenna the answer is easy, set it to 50. I use 50.
My notes from 8/29/13 reveal setting Z to 200 reduced the sensitivity by 1db. Enabling LNA sensboost increases sensitivity by 3db.
john
Thanks for the responses, John and Felix. Maybe it would be worth clearing bit 7 of REG_LNA in the initialize() function of the library if the out of the box Moteino performance ends up a bit better. Just a thought; it's always possible to tweak things using writeReg(), of course.
Quote from: john k2ox on March 14, 2014, 09:03:38 AM
....I've often wondered why they default to 200 ohms. I should look at it again now that I understand the differences in the radios. It might be different between the 'w' and 'hw' radio's.
....The biggest problem is knowing the impedance (Z) of your antenna. If it is 50ohms then the LNA Z should be set to 50. If the antenna Z is 200, then 200....
Hi Felix, John and other fellows! Thanks for posting so much useful information available online.
My quess is that LNA impedance is set to 200 ohms for better noise performance. 50ohm matched load yields the best SWR but noise optimum might be with different LNA impedance. SWR of 2:1 decreases received signal power ~0.5dB so LNA should decrease the noise level >0.5dB to gain any benefit. Depending on the application, antenna type, objects near antenna etc. the antenna impedance is varying and I think this might cause some variation in sensitivity results between 50ohm and 200ohm (also interference from other users is affecting).. It could be measured more accurately in coupled by feeding signal with signal generator and reading the RSSI from the receiver.
Quote from: john k2ox on March 14, 2014, 09:03:38 AM
..The biggest problem is knowing the impedance (Z) of your antenna....
I'm using Linx planar antennas (two different types from Mouser) with PI-circuit for impedance matching. I did get those tuned pretty well with network analyzer and I'll post those results later on to the antenna topic with example layout.
Quote from: Nuudeli on March 15, 2014, 04:39:10 PM
I'm using Linx planar antennas (two different types from Mouser) with PI-circuit for impedance matching. I did get those tuned pretty well with network analyzer and I'll post those results later on to the antenna topic with example layout.
Hey that would be great, thanks for joining the forum.