search

jeelabs / jeelib

JeeLib for Arduino IDE: Ports, RF12, and RF69 drivers from JeeLabs
https://jeelabs.org/202x/sw/jeelib/
The Unlicense
490 stars 215 forks source link

Change RxBw to proper value for 90KHz TX-dev. #103

Open mvdswaluw opened 6 years ago

mvdswaluw commented 6 years ago

0x94A2 means 134KHz RX-bandwidth. With +/- 90 KHz FSK deviation 200 KHz is a better choice. Else modulation is lost in the RX-filter.

mvdswaluw commented 6 years ago

Please define "interwork"? Do you mean TX deviation should be matched to RX bandwidth and vice versa? If so, the answer is yes.

JohnOH commented 6 years ago

Will a node configured in the old style be able to successfully TX & RX with a node configured in the proposed new style. In essence do I to reprogram all my existing nodes to maintain connectivity?

jcw commented 6 years ago

I think there's some confusion about how the RFM69 interprets that b/w setting - I'm not 100% sure, but it might be that the 125 KHz config is interpreted as ±125 KHz, i.e. 250 KHz swing. Maybe there is some way we could check this?

jcw commented 6 years ago

reprogram all my existing nodes to maintain connectivity?

It's an RX-only setting, and if Milan is correct, it would merely increase RX sensitivity slightly.

JohnOH commented 6 years ago

I have made the change and see nothing unwelcome. Is there an equivalent change to the rf69 to consider? Perhaps the comment could be updated to 200kHz?

mvdswaluw commented 6 years ago

Only reception is degraded, but it will work if nodes are close to each other. Thankfully the filters will let the +/- 90 KHz pass, but attenuated. I found this out because some of my nodes were a bit off frequency and de-tuning the RF12demo resulted in loss of reception quicker than I expected. So I started looking into the RF-settings. This one was the first obvious.