I have three custom boards with RN2903 modules on them. They are all running the exact same firmware and if I let the nodes go for a while, the 3rd one is always 20% slower in running. I checked after an hour, 6 hours, overnight, and it is always 20% behind. The problem seems to be when I send my unconfirmed uplink from the 2903, it is much slower in getting a mac_tx_ok packet than the others. Why is that? It seems to happen on every single transmission, not random at all.
(I couldn’t figure out how to edit, so I will just reply with more information)
What I kind of see happening is ChirpStack is receiving the LoRaWAN packet from the device and says that it is channel 8 and dr 4. I am only transmitting between 0-7, and should be using dr 0. I check, and chan 8 is turned off, and even if it was on, it has a min/max dr of 0/3. Looking on a spec an, The device does appear to be transmitting every time at 903MHz, which is what ChirpStack is reporting as well (what is weird about that is that chan 8 should be at 902.46MHz and chan 35 should be at 903).
Because the gateway is expecting the channels to be between 0 and 7, it is responding with a command to have the device adjust itself, and I think that that is what is taking the extra time.
So does anyone know why an RN2903 would transmit on a channel that is disabled and at a dr that isn’t allowed?
I’m afraid we don’t have an answer to your question, the best place to find an answer would probably be to contact the manufacturer (Microchip) to see if they know. The link below will get you too Microchips website.