In the Internet of Things, Packet Delivery Ratio and Time on Air are two predominant characteristics for both applications and operators, especially while using transmissions over Low Power Wide Area Networks such as LoRa T M. Our channel coding approach aims to improve these Quality of Service characteristics for LoRaWAN T M networks. Our CCARR protocol uses Reed-Solomon FEC and structures successive frames into segments. A completion acknowledgements dynamically controls the amount of FEC overload. We estimate the potential gain of CCARR with a proba-bilistic analysis. Simulation and of-the-shelves testbed experiments of the protocol corroborate analysis trends and show a large Packet Delivery Rate improvement over LoRaWAN T M and the literature with a controlled Time on Air increase due to optimized FEC overload.