This project addresses the problem of correct diagnosis of compressors for air conditioning and refrigeration systems, using instrumentation easily found in the global market in order to be low-cost, accessible and easy to assembly and diagnose, for all kind of users. For this work, the use of an Espressif ESP32 controller compatible with Arduino language and connected to Microsoft Azure is proposed, as well as algorithms based on trends that provide to different users, basic and advanced, with real-time knowledge of relevant variables and failures detected in the refrigeration process allows anticipating them under an enhanced condition-based maintenance scheme,reducing incorrect diagnoses and system downtime while improve capabilities for a better failure troubleshooting, control over spare parts and the reduction of spoiled product. Also, in order to massify this solution, this device can be used as an input with system data to generate relevant information to a recurrent neural network which diagnose, not only compressors but thecomplete system as well, under two signifcant variables which are coeffcient of performance and mass flow.