<p>The issues of increasing controller resistance, reducing convergence time, and reducing the chattering phenomenon are among the important challenges in controlling push–pull converters connected to photovoltaic systems. To address these issues, in this article, an independent model of the output load is first extracted, taking into account the uncertainty of the model parameters. While modeling the desired system, disturbances such as changing radiation and cloud conditions are considered. Furthermore, considering the aforementioned factors, a dynamic uncertainty bound is considered for the desired model and the controller is developed based on it. In the next step, a new controller from the sliding mode control family is proposed that has finite time convergence and also greatly reduces the chattering phenomenon. This controller has these properties based on the arctangent function and the saturation function. After designing the controller, its convergence time is derived based on the controller parameters, giving the designer the ability to determine this convergence time. Furthermore, the controller’s stability (finite time) is proven by Lyapunov theory, in which the mentioned properties are guaranteed for it. It should be noted that this controller has the ability to be applied to other applications, and the properties mentioned on it can be effective for these applications. To examine the proposed controller’s performance, laboratory experiments were conducted in the laboratory in 3 different scenarios, in which practical implementations were considered. The results obtained confirmed the method’s effectiveness in reducing convergence time, increasing accuracy, and reducing chattering.</p>