Thermodynamics is hard to master for novices and is one of the more challenging topics for chemical engineering undergraduate students. Students need a lot of problem solving practice so that they can better understand and apply the theory. Educational research has shown that the best learning environment is one-to-one with an expert human teacher (Bloom,1984). However, the current situation in education falls far short of this ideal and, even in the best funded institutions, the student-to-teacher ratio is much higher. Intelligent Tutoring Systems (ITS) are knowledge-based systems that simulate the behaviour of human teachers. ITSs aim to provide an experience similar to personal tutoring without the need for human intervention. We present the design and an evaluation of Thermo-Tutor, an ITS for teaching First Law analysis of thermodynamic cycles on a closed system using an ideal gas as the working fluid. Initially, students must draw a thermodynamic cycle on pressure-volume coordinates working from the problem description. The possible cycle steps are adiabatic, isothermal, isochoric and isobaric transitions between states. Students must then solve for unknown state properties and use the First Law of thermodynamics to find the internal energy changes and heat and work transfers associated with cycle steps. When a student submits a solution, Thermo-Tutor analyses it and provides appropriate feedback, identifying and explaining any errors made. We discuss the support for problem solving, and the student model the ITS maintains. An initial evaluation of Thermo-Tutor was performed at the University of Canterbury by second year chemical engineering students. The findings show that the ITS supports student learning effectively.