The International Maritime Organization, through its correspondence group on fire safety of ro-pax ships, has underlined the need for more scientific studies regarding the performance of boundaries in case of a ro-ro space fire, especially to prevent fire and smoke spread to accommodation spaces. Following these discussions, Swedish Flag State has underlined the issue of the smoke tightness of doors in A class divisions. While smoke tightness is a requirement for A class divisions, the fire resistance test method in the FTP Code is not designed to evaluate hazards associated with smoke spread.RISE has carried out the RoBound project to meet this need.To increase the understanding of this weakness in the FTP Code, RISE has performed experimental tests of two almost identical doors. The only difference between the two doors was the presence or not of an intumescent joint between the leaf and the frame of the door, intended to prevent the passage of smoke. The doors were exposed to the test for fire boundaries in Part 3 of the FTP Code, which exposes specimens to a simulated fire by a temperature increase according to the standard fire curve ISO 834.A modification of the standard experimental rig was added and consisted of the addition of a canopy above the tested doors to gather and measure the rate of carbon dioxide to quantify the amount of smoke leaking from the doors. This set up of canopy and measurement rig was taken from the standard EN 81-58 which is applied for elevator doors acting as fire barriers.The results of the tests showed that both doors marginally failed the A-60 integrity criteria since there was presence of a sustained flame at the unexposed side before 60 minutes of test. However, both doors satisfied to the insulation criteria by maintaining a rise of temperature lower than 140 °C in average at the unexposed side. The main difference between the doors was that the door with the intumescent joints presented a rate of smoke leakage which was almost half of that of the fire door without intumescent joints.This result clearly shows the importance of evaluating the smoke tightness of A class doors during testing and the need