Damages, caused by moisture and mould, in cold attics have increased in Sweden during the last decades. The main reasons are thicker thermal insulation between the living space and the attic and lack of the heat to the attic from a warm chimney and also ventilation effects to the living space which to some extent also originate from a warm chimney. With lower temperature in the attic it is now more important than earlier to prevent moisture from the living space entering into the attic through diffusion and/or air leakage. One reason for ventilating attics is to reduce attic heating from habited space and thereby the melting of snow and formation of icicles. Using traditional ventilation the climate in the well insulated attic becomes close to the outside climate. The original main purpose of the ventilation is therewith not valid any more. There can be other reasons to ventilate the attic, for example to take away the humid air that can be transported to the attic from the living space trough diffusion and air leakage. Also water from small and occasional water leakage can be removed from the attic by the ventilation. There can also be disadvantages with ventilation. The wind pressure against the building can sometimes cause a negative pressure in the attic, which increases the air leakage from the living space into the attic. Night-time, clear sky radiation during calm and warm conditions can reduce the temperature of the roof construction below the dew point temperature of the outside air brought into the attic. This will in turn, increase the moisture content of the roof construction. With very well insulated loft ceiling beams and the lack of a warm chimney we have to look into new ways of designing cold attics. One possible way is to reduce the ventilation of the attic and combine this with new types of underlay that has a much lower resistance to moisture than traditional underlay materials. With this design it might be possible to avoid the disadvantages with traditional ventilation with the possibility for incoming water to leave the attic through diffusion by the underlay vapour retarder. The aim of the ongoing project is to measure moisture- and temperature conditions in a test-house with two different designs. The test-house is situated at LTH in Lund in the south of Sweden. From the outside the roof consists of concrete tiles, batten, counter batten, ordinary underlay felt or a vapour retarder and a tongued and grooved board. In the middle the attic is divided in two halves by a wooden wall. For about a year we measured every 30 minute in the underlying heated space, at two different locations in the attic, at twelve different points between the tongued and grooved board and the underlay felt, at eight different locations between the underlay felt and the concrete tiles and outdoors. Every second fortnight the moisture ratio was measured in ten different points. During the first year the only difference between the two halves was the underlay material on top of the tongued and grooved board. After this calibration period the ventilation was considerably reduced in the half attic with vapour retarder on the tongued and grooved board. The measurements are still going on. Preliminary results and conclusions from the case study will be presented in the paper.