Stainless flue pipes are used since the 50ies; smokes temperatures were high enough to avoid condensation in pipes. With oil crises in 1973 and 1979, and in order to save energy, smoke temperatures have been lowered so that condensation appeared in domestic heating systems. lit these conditions, it was asked to make the right choice of stainless steel for flue pipes. A lot of studies were done in all countries, however, no head lines were found. This paper shows that all physical and chemical phenomenon can be explained by use of distillation diagrams at low pressure according to the amount of water in smoke. Usually, this amount in gas or oil burning is about 10%; in wood burning, it may rise 100% when it is burned shortly after felling and when the natural draft during burning is very bad. In gas burning : the main corrosive product is nitric acid with some hydrochloric acid. Diagrams indicate that the nitric acid content in condensates can rise up to 70%. So, its boiling temperature is lower than 70 degrees C. Usual stainless steels such as stabilized ferritic stainless steels are suitable. This point had been confirmed by experiments in CSTB in 1986-1988 and by informations coming from <<the field>>. In oil burning: the situation is more complicated. With smoke temperature lower than water dew point, i.e. around 50 degrees C, no problem can occur. With smoke temperature higher than acid clew point, only oxidation can happen. Between them, acidic condensates are very corrosive; corrosion depends on condensation flow condensation is controlled by: heat exchange through flue liner wall, thermal capacity of the liner wall, gas velocity, insulated stainless steel chimney seems to be the best compromise. Simulation rests are carried out in 1992-1994 in CSTB to verify these points. In wood burning : complexity is due to heterogeneous burning : wood can be dry or wet, combustion can be controlled by high or little airflow rates. However, it appears that the most corrosive agent is acetic acid whose boiling point is 80 degrees C when the water content in smoke is 30%; 115 degrees C when it is 100%. Obviously, 100% seems to be difficult to rise, even with wood burned shortly after felling, in recent heating systems, so, molybdenum ferritic stainless steels such as F18MT or austenitic stainless steels are suitable. In spite of good informations feeding back from <<field>>, new studies are intended to control all these points and to be confident in these guide lines.
