SO2 is enriched in wry-fuel combustion due to flue-gas recycle, and a significant higher SO3 concentration can be expected compared to air-firing. Since SO3 can cause high and low temperature corrosion, it is important to measure the SO3 concentration under oxy-fuel fired conditions. However, measurement of SO3 is not straightforward, since SO3 is a highly reactive gas. This paper presents an experimental study in the Chalmers oxy-fuel test unit, comparing different SO3 measurement techniques applied during oxy-fuel and air combustion. Propane (60 kW(th)) was used as fuel and SO2 was injected in the oxidizer to generate a controllable amount of SO3. The SO3 concentration was measured with four techniques: the controlled condensation method, the salt method, the isopropanol absorption bottle method, and with the Pentol SO3 monitor (previously: Severn Science analyzer). The controlled condensation method was used as the standard for comparison. Additionally, the acid dew-point temperature was measured with a dew-point meter. The controlled condensation and the salt method gave comparable results, and the repeatability with these methods was good. The SO3 concentrations measured with the Pentol SO3 monitor differed in average less than 20% from the SO3 concentrations obtained with the controlled condensation method. With the isopropanol absorption bottle method, a large amount of the SO2 was absorbed in the isopropanol solution, which gives a positive bias if the SO2 is oxidized to sulfate in the isopropanol solution. This was minimized by reducing the measurement time, bubbling argon through the absorption bottles after the measurement to force the SO2 out, and analyzing the solution immediately after the measurement. No principal differences between measuring the SO3 concentration during oxy-fuel combustion and air-firing were obtained. However, a correction factor for the mass flow meter of the Pentol SO3 monitor has to be used because of the high CO, concentration during oxy-fuel operation.