•The piston provers are in general susceptible to non-stable ambient temperature.•The temperature inhomogeneity in the piston prover is related to the mass flow errors.•The magnitude of heat transfer effects decreases with the gas flow rate.•Covered flow cell shows systematic flow errors at stable temperature conditions. This paper deals with the quantification of ambient temperature effects on the measured mass flow rate of a gas using a clearance-sealed piston prover. The variations of temperature were simulated by placing the piston prover in a climate chamber. The tests were made for flow rates of air between 13 mg/min and 15 g/min for three different flow cells. The temperature in the chamber and the piston prover, and the deviations of the piston prover’s readings from the flow source, were recorded. The results confirm that the temperature inhomogeneity in the piston prover can be related to the observed mass flow errors, which increase with the magnitude and the rate of temperature change and decrease with the gas flow rate. Placing the measurement cylinder under a protective cover reduces the sensitivity to variable ambient conditions, but introduces a systematic mass flow error, even in the case of a stable temperature.