A signifi cant limiting factor that affects tube life in fossil fi red steam boilers is the growth of iron oxide scale (magnetite) on the inside and outside tube surfaces. The oxide scale, which is formed under long term exposure to very high temperatures, acts as a thermal insulator. While external scale limits heat transmission into the tube and reduces boiler effi ciency, internal scale build-up represents a potentially more serious problem and reduces the boiler life. While very high frequency ultrasonic gagging techniques have been available for this measurement for a number of years, they involved the use of cumbersome and non portable instrumentations. In addition, all the methods used up-to-date rely on the ability of high frequency ultrasonic wave to yield better resolving signals that can be used to measure smaller oxide thickness. However, high frequency ultrasound is prone to higher level noise, higher attenuation, and requires special surface preparation. In this work, we will investigate the possibility of formulating the acoustic path inside the oxide scale to be as a convolutional model, and thus, the measured A-scan signal would be a result of a convolution between the A-scan at the steel/oxide interface and an “impulse response” (IR) of the metal/oxide and oxide air (MOOA) interfaces. Extraction of the IR of the MOOA would be done through a deconvolution operation which is expected to have higher resolving power than the original A-scan signal used conventionally to measure the thickness of the oxide scale.

Related Publications