本研究分別使用交叉陣列熱線探針與裂膜探針搭配恆溫測速儀並且以構造相對簡單的單熱線探針為基準，比較分別在紊流量測上的量測差異。
受限於熱線/熱膜探針必須建構在一空間與時間上進行量測，然而紊流擾動對於空間與時間有其相關性存在，若將使用在量測簡單流場上的假設於紊流流場量測上並不完全成立，因此會產生空間解析度與時間解析度(反應時間)上的誤差。
裂膜探針因為體積比熱線探針大，相較之下感應器每單位體積所具有的表面積相對就變小，對於以對流熱傳為量測原理的熱線/熱膜測速儀而言，其反應時間較熱線探針來得長，特徵反應頻率僅有數百赫茲，所以在紊流量測上會造成紊流擾動量不同程度的低估；另外，交叉陣列熱線探針因為是由兩支熱線探針所構成，其量測空間相對較大，在紊流量測上已不可忽略第三維方向擾動，因此會造成紊流擾動量不同程度的高估。本研究利用圓柱模型產生圓柱尾流紊流流場，分別在圓柱模型下游不同截面，針對高雷諾數(Red=9566)和低雷諾數(Red=3826)進行實驗量測，以合成速度的方式觀測交叉陣列熱線探針與裂膜探針在量測上所造成的差異，並提出合理解釋。
本研究透過實驗的方式得知：裂膜探針越往下游，在低雷諾數時，使用裂膜探針量測圓柱中心(y/d=0)附近的平均速度越往下游其高估量越來越大(大約5-7%)，然而在高雷諾數時，平均速度的量測差異則較小(大約1-2%)。擾度速度方均根值的部分，在剪流區內裂膜探針的量測值則全部低估於單熱線探針的量測值，且此低估量在同一截面時越接近圓柱中心差距越大(大約低估40-50%左右)，但在越往下游，其低估量有變小的趨勢；在交叉陣列熱線探針的部分，平均速度的量測值普遍高於單熱線探針的量測值，且在剪流較大的區域交叉陣列探針高估的量值較其他位置高，到了圓柱中央(y/d=0)兩者差距縮小(大約1-2%)。擾動速度方均根值在上游時交叉陣列熱線探針的量測值低估於單熱線探針，但隨著越往下游慢慢高過單熱線探針的量測值。

Measurements using the X-array hot wire probe and the split fiber film probe collocated with constant temperature anemometry (CTA) are compared with those using the single hot wire probe to investigate the errors of turbulent statistics and energy spectrum of the measurements made by these instruments.
For the turbulent flow characteristics, there exists spatial and temporal correlations. However, due to the layout/size required in the hot wire probe and the split fiber film probe, respectively, direct application of the conditions which are drawn from the measurements of the hot wire probe/split fiber film probe in simple flows is not always appropriate in the measurements of complex turbulent flows.
The measurement principles of both the hot wire/film anemometries are based on the convective heat transfer between the wires and the fluid flow. The split fiber film sensor, which has larger size as compared to the hot wire sensor, is associated with much longer response time (typically with hundreds Hz of frequency) as compared to that of the hot wire sensor. It, thus, results in underestimation of turbulent fluctuations in the measurements. The X-array hot wire probe is constructed by two separate hot wires. There exists distant space between two wires, which may ignore the small sizes of turbulent fluctuations. In additional, the third-dimensional fluctuations become significant and cannot be negligible. Accordingly, it may induce some errors in the measurements of complex turbulent flows. This research uses a cylinder model to induce turbulent wake. Measurements are made with two conditions of high Reynolds number (Red=9566) and low Reynolds number (Red=3826) at downstream sections in different distances from the cylinder model. Errors in the measurements of X-array hot wire probe and split fiber film probe are investigated by comparing their resultant velocity with that measured by the single wire anemometry.
In the low Reynolds number case, mean velocities measured by split fiber film probe near the center of cylinder (y/d=0) are overestimated from upstream to downstream about 5-7%; while in the high Reynolds number case, the difference in mean velocities becomes smaller (about 1-2%). Root mean square in the part of fluctuation value in shear layer measured by split fiber film probe is underestimated entirely. Furthermore, the underestimation is larger (about 45%) from freestream region to center of cylinder in the same test section, but the underestimation tends to be smaller in the downstream. In the part of X-array hot wire probe, mean velocity is higher than that measured with the single hot wire probe everywhere. The overestimation is large where in the most shear layer region, but becomes smaller in the center of cylinder (about 1-2%). The root mean square of velocity fluctuation measured by the X-array hot wire probe is underestimated in the upstream region, but overestimated in the downstream region.

[1]Barlow, Jewel B. (1999). Low-speed wind tunnel testing. New york, John Wiley & Sons.
[2]Bruun, H. H. (1995). Hot-Wire Anemometry. New york, Oxford university press.
[3]Bruun, H. H., Nabhani, N., Fardad, A. A., & Alkayiem, H. H. (1990). “Velocity component measurements by x-hot-wire anemometry”. Measurement Science & Technology, 1(12), pp. 1314-1321.
[4]Chen, C.J., L.-D.Chen, & F.M. Holly, Jr. (1987). Turbulence measurements and flow modeling. Washington, Hemisphere Pub. Corp.
[5]Cutler, A. D., & Bradshaw, P. (1991). “A crossed hot-wire technique for complex turbulent flows”. Experiments in Fluids, 12(1-2), pp. 17-22.
[6]Fransson, J. H. M., & Westin, K. J. A. (2002). “Errors in hot-wire X-probe measurements induced by unsteady velocity gradients” Experiments in Fluids, 32(3), pp. 413-415.
[7]Freymuth, P. (1968). “Noise in hot-wire anemometers” Review of Scientific Instruments, 39(4), pp. 550-557.
[8]Freymuth, P. (1977). “Frequency-response and electronic testing for constant-temperature hot-wire anemometers” Journal of Physics E-Scientific Instruments, 10(7), pp. 705-710.
[9]Freymuth, P., & Fingerson, L. M. (1997). “Hot-wire anemometry at very high frequencies: Effect of electronic noise” Measurement Science & Technology, 8(2), pp. 115-116.
[10]Heenan, A. F., & Morrison, J. F. (1998). “Split-film probes in recirculating flow” Measurement Science & Technology, 9(4), pp. 638-649.
[11]Hirota, M., Fujita, H., & Yokosawa, H. (1988). “Influences of velocity-gradient on hot-wire anemometry with an x-wire probe” Journal of Physics E-Scientific Instruments, 21(11), pp. 1077-1084.
[12]Jorgensen, F. E. (1971). “Directional sensitivity of wire and fibre-film probes” DISA Info., 11, pp. 31-37.
[13]Jorgensen, F. E., & Hammer, M. (1997). “Hot-wire anemometry behaviour at very high frequencies” Measurement Science & Technology, 8(3), pp. 221-221.
[14]Kunkel, G. J., & Marusic, I. (2003). “An approximate amplitude attenuation correction for hot-film shear stress sensors” Experiments in Fluids, 34(2), pp. 285-290.
[15]Ligrani, P. M., & Bradshaw, P. (1987). “Spatial-resolution and measurement of turbulence in the viscous sublayer using subminiature hot-wire probes” Experiments in Fluids, 5(6), pp. 407-417.
[16]Ljus, C., Johansson, B., & Almstedt, A. E. (2000). “Frequency response correction of hot-film measurements in a turbulent airflow” Experiments in Fluids, 29(1), pp. 36-41.
[17]Lomas, Charles G. (1986). Fundamentals of hot wire anemometry. New York: Cambridge University Press.
[18]Muller, U. R. (1992). “Comparison of turbulence measurements with single, x-wire and triple-hot-wire probes” Experiments in Fluids, 13(2-3), pp. 208-216.
[19]Perry, A. E. (1982). Hot-wire Anemometry. Clarendon Press, Oxford.
[20]Saddoughi, S. G., & Veeravalli, S. V. (1996). “Hot-wire anemometry behaviour at very high frequencies” Measurement Science & Technology, 7(10), pp. 1297-1300.
[21]Talamelli, A., Westin, K. J. A., & Alfredsson, P. H. (2000). “An experimental investigation of the response of hot-wire X-probes in shear flows” Experiments in Fluids, 28(5), pp. 425-435.
[22]Uberoi, M. S., & Freymuth, P. (1969). “Spectra of turbulence in wakes behind circular cylinders” Physics of Fluids, 12(7), pp. 1359-1363.
[23]Vukoslavcevic, P. V., Beratlis, N., Balaras, E., Wallace, J. M., & Sun, O. (2009). “On the spatial resolution of velocity and velocity gradient-based turbulence statistics measured with multi-sensor hot-wire probes” Experiments in Fluids, 46(1), pp. 109-119.
[24]Vukoslavcevic, P., & Wallace, J. M. (1981). “Influence of velocity-gradients on measurements of velocity and streamwise vorticity with hot-wire x-array probes” Review of Scientific Instruments, 52(6), pp. 869-879.
[25]Yoshie, R., Tanaka, H., Shirasawa, T. (2007). “Technique For Simultaneously Measuring Fluctuating Velocity, Temperature And Concentration In Non-Isothermal Flow” ISME.
[26]Zdravkovich, M. M. (1997). Flow Around Circular Cylinders (Vol. 1). New york, Oxford university press.
[27]Zhu, Y., & Antonia, R. A. (1996). “The spatial resolution of hot-wire arrays for the measurement of small-scale turbulence” Measurement Science & Technology, 7(10), pp. 1349-1359.