XIX Biannual Symposium on Measuring Techniques in Turbomachinery Transonic and Supersonic Flow in Cascades and Turbomachines PROBES IN TRANSONIC FLOW

A constraint exists for intrusive probes in transonic flow which leads to a complete insensitivity to static pressure at Mach number Ma=1. Nearby Mach number unity different types of probes are affected more or less by the reduced sensitivity to static pressure or Mach number. In this paper three types of probes are compared and it is shown that the reduced sensitivity near Ma=1 causes large errors when the flow Mach number is determined by a conventional probe. A closer look onto the pressure distribution of a blunt body reveals that only the front part of a probe is subject to the constraint while base pressure is not affected. Therefore a base pressure tapping was added to a conventional Cobra type probe. The new probe was calibrated in the Probe Calibration Facility of DLR Göttingen at a wide range of Mach numbers and at two Reynolds numbers. INTRODUCTION Probes are still an indispensable tool to determine flow values in turbomachines. In most cases they have to be inserted radially and then especially the probe stem causes some disturbances to the flow. This is normally accepted, but in transonic flow the disturbance by an intrusive probe makes it impossible to determine the flow Mach number at all. It was already mentioned by Shapiro [1] that the pressure distribution on a body is independent of Mach number near Ma=1. The effect on probes was described by Hancock [2] who showed that in principle the sensitivity of any intrusive probe to static pressure must be zero at Mach number unity. This constraint is due to the detached shock standing ahead of a body in supersonic flow. When beginning from subsonic conditions the flow Mach number is increased, a shock appears at Ma=1 standing infinitely far ahead of the body. Downstream of the detached shock the flow is still subsonic and therefore the tappings on the body still sense subsonic conditions. Increasing the Mach number leads to a movement of the shock closer to the body but still subsonic flow conditions exist at the front of the body. At a sharp-nosed body the shock finally attaches and becomes an oblique shock downstream of which the flow is fully sensitive to upstream Mach number variations. Blunt bodies always develop a detached shock in front of the body, but the subsonic region at the nose diminishes in size with increasing Mach number. It is therefore clear that different types of probe shapes generate varying magnitudes of insensitivity to static pressure in the vicinity of Mach number unity. In DLR’s Probe Calibration Facility (see Gieß et al. [3]) several nozzles allow the calibration of probes in the Mach number range of 0.2 to 1.8. A specially designed slotted nozzle enables the calibration in the transonic range. A lot of different probes have been calibrated and the results of three of them are presented here. RESULTS AND DISCUSSION A Mach number coefficient, CMa, is derived from the probe pressures by calculating a Mach number from the ratio of the averaged angle pressures, p = (psl+psr)/2, (see Fig. 4) to the central pressure p0 = p0s. 1 0 0 2 1 1 Ma p p C Ma p p γ γ γ − ⎡ ⎤ ⎛ ⎞ ⎛ ⎞ ⎢ ⎥ = = ⎜ ⎟ ⎜ ⎟ − ⎢ ⎥ − ⎝ ⎠ ⎝ ⎠ ⎢ ⎥ ⎣ ⎦ , with γ = 1.4 In Fig. 1 the Mach number coefficient CMa plotted versus Mach number is shown for three probes. One clearly observes that the CMa-curves display a flat part near Mach number unity. The gradient of the CMa-curve determines the sensitivity of the probe to flow Mach number variations and therefore the gradient is plotted in the next Fig. 2. The gradients are calculated from the difference of the measured points therefore the plotted gradients are not very exact but nevertheless give an adequate impression. According to Figure 2 the cylindrical probe head is superior at subsonic flow, but really bad near Mach number Ma=1.1. The Cobra probe is the worst one. The wedge probe is aligned to the flow (like a lance) because it is a probe used in DLR’s Straight Cascade Facility where enough space enables such a solution. The wedge probe results are only shown for comparison as such a probe stem cannot be realized in a turbine stage. One may conclude that probe configurations which can 1 Rhode-St-Genèse, Belgium April 7-8, 2008 The 19th Symposium on Measuring Techniques in Transonic and Supersonic Flow in Cascades and Turbomachines be used in a realistic turbomachine geometry are inevitably insensitive not only at Ma=1, but also in a Mach number range from 1 to 1.3.