First measurements with boundary mouse

[j...@oxaero.com]

My interest has turned to examining the effects on the boundary flow from the leading edge tapes by measuring boundary flow profiles.  For this I have constructed the probe shown below in Fig. 1.  The location of the probe is shown in Fig. 2.

Tube ODs are .065", wall thickness is .005" and vertical spacing is .050".  The lower 4 tubes were selected for the first measurements because the because the should be very thin at .125 chord, perhaps too thin for the diameter of the tubes and their spacing.

The four selected tubes are plumbed to a custom sensor board taped to the under side of the wing near the trailing edge.  This is pictured in Fig. 3 below.  These attach through a flat cable to a WayTrac data logger that samples four 0-3.3V analog inputs at 1 Hz.  It also logs glider airspeed, which is used to compute probe/glider (U/Uinf) ratios.

Measurements were taken at 40, 45, 48, 49, 50, 51, 52 and 55 KIAS, with and without 2" wide tape around the leading edge in front of the probe.

Probe/glider airspeed ratios generally followed pattern shown in Fig. 5.  The shape of the curve is a surprise since it shows the flow showed to a constant value less than half Uinf over the first three tubes (.005" to .160"), then a sudden change to about 1.6 Uinf at the next level (.170" to .225").  Both the thickness of the boundary flow at .125c and the profile shape are a surprise.  

No significant change in airspeeds occurred at any speed from adding the tape.  This could be from not taking precautions to locate the tape precisely as before, since the position measurements were not available when the probe was installed.  If so, then the wrong speeds may have been measured.  Or, the change looked for may lay higher than the 4th tube.

Perturbations in the probe measurements were observed, as seen in Fig. 6 below.  Frequencies are not knows, as sampling at 1 Hz is too slow.  But the amplitudes of the perturbations are fairly well defined.  Plotted as RMS values reveals some interesting results.  The amplitudes of perturbations at the f4th (highest) level are much higher at 45 and 52 KIAS than the lower three levels.  Adding the tape, suppressed these differences entirely!

Also, discontinuities appear to occur at all probe levels for the middle speeds, where the 4th level naturally has a suppressed level.

The 4th level shows the highest speed boundary flow and more stable flow with the tape present.  These suggest looking higher for effects that may correspond to performance changes.

One last nuance.  Figs. 8-9 show an apparent correlation in the perturbations at the lowest two levels.  The higher levels show no such correlation.  This indicates the scale of the instability producing the perturbations, about .115".  Whatever the frequencies involved, there is no difference over the lower two levels at 55 KIAS.  At 52 KIAS, some of the correlation is lost.  

At other speeds, no apparent correlations are seen, but a little can be seen between levels 1 and 3 for 49 KIAS in Fig. 10.  What that might mean physically, eludes me.  I plan to compute actual correlations in all cases to look for patterns.

Jim Hendrix

Figure 1. Close-up of business end of probe

Figure 2. Location of probe at .125 chord on top surface, 52" from wing root

Figure 3. Sensor board

Figure 4. WayTrac flight data recorder samples four analog inputs at 1 Hz

Figure 5. Probe/Glider airspeed ratios for 50 KIAS with and without leading edge tape

Figure 6. Raw airspeed probe data for 52 KIAS with taped leading edge

Figure 7. RMS perturbation amplitudes with and without leading edge tape

Figure 8. Apparent very close correlation of perturbations in tubes 1 & 2 at 55 KIAS

Figure 9. Apparent near correlation of perturbations in tubes 1 & 2 at 52 KIAS

Figure 10. Apparent partial correlation of perturbations in tubes 1 & 3 at 49 KIAS