|
Group
Gap Seal Tests
Background
Every glider owner wonders about the best way to treat control gap seals. Conventional wisdom
says that the main thing is stopping flow through the gap because it spoils lift and creates drag.
The next most important thing is ensuring that the air sees a smooth transition from the wing to the
control surface. This is obvious enough. But tapes and mylar strips of various dimensions
are available, and hard data about their effectiveness not readily available.
So, since my glider, Standard Cirrus #60, was equipped for taking two drag measurements at a time, I thought I'd piggy back some control seal tests onto SinhaFCSD test flights that only use one drag probe. Also, I could take data at the end of normal soaring flights, late in the day when the air is smooth and I have altitude to burn. Bottom line, I'm doing this on the cheap.
My intention is to take boundary layer drag measurements of various tapes and mylar treatments of various thicknesses and report the results here.
Methodology
I'm using a Johnson style drag probe for these measurements. Only
a single surface is measured at a time. Pitots on the other side are taped over. Although this device has limitations,
as pointed out by Dr. Althaus,
it's fine for simple comparative studies like this one.
For those not familiar with drag probe measurements, I offer this simple explanation. Flow in the boundary layer is reduced by its proximity to the wing surface. If you put an array of Pitot tubes in the boundary layer and plumb them together, the combined stagnation pressure will be an average of the pressures seen by the tubes individually. Thus it represents the average flow rate in the boundary layer. Now connect a differential pressure sensor (an ASI really) between the aircraft Pitot and the drag probe. It will see the difference between the two dynamic stagnation pressures. The greater the difference, the greater the drag. The maximum difference theoretically possible will be just the pressure the aircraft's ASI sees. That would correspond to a "brick wall" wing that absolutely stops all flow in the boundary layer. The minimum difference theoretically possible would be zero, corresponding to a perfect wing with no boundary later, only free stream flow right down to the surface. The real world lies somewhere in between.
For these tests I'm using one of my digital differential pressure sensors. They give a digital voltage reading of the pressure difference. In smooth air, the readings, are repeatable to within 0.01 Volts, which corresponds to 0.01" H2O. For each data point, I take three readings and calculate the average.
Since my wings have S-seals inside and mylar beneath, I decided to test only for gap smoothness on the upper surface. Measurements, of drag from flow through the gap are not in the scope of this study. The initial baseline, open gap, data shown below (black line) was taken in less than desirable conditions. There was enough turbulence to make the ASI and digital pressure readings a bit iffy. Later I will retake that data, but for now here it is. I think it's good enough for reasonable conclusions.
Fabric Tape - Tesa 4651
On October, 16, 2004 the first comparative data was taken. On this flight the gap was sealed with Tesa 4651
fabric tape of width 1 9/16" (40 mm) and thickness 0.012" (.3 mm). See the image below.
The result of the test is shown in the graph below as voltage (proportional to the differential pressure) vs airspeed in kts.
The black curve is open-gap data and the orange solid curve is from the taped gap. The left vertical scale applies
to these curves. The orange dotted line is the percent change vs airspeed in kts plotted against the right vertical scale.
These data put numbers on the conventional wisdom about covering the gap. Remember that the baseline (black) curve has some scatter because conditions were marginally turbulent. I suspect that the dips in the baseline curve at 50 and 70 kts are exaggerated by data scatter. They also appear in the more reliable orange curve, but much less exaggerated. I believe that better baseline data will yield a more consistent trend across the airspeeds. A given error affects the low speed percentage change much more than the high speed. Thus the results become more reliable as airspeed increases.
Assuming no flow through the gap, taping the gap with .012" thick tape on the upper surface produces a reduction in boundary layer drag that increases with airspeed. The improvement is about 2.5% at normal cruising speeds, and increases by about three times to 7% 90 kts. I suspect that this trend continues all the way to Vne. If so, the improvement is about 10% at 100 kts and between 12% and 15% at 119 kts (Vne).
The increased improvement with airspeed is probably because as airspeed increases the angle of attack becomes more negative and the boundary layer becomes thinner as the rear of the wing faces more into the flow. A perturbation on the surface will have a greater effect in a thinner boundary layer, and so would a reduction in the perturbations.
Conclusion. Don't leave your aileron gaps open. Use tape, even thick tape, if you don't have something better.
It should be pointed out that boundary layer flow measurements like these do not represent the drag coefficient of the wing. they do not tell us what's happening on the other side of the wing or how it all comes together in the turbulent wake behind the wing. But they are adequate for comparing the relative benefits of different gap treatments on the measured surface.
|
