• Contact
• Groups
• Participants
• Files
• Archive
• Articles
• Summary

It's Deturbulation Time Again (10/29/2005)

Figure 38. Performance changes over time. (click for larger image)

Figure 39. 10/8/05 and 10/12/05 sink rate measurements. (click for larger image)

Figure 40. 10/8/05 and 10/12/05 L/D measurements. (click for larger image)

We have worked hard to understand this and we think we know what is happening. We have looked at several variables affecting loss in performance ranging from the location of the deturbulator on the wing to the details of its construction and installation. A theoretical analysis of the problem has revealed the necessity of maintaining closer tolerances on the inner details of the deturbulator even when external conditions change. We are in the process of evaluating possible engineering solutions to the problem.

To better appreciate the problem we are dealing with, as well as the potential for deturbulation to improve aircraft performance, I refer you to the graph in Figure 38. These are six crude polars that we took for the purpose of seeing whether or not the deturbulator was working. The plan was to arrive at the best practicable configuration for the Standard Cirrus and then take a carefully measured polar, preferably by parallel flying against another glider of known performance. Therefore, we have not measured as many airspeeds as a good polar should have. I say this so you will not read too much into the precise shape of the curves. These are just Excel's polynomial curve fits. No doubt, actual performance varied significantly from the curves connecting data points. The curves serve to make the data more readable and to roughly indicate what is happening between data points.

Notice how performance faded with time, after the initial application of deturbulator in February, and then rebounded in October. Click the image for a larger view and you will see this more clearly. The asterisk in the legend by the first polar on 2/26/05 indicates that this was a partial deturbulation of only the inboard 60% of the upper wing surfaces. The remaining measurements were all taken with a full span treatment on the upper surface. Look closely at the 80 kt points. You can see that the 60% deturbulation gave a nice improvement at that speed and that it increased with the full span treatment on 3/18/2005. However, subsequent measurements declined until 5/21/05. We have more data after that, but showing it all would clutter the graph unnecessarily. Also, I have chosen to show dates when there is not much scatter in the data.

Also, notice the clustering of data points at 40 kts, and at cruising speeds to about 90 knots, and that that the general shape of the curves is consistent. This is convincing evidence that we are seeing a real effect and not merely aberrations from air movements. It is my opinion, after seeing all of the data, that the red curve (10/12/05), most faithfully represents the actual performance of my glider with the present deturbulator configuration.

The next two graphs, Figures 39 and 40, show sink rates and L/D measurements for two recent dates -- 10/8/2005 and 10/12/2005 -- after deturbulation started working again. The orange curves are unbelievable, and, in fact, I don't believe them. It was a strange day, not one that you would ordinarily choose for testing. However, it is noteworthy that the L/D curves are parallel. This indicates that the deturbulator was working essentially the same on both days and that neither curve is corrupted by random data scatter. Since the 50 kt point and the high speed crossover for the red polar curve are consistent with many prior measurements, I think that the red polar is reasonably accurate and that the orange one reflects unusual atmospheric conditions. For this reason, I did not show data from the 10/8/2005 measurements in Figure 38. Looking at the red L/D curve (Figure 40), we see an 11% improvement at 40 and 60 kts and a 21% improvement at 80 kts. These are the magnitudes of improvement that deturbulation can achieve on a glider. And, if I am wrong about the orange L/D curve, then even more improvement may be possible. Greater improvements may be possible, in any case, since the configuration tested has not been optimized to maximize results and the quality of most of the deturbulator is very poor compared to the stuff Dr. Sinha is producing now. One can only wonder what might be possible with a wing specifically designed for deturbulation.

So, what about the odd "hump" at 50 kts? This has been a consistent feature of deturbulated polars, as you can see in Figure 38. At extremely low airspeeds the angle of attack is high and the ability of the deturbulator to delay flow separation on the upper surface of the wing increases lift and reduces induced drag and form drag. The contributions of induced drag and form drag reduce with increasing airspeed while the contribution of skin-friction drag increases. At 50-kts the sum of induced and form drag equals skin friction. Hence, we have a maxima at 50-kts. Within the limits of experimental error this does not represent an increase in sink rate at 50-kts; simply a return to the clean wing baseline.

There is a story about the improved low speed sink rate. On 10/8/2005, the day that gave the exaggerated data in Figures 39 and 40, after taking my data, I cruised home from 8 nm out while playing with airspeeds to find the one that gave the best glide into a 19 kt headwind. While watching the differential altitude (altitude above glide slope) I gained on the glide slope at the fastest rate at 40 kts airspeed. 50 kts gained at a lower rate and 60 kts pretty much held the glide slope. Higher speeds fell below the glide slope. It was a blue day, except for some light high could cover, and it was around 5 pm with very little turbulence. I think this corroborates the unusual 40-50 kt feature of the deturbulated polars.

A Positive Note: Since the deturbulator is a delicate device, people often ask how it holds up over time. It is noteworthy that all of the deturbulator, with the exception of two short patches used for drag measurements, has been installed for nine months and is now performing as well as ever. It shows no sign of deterioration despite having been covered with hangar dirt and desecrated by birds. Just clean it off carefully and it's as good as new.

Bottom Line: We are dealing with a consistency issue that will take time and money to solve, but these data demonstrate convincingly the potential of this new technology. I believe that the consistency issue can be solved with today's capabilities in materials manufacturing and that this technology will become a viable product in the not too distant future.