Zeeoo,
Turbulators are essentially boundary layer trips, which are used to induce laminar to turbulent transition further forward than it would have occured naturally. Boundary layer transition is a very complicated process for which there is no known general theory - even for simple geometries like a parallel flow flat plate case. A number of empirical relations are available but they have very limited applicability and cannot be applied willy-nilly.
The most important point to understand here is the way in which transition occurs on a 2D steady aerofoil at different Reynolds numbers. Let's take for example the venerable NACA0012 aerofoil, with which I have a lot of experience.
Firstly, for a medium size helicopter the tip Reynolds number in the hover is around 3,000,000. Reynolds number is simply the ratio of inertial and viscous forces in the flow. At these sorts of Reynolds numbers transition on isolated steady aerofoils generally occurs as the result of the Tollmien-Schlichting instability up until relatively high incidence when separation can become important. The NACA0012 does have a small separation bubble under these conditions, but it usually plays no part in the stall process. (See NASA TP1100) In addition for this Reynolds number and assuming a very low speed, say Mach 0.15, the transition on the upper surface will move from 0.21c to the leading edge between 3-12 degrees, much further forward than the low Reynolds number cases on the web site linked.
Now, at very low Reynolds number such as those found on low pressure compressor and turbine blades, glider wings and model aircraft, things change considerably. At very low Reynolds numbers transition would naturally occur a very long way aft. This results in the existence of a large region of laminar flow. As has rightly been pointed out; laminar flow is relatively unstable and cannot tolerate adverse pressure gradients or surface roughness. Therefore, a laminar separation often occurs. When this happens two events can follow. If transition occurs in the free shear layer, then the resulting turbulent boundary layer will be able to reattach creating a laminar separation bubble - as described in the web page zeeoo linked to. If transition, doesn't occur early enough then the flow will remain completely separated. This highlights the key difficulty in producing efficient aerodynamic shapes at low Reynolds numbers and is one of the key reasons why small scale gas turbine engines are inefficient and why model aircraft require such high power-to-weight ratios to make them work effectively.
In low Reynolds number flows, the turbulator provides a boundary layer trip that causes transition to occur much earlier than it would have done naturally. By triggering transition early, the laminar separation cannot occur and so the flow behaves much more like a high Reynolds number flow.
Clearly then, this technique can only add value in situation where the rotor is operating at very low Reynolds numbers, such as a Martian rotorcraft, and has been designed with an aerofoil that is completely inappropriate for that aerodynamic condition. At low Reynolds numbers, aerofoils must be carefully designed to control the large laminar region without triggering large separations. At high Reynolds numbers these problems seldom occur because transition occurs much closer to the leading edge (on the upper surface) and so transition occurs before laminar separation anyway.
In my opinion, turbulators are not appropriate for real helicopter/gyro plane flows, because the Reynolds numbers and Mach numbers will be too high for laminar separation be a serious problem - provided the aerofoil choices are sensible. At low Reynolds numbers they are a bit of a 'bodge-fix' and are a kind of admission of defeat in terms of designing a suitable aerofoil for the flight condition...we couldn't do it properly so we tripped the flow early to avoid the problem. Certainly not elegant engineering!
The other important point as Nick rightly pointed out is that there is a whole world of difference between the transition behaviour in steady and unsteady flows. I can tell you for certain that at typical helicopter Reynolds numbers and flight conditions that the transition behaviour in the leading edge region of the upper surface of common rotor aerofoil has a dramatic effect on the dynamic stall behaviour and the unsteady loads produced...but that's a heck of a lot more complicated!
So in answer to your question; has it been experimented with…yes, but not for the reasons you want. Will it help? No, not unless you are building a model gyrocopter! Will it help control stall? No, because stall is not a result of premature laminar separation on most helicopter aerofoils at realistic Reynolds numbers.
Hope this helps
CRAN
