In modern gliders, active tripping of the laminar boundary layer helps to reduce or avoid a laminar separation bubble. There are three ways of doing this:
- Noppenband (dimple tape):
Close-up picture of a spool of Noppenband (picture source)
Spool of zig-zag tape (picture source)
- Blowhole tripping: a line of tiny holes (0.8 mm diameter, usually with a steel liner) along the span which are fed by dynamic pressure, usually created from a small scoop or a NACA inlet (sorry, no picture). Note that while you can add the first two turbulator types at any time, blowhole tripping must be designed into the wing from the start by placing a duct inside and adding the scoop.
The order in which they are listed is no accident: Robustness decreases and effort for production increases as you move down the list. The pimples are like bugs and trip the boundary layer directly. This is fine if you fly at one angle of attack and one speed only, because then the chordwise location of the separation is fixed. Just place the Noppenband 2-3% of chord ahead of the separation point and it will effectively avoid laminar flow separation by tripping the boundary layer.
If you vary both speed and angle of attack, the location of the laminar separation bubble on the upper surface will vary over a wide distance. If you try to optimise the transition for slow flight (= low Reynolds numbers and high angle of attack), you need to place the turbulator tens of percent ahead of where it would be best placed for fast flight, thus adding a lot of unnecessary turbulent boundary layer when friction drag counts most.
On the lower side of the airfoil, the effects of Reynolds number and angle of attack work against each other, and the separation bubble changes its position only slightly over the speed range of the glider. Now it makes sense to actively trip the boundary layer, but you still get some additional turbulent length when using Noppenband.
Zig-Zag tape will merely prepare the boundary layer for transition, so laminar flow continues for a few percent downstream until the local pressure gradient does the tripping. Now the variation in the separation point will not result in extra turbulent length. Normally, Zig-Zag tape is placed 2% ahead of where Noppenband would go for best effect.
Blowhole turbulators can even be placed inside the laminar separation bubble and will destroy it when positioned a few percent past the actual separation line. By blowing into the eddies of the separated flow, the tripping will alter the pressure distribution such that the laminar bubble is either much smaller or even disappears altogether. Now the turbulator can be placed further back and will add even less extra turbulent boundary layer.
With blowing also the upper separation bubble could be avoided if several rows of holes are placed along the span and the pilot can control to which one the airflow from the scoop is directed. To my knowledge, this has never been used in real life, though.