Do vortex generators really work.

[planesmaker]

Dafydd, thanks for sharing your knowledge and experience with us all, I find it very interesting and educational. I have thought about using vortex generators on a project I have, however the marketing hype has now been put into perspective. Thanks. Tom

 

 

[Dafydd Llewellyn]

Dafydd, thanks for sharing your knowledge and experience with us all, I find it very interesting and educational. I have thought about using vortex generators on a project I have, however the marketing hype has now been put into perspective. Thanks. Tom

Thanks, Tom. I'd use VGs on my own aircraft - but set up for the improved stall behaviour, not because of the numbers. What they do, is to make the full speed range usable. One of the demonstrations I have used in the Seeker, is to simulate a violent evasive manoeuvre - such as one might make if somebody were shooting at you - from 65 knots (the flaps-up stall speed being 56 knots): Boot full rudder, full aileron, stick on back stop and full power, to haul the thing into a tight turn. Any normal aeroplane would flick-roll from such treatment. The Seeker simply turns and keeps flying, with a barely-perceptible buffet from the stalled centre-section. The people marketing VGs are pushing the wrong aspect of them, in my view; used properly, they are a magnificent safety device. Used wrongly, they are bloody dangerous. I just wanted to explain that to people.

 

 

[Head in the clouds]

Yes planesmaker, I add my thanks to Dafydd for the valuable insight into the whole VGs subject.

 

And Dafydd, I'd particularly like to say what an ingenious set of solutions you devised for the Seeker. This is something that designers have been trying, with very little success, to accomplish for decades. Probably the Ercoupe was the first to make some inroads into preventing people from killing themselves from stall/spin events in the 'pattern'. For those who don't know the Ercoupe, the first versions of it had rudder coupled to aileron on the stick and no rudder pedals. Additionally the elevator didn't have sufficient authority to stall the wing in a progressive stall. The result was that in the circuit you couldn't end up with crossed controls and then stall it trying to line up on final. It saved a lot of lives probably but was quite a kindergarten way of going about it, although I guess it suited some pilots with dead feet. Ercoupe pic below -

 

 

Dafydd's concept of modifying the elevator effectiveness during the initial part of the stall by killing the downwash from the mainplane at the critical time is, in my mind, truly visionary.

 

To hypothetically take the 'Airflow' process one step further, if one was to add slats to a Seeker (or Champ, or whatever) I'd assume that to prevent them interfering with the nicely 'developed' centre section stall and vortex characteristics the slats should only be between a little outboard of the fence and the wingtip?

 

Theoretically that should make it even more spin resistant (if that's possible) and give it the ability to reach a higher alpha/lower speed before stalling by overriding the centre section's control of elevator effectiveness when required for STOL operations, by using high engine power to energise the HS/elevator similar to the SuperSTOL in the video below? (See series of high powered landings from 1:30)

 

If one went down that road it would probably remain very reluctant to spin but if it did do so what effect do you think the slats would be likely to have on the recovery, based on the Easy In, Easy Out, Hard In, Hard Out notion of spin characteristics?

 

 

 

[Oscar]

From all of the above, it occurs to me that one should be very prudent about flying in a 19 (or CH-exp) reg aircraft that has had vg's attached - because the range of effects may not have been properly tested. It is, I think, one of those areas where a little knowledge can be a dangerous thing, and there is a bit of a perception that vg's are a marvellous cure for any sort of undesirable stall behaviour, or just a silver bullet for improving stall speed.

 

Human nature being what it is and vg's being cheap to buy and easy to attach, inevitably some people will adopt the 'more the merrier' approach, and also assume that what they see working on one aircraft will translate to the same effects on another. We've all seen that syndrome over and over - remember the fad for enormously wider tyres than standard on cars years ago? Many people assumed that more rubber meant more cornering power, and then found that things like camber change meant that the tyres would suddenly lose huge amounts of grip as the tread became removed from the road, that they aquaplaned like curling pucks on spit or simply slid uncontrollably on gravel patches etc.

 

I think it would be useful information if RAA were to produce a 'beginner's guide to vg's' that would give people some sort of idea of what they are looking at before they jump into a modified aircraft - and perhaps some idea of the sort of questions they should ask of the owner of the aircraft re the testing that's been done. Just because something looks like an excited hedgehog obviously doesn't mean it's necessarily going to fly like a cuddly puppy...

 

 

[facthunter]

The swept back tail is one of my pet hates. Styling should be the last consideration AFTER all necessary or achievable control capabilities have been satisfied. There appears to be a tendency to make something that would look pretty as a model on a desk, and not fly well. Style sells but is a trap sometimes.. Nev

 

 

[Dafydd Llewellyn]

To hypothetically take the 'Airflow' process one step further, if one was to add slats to a Seeker (or Champ, or whatever) I'd assume that to prevent them interfering with the nicely 'developed' centre section stall and vortex characteristics the slats should only be between a little outboard of the fence and the wingtip?

 

Theoretically that should make it even more spin resistant (if that's possible) and give it the ability to reach a higher alpha/lower speed before stalling by overriding the centre section's control of elevator effectiveness when required for STOL operations, by using high engine power to energise the HS/elevator similar to the SuperSTOL in the video below? (See series of high powered landings from 1:30)

 

If one went down that road it would probably remain very reluctant to spin but if it did do so what effect do you think the slats would be likely to have on the recovery, based on the Easy In, Easy Out, Hard In, Hard Out notion of spin characteristics?

 

Hypothesizing - The elevator-inhibiting effect of the reduced downwash from the stalled centre-section is limited; it cannot be worth more than about half the angle of attack at which the centre-section stalls - maybe around eight degrees, at a guess. That was barely sufficient, in the Seeker, to maintain the elevator inhibition over the aircraft's CG range of 11% MAC; without the 3rd stage supplied by the leading edge notch at the inboard end of the ailerons, it was possible to occasionally get the second-stage stall at the aft CG limit, which gave a small wing-drop, but the VGs were keeping the ailerons effective, so one could easily catch it with aileron. When the leading-edge cuffs were added, with the VGs on top of them, it became impossible to get any wing-drop at all. So, firstly, the secret to spin-resistance is to use a small number of discrete devices that generate powerful chordwise vortices that rotate the opposite way to the tip vortices, as soon as the stall propagates to them from inboard. The fences and stall strips do that for the first-stage, and the sharp inboard end of the leading-edge cuff does it for the second stage. And at each step, you want a graduated increase in the stall angle.

 

My guess would be that you could make this principle work to higher lift coefficients by a suitable application of slats. Quite how you would work the second major chordwise vortex at the inboard end of the ailerons, eludes me.

 

Anything approaching full-span slats raises an enormous problem with the undercarriage, though; Handley-Page demonstrated that with his Gugnunc (see photo) and the Alaskan super-stol gadgets also give ample evidence of that. Also, slats - even retractable ones - always cause substantial increased drag. I flew the sole example of a Rallye Commodore in Australia a bit, and I'll never forget landing it on the 06 runway at Camden, at light weight one afternoon, into about a ten knot breeze - it stopped literally within the length of the piano keys; the student waiting to line-up in a 150 gave me a staggered look as I turned off under his wingtip. One of those moments, it was . . . However, I came to agree with Randy Green, who had done the first-of-type testing of the first Rallye into Australia, that it was essentially a landing-and-taking off device, rather than a thing for going places. I never tried spinning it; but I suspect the designer's strategy was to make the wing stall angle so great that you could never reach it. VGs are less spectacular, but they have very little if any adverse effect on the cruise or climb performance, so are more useful in my view, unless "super-STOL" is your prime aim.

 

 

 

[Oscar]

I flew the sole example of a Rallye Commodore in Australia a bit, and I'll never forget landing it on the 06 runway at Camden, at light weight one afternoon, into about a ten knot breeze - it stopped literally within the length of the piano keys; the student waiting to line-up in a 150 gave me a staggered look as I turned off under his wingtip. One of those moments, it was . . . However, I came to agree with Randy Green, who had done the first-of-type testing of the first Rallye into Australia, that it was essentially a landing-and-taking off device, rather than a thing for going places.

 

If that was the same Rallye I flew in from Canberra to Tocumwal with you, it was absolutely a horrible thing for cross-country - too directionally stable and weaved everywhere every time it hit a gust from one side. The nearest thing I've ever been in for an unpleasant travelling motion was a Spray-inspired yacht with a vastly too small rudder and a steering wheel that was slipping on the shaft in a nasty following sea with added side-chop. We did several 360's from broaching between Sydney Heads and Barrenjoey Head, and that was with one person playing the main sheet and another playing the jib - and the mainsheet hand was a highly-experienced racing yacht crew member!

 

 

[Dafydd Llewellyn]

If that was the same Rallye I flew in from Canberra to Tocumwal with you, it was absolutely a horrible thing for cross-country - too directionally stable and weaved everywhere every time it hit a gust from one side. The nearest thing I've ever been in for an unpleasant travelling motion was a Spray-inspired yacht with a vastly too small rudder and a steering wheel that was slipping on the shaft in a nasty following sea with added side-chop. We did several 360's from broaching between Sydney Heads and Barrenjoey Head, and that was with one person playing the main sheet and another playing the jib - and the mainsheet hand was a highly-experienced racing yacht crew member!

 

Yep - that's the one.

 

 

[Old Koreelah]

...My guess would be that you could make this principle work to higher lift coefficients by a suitable application of slats. Quite how you would work the second major chordwise vortex at the inboard end of the ailerons, eludes me.

We humans must be getting close to the limits of what we can do with conventional materials and designs. Birds approach these problems from a different angle; with individually-controlled, self-repairing surfaces.

 

 

[Dafydd Llewellyn]

Garn . . . They use retractable slats, variable geometry, and VGs too.

 

 

[Head in the clouds]

We humans must be getting close to the limits of what we can do .....

Somewhere around 1908 the Director of the US Patent Office reputedly said that the Office would be closed down in the near future because there was nothing worthwhile left for man to invent.

 

Whether the utterance is purely anecdotal or not I don't know but in most years since then more inventions have been registered than in each previous year because humans instinctively improve upon what they already have. Hence the emergence of the Innovation Patent in some parts of the world, which I expect will become more widespread in due course.

 

A good example is Dafydd's trick Airflow kit which certainly appears to be a complete solution. But then it started me thinking about how one would take it a step further ...

 

 

[Oscar]

There's always room for improvements, but don't forget that the Seeker work was accomplished over a considerable period of time and effort and flown by an experienced (and qualified) Test pilot with all the necessary safety gear attached - especially a carefully-developed spin chute that had itself been tuned to the specific aircraft. Plus, a certified TAS probe to ensure that results were real.

 

It certainly appears that the use of vgs should be considered always as part of a 'system', where the various interactions of components that affect the aerodynamics are very carefully considered. As Dafydd has pointed out, it's not just all about the wing and separation being controlled in a more desirable way than many designs deliver. How and what the tail feathers do downstream of the airflow off the wings is important.

 

While pretty much all of the reporting on the 'aftermarket' attachment of vgs is generally positive, I suspect that there could be a world of difference between a conservative use of them to 'smooth out' nasty sudden separation characteristics and deciding that a large bucket of them and some glue can change your lead sled into a Piper Cub. I personally wonder - as a complete non-engineer - what structural effect on the wing spar loading you might get from considerably increasing the lift available outboard while the inboard section unloads. I'd be a tad unhappy (if only very briefly) if my strut-braced thingywot with a full load of fuel in the wing tanks decided to give up the outboard section when hauling out of a tight site because I was taking full advantage of my new 'stol' capability..

 

 

[Old Koreelah]

Garn . . . They use retractable slats, variable geometry, and VGs too.

So who holds the patent on those?

 

 

[Keenaviator]

I noticed while building my Jabiru UL450 that the wings were absolutely straight - no washout. The rigging of the flaps appears to be how Jabiru get the centre section of the wing to stall before the tips. The flaps are rigged a couple of degrees down while the ailerons are set at zero degrees. They are also quite specific in the rigging of the elevator, 18 degrees up and 6 down (limiting power to achieve very high angles of attack?). Do you have any experience with these Dafydd? Laurie

 

 

[Dafydd Llewellyn]

There's always room for improvements, but don't forget that the Seeker work was accomplished over a considerable period of time and effort and flown by an experienced (and qualified) Test pilot with all the necessary safety gear attached - especially a carefully-developed spin chute that had itself been tuned to the specific aircraft. Plus, a certified TAS probe to ensure that results were real.It certainly appears that the use of vgs should be considered always as part of a 'system', where the various interactions of components that affect the aerodynamics are very carefully considered. As Dafydd has pointed out, it's not just all about the wing and separation being controlled in a more desirable way than many designs deliver. How and what the tail feathers do downstream of the airflow off the wings is important.

 

While pretty much all of the reporting on the 'aftermarket' attachment of vgs is generally positive, I suspect that there could be a world of difference between a conservative use of them to 'smooth out' nasty sudden separation characteristics and deciding that a large bucket of them and some glue can change your lead sled into a Piper Cub. I personally wonder - as a complete non-engineer - what structural effect on the wing spar loading you might get from considerably increasing the lift available outboard while the inboard section unloads. I'd be a tad unhappy (if only very briefly) if my strut-braced thingywot with a full load of fuel in the wing tanks decided to give up the outboard section when hauling out of a tight site because I was taking full advantage of my new 'stol' capability..

Good point - especially with leading edge slats, which can increase the wing local lifting capability by a lot more than 16%. The change in spanwise lift distribution for a setup like the Seeker's is not really any more structurally significant than applying full aileron at Va; and one does not normally expose the aircraft to the full limit-load dynamic stall situtation, so it's unlikely to have any real effect on the wing fatigue life or its static strength. People normally use the STOL capability at essentially one G, but if you're doing combat turns relying on the leading edge slots a la Me109, then the effect had better be taken into account in the aircraft's structural design.

 

You are also entirely correct that this sort of system needs to consider the aircraft as a whole; the Seeker also got a set of VGs up the sides of the fin, to keep its crosswind capability in step with its reduced stall speed.

 

The "super STOL" thing seems to ignore a number of the collateral effects of increasing the maximum lift coefficient of the aircraft; doing so greatly increases the liklihood of getting caught "on the back side of the power curve" - for example, an increase of 50% in the maximum lift coefficient carries with it a 125% increase in the induced drag; and as soon as the speed reduces below the minimum drag speed - at which the induced and parasite drags are equal - it becomes speed-unstable; and that's a slippery-dip; as you slow down, the drag increases exponentially; and unless you have a LOT of power available, the thing will fall out of the air simply as a result of the drag increase. And if you DO have lots of power available, you risk a torque-roll. Going to extremes with high-lift systems has a LOT of problems. It tends to make an aborted landing quite exciting, too. Certification standards address aborted landing issues, and that has the result that most aircraft end up with a reasonable all-round balance - and they do not normally, in the GA world, go as far as double-slotted flaps. Swept-wing jets have to try much harder, because of the fundamental limitations of swept wings at low speed - but that's really another world.

 

As a matter of interest, the Ercoupe safety record has been the subject of a number of studies - and it was not the great step forward that Fred Weicke had hoped. It appears that you do NOT increase safety by restricting the controllability available to the pilot. It also appears that every fool-proof device breeds a new evolution in fools; "fool proof" seems to be an oxymoron; I prefer the term "idiot-resistant".

 

The Helio Stallion would be an interesting study, I suspect. However the Dornier 28 seems to have made a significant gain in safety by the use of full-span slats; but it also has other features that give it a minimum control speed even lower than its stall speed.

 

 

[Dafydd Llewellyn]

I noticed while building my Jabiru UL450 that the wings were absolutely straight - no washout. The rigging of the flaps appears to be how Jabiru get the centre section of the wing to stall before the tips. The flaps are rigged a couple of degrees down while the ailerons are set at zero degrees. They are also quite specific in the rigging of the elevator, 18 degrees up and 6 down (limiting power to achieve very high angles of attack?). Do you have any experience with these Dafydd? Laurie

I was not involved in the flight testing of any Jabiru after the ST (LSA 55 with the 2200 engine); I've briefly flown the 230C and Rod Stiff's personal 450, but certainly not to the extent of exploring their stall characteristics. However, the original LSA 55 stall behaviour at one stage necessitated very large stall strips. I think they were eventually rendered unnecessary by very tight control over the wing twist that was induced if the feet of the wing mould were not accurately levelled; the same "wandering stall bubble" behaviour as for the Seeker and the Skyfox was noticeable on the LSA 55. It uses a very different airfoil, but still an aft-stalling one - and any aft-stalling airfoil on a rectangular wing will show this behaviour, I expect.

 

However, the flap and elevator control-stop rigging is almost certainly there in order to control the stall behaviour; and also quite probably the spin recovery. They should be scrupulously followed, because errors there are likely to have quite considerable effects on those aspects of handling.

 

 

[Oscar]

So, one presumes, it would be a very necessary piece of knowledge to know the actual wing section and the characteristics of the stall behaviour before just deciding that 'it look the same as' and bunging on some vgs and expecting the results to be similar to that experienced on something else?

 

This is needed information; since becoming involved in RAA-class aviation I've noticed a worrying propensity on the part of some people to rely on 'what some bloke in the clubhouse told me' combined with the ever-present 'we don't need no engineers' syndrome - and way too much of that is treated as gospel because it comes from the mouths of apparent 'experts', some of whom are L2's venturing into areas in which they have no actual competence.

 

 

[Head in the clouds]

So, one presumes, it would be a very necessary piece of knowledge to know the actual wing section and the characteristics of the stall behaviour before just deciding that 'it look the same as' and bunging on some vgs and expecting the results to be similar to that experienced on something else?This is needed information; since becoming involved in RAA-class aviation I've noticed a worrying propensity on the part of some people to rely on 'what some bloke in the clubhouse told me' combined with the ever-present 'we don't need no engineers' syndrome - and way too much of that is treated as gospel because it comes from the mouths of apparent 'experts', some of whom are L2's venturing into areas in which they have no actual competence.

Probably fair comment but also worth keeping in mind - where VGs are concerned - is that VGs have been around a long while now and amateurs all over the planet 'without expert engineering knowledge' have been bunging them on their planes, right place, wrong place, too many, too few and all that for decades and I haven't yet read a crash report that made mention of the VGs, let alone blamed them for the event. Not to say it hasn't happened but just possibly willy-nilly addition of VGs hasn't done as much harm as might be imagined.

 

 

[Dafydd Llewellyn]

Well, if it's an LSA factory-built aircraft, one cannot go bunging VGs on it without the manufacturer's consent. If it's a Type-certificated aircraft, you don't need the manufacturer's consent, but you do have to do it in accordance with approved data - which could be an STC or a Part 21M approval. If it's a homebuilt or an experimental aircraft, you can do what you like, subject to the rules re operation in a designated test area - but unless you don't mind possible unpleasant surprises, it would be at least prudent to know enough about the aerodynamic design of the aircraft to have some realistic idea of the likely effects. Having said that, I'd expect the Seeker-type solution to work pretty much the same on most high-wing aircraft that do not already have leading-edge devices, except possibly those with NACA five-digit wing airfoil sections - provided you adjust the stall strips in the manner I described.

 

If the aircraft already has leading edge devices, then you would be in an unknown area.

 

 

[Dafydd Llewellyn]

Probably fair comment but also worth keeping in mind - where VGs are concerned - is that VGs have been around a long while now and amateurs all over the planet 'without expert engineering knowledge' have been bunging them on their planes, right place, wrong place, too many, too few and all that for decades and I haven't yet read a crash report that made mention of the VGs, let alone blamed them for the event. Not to say it hasn't happened but just possibly willy-nilly addition of VGs hasn't done as much harm as might be imagined.

Fair comment - but the absence of mention in crash reports may well be due to the fact that the possible effects are not generally well understood - and also to the fact that ATSB does not investigate crashes of experimental aircraft.

 

 

[Old Koreelah]

... fit small leading-edge "fences" to the wing leading edge, in line with the tailplane tips - i.e. a couple of inches inboard of the first VG. See photo. The fence need only go to about 20% of the wing chord, on the top surface, and it need only be around 60 mm high...

Dafydd I am snookered; when I fold up my Jodel the carrier mechanism only just clears the right wing where I need to install the fence. I guess a 10mm high fence is a waste of time. I can install fences about 150mm further outboard, but I presume their vortices would miss the tailplane. Any suggestions?

 

 

[Dafydd Llewellyn]

I think you will find that they'll work just as well that bit further out; the main thing is that the tailplane is wholly in the wake of the part of the wing between the fences. You may lose a couple of tenths of a knot of the reduced stall speed, but the spin-resistance will still work. It's not critical that the vortices hit the tailplane.

 

 

[Old Koreelah]

I think you will find that they'll work just as well that bit further out; the main thing is that the tailplane is wholly in the wake of the part of the wing between the fences. You may lose a couple of tenths of a knot of the reduced stall speed, but the spin-resistance will still work. It's not critical that the vortices hit the tailplane.

Thanks Dafydd. " a couple of tenths of a knot reduced stall speed". That's a worry; my landing run might be about 5 metres further. The things we have to sacrifice to avoid falling out of the sky.

 

 

[Dafydd Llewellyn]

Yep; silly, ain't it? Oddly, I found I could consistently do shorter landings in a Beech V35B (78 knots over the fence) than I could in my Cherokee 140 (60 knots over the fence) simply because you could always pick exactly when the Bonanza would stop flying, whereas the Cherokee was inconsistent. May have been the braking effect of the CS prop in fine pitch, but it showed me that controllability was at least as important as a slow approach speed. I've not flown a D9, only a flapped D11 (once), but I expect they're not to dissimilar in general behaviour - and it felt pretty good in that department. The correct use of VGs does help the controllability - that's its principal advantage, in my view.

 

 

[paulh]

Theres a D10 on my current work site but try as much as they can it just won't lift off with or without flap (maybe the problem is the scraper in front that it's pushing?)

 

Sorry been a long day