PDA

View Full Version : One for the genuine aerodynamics experts

RMC
12th May 2018, 18:45
we have a general lack of awareness of the impact on fatigue life of flaps and associated structures of routinely operating close to the flap placard speeds.

The manuals are clear on a TYPICAL radar vectored approach... don’t use flaps as speedbrakes, descent clean where poss, second stage of flap a few miles before the loc etc.

I have calculated the basic loading on a retracted 787 flap system to be 20 Tonnes at 170 Tonnes landing weight. How do I calculate the increased loading at flap 30 / 170 knots. Rule of thumb is fine as I suspect this is hugely complicated in terms of certification type calls. Thanks in advance

scifi
12th May 2018, 20:17
Hi RMC, How have you calculated the 20 tonnes at 170 tonnes...?
If they are anything like ailerons, then if an aileron becomes decoupled from its actuation mechanism, it will lift up about a further 3 degrees then obviously would not have any loading either positive or negative.
So again how do you calculate the retracted 20 tonnes...?

It might be possible to guestimate the amount of drag the flaps produce, by seeing how much the glide ratio is increased then using the formula Glide Ratio = L/D, where you know L = 170t so can then find D. (assuming all other controls are not adjusted.)
Putting in some guess figures, If you apply 30 deg flaps and the plane goes from level flight to a downward 10:1 glide slope, then L/D =10 ... L=170 so D=17.. yes...?

sheppey
13th May 2018, 13:19
don’t use flaps as speedbrakes, descent clean where possible
My guess is that, regardless of the manufacturer's recommendation, the vast majority of pilots aim to use flaps as speed brakes in their initial flap selection during descent, instead of planning to extend them in level flight or close to it. This may be due to their impatience to reduce speed while descending. Hard to do in some aircraft. There are always seemingly plausible excuses for the habit; one of which is that ATC might notice a momentary marked reduction in flight path angle as the aircraft is trying to slow up before initial flap extension?.

VinRouge
13th May 2018, 14:18
we have a general lack of awareness of the impact on fatigue life of flaps and associated structures of routinely operating close to the flap placard speeds.

The manuals are clear on a TYPICAL radar vectored approach... don’t use flaps as speedbrakes, descent clean where poss, second stage of flap a few miles before the loc etc.

I have calculated the basic loading on a retracted 787 flap system to be 20 Tonnes at 170 Tonnes landing weight. How do I calculate the increased loading at flap 30 / 170 knots. Rule of thumb is fine as I suspect this is hugely complicated in terms of certification type calls. Thanks in advance

Hmm, complicated. Could possibly calculate by working out the change in stab angle at set points and change in pitching moment about CofG. Suspect the data you will need will include CofG, stab angle and TAS clean, environmental data at the data point, aircraft weight.

You can then calculate the change in pitching moment by using a second data set as above, aircraft clean, settled, flaps posn x. You could then look at data to come up with a rough idea of how muc hof this pitching moment change is caused by the flap, although you will need pressure distribution graphs for the aerofoil with the flap deflected.

However, I would be careful stating that this is evidence increased likelihood of fatigue, which is presided upon cyclic load frequency often more than the scalar of force. Have a look at the Tacoma Narrows Bridge collapse to see what Im talking about. To figure out this, you would need to be a structural engineer, with access to the aeroelastic model they have for the aircraft as well as detailed composite layup data for the likes of flap panels and tracks. I would hazard a guess that there may well be a particular speed, which may well be lower than flap extended limit, which may cause the greatest fatigue damage due to the likes of vortex shedding from nacelles/vortilons or the complex interaction of other aerodynamic components. In short, this is not trivial and from my time in industry, aero(servo)elastics are the most difficult of problems to solve or even accurately predict, particularly with the massively flexible structures now in use to generate the largest Aspect Ratio possible.
However, this is well accounted for within the design spec and will be factored into aircraft cycle limits. With it being a new aircraft, Boeing may well have strain sensors on redundant load bearing components which will monitor usage to ensure that the cyclic loads are within predicted limits.

Final point, fatigue is usually not as big an issue in composite structural components as the likes of cut aluminium or titanium spars (as long as an ultimate load factor is not exceeded). One of the benefits of this material. Usually, degradation in resin will be the limiting factor in material life.

My other observation would be that jets like the 78 are so slippy that small flap deflections are useful (as long as you dont end up dragging her in at the bottom). Most of the chaps I know on the 78 have stated how slippery she is. Speed brake from my time in the boot is noisy, so I would argue, why not?

RMC
13th May 2018, 16:32
Vin rouge......I said I thought it would be hugely complicated.....so thank you for not disappointing. In layman’s (ie simple pilot terms) would it be even possible to hazard a rough guess as to how much more stress deployed flaps would be under at 5 degrees or 30 degrees (I have no idea if it is 10% or 50%). Take your point though that it is cycles that are the primary factor.

scfi, I used to work at Broughton many moons ago. I also said my calc was basic. The 20 Tonne figure was essentially the exposed flap surface as a percentage of the 350 square metres of lifting surface. Effectively half a tone per square metre.Dont quite get your “aileron decoupled” comment the clean scenario in a non gust load alleviation scenario.

VinRouge
13th May 2018, 21:10
Its more than just flap cycles... the flow over the flaps from other sources will be unsteady and as such be creating cyclic forcing of the flap. This is why I say the force magnitude is probably irrelevant in this case. Vortex shedding is again highly complex and as they discovered at Tacoma Narrows, frequency is dependent on flow velocity.

You could look at typical pressure distributions for similar sections clean / flaps extended to gain a view of what the typical increase in static force will involve. But I suspect it is the dynamic case that will be more relevant here.

An example graph of Cp for varying flap deflections:

https://ars.els-cdn.com/content/image/1-s2.0-S1270963815003338-gr013.jpg

Jwscud
14th May 2018, 13:55
No quantitative data I’m afraid, but I carried the 777X chief tech pilot to Seattle a few months back and he was talking about design life issues and component changes required due to the way we are flying the aircraft.

The way stable approach requirements and speed control are interacting, Boeing are concerned that routinely taking landing flap and flying down the ILS within 10 kts of the limiting speed is invalidating a number of the assumptions used to size the components in the system. I imagine airlines engineering departments would have data on component replacements vs expected life and such like but sadly they’re unlikely to share it.

14th May 2018, 23:45
I'm not an aerodynamics expert. Don’t have to be. I just follow my company’s SOP.

We’re allowed up to ten knots above the proper flap extension speeds. Beyond that, the QAR is used as evidence against me when I’m in the Fleet Manager’s Office.

I could offer my opinion as to why our SOP is correct, but my opinion makes no difference. Only the SOP counts.

Fly safe,

Volume
15th May 2018, 13:05
As a very generic statement it can be said, that flaps are the item on most aircraft which are routinely operated closest to thier strength limit. Aircraft are very often flown only a few kts away from the maximum speed for the selected flap setting, while the overall wing is rarely flown close to its limits, neither in g´s nor in speed.
Additionally flaps are subjected to high relative motion due to wing bending and to high vibration from the engine jetblast, reverser turbulence or spoiler turbulence.
No designer dares to design flaps just strong enough to meet the requirements, typically you have to add quite an additional strength margin to keep the operators (reasonably) happy.

There are several NACA reports with detailed pressure distribution or flap loading measurement available on the web.

RVF750
15th May 2018, 21:29
Our lot are quite strict as well. The problem is it's virtually impossible to get the damn ship down if you're told to slow down to a set speed, say 220kts. The Speed brake is ineffective at slower speeds, hence the clue in the name. Taking gear is noisy and that really only leaves configuring early and using a "few knots" more of speed to help get it down where you want. For me, I prefer to get slowed down level, to a good amount less than the limit, then pull the flaps then hold the higher speed to increase descent. 220kts maximum where the limit is say, 250kts. I won't go higher or allow my F/Os to. If I can 210 or 200kts is almost as effective, depending on weight.

16th May 2018, 02:44
Many years ago (early nineties, maybe), Boeing published an article on this very topic. As I recall, they ran the numbers on different techniques to “come down and slow down.” The results were interesting.

What sticks in my mind is that slats/flaps, unless you throw out a lot of them, provide little drag. At slow speeds, speed brakes provide little drag, as well, but more drag than the first or second increment of flaps/slats.

If my memory serves me (which is questionable), the “experiment” involved the 737....an airplane which has a reputation of not wanting to slow down and come down simultaneously. One scenario was throwing out slats/flaps at Vfe. Another was speed brake, alone. One was gear alone. Finally, speed brakes and gear. The conclusion was that the first increment or so of slats/flats, even at Vfe, provides less drag than other options.

In my career, I have flown for three airlines. The first was bought by the second. Retired from there to fly for the third. In all three cases, the SOP was to extend slats/flaps on a specified schedule. My first two airlines were union; the third not. It was the third where some guy with thick glasses looked at QARs for any and all “deviations” from SOP, including deviations from the slat/flap schedule.

Fly safe,

RMC
19th May 2018, 15:05
Good food for thought guys .....thanks for the input.

ManUtd1999
19th May 2018, 16:17
As others have said, I'd say that accurate prediction of the effect on fatigue life will be close to impossible without access to detailed loads information. You'd need a histogram of the different cyclic loads and their frequencies (not easy to derive, even with CFD and FE analysis). That said, I doubt the life of the main flap body is limited by fatigue, especially if it's composite. IT might well be designed to the "endurance limit", below which stress can be assumed not to cause fatigue damage accumulation.

scifi
20th May 2018, 22:02
The flaps are not just one component, but multiple elements all working together. So it is hard to say when 'the flaps' will fail. It could be that the hinge pins will fail first, or some part of the track mechanism, or some bearings in the hydraulic actuators. Each one of these multiple elements will wear at a different rate.

Owain Glyndwr
21st May 2018, 08:42
The flaps are not just one component, but multiple elements all working together. So it is hard to say when 'the flaps' will fail. It could be that the hinge pins will fail first, or some part of the track mechanism, or some bearings in the hydraulic actuators. Each one of these multiple elements will wear at a different rate.

Agree with those sentiments. Flap loads (on all parts) will increase as square of speed with a small correction for the effect of the aoa reduction that goes with the extra speed.
There isn't much out there on that subject - in fact the only thing I found was NACA TR 633, but that was published in 1938!
For what its worth that gives about 3% reduction in flap normal force coefficient for a 10% speed increase, so deploying flap at 160 KIAS instead of 145 KIAS would increase the loads by about 18%.
The effect of that on damage tolerance (and hence inspection intervals) would depend on the design, but is not negligible

Chris Scott
21st May 2018, 11:27
Forgive me raising a tangential issue: the aforementioned use of landing gear as an air-brake - considering the limited efficacy of air-brakes (sorry, "Speedbrakes"). Extending the gear was remarkably effective on all the jets I flew. Particularly, of course, during gear transit, when the doors were open. Unless ATC had presented you with a near-impossible "go-down-slow-down" requirement, however, there was a feeling of having screwed-up. Because it was such a rarely-used expedient, it was quite difficult to judge how early to do it without over-kill. Subsequently, one could end up having to drag the a/c on to the approach with a lot of thrust, adding to the general noise and vibration. To make matters worse, sod's law sometimes involved ATC taking you around a leisurely circuit to fit their congested pattern. The dilemma then was: shall we raise the gear temporarily? Reckon I only saw that done once during over 30 years on jets. There seemed to be an unwritten taboo on it, and from the 1990s we were contemplating some sort of flag being raised on the post-flight QAR analysis (such as SESMA).

As with slats and flaps, obviously we have to be aware of the fragility of gear doors operated close to the limiting speeds, but would the occasional cycling of the legs themselves represent a problem in terms of wear and tear? I'd be interested in the comments of pilots and engineers on the above.