Trailing edges

[Mike J Lawrence]

Hi all

A friend of mine has a Exitation.flys very well. All the control surface trailing edges are finnished square about 1/4" thick neither of us know why. Is it to stop flutter or what.

Thx Mike:confused:

[davidfee]

I think that's the size the balsa was... so they left it that way. ;)

-David

[Ollie]

The main consequence of the blunt trailing edge is drag from the turbulent wake that it leaves. Unless the plane would benefit from some reduced drag because it is under powered or its top speed isn't high enough or its fuel consumption is excessive, there is little point in reducing drag. For an aerobatic plane, the added drag can reduce acceleration on the down line and give slightly better speed control.

[Bill Glover]

Square cut TEs are certainly supposed to reduce the chance of control surface flutter.

[davidfee]

Please explain to me how increasing turbulent airflow (square, thick TE) results in decreased tendency for flutter. It makes no sense.

I would argue that any observed decrease in flutter is a result of the increased stiffness due to the thicker cross section of the control surface.

-David

[Ollie]

There are many possible modes of flutter. One of them is flutter about the hinge line. The extra mass of the untapered trailing edge actually lowers the airspeed at which that particular mode of flutter will set in.

[Bill Glover]

Originally posted by davidfee
Please explain to me how increasing turbulent airflow (square, thick TE) results in decreased tendency for flutter. It makes no sense.

I'm only guessing, but perhaps in some situations a control surface with parallel upper and lower surfaces can be less prone to airflow separation than one that tapers in thickness towards the TE? Of course any turbulent airflow with a square-cut TE occurs behind the control surface.

It's usually only relatively slow flying models with extreme control throws that specify a blunt TE on the surfaces. In high speed flight at low AOA a finely tapered TE is always used, I used to race IC pylon where flutter was generally catastrophic so great care was taken with hinges and linkages, as well as control surface stiffness.

[davidfee]

Originally posted by Ollie
The extra mass of the untapered trailing edge actually lowers the airspeed at which that particular mode of flutter will set in.

A very good point, Ollie. It's another reason why the two probably most significant design criteria for preventing flutter are that the surfaces should be both light and stiff. It's my understanding that the resonant frequency of the surface should be as high as possible to prevent flutter.

I can see using the drag from thick TE's as a speed control as Ollie suggested, but that's about the only positive thing I see. Yes, it's easier than designing and building a proper control surface, but... ;)

I only have one aerobatic model with these blunt TE's... so there may be something I'm missing.

-David

[capncrunch]

I've read/heard anecdotal reports about the benefits of leaving blunt TEs for 3d maneuvers.

perhaps it shifts the center of drag rearwards, helping stabilize a plane with its CG pushed pretty far back.

otoh, a flat control surface would swing a few degrees wider each way (at the surface) than a tapered surface would.

just thoughts,

barrett

[Bill Glover]

Originally posted by davidfee
I can see using the drag from thick TE's as a speed control

Check out the fin TE on this X-15!

[Ollie]

The aerodynamics near, at and beyond the speed of sound are not the same as what our models experience.

[Bill Glover]

Quite true, but if a square-cut TE generated the kind of drag that's being suggested then the X-15 might have had a tough job getting to the speed of sound in the first place!

Coming back to models, wouldn't the airflow typically be detached long before it reached the rear edge of the TE? Doesn't that make it a bit academic whether it's left square or rounded/tapered?

[Sparky Paul]

I have all manner of trailing edge treatments, from needle-sharp tapered to squared-off blunt, some on the same plane, and have yet to find "flutter" is caused by the shape of the trailing edge.
A loose control system or a surface with a lot of mass hanging off the hinges will flutter.

[davidfee]

I did some polar calculations using xfoil a while back to see the predicted change in drag on common racing airfoils with varying TE thickness, from an ideal infinitely tapered TE to a max thickness of 2mm. The attached polar shows the result... a dramatic increase in drag. I don't know enough about the theory of the calculations to know how accurate this is, but it is consistent with intuition.

About the X-15 fins... I think Ollie addressed that. Keep in mind, it was a rocket and thrust was not in short supply. As for the drag, the Fiddler's Green website (http://www.fiddlersgreen.net/aircraft/jets/X-15/X-15info/X15-info.htm ) (they unfortunately don't cite the source) states:
"The fins also had a wedge cross-section with a broad, flat, chopped-off trailing edge. The thick fuselage side strakes that contained the fuel tanks were similarly chopped. This reshaping improved hypersonic stability, but it also caused a lot of drag, particularly subsonically. It was estimated that the base drag caused by the aft faces of the fins and strakes was as much as the drag of an entire F-104 airplane ."

-David

[Bill Glover]

Originally posted by davidfee
I did some polar calculations using xfoil a while back to see the predicted change in drag on common racing airfoils with varying TE thickness, from an ideal infinitely tapered TE to a max thickness of 2mm. The attached polar shows the result... a dramatic increase in drag. I don't know enough about the theory of the calculations to know how accurate this is, but it is consistent with intuition.

It would be interesting to repeat that with data (Re number and airfoil) more relevant to the type of model that is claimed to benefit from square cut TEs. A thick-ish symmetrical profile, and maybe Re 350,000 ?

Edit - and maybe TE thickness from 2mm to 5mm?

I know nothing about the theory!

Interesting about the fin drag on the X-15, just shows what you can do with a lot of thrust!

[davidfee]

I can't find any of these funfly/3D type airfoils (with the plank ailerons) in the database, so I'll have to draw one based on a normal airfoil... then I'll see what the polars look like. It'll have to wait until tonight...

-David

[BMatthews]

Bill, you're under the mistaken impression that drag on those fly fly models actually matters. In fact they WANT lots of drag to control the speed and the model flits about. If these types of models fly too fast they WILL flutter themselves to peices so the extra drag is actually a blessing.

For the same reason control line stunt models use much the same airfoils so the speed in the verticals is matched closely to the horizontals. And a few years back the FAI RC pattern models used thick airfoils for the same reason. I'm out of touch with the current philosophy in Pattern but I doubt things have changed much.

It's only the glider and electric guys that either don't carry up our fuel supply or it's in a heavy storage medium that care about such matters.

[Ollie]

For most prop driven aircraft there are two things that dominate performance improvement and they are weight reduction and thrust increase. Only pylon racers and like models have drag reduction as a high priority.

For most sailplanes, the two most important performance enhancements are weight reduction and drag reduction. Usually only pylon or cross country racing sailplanes benefit from weight increase.

[Bill Glover]

Originally posted by BMatthews
Bill, you're under the mistaken impression that drag on those fly fly models actually matters.

I'm not under any mistaken impressions, in fact I competed in fun-fly for many years ... so I'm really quite familiar with that type of model ;)

I'm simply querying whether there's any evidence that the common practice of leaving control surface TEs square (rather than tapering to say 1/16") significantly affects drag on a relatively slow-flying plane. My guess is that it doesn't.

[Ollie]

You may not notice the extra drag of a thick, blunt trailing edge when all it takes to overcome it is a click or two of throttle advance, but if you followed that practise on a glider I grarantee that you would notice it.

[davidfee]

Well, I drew up a plank-aileron style airfoil based on the NACA 0018, since it was easy. However, I have been unable to get xfoil to spit anything out. I think it doesn't like the kink in the airfoil... I think xfoil requires a smooth curve.

So, instead, I used the xfoil TE thickness feature in profili to modify the TE of the NACA 0018. Obviously this is not the same... but it should give an indication (as if we needed it). I used Re=150,000 for the plot, because anything lower just gives ugly results (no big surprise there either). So what you have to ask yourself is, what is the Re of a 3D model in a hover with zero airspeed? ;) Drag is nearly meaningless in that case... so who really cares. Drag is probably useful for limiting speed, as others have already mentioned.

So, to be done with this (I hope), here are some polars:

-David

[Bob Chiang]

I thought I read that a thick but sharp-cornered trailing edge can actually cause less drag than a rounded trailing edge (a finely tapered edge causing the least drag). Isn't that the idea of the Kamm back cars?

So maybe the thick trailing edge is a compromise between low drag, and a fragile wing.

[Ollie]

"So maybe the thick trailing edge is a compromise between low drag, and a fragile wing."

Very true if you use fragile materials. You can greatly strengthen the trailing edge and make it fairly sharp by using less fragile materials. You could glue a piece of 1/16 X 3/16 spruce or bass wood to tapered balsa trailing edge stock and continue the taper to about 1/64 of an inch thickness at the extreme trailing edge. If the trailing edge were sheet balsa top and bottom, you could bevel the balsa and sandwich a strip of 1/64 birch ply between the top and bottom bevels. Another option would be to sandwich a strip of 0.007 precured carbon between the top and bottom trailing edge sheets. The harder materials at the trailing edge will be at least as hanger rash resistant as blunt balsa trailing edges and a lot more aerodynamic. That's assuming that reducing drag would be beneficial. It may be advantageous to have lots of drag on aerobats to limit acceleration on the down line.

[Sail 'n Soar]

I did some polar calculations using xfoil a while back to see the predicted change in drag on common racing airfoils with varying TE thickness, from an ideal infinitely tapered TE to a max thickness of 2mm. The attached polar shows the result... a dramatic increase in drag. I don't know enough about the theory of the calculations to know how accurate this is, but it is consistent with intuition.

-David

Are the XFoil calculation assumptions valid with blunt TE's?

[Sparky Paul]

I've used 1/2" fiberglass tape between sheets on trailing edges... and before doing anything at all I use a Magic Marker™ to put a stripe right along the edge of the sheeting, on the inside, about 1/4" wide. Then I sand until I see this stripe.
Makes for a stiff and light t.e..

[davidfee]

Are the XFoil calculation assumptions valid with blunt TE's?
That's a very good question, one which I have often pondered myself. I have not read an answer in any of the "help" files I've read, but I honestly haven't done much research.

For what it's worth, the TE thicknesses in the polars above were adjusted from within XFoil, using XFoil features (using Profili as an interface). I doubt XFoil would have those features if the results made the calculations invalid... but only Dr. Drela would know best.

-David