Intersection drag of wing/fuse joint

[SoarNeck]

Hi all,

After a brief hiatus, I've picked up my F3B design again (that's what, six months of design and nothing actually built yet :) ). I've finished all the aerodynamic and stability design work, and now I've started to flesh out the details so I can generate some working plans. Basically the only thing that's left is to make the stab tips a bit more elegant.

Right now I'm into "drag reduction" mode, and hence my question for today.

Basically, I need some help with the wing saddle. The cross section of the fuse in front of the leading edge and under the wing saddle is largely circular, and it tapers to a vertically elongated oval towards the tail. The wing is set down into the fuse by an inch or so, which means that the side of the fuse might meet the bottom of the wing at an acute angle if I don't fair it in some way. The deepest I'd like to set the wing into the fuse would be to have the highpoint of the airfoil flush with the top of the fuse, assuming the fuse was a complete circle.

I remember reading Dr. Drela's recommendation against filleting that joint, but I can't remember why he said not to do so - I imagine it has something to do with interference drag.

Is the acute angle okay, or should I bring the sides of the fuse up at 90 degrees to the bottom of the wing? That would require a smooth fillet at the LE, and one to "tuck the contours" back into the fuse behind the wing?

Ideas would be helpful (including any other drag reduction tidbits). I've attached a drawing that hopefully helps (you're looking down the fuse, from the nose).

Cheers,
Adam

[Salto]

Adam,

In a past life I made several "mileage marathon" vehicles - the ones that carry one person around a race track using as little fuel as possible. We got about 4,000 miles per gallon, so they worked OK.

Apart from the engine technology, and reducing rolling resistance, the big emphasis in the design was on aero drag reduction. The geometry you mention above was a typical one in these vehicles where a beam axle between two streamlined bicycle wheels passed under or through a teardrop body fairing.

All the reference texts we used warned against having an acute angle there. The theory was that pinching the air into the narrow Vee shape would create eddies and seperation. The lowest drag configuration will be when the axle/wing/tail meets the body at 90 Deg., and is generously filletted. This is exactly what full size sailplanes do.

We made versions both with and without an acute intersection, and whilst we weren't able to do any direct drag measurements, we did do extensive wool tuft testing which clearly showed the bad effects of an unfilletted acute intersection.

So my gut feeling would be you should bring the fuse up vertically to meet the wing, and also have some sort of fillet. Or better still, have the wing down at the centre of the fuse. Another consideration would be to minimise the surface area around that area, and it's neat that a filleted geometry can have less area than the bare acute intersection.

I'm sure there are many more subtlties and trade-offs to do with the flow under and over the wing and how that effects lift and drag of the wing, but I'm not qualified to comment on these.

BTW, the best reference I found for this sort of stuff was Fluid-Dynamic Drag by S.F. Hoerner 1965. It's largely a collection of wind tunnel tests, some dating back to the 1930s in Germany. Wonderful empirical stuff! I have a copy, but I've no idea whether it's still available. I think it was self published in the US.

Graham.

Edit: I'd really like to hear why Dr Drela recommends against filleting that area.

[SoarNeck]

Thanks for the info, Graham. Is that the SAE Supermileage competition you're talking about? I did the Heavy Lift competition for more years than I care to remember as well!

Do you have any other drag-reducing tips you'd like to share?

I tried to hunt for posts on the subject of fillets, but couldn't find any. I'll keep looking.

[SoarNeck]

BTW - the wing can't sit too low on the fuse because the ballast tubes and towhook get in the way. Also, the notch in the fuse is a stress riser as it is, and I need to keep this airframe as stiff and strong as possible.

[SoarNeck]

Here's a quick 3-view, in case you're interested. The stab tips are getting re-shaped, but this should be close otherwise. My last change is to decide whether fuselage fillets are worth the effort, since it would complicate the build considerably.

Any other drag-reducing tips would be appreciated (not changing the planform or tip shape).

[SoarNeck]

Jeez that looks bad...oh well, just a screen capture!

[Salto]

Adam,

Nice design. Please keep posting your progress here.

I guess your SAE Supermileage comp is the same thing. Here, it was sponsered by the Shell oil company, and was called the Shell Mileage Marathon. Unfortunately it no longer runs, but I've still got some nice 2.5m long teardrop moulds. One day, I'd like to make a low drag roof top model transporting box.

I've just re-read the relevant parts in Hoerner and also in Martin Simmons Model Aircraft Aerodynamics (highly recommended). To paraphrase these, here's a few more ideas.

1. The front of the fuse, back to the wing, has potential for some laminar flow, so it's worth using a 60% laminar flow low drag body shape. To maximise this laminar flow extreme care should be taken to ensure an accurate shape with no protuberances or gaps, so a slide-on nose cone with an accurately made joggle joint will be a good approach.

From the wing LE back, laminar flow will certainly be lost, so the shape from there back should be determined by minimising wetted area, structural considerations, and avoiding sharp contractions in cross section area that may produce seperation.

2. Fillets should be used at the fuse/wing intersection. These fillets can be smaller (less additional skin area) if the wing is mid mounted. Hoerner suggests the optimum fillet radius be 4 to 8% of chord.

3. At thermalling speed the wing induced drag dominates, so fuselage parasitic drag is less important. But at higher speeds parasitic drag is much more significant, so your design in terms of profile and parasitic drag issues should be optimised for high speed flight, ie. fillets should be designed to work correctly at low (wing) angles of attack. If I understand F3B correctly, weight is not a major issue, so a mid mounted wing at the expense of some added structural complexity is probably worth looking at.

If the wing is mid mounted, in order to avoid adverse affects on wing lift and wing drag generation it will be very important to get the top fillet right. Depending on how much development time is available, this may dictate going to a more conservative high wing configuration.

4. The angle of incidence of the fuse should be set to achieve lowest drag at the high speed (low wing angle of attack) condition, but check that the low drag shap you use also stays in it's low drag area at thermalling speed.

5. There are similar interference drag issues at the tail/fuse intersection. Simplistically, the more intersections you have, the higher the interference drag will be. So a Vee or a Tee will have lowest interference drag. The Tee imposes higher structural requirements, so maybe a Vee is the optimum since weight at the very back of the fuse can be a significant issue. Needless to say, all tail/fuse or tail/tail intersections should be filleted.

6. Other drag reducing ideas (which I'm sure you've already considerd): Use live hinges for ailerons etc. Seal everything against air leaks. Keep aileron linkages etc. below the surface, or well faired. Fair the tow hook. Keep the antenna inside. Use a thin airfoil section. Keep the fuselage cross section as low as possible. Make all surfaces smooth, without waves or bumps. Minimise panel joints and gaps.

Hope this helps.

Graham.