Hazykid

Hi All.

After lurking for a while, this is my first post, so here goes!

I need some advice - I'm tackling a scratch build - a MiG 15 approx 1/6 scale IC Ducted fan. I know a lot of people will say DF is washed up, and this is a big undertaking, but this is a 'special' project that I want to complete.

Thing is, I'm stuck on what airfoil to use for the wings. I took a look at some online databases and I see that the full size MiG 15 used TsAGI S-10/TsAGI SR-3 airfoils.

I know enough to realise that a slow flying radio control scale model would not use the same airfoil but I don't know enough to choose the correct airfoil for this application. I took a look to see if I could find what airfoils similar scale models use, but just can't find the information anywhere. Does anyone have any advice they could give me?

I also note that the MiG 15 had a degree of anhedral - should I build this into the model as well?

Any help that anyone could give would be greatly appreciated.

All the best,

Richard.

BMatthews

For your needs any symetrical airfoil of the right thickness would be just fine. Especially if you will be including the use of functional flaps for takeoff and landing.

I would include the slight anhedral since it does add something to the model when seen on the ground and in flight. Aerodynamically it won't make a bit of difference due to being so slight an amount.

And DO include the wing fences. It was the Mig's way of controlling the span wise airflow and including them should reduce the need for washout in the wing tips. However I would still include a couple of degrees of washout. Swept wings can do funny things near the stall without such tricks.

Hazykid

Hi,

Thank you for the advice. I had wondered if a symetrical airfoil was the way to go. I feel more confident now.

One more question though You say 'of the correct thickness' - Is it acceptable design practice to find the scale thickness then choose a symetrical (perhaps NACA?) airfoil that fits the bill, or are there design constraints that mean the actual wing I build will need to be thicker than a scale wing?

Once again, many thanks,

Richard.

soarrich

My Feedback: (98)

This is just my opinion, but here it goes.

The symmetrical airfoil will work, but your plane won't be flying upside down half the time so I wouldn't use one. My friend did a big gasser Japanese Cydie for a magazine and used a symmetrical airfoil it took about 100m to get airborne, and took full power most the time to fly. He later destroyed the wing and had to make a new one, I printed him up a set of Clark Y ribs for the same wing platform, it was a totally different plane, it took off in about 30m and flew around on 1/2 throttle most the time.

If I where building the Mig 15 I'd use a NACA 2412 or a Clark Y, they are really good airfoils. If you need thinner airfoils I'd go with an AG 40.

Leave the anhedral in, every 3* of sweep acts like 1* of dihedral, so it makes it easier to roll.

Hazykid

Thanks for your input Rich.

It does make sense - I don't see myself flying the MiG upside down too often.

Can I ask you another question - The NACA 2412 looks slightly undercambered, but the Clark Y is flat bottomed. What reasons would one choose the NACA over the Clark or visa versa?

Many thanks,

Richard.

soarrich

My Feedback: (98)

You might chose a airfoil to make the wing look right, if the max thickness of the MIG's airfoil was a 40% then a airfoil with the max thickness at 25% might make it look strange.
The Clark Y is flat from about 30% back, that makes it easier to build if you are using ribs. If you are using foam I would pick the airfoil to make it easier to fit your wing tube or landing gear in. The Clark Y has ~3.5* of camber, the has 2* of camber so the Clark Y will make more lift at a given speed, so it would slow fly better. The 2412 with less camber might be easier to make go fast though, and the MIG 15 being a jet you may not want it to float around like a Cub

BMatthews

By correct thickness I meant to pick a percentage thickness that matches the original. There's no real need to make the airfoil thicker than scale unless it turns out that it'll be something rediculous like 3/4 inch thick. Then structural issues would occur and for reasons of making it easier to build strongly enough I'd suggest making it more like 1.25 inches thick at a minimum for the size you're looking at.

I suggested a symetrical airfoil due to the speed your model will be flying at and not because I thought you'd be doing a ton of aerobatics. At higher speeds the lift coefficient becomes so small that even a small amount of camber produces a situation where the model will be dragging the airfoil through the air at a negative angle of attack. Yes, this can and will happen because cambered airfoils produce lift even at small negative angles of attack. Flying under such conditions can often result in unwanted drag.

Now if you could find a near symetrical airfoil that has only a small amount of camber, like around 0.5 to maybe 1%, then you may well gain from this and find that you can fly at high speed cruise with a zero angle of attack and rely on the camber lift to support the model for a slightly lower degree of drag. But such airfoils can become pretty difficult to build to the degree of accuracy they require.

Or I suppose another option would be to use a NACA airfoil generation program to produce a custom airfoil of the exact thickness and camber you want. Enough thickness to meet the scale thickness or 1.25 inches, whichever is greater and maybe .5 to 75% camber.

Certainly using a 2412 will result in the wing flying at a negative angle of attack and higher drag. And I strongly suspect that the 12% thickness will result in much more thickness than what is scale.

You can check the angle of attack vs camber thing for yourself. Set up your "wing" to a rectangular equivalent in span and chord and then set the speed to what you expect for your high speed cruise. Then alter the angle of attack until your wing is lifting by the same amount as you expect your model to weigh. Then play with the camber percentage and alter the angle of attack untl the lift is back to the weight of the model. I think you'll find that you're working with some extremely small angles of attack and that even .5% camber results in a zero or maybe even negative angle of attack. Either way this program is an eye opener and highly educational. Note that you can also choose between lift amount and lift coefficient. Again a way to determine a typical lift coefficient range of operation for a given model.

www.grc.nasa.gov/WWW/K-12/airplane/foil2.html

soarrich

My Feedback: (98)

I agree with what BMatthews wrote which is why I mention the 2412, which has half the camber of the Clark Y, I just don't think at this scale and speed you will be flying at negative angles of attack. The 2412 is zero lift at about -2* so assuming you got to make some lift it will be flying at ~0* to 1* which would make it look correct in the air.

Have you checked what airfoil other jet models this size are using, that might get you close to what you need.

soarrich

My Feedback: (98)

You might email these guys and ask for airfoil info from them. http://www.bvmjets.com/Pages/kits/mig-15site.htm#AD

I check my Nitro Bobcat and it has a 2412ish airfoil, not a symmetrical airfoil.

eddyc

My Feedback: (11)

Michael Selig developed a bunch of airfoils for scaled down models and they are widely used on the Great Planes airplanes. One of his airfoils the S8038 was used by David Ribbe for a Top Gun entered Mig 15 that he built. It is semi symmetrical 12% thick.

soarrich

My Feedback: (98)

That's good info. That airfoil has about 1* camber and zero lift at -1*

Michael knows his stuff, I made airfoils for him for wind tunnel test long ago, FX60-100 & E-214 modified, when he was at Princeton.

Hazykid

Hi Soarrich, BMatthews and Eddyc.

Thanks so much for your input - I have so much more to go on now.

Eddyc - I have the S8038 in the Profili2 database, so it's a definite option for me.
BMatthews - I've been looking at the FoilSim program and I'm really impressed. What is clear to me is that I need to think about this a little more carefully than I had previously thought. You are right that very small angles of attack are needed. It is indeed educational and eye opening - I only wish I had known about it sooner. Many thanks for the link.
Soarrich - thanks for the BVM link. I did know about Bob Violett but hadn't thought about asking them directly for airfoil info. I'll let you know how I get on.

Another question has occured to me whilst I've been thinking about this. What about tail surfaces? Do I stick with the same airfoil for the tailplane? How critical is the fin thickness?

I'm trying to juggle this design with completion of a 1/5 scale Sukhoi 26 and at the moment both are coming second to work commitments. The weekends are very short, alas. I hope I'm not taking on too much for a first time design/scratch build. Hopefully this weekend I'll get a chance to crunch some numbers and come up with a decision. Your excellent advice has really narrowed down the field for me and I am ever so grateful. It's the support of guys like you that make this forum so good.

All the best!

Richard.

BMatthews

I would stick with the scale airfoil thicknesses on the horizontal and vertical tails. The horizontal will likely come out quite thin and that's fine. For the vertical you WANT the thickness from a purely structural standpoint. With the stabilizer being mounted so high on the fin there's a real need for structural strength in the fin. You're going to want to treat this part of the design as if it was a load bearing wing with suitable structure and a definete tie in to the fuselage. A couple of plywood fuselage formers that have vertical stubs that extend up into the fin to tie into the fin spars would be appropriate. You want to keep things light back there but not at the expense of structural strength. Ribs, spars and a structural skin such as 1/64 or maybe 1/32 ply for the lower section of the fin up to the stabilizer mount would not be misplaced. Or if you make it from foam then the stubs should extend up into the foam and be formed so they are at the outer surface so the plywood or balsa skin can bond to them. For a foam fin 1/16 ply for the portion at least up to the stabilizer would be ideal. Light but far stronger than a balsa skin at carrying any side loads you'll encounter during flight or less than perfect landings.

12% seems pretty thick from the looks of the pictures I'm finding. I can't seem to find any good drawings of the Mig so I can't measure anything. But by eye it LOOKS like it's down around 7 to 9% thick.

BMatthews

Rich, for curiosity I inputed a 5 by 1 foot wing into Foilsim and ran it for a 12% thick airfoil with 2% camber. At 100 mph (should be about right for a hotter ducted fan I would think) the angle of attack to generate 7 lbs of lift comes out to -1.64 degrees. If I reduce the camber to 1 degree I hit 7 lbs of lift at 100 mph at around -0.62 angle of attack. If we go for zero camber the resulting angle of attack is a paltry 0.4 positive degrees. Even at 80 mph the two degree camber wing requires a -1.4 angle of attack. A symetrical airfoil at +0.4 angle of attack is going to have less drag than a cambered shape of the same thickness at -0.62 even in the case of the 1% camber.

So all in all this is why I tend to prefer a straight old symetrical airfoil for a fast flying design. Or at least one with minimal camber such that the lower drag "vertical" portion of the Cl vs Cd curve crosses the zero lift coefficient line. From having looked around at a lot of airfoil curves I've found that anything much over 1.5% camber typically won't manage a curve of that sort. For something like a pylon racer where that last little bit counts I'd be all over this second option. But for something like a fast and fun flying Mig 15 just going with a nice simple symetrical airfoil that doesn't come with a critical to produce shape makes more sense to me. And in the case of the Mig I'd suggest using flaps for adding a bit of camber for takeoff and a lot of camber along with a heap of drag for landing.


EDIT- For giggles I set the angle of attack to 0, speed at 80 and added camber until I hit 7 lbs of lift. The resulting "ideal" camber value came out to 0.6%. In light of such a low value it comes down to the designer and builder to determine if it's worth the trouble. If this was a record setting model there may well be some value in it. But for just cutting up the sky to have fun my thoughts are that it's just not worth the fuss.

soarrich

My Feedback: (98)

I think you're right symmetrical may be the way to go on this one.

I like to error on the side of letting the plane fly on the wing rather than the prop. I see so many planes that are just stones because the guy used too of thin or small of a wing. My friend's Clyde couldn't go 100 mph if the wing fell off, it flies at about 40 to 50 mph max the symmetrical airfoil there was such a mistake.

BMatthews

I know that up until a couple of years back when another speed model thread sent me scurrying to Foilsim to see where things fall I would have been firmly in line with your thoughts on using a cambered section. It was quite the eye opener to realize just how little speed is needed to make a symetrical or super low camber section the weapon of choice. These days I tend to spout it off like it's common knowledge to everyone. Sorry if I came across like that.

At least now a few of you know why I LOVE that Foilsim app. Sure it's rather limited but it makes a superb tool that does all the number crunching for us when we want to play with cambers, high speeds, stall speeds and quickly get an idea of lift coefficients to determine just what zone of numbers our models are flying within. I've even gone and found the downloadable version and have it saved on a few computers just in case a NASA budget cutback results in it being yanked from their website.

ARUP

If you understand the concept and definition of 'cambered airfoils' you will find the Clark Y is also undercambered! A fully symmetrical airfoil has zero camber. Plot equidistant points between top and bottom surfaces to visualize the camber line. Hope this helps. ARUP

da Rock

The Clark Y is cambered and most would say that it's not undercambered.

It's bottom line curves down from the LE until it reaches it's lowest point between 20C and 30C and remains at that 0 coordinate to 100C. There really isn't any undercamber down there.

There are many cambered airfoils that have no undercamber. The path the lower line of the airfoil takes determines whether or not there is undercamber, not the path the camber line takes.

So I guess I don't really understand what you're saying.

Himat

Ideal chamber for MIG 15

I have not worked the numbers, but I think there is one consideration left out with the 0.6% camber airfoil. It's based on straight and level flight. The flight pattern of a poweredradio control modell is mostly turning and looping. Apattern and a 3D ship either way up, a sport or scale model probably the up side up most of the time. The turning add load to the wing, the weight is still7lbs, but the force is higher due to the Gloading in a turn/loop. The airfoil might then be selected with a greater camber optimising it for performance in a turn. This would help low speed hanling too, at the expence of some straight ahead speed. I do think I would have traded the last little in top end speed for better turning capability.Probably I would chose something with a 2% camber like theNACA 2412.

Edit:Corrected wording, thank's BMatthews

BMatthews

You're quite right Himat (and a "chamber" is a room or large container. An airfoil has a "camber" line. Just trying to help out for next time.)

What is really needed is to look at airfoils where the Cl vs Cd has a lower drag bucket that crosses the Cl=0 axis of the graph. But choosing one that has too much of a shift to the positive side will generate more drag if loaded in the negative G direction. However if the emphasis will be on turns and positive G maneuvers there's some advantage in looking at a compromise section that still offers minimal drag at a CL=0 point while extending that minimal drag value up to a CL of perhaps .4 to .5.

Either way the camber would still be very low. Perhaps 1 to 1.5% at a maximum. From looking at a lot of lift vs drag graphs airfoils with even 1.5% camber often don't have a minimal drag bucket that extends to CL=0. And even at 1% for what it provides the builder would have to look at it and decide if the gains are worth the effort. The upper end shift of the minimum drag portion may only move up a little.

P-40 DRIVER

NACA 0015 root, NACA 2412 tip with 2 degrees of washout.

vertical grimmace

My Feedback: (1)

Maybe take a look at the Eppler 203. Those worked well for us on heavy 1/12th scale warbirds. Somewhat symmetrical for the first 2/3 of the airfoil but then a cambered drooped tadpole shape. Worked well with a small amount of washout and works well thin.