D Hillberg
Well-Known Member
Make a model for flight test ... Looks interesting
Help Support Homebuilt Aircraft & Kit Plane Forum:
Joined wing designs... UGH!
Much written about.
First are biplane issues. If the wings are stacked one above the other, you are trying to simultaneously make low pressure above the bottom wing and high pressure below the top wing - in the same space. No problem, put in at least 1 MAC of stagger and 1.5 MAC of wing separation, and things are better. Ugh.
Then we go and join the wings, and now the low pressure along the top of lower wing bleeds off high pressure from the bottom of the top wing. Ugh.
You can do the box wing thing with the upper slid back far enough to be the horizontal tail. Then the low pressure on the top of the bottom wing is the low pressure on the bottom of the horizontal tail. Trouble is you get a lot of wetted area in the join, the horizontal tail is big with a lot of wetted area too, and the tail arm is short, which when it gives you adequate horizontal tail volume, it tends to be short on pitch axis damping, and the large tail with a short arm requires big downward lift and thus a lot of induced drag increase. Then the pressures on the wing and the pressures on the tail become linked, which can make for difficult control. Ugh.
Then as a seaplane, the lower plane is deep in ground effect during takeoff and landing while the upper plane (however it is working) is also in ground effect, but not as deeply. Ugh.
In the end, you want the tail mounted further back and not aerodynamically coupled to the wing pressures. To bring the tail forward and couple things up compromises the heck out of lift, control, and damping. UGH.
Now if you just gotta have it, go for it. And make the lifting surfaces oversized, or you will not get it unstuck...
Billski
Much written about.
First are biplane issues. If the wings are stacked one above the other, you are trying to simultaneously make low pressure above the bottom wing and high pressure below the top wing - in the same space. No problem, put in at least 1 MAC of stagger and 1.5 MAC of wing separation, and things are better. Ugh.
Then we go and join the wings, and now the low pressure along the top of lower wing bleeds off high pressure from the bottom of the top wing. Ugh.
You can do the box wing thing with the upper slid back far enough to be the horizontal tail. Then the low pressure on the top of the bottom wing is the low pressure on the bottom of the horizontal tail. Trouble is you get a lot of wetted area in the join, the horizontal tail is big with a lot of wetted area too, and the tail arm is short, which when it gives you adequate horizontal tail volume, it tends to be short on pitch axis damping, and the large tail with a short arm requires big downward lift and thus a lot of induced drag increase. Then the pressures on the wing and the pressures on the tail become linked, which can make for difficult control. Ugh.
Then as a seaplane, the lower plane is deep in ground effect during takeoff and landing while the upper plane (however it is working) is also in ground effect, but not as deeply. Ugh.
In the end, you want the tail mounted further back and not aerodynamically coupled to the wing pressures. To bring the tail forward and couple things up compromises the heck out of lift, control, and damping. UGH.
Now if you just gotta have it, go for it. And make the lifting surfaces oversized, or you will not get it unstuck...
Billski
This is the absolute perfect case for sub-scale model testing, as has already been mentioned. You will have a fantastic experience building a 1/5 scale model, a fantastic experience troubleshooting all the little issues and getting it to fly. Then the results of the 1/5 scale model will tell you whether your idea was good enough to build a 1/3 scale model. If you can get the 1/3 scale model to fly well, with higher and higher weight, then you would be in the best position to determine if a full-scale prototype is worthwhile.
When building scale models for testing, I am curious how power is scaled.
A 1/5th(0.2) scale model is really 1/125th(0.008) the volume(and if scaled for weight mass as well?), correct? (scale cubed, aka in all 3 dimensions)
So a plane designed for 120hp would need to be limited to less than 1hp for the 1/5th scale model? or are there other aspects to consider as well? (or I am completely wrong....)
curious how that all works together, as I haven't done any modeling or even know what is available for realistic testing purposes (as opposed to just for fun)
A 1/5th(0.2) scale model is really 1/125th(0.008) the volume(and if scaled for weight mass as well?), correct? (scale cubed, aka in all 3 dimensions)
So a plane designed for 120hp would need to be limited to less than 1hp for the 1/5th scale model? or are there other aspects to consider as well? (or I am completely wrong....)
curious how that all works together, as I haven't done any modeling or even know what is available for realistic testing purposes (as opposed to just for fun)
TFF
Well-Known Member
This is for models as physical not models as theoretical. Most models are overpowered. Power is easy. When you build a scale model, you can just about make anything take to the air with power. How is this a help? You can throttle back. You essentially keep throttling back during the tests and at some point you know it’s flying because of lift and not horsepower. If you search far and wide, you can find videos of someone attaching a model engine to a towel and sending it off. You can try all the calculations to match a scaled power but usually it is more complicated. You have to be honest with model building. You can make some one pound piece of foam fly easily. Make a 10-12 pound minimum model fly. You know if it flies or not. 20 lb and you will know if your physics are working. If a 20 lb model flys like a rock, the real one will fly like a bigger rock. Once you get into model planes over 50 lbs, you start getting into lanes being stuck with larger steps between available engines.
About right. A common .40 glow engine is about 1 hp.
Acknowledged that this design demands compromises. Was planning to move the tail further back.
Last edited:
I was planning on a 1/4 scale. for no other reason than I thought it would make the calculations easier. I know this all has to do with Squared Cube Law in some way or another in that the dimensions of a model plane scale perfectly but the weight and horsepower do not. But I'm really glad you brought this up because my intent was to pursue the use of materials that would be a close approximation, at scale, for the real world construction materials. So is it safe to say that an SLA Amphibian that is permitted to weigh 1430 pounds total would translate to a model weighing in at 11.44 pounds?
Last edited:
The very best idea, used by many of the greatest designers in history, is to make small hand-toss gliders first, using sheet balsa. Something with 12 or 14 inches of span. Balsa sheets and lumps of clay. You cannot imagine how much you will learn from just this! Then you can progress to something radio controlled with three feet span, learn everything from that, and then progress to six feet span, learn everything from that, and then finally to nine feet span. This exercise will move you forward faster, better, smarter, safer, and cheaper than any other method. All the computer modeling and CFD in the world cannot give you as good information as this simple and old-fashioned method.
- Joined
- Jul 16, 2019
- Messages
- 186
There's a fellow in Henrietta Okla that built a Prototype of what I see in your picture . It sets on a pedistal on main street . You might call the Henrietta Chamber of Commerce and see if they can get you a picture of it . Don't know the guy personally but a couple of friends and I were returning home from an EAA Chapter meeting and stopped in a small cafe 40 or so years ago and this guy was having coffee late that evening and met him . He drew this craft on a napkin . We didn't take him too seriously but one day when passing through , I saw it there . Can't tell ya if it ever flew or not but was very interesting . He most probably has passed on by now as he was older at that time . It was definately his dream of flying .
Hi Dennis. I attempted a drive down Main Street via Google Maps but I was unable to find said prototype. Do you recall any additional information as to it's location? Was it in town, or down the way. Main street dead ends at one point to it may just be a matter of following it away from town for a while. What do you remember? If it's still there I'd like to see it. Edit. I went ahead and did as you suggested and wrote to the Chamber of Commerce. Hopefully it's still there and they cooperate. Even better if they have the designer's name. If I borrow anything from him I want to give him proper credit.
Last edited:
Sounds like a plan. I'm going to have to be certain about these Cube Square calculations though.
Airplane performance can be pretty well summed up, regardless of size, by looking at loadings and dimensionless coefficients. Loadings:
Power loading - How many pounds is each horsepower hauling around? This relates directly to rate of climb and acceleration;
Wing Loading - How many pounds is each square foot of wing lifting? This relates directly to stall speed and takeoff/landing performance, as most wings have about the same max Coefficient of Lift without flaps;
Span Loading - How many pounds is each foot of wing span lifting? This relates directly to induced drag of the wing, climb rate, and maximum altitude;
Horizontal Tail Volume Coefficient - This is a dimensionless coefficient looking at the ratio of horizontal tail area times tail arm normalized by wing MAC and wing area. This mostly pertains to adequacy of control in the pitch axis, and the range of needed values is determined empirically. The authors on the topic (Pazmany, Roncz, and Thurston are easily found) have tabulated this data and recommended ranges.
Vertical Tail Volume Coefficient - This is a dimensionless coefficient looking at the ratio of horizontal tail area times tail arm normalized by wing area and either MAC (Thurston) or wing span (Pazmany and others). This mostly pertains to adequacy of control in the yaw axis, and the range of needed values is determined empirically, with different values applied depending upon which equation you use. The authors on the topic have tabulated this data and recommended ranges.
Static Margin - this how far the CG is ahead of the Neutral Point normalized by dividing by MAC. This is usually talked about in the pitch axis as this where airplane usually become uncontrollable. It can also be calculated in yaw axis, and can also be problematic. This is why a balsa glide will tumble instead of glide until enough clay is added to the nose.
Reynolds Number - this is the ratio of dynamic forces to viscous forces, and dynamic forces go up with velocity and wing chord. At low Re, like in many RC models, separation of flow and thus stall behaviour can be quite benign, while at Re typical for human rated airplanes, the same foil can have dangerous stall behaviour. The only way to have correct Re in subscale flown in the lower atmosphere is to fly much faster.
Then there are all of the airfoil coefficients. If you do not already have them, get a copy of Theory of Wing Sections and Light Plane Design. Chapters 1,6,7 of TOWS should be read through for content, then you can look at the catalog of airfoils for yourself. ALL of Paz is pertinent.
Now that you know about these things, what are you trying to learn from your model? If you want full size to be as controllable, you need to scale plan view and profile view directly. Stability as good, you need plan view and profile view and static margin to scale directly. Proportions must all hold.
Want to demonstrate stall speeds? Loaded question. If you want the same stall speeds in model and human rated, you need the same wing loading. This is not usually done. You can find stall speed and stall AOA in the model, then using your knowledge of wing loading and dynamic pressure, go to wind tunnel data to estimate stall AOA at the human rated airplane, then compute stall speed in the human rated size.
Takeoff and landing performance? Most models have such short TO and landing runs, they are hard to measure accurately, but maybe they can be useful. Climb out and approach speeds are about 1.3 times stall speed, and runaway performance is just Newtonian mechanics to get between zero and those speeds. Brakes in airplanes are not good for more than about 1/4 g. You may be able to do all of this more accurately with calculations than by flying models...
Top speed and climb rates? Models usually are way fat on power, so this can be tough to really apply too. Calculation are your friend here too.
Have fun.
Billski
Power loading - How many pounds is each horsepower hauling around? This relates directly to rate of climb and acceleration;
Wing Loading - How many pounds is each square foot of wing lifting? This relates directly to stall speed and takeoff/landing performance, as most wings have about the same max Coefficient of Lift without flaps;
Span Loading - How many pounds is each foot of wing span lifting? This relates directly to induced drag of the wing, climb rate, and maximum altitude;
Horizontal Tail Volume Coefficient - This is a dimensionless coefficient looking at the ratio of horizontal tail area times tail arm normalized by wing MAC and wing area. This mostly pertains to adequacy of control in the pitch axis, and the range of needed values is determined empirically. The authors on the topic (Pazmany, Roncz, and Thurston are easily found) have tabulated this data and recommended ranges.
Vertical Tail Volume Coefficient - This is a dimensionless coefficient looking at the ratio of horizontal tail area times tail arm normalized by wing area and either MAC (Thurston) or wing span (Pazmany and others). This mostly pertains to adequacy of control in the yaw axis, and the range of needed values is determined empirically, with different values applied depending upon which equation you use. The authors on the topic have tabulated this data and recommended ranges.
Static Margin - this how far the CG is ahead of the Neutral Point normalized by dividing by MAC. This is usually talked about in the pitch axis as this where airplane usually become uncontrollable. It can also be calculated in yaw axis, and can also be problematic. This is why a balsa glide will tumble instead of glide until enough clay is added to the nose.
Reynolds Number - this is the ratio of dynamic forces to viscous forces, and dynamic forces go up with velocity and wing chord. At low Re, like in many RC models, separation of flow and thus stall behaviour can be quite benign, while at Re typical for human rated airplanes, the same foil can have dangerous stall behaviour. The only way to have correct Re in subscale flown in the lower atmosphere is to fly much faster.
Then there are all of the airfoil coefficients. If you do not already have them, get a copy of Theory of Wing Sections and Light Plane Design. Chapters 1,6,7 of TOWS should be read through for content, then you can look at the catalog of airfoils for yourself. ALL of Paz is pertinent.
Now that you know about these things, what are you trying to learn from your model? If you want full size to be as controllable, you need to scale plan view and profile view directly. Stability as good, you need plan view and profile view and static margin to scale directly. Proportions must all hold.
Want to demonstrate stall speeds? Loaded question. If you want the same stall speeds in model and human rated, you need the same wing loading. This is not usually done. You can find stall speed and stall AOA in the model, then using your knowledge of wing loading and dynamic pressure, go to wind tunnel data to estimate stall AOA at the human rated airplane, then compute stall speed in the human rated size.
Takeoff and landing performance? Most models have such short TO and landing runs, they are hard to measure accurately, but maybe they can be useful. Climb out and approach speeds are about 1.3 times stall speed, and runaway performance is just Newtonian mechanics to get between zero and those speeds. Brakes in airplanes are not good for more than about 1/4 g. You may be able to do all of this more accurately with calculations than by flying models...
Top speed and climb rates? Models usually are way fat on power, so this can be tough to really apply too. Calculation are your friend here too.
Have fun.
Billski
jedi
Well-Known Member
You just described my experience with the Sea-Era (
Paul has a lot of good ideas and did all the model testing VB suggests. He has several follow on design variations he would like to see someone develop and build.
Your configuration is interesting and useful. I know of one low span design study that would allow road transport without folding wings.
Thurston says add the power loading and wing loading. The combined total should be around 25 for a sporty plane or 30 for the average four seater. The Cherokee 140 is 20 solo and 28 at gross.
My JMR has a combined total at gross weight of 22.72
Power loading at GW== 12.9
Wing loading at GW===9.82
Total === 22.72
Loading with a 170lb pilot and full fuel
Power loading === 11.17
Wing loading====8.48
Total === 19.65
Power loading at GW== 12.9
Wing loading at GW===9.82
Total === 22.72
Loading with a 170lb pilot and full fuel
Power loading === 11.17
Wing loading====8.48
Total === 19.65
It has been a long time since I went through his books, where does he say that?
- Joined
- Jul 16, 2019
- Messages
- 186
Hi there bosco . It was setting on the main street , not much more than 3 or 4 blocks off of 75 headed West , on the right hand side right next to the road . I don't think it would have been hard to see from the red light as you turn off of 75 . (if you knew where to look ) but would be hard to miss if on the main street . It was kind of a land mark and I'm sure it got alot of attention . I would almost bet that someone at the Chamber will be able to come up with a name for you . I get up that way regularly and I'll look and ask around . Will let ya know if I tag on to any info .
Thank you for sharing this video about Paul Weston's work. I 've seen a short clip of the "crash" but did not know the name of the plane nor it's creator. It was gratifying to hear Paul say that he had designed the delta hull with the intent of it being a lifting body. That's what I had intended when I came up with my design. Though to be certain, I don't know sucessful I was with that as I was only working intuitively in an attempt to offset energy losses that would otherwise be unproductive drag.
challenger_II
Well-Known Member
On something like this, I would build a 1/12 scale chuck glider, first. Work out the CG, and trimming issues, before moving up in scale (and expense, time-and $$$$-wise). But, that's just me being cheap. 
