Why do foamies Deltas are so poorly efficient? - Page 5 - RC Groups
Thread Tools
Apr 20, 2018, 08:58 PM
flyin' fool
goldguy's Avatar
Everything, and I do mean everything, with flying using electrons is a compromise.

A wattmeter used to find the right propeller is a good investment. It can tell what your power system is actually doing and expected flight time.

It's the very first tool you should buy.
Last edited by goldguy; Apr 20, 2018 at 09:05 PM.
Sign up now
to remove ads between posts
Apr 20, 2018, 09:15 PM
Registered User
Don Stackhouse's Avatar
OK, a LOT of misinformation floating around here. Don't feel bad, when I started out in model airplanes (a VERY long time ago), I went through a similar ordeal myself, sorting out all the useful data from all the misinformation. I eventually sorted it out, at least for the most part. Now it's my turn to share, and maybe save you some of that pain.

BTW, "Extreme Sports", although I disagree with some of your comment details, overall I'm impressed with most of your input. And "Whiskers" I've seen your comments before in other forums as well as this one, and they generally tend to be pretty reliable.

Let's go back to fundamentals and see if we can clear up some of this.

First of all, two kinds of drag - induced and parasite (of which "profile" drag is just one kind). Induced drag is the drag that is the by-product of making lift. Parasite drag is any and all drag that is NOT the by-product of making lift.

Let's start with induced drag, beginning with the old "Bernoulli vs Newton" debate. In actuality, there is no conflict between the two. A wing makes lift by grabbing chunks of air and shoving them down, action and reaction (Newton). Bernoulli and all that low pressure on top and high pressure on the bottom simply explains HOW the wing manages to get a grip on something as slippery and insubstantial as air, so it can then shove it down. Bernoulli is just one part of the overall process.

To make a given amount of lift, you can take a small chunk of air and give it a violent shove, or grab a big chunk of air, and give it a gentle push. It should be intuitively obvious which of those two options is more efficient.

So how big are these chunks of air? I've attached a pic from a NASA study showing colored smoke circulating around the wing of a small plane. It shows how the height of the flow field around the wing is roughly equal to the span of the wing. Note, in the picture the center of the tip vortex behind the wing is a bit higher than the wing tip. This is because the plane in the pic is flying in ground effect. In an airplane out of ground effect, the tip vortex would be roughly centered on the wing tip at the tip, and then go downhill behind the wing. The energy required to grab and move all of that air is what the plane sees as induced drag.

The details are more complicated, but you can visualize one of these "chunks of air" as a horizontal cylinder, with its diameter defined by the tips of the wing, and its length equal to the distance the plane flies in one second. The mass of the air inside this cylinder is a representation of one of these chunks of air.

Note, there are two parameters that control the volume inside this cylinder - the span (or more correctly the square of the span), and the length (which is proportional to airspeed) Therefore, if we want to make the size of our chunk of air bigger (and therefore reduce the induced drag), we can either increase the span, or increase the airspeed. Even if you have short, stubby wings, you can get reasonable induced drag if you fly fast enough, making that cylinder long enough (of course then you have to deal with all the parasite drag from all that speed).

Note, nowhere in there did I mention "aspect ratio". Yes, if you have picked a certain amount of wing area, and then increase the aspect ratio, the induced drag will go down. But, that is because when you increased the aspect ratio (while holding area constant), you increased the span. It's the increase in span that actually reduced the induced drag.

Furthermore, it's not just the span that matters, but rather the "effective" span. If you have a bunch of (geometric) span, but it is not used effectively, then the effective span is less than the geometric span, and the induced drag is worse because you are not using your span as effectively as you could.

Ideally, for a given wing span, the lowest induced drag happens if the lift is distributed along the wing in an elliptical pattern. Note, there are various ways to achieve this; you can have an elliptical "span loading" without the wing actually having an elliptical shape. Things like twist (washout) and different airfoils along the span can be used to get an elliptical lift distribution with even a constant-chord "Hershey bar" wing.

However, the problem with deltas is that they are too far from having a decent effective span to be able to help them much with twist changes and airfoil tricks. The root is far too wide, the tips are far too narrow, the aspect ratio is low (so for a given amount of wing area the span is pretty short and stubby), and with those tiny tips you really can't do much work with them, so the effective span tends to be much less than the already horrible geometric span. The net result is you're making your lift by grabbing tiny chunks of air and giving them really, really violent shoves.

Not only that, but a delta tends to have gobs of wing area for the minuscule amount of span it does have. You can't use that area as effectively for making lift, but it's still there, and still making "profile drag" (basically "skin friction"), which is a major form of parasite drag. You end up with lots of induced drag in particular when you are flying slow, near stall or when trying to make a tight turn or a maneuver like a loop,, and at high speed all that skin friction eats up your performance. Yes, it has some advantages for supersonic flight, but not at any of the airspeeds we typically deal with.

Can you make a delta that is capable of flying slow? Of course. How slow you can fly is strongly influenced by wing area. If you make a super-light airplane with ridiculous amounts of wing area, it should be capable of flying slowly. However, unless you use that area effectively and with enough span, the induced drag and the parasite drag will both be high, and the power needed to overcome that drag and sustain flight will be very high. You can make anything fly if you put enough power in it, but the endurance is likely to be very poor (like maybe 8 minutes instead of 22).

Now, yes, you can help the induced drag on a delta by flying faster, making that pathetically skinny cylinder of air longer, giving it more volume. However, now you have the higher parasite drag at that higher airspeed, which increases the required power to sustain that airspeed (power required is proportional to the CUBE of the airspeed). If you want to stay up a long time on a limited size battery, that generally requires keeping the speed down.

So, the bottom line on deltas in our typical model airplane world is if you're trying to be efficient, you "can't get there from here" with a delta. Which is precisely what your experiments so far have taught you. You can take comfort from the fact that aerodynamic theory agrees completely with your experimental results.
Apr 20, 2018, 10:34 PM
Registered User
Don Stackhouse's Avatar
OK, so what about flying wings?

In general there are two types, "plank" and swept. in general, a plank uses reflex in the trailing edge for stability, while a swept flying wing uses twist ("washout").

One (very simplistic) way to look at it is a plank uses about the last 20% of its chord to act as its horizontal tail, while a typical swept flying wing uses the tips as its horizontal tail. In general, a plank needs reflexed airfoils, but a swept flying wing can get away with conventional, non-reflexed airfoils. However, getting good span efficiency from a swept flying wing can be a challenge, while getting decent yaw stability from a plank, as well as a practical C/G range, can be a problem.

It's not terribly difficult to design a flying wing that outperforms a conventional layout at one airspeed. However, achieving that over a reasonably wide range of airspeeds is substantially more difficult, although it is most definitely possible if the designer knows their business and does their homework.

The other thing that typically comes up at this point in the discussion is "BSLD's" (Bell-Shaped Lift Distributions), and the work of Ludwig Prandtl, and the Horten Brothers. Note, there was a "Horton" who experimented with flying wings, but it's the HorTEN brothers whose work is really the foundation of some of the best flying wing technology we have today.

In 1914 the great Ludwig Prandtl figured out that the most efficient way to distribute the lift along the span (i.e.: the lowest induced drag for a given span) was to distribute it in the form of an ellipse. This is the method shown in all the aero text books, and the method used in the vast majority of the planes flying today.

However, Prandtl was not sure if he was asking the right question. Maybe span was not the criterion to optimize around. One of the heaviest items in an airplane's structure is the wing. What if instead of optimizing the use of the span, we optimize for the least wing weight needed to get the job done most effectively?

It took him until 1932, but Prandtl finally figured it out. He noted that the weight of the wing structure was proportional to the bending moment at the wing root, and the lowest wing root bending moment for a given amount of lift occurred if the lift was distributed in the shape of a bell curve, rather than the shape of an ellipse. By using a bell curve, he could make a wing that had 22% longer span but still had the same weight, and its induced drag would be 11% LESS than an elliptical-shaped lift distribution wing ("ESLD") of the same weight.

So, two different criteria, each valid for specific circumstances. If you are designing for a situation where span is limited, for example a model glider for a "2-meter" competition class, you would most likely be better off with an elliptical lift distribution. However, if there is no limit on span, but your structural weight is the main concern, you could be better off with a BSLD.

In addition (and this is where the Horten Brothers, Reimar and his older brother Walter, really become important), using a BSLD can eliminate the need for a vertical tail.

Normally, deflecting an aileron downward on one side increases its drag more than the up-deflected aileron on the other side. This causes "adverse yaw"; if you try to roll to the left using ailerons (or elevons) alone, the plane will also yaw to the right ("adverse yaw"). We normally deal with this by also adding enough rudder deflection to counteract the adverse yaw.

However, with a BSLD, the outer 20% or so of the wing tips are actually making THRUST, not drag. If the ailerons (or elevons) are within that zone near the tip, they make PROVERSE yaw; if you try to roll to the left, the plane will also yaw to the left. If you get the size and location of the ailerons/elevons just right, there will be zero yaw. With no yaw, it may be possible to get by without a vertical tail at all.

This can even work for plank-type flying wings. Some friends of mine recently designed and built an electric-powered plank-type flying wing with a BSLD, and no vertical surfaces at all. The plane flies and handles just fine, with no problems with yaw or yaw stability.

So, as I said earlier, it is possible to design a flying wing for at least certain applications that out-performs a conventional tailed layout. However, the design "homework" required is generally significantly more than for a conventional layout, in part because we have all sorts of "rules of thumb" to make designing a conventional layout easier. However, as with anything in the general subject of airplane design, good things come to those who do their homework. A superior design requires putting in superior effort, regardless of what the plane looks like.

There are a lot more details we can get into, but this will do for a start. As far as a lot more info and pictures of flying wings of all types, pay a visit to the Nurflugel website ("nurflugel" is German for "only wing" or "flying wing") www.nurflugel.com
Apr 20, 2018, 10:44 PM
Registered User
Don Stackhouse's Avatar
Regarding propellers, that subject gets REALLY complicated, but generally speaking, for the highest propeller efficiency you need lots of diameter and LOW RPM, not high. I have designed props for special applications that had efficiencies around 90% or so, but generally speaking the majority of off-the-shelf model airplane props have truly AWFUL efficiency. The very best tend to be around only 60%, and some of them are only about 25-30% efficient.

Part of this is because getting really good efficiency requires closely matching the prop's details to the specific application, operating envelope and mission profile. It also tends to result in high efficiency over only a narrow range of airspeeds and power settings. That level of optimization just isn't practical for something you're going to select off the rack at your local hobby shop. However, for something such as a long-endurance UAV, where prop efficiency has a major impact on the plane's overall performance, especially if the plane spends most of the flight at one airspeed and power setting, designing and building a custom prop for that specific application can be more than worth the trouble.
Apr 21, 2018, 01:01 AM
Registered User
Extreme Sports's Avatar
Thanks Don for getting this thread back on track. The last few posts have certainly helped me put in place a few of the pieces of the aerodynamics puzzle I could not quite get to fit. Very constructive and helpful.
Last edited by Extreme Sports; Apr 21, 2018 at 01:06 AM.
Apr 21, 2018, 03:44 AM
flyin' fool
goldguy's Avatar
In our model airplane world drag is not always bad. We use it to our advantage in some applications, like flying 3D. Just look at all the devices added to create drag, which we then use to our advantage.
Apr 21, 2018, 04:30 AM
flyin' fool
goldguy's Avatar
Not too many designs here on the scratch build forms follow the full sized rules or something like this would not have happened because it would have thought to be impossible. .........................

https://redirect.viglink.com/?format...%20EPP%20Delta

I think the 'chucks of air' our foam wings grab with our blunt LEs and KF airfoils are not 'pushed around' with the same force as with full sized, mostly because of the drag, physical size, weight, wing loadings and speed differences. The only application where I do notice a similarity is with my slopers.
Last edited by goldguy; Apr 21, 2018 at 05:05 AM.
Apr 21, 2018, 09:33 AM
Registered User
Wow, cool stuff there don. This thread turned out to be darn interesting even if it wandered.

I don't see a lot of "misinformation" here, aside from people saying the bernoulli effect is how wings generate lift. Again, just my "opinion". And most people here are probably not "starting out" but started out a long time ago as you described yourself. Most of the stuff you posted I already knew or read about, and those topics are wide-ranging.

As an aside (wish I could find the page, but the poor guy had so much negative feedback from the guys spouting bernoulli stuff he probably took it down)... I remember him concluding that lift is all about pushing air down, and for some reason I think it he said was mostly happening at the trailing edge, but then again he used a standard airfoil clark Y I think, and I have a hard time seeing a flat plate do that.

In my simple minded thinking I see "lift" almost identical to skipping a rock on water, sure air is far less dense, but propellers and momentum take up the slack. Look at 3D planes and people that know how to fly them, I would say half the time the wings are doing nothing, just a propeller being guided by control surfaces.

You'd have to explain to me how the Bernoulli effect helps wings "grab" slippery air, and somehow rectify the basic observation that a flat plate wing, at least in our models, provides lift... I just don't see it, or I don't have enough information.
Last edited by rotagen; Apr 21, 2018 at 10:37 AM.
Apr 21, 2018, 09:40 AM
Registered User
Just one thing on wing weight. I noticed pretty early on how important that was in how much "fun" my planes were. Fun to me is proportional to how floaty and stable they are. Anyways, the ones that were the most fun were the ones where I used a full airfoil with a smooth upper surface made by simply using 1mm depron or even colored packing tape draped over a lightweight longitudinal foam spar. I weighed these wings vs a conventional KF , double-layered foam or even epp wing and they were always lighter, sometimes less than 50% the weight.

Bird wings are way lighter than their bodies.
Apr 21, 2018, 10:33 AM
Registered User
Extreme Sports's Avatar
Don, you may have opened up a can of worms hungry to improve their understanding.

The one thing that stuck out for me in your explanation was that induced drag was independent of chord - that the span and not the aspect ratio was the right parameter. Looking at the formula for the ID co-efficient, I see it has the AR in the denominator, but the formula for the drag itself only has span and speed. So that all makes sense (although the equation also has a factor "L" (lift) and I've not dug deep enough to understand if that has any dependency on chord.)

But two questions immediately jump out, and some internet searching today has not given me the answers:
  1. Why then do gliders have such high AR wings wings? Is it purely a matter of solving for a given wing loading - i.e to maximise the span, you must minimize the chord? I see from Wikipedia that parasitic drag actually goes up slightly with a thinner chord (for a given span), so I guess that rules out minimizing parasitic drag as the reason. Is there another reason I'm missing?
  2. If wingspan and not AR is the major driver of induced drag, and ID is inversely proportional to the square of speed, it would follow (logically?) that our models with their small spans and low speeds (relative to say a B777) would have proportionally massively higher induced drag. Is this true, or are there other factors at play that my afternoon of googling has missed?

Happy to just be pointed in the right direction - no need to waste your time spoon feeding.

Definitely some interesting things coming out of this thread!
Apr 21, 2018, 10:36 AM
Registered User
Extreme Sports's Avatar
Quote:
Originally Posted by rotagen
Just one thing on wing weight. I noticed pretty early on how important that was in how much "fun" my planes were. Fun to me is proportional to how floaty and stable they are. Anyways, the ones that were the most fun were the ones where I used a full airfoil with a smooth upper surface made by simply using 1mm depron or even colored packing tape draped over a lightweight longitudinal foam spar. I weighed these wings vs a conventional KF , double-layered foam or even epp wing and they were always lighter, sometimes less than 50% the weight.

Bird wings are way lighter than their bodies.
I assume that has to do with minimizing angular momentum. A light wing should be more more agile that a heavy one, in the same way that, for a given area, a low aspect ratio wing should be easier to roll than one with a long span. All other factors remaining the same, of course. Probably why we all love our deltas
Apr 21, 2018, 10:58 AM
Registered User
Right, less to move... aside from the obvious, just lowering the total wing loading.
Apr 21, 2018, 11:55 AM
Potato003
My dear friends I still didnt have enought time to read all the valuable comments you added to this thread (I will do that today night!) but I just would like to give you some feedback on my progress into getting the longest flight as possible. So I increased the wing of my plane by A LOT just like the many sailplanes/gliders you told me! With the smaller wing (about 50% smaller than the one in the video below) I could get 22 minutes of flight and today with the big wing I got impressive 34 minutes!!!

See the video of taking off (I know, I know, the place I found to fly is not good to test a "first time wing" but I had to walk around with the plane on my hand for 2 or 3 miles to get to the place at the video below. The two bus drivers that I asked for to get inside with the plane didnt allow me, one of them said I would annoy other passengers with the plane; that's why I couldnt go to the soccer field to test the plane ). I must say I already waterproofed the electronics (even the battery header with nail polish to not damage it if I fall on the water.

https://photos.app.goo.gl/V5NH2JfSY3wYR1EF2

Here is the landing:

https://photos.app.goo.gl/WvPV4NF0K6FqoBcR2

I managed to get incredible 34:50m of flight, almost 35! I am really amazed how well this is compared to the delta. I am also very impressed how a poorly build plane got so much flight time!! I think the gains from here will be marginal cause I dont think it's gonna be easy to improve this time too much more.

I must say that after wathcing lots of videos of sailplanes/gliders that you guys nicely provided me, most of them had the wing angled at the middle (to add dihedral). I never did that cause I only add the dihedral at the tips (which helps with induced drag). But this time I went ahead and experimented with polihedral and the result was impressive! Even with a small rudder I got lots of maneuvabilitty with this plane!

THE delta that I built, with the same motor, same battery and same ESC coudl not even get to 9 minutes! And the delta was weighing about 340g while this plane is wheighing around 390g! In my opinion if you want a long flight it does not worth spending time with deltas.

I think I could have gotten an even longer flight cause I ended the flight with the battery at 3.7v! I usually fly untill it reaches 3.5v but since this battery that I am using was a gift from @balsa or carbon I dont want to have the risk of damaging it or making it lose its capacity, so I am pretty sure I could get more 2 or 3 minutes of flight!

The wing ended up a little big but I will still build an even larger wing!!! I liked it so much! MY MAIN QUESTION REGARDING THIS PLANE IS: by how much could I reduce the chord size in order to get even longer flight? If I keep this wingspan the same and cut the chord by 50%, do you think I could have an even longer flight?
Last edited by batata003; Apr 21, 2018 at 12:03 PM.
Apr 21, 2018, 01:16 PM
Registered User
I too went down the path of trying to achieve long flight times.
By far the most important element is structural efficiency. You are carrying around the air frame all the time so it needs to be as light as possible for any given set of dimensions.
As Dan suggests motor/prop efficiency is vital. Its not the efficiency at full power that matters but rather its efficiency at the power level required to sustain flight.
The power required to fly is the product of the gliding sink rate and the planes weight.
The result?
A relatively broad chord wing to give a good strength to weight ratio.
A minimum fuselage just big enough to house the required components
The lightest possible electronic components.
Just sufficient 'built in' aerodynamic stability to maintain level flight with the minimum of control inputs.
Fly slowly just above the stall speed to give the minimum sink rate.
Name: 18Sep17.JPG
Views: 12
Size: 108.2 KB
Description:
A 'pusher pylon' configuration may be considered rather unlikely for a maximum endurance plane but it does have some advantages.
Endurance development requires a lot of test flying. A fully protected prop does limit repairs.
The low mounted battery (by far the heaviest single component) gives an element of pendulum stability.
The pod and boom layout keeps the fuselage surface area to a minimum. At very low speeds 'skin' drag is a major factor.
A pusher prop creates additional side well area behind the CofG allowing for a smaller fin for the same degree of lateral stability.

With a 5000mAh 2s battery it can sustain level height drawing just under one amp (8W) so it has a potential endurance of 5 hours.
It weighs 20 oz so it flies using just10W/lb yet at full power it has just over 100W available or ten times the minimum cruise power.
Exciting to fly? no. Boring? very.
The longest I have flown it is 2 hours non stop and I really don't want to do it again!
Apr 21, 2018, 01:58 PM
Registered User
Extreme Sports's Avatar
Excellent!! I must say your building skills have come a long way too - that new wing looks quite impressive, considering its trash can origins

Can't add much to what Quoreng has suggested - except, possibly think about a tractor prop. You'll have to decide on whether this is safe where you fly, but looking at how slow this plane flew, and how few people were on the beach, it would be worth considering. Especially if it is on a streamlined pylon.

Just a data point on motors: I've been doing exactly that the last few weeks and I've managed to get a 20% reduction in power at around my cruise thrust by selecting a better quality motor (albeit still a very cheap one) and then rewinding it to add more copper. That should translate into 20% more endurance/ range. Not sure what motor you've got, but 20% would take you to almost 45 mins. I bet you could get to an hour if you really wanted, though as Quoreng pointed out, it won't be quite the same experience as flying your wing.

Good luck, you've come a long way from those first rather disastrous attempts at flying!!


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Why do so many people look down against illegal aliens when they would do the same ? Goonie Works Life, The Universe, and Politics 296 Feb 14, 2018 10:23 PM
Discussion Mark Cuban . so........why do people think he is so smart ? rockyboy2 Life, The Universe, and Politics 11 Dec 27, 2016 01:28 AM
Discussion Why you are poor! kenpoprofessor Life, The Universe, and Politics 20 Jul 14, 2016 10:42 PM
Discussion Why are R/C plane kits made so poorly? Flexserve Electric Plane Talk 50 May 04, 2014 05:53 PM
Help! Why are my AH F18's flying so poorly? aeronaught Micro Ready-to-Fly 6 Jul 15, 2009 09:02 AM