By continuing to use this site, you agree to our use of cookies. Find out more
Forum sponsored by:
Forum sponsored by CML

What are the rules?

Plans

All Topics | Latest Posts

Search for:  in Thread Title in  
eflightray21/02/2018 14:44:23
avatar
435 forum posts
97 photos

TN Spitfire 72" span, free plan in RCME a while back, showed balsa build with IC engine and expected flying weight of 15lbs.

I built it in Depron foam, powered with electric, 6s Lipo, flying weight 7Lb - 2oz, includes retracts and flaps. Fully aerobatic.

That's why I don't bother with the maths. It's a model plane, it's built from experience of what works, and for the fun of flying.

Ray. wink

supertigrefan21/02/2018 15:10:59
avatar
119 forum posts
1 photos

Thanks for the response fellas.

Wow! This has developed into something more than I expected but none of it is irrelevant.

I'm now questioning myself as to what I was actually asking for! I guess what I really should have asked was for a 'Recipe' or 'Guidelines' to follow in an order of some kind however the links and discussion has added some meat to it all. I'm not a mathematical bloke, I try to avoid it when possible but I do enjoy applying the formulae to produce something practical, I completely get the TLAR, it's been proved successful over and over again but you only know if it looks right once you've produced a couple of designs.

I think the maths is needed more when you modify a design, producing a scale design is 'simply' (I use that adverb extremely loosely) a matter of re-sizing the plan as the design magic has already been done.

Would it be a fair statement to say that TLAR will produce a working, flying model but the maths will produce a working model that will fly in a scale like manner?

Edited By supertigrefan on 21/02/2018 15:12:41

Edited By supertigrefan on 21/02/2018 15:14:37

Nigel R21/02/2018 15:12:17
avatar
1219 forum posts
275 photos

As David alluded, 'what works' has to be established somehow, and yes, most of us on here know what "about right" looks like.

But, nothing wrong with digging a little deeper.

 

"Would it be a fair statement to say that TLAR will produce a working, flying model but the maths will produce a working model that will fly in a scale like manner?"

At this point in time, with RC aircraft being the mature thing that they are - first part, yes, second part, the maths just helps remove a little guesswork.

 

What sort of recipe are you looking for?

Pete's book linked earlier in the thread is good. Also, Dave Boddington's 'radio control primer' (old, but second hand on ebay or similar) has loads of designs in it, the proportions and structures of which can easily be borrowed and recycled.

 

Edited By Nigel R on 21/02/2018 15:15:53

Denis Watkins21/02/2018 15:30:53
2725 forum posts
137 photos

There are Professors in Aerodynamics on here Nigel, and advisers to the Ministry

I am non of those, and not an academic of Aerodynmics either

But in many years of dipping into books, have never found " just one way " of doing things

And do not even think of propeller design, do not even go there

David Mellor21/02/2018 15:48:27
avatar
673 forum posts
230 photos
Posted by Nigel R on 21/02/2018 14:40:36:

"Surely the 'half the value' must be a continuous function of wingspan,"

My guess is it is a function of the Reynolds the wing flies at. But that is only (indirectly) related to wingspan.

David, I think your theory needs some adjusting...

 

Nigel, its not really a theory (that sounds very grand). It is a rough and ready way of describing one part of the aero-modelling real world without violating the underlying physics too much.

I've tried to avoid tackling Reynolds number head on because it isn't a popular subject. The two main aspects of wings that concern us really are chord and speed - span itself isn't relevant.

You've picked two very different aircraft to compare, not just because they cruise at very significantly different speeds and have very different chords, but also because they cruise at very different altitudes.  This means that the air temperatures and pressure they experience are very different too. Unsurprisingly, therefore, their working Rn ranges are very different. But you can't expect a simple design short cut to apply to everything we do......

Having said that, the rough and ready method will work just fine if you have seen some model that you like and have basic data for (mainly weight and wingspan). And that includes the 777, but you'd have a slightly different correction gradient to that shown in my graph.

Perhaps we should take a moment and step back to see just how breathtakingly massive and beautiful the scale of sizes and shapes is for flying objects in general - including birds, bees, pterosaurs and airplanes.

Between a fruit-fly and an Airbus A380, for example, there are 12 orders of magnitude of mass, yet only 2 orders of magnitude of cruising speed. There are no phenomena in the field of engineering in general (with the possible exception of quantum computing) or elsewhere in nature on earth that span this range. You have to look to cosmology to find such big ranges, and even in cosmology most of the equations are surprisingly simple by comparison with those necessary to tackle flight through the complex fluid we call air.

I mention this just to point out how extraordinarily lucky we are to get stuff to fly at all!

 

 

Edited By David Mellor on 21/02/2018 15:56:16

The Wright Stuff21/02/2018 16:09:02
avatar
1160 forum posts
225 photos
Posted by David Mellor on 21/02/2018 15:48:27:
There are no phenomena in the field of engineering in general (with the possible exception of quantum computing) or elsewhere in nature on earth that span this range.

Electrical resistance is the obvious exception, but yes, I take your point!

Nigel R21/02/2018 16:12:04
avatar
1219 forum posts
275 photos

Sorry, that was a bit naughty of me, I did it deliberately. devil

The rough and ready calculations will work pretty good for anything lightplane/warbird type size and shape. Which probably accounts for at least 50% of what we fly. Anything else needs a bit of wet finger in the air or more complex analysis! As you've stated, the air that a warbird and an airliner fly in, and the speed they fly at, and their chord, are all different. smiley

David Mellor21/02/2018 17:27:49
avatar
673 forum posts
230 photos
Posted by supertigrefan on 21/02/2018 15:10:59:

Thanks for the response fellas.

Wow! This has developed into something more than I expected but none of it is irrelevant.

I'm now questioning myself as to what I was actually asking for! I guess what I really should have asked was for a 'Recipe' or 'Guidelines' to follow in an order of some kind however the links and discussion has added some meat to it all. I'm not a mathematical bloke, I try to avoid it when possible but I do enjoy applying the formulae to produce something practical, I completely get the TLAR, it's been proved successful over and over again but you only know if it looks right once you've produced a couple of designs.

I think the maths is needed more when you modify a design, producing a scale design is 'simply' (I use that adverb extremely loosely) a matter of re-sizing the plan as the design magic has already been done.

Would it be a fair statement to say that TLAR will produce a working, flying model but the maths will produce a working model that will fly in a scale like manner?

 

I thinks the maths is useful for swapping between the variables that you are interested in during the design phase.

It is really only "back of a fag-packet" stuff that takes minutes (at most). My explanations have been wordy only because I know you lose people unless you present the reasoning step-by-step. It isn't that it is "hard" but that the concepts may be unfamiliar.

So the maths enables you to answer "what ifs" such as "if I increase the wingspan by X, what will that do to the weight?" And so on.

So it is very good for basic design from scratch right from the get-go. Its also very good for adapting existing designs to suit you.

Like Ray (eflightray) I too build in Depron because it gives much lower cubic wing loadings. Unlike Ray, who probably has more experience than I have, I run the calcs first. That takes me 5 minutes and saves me hours of faffing about later.

You asked about flying in a scale-like manner. This is actually a very tricky subject because a lot hinges on human perception and the evidence is that not everyone shares the same opinion (or perception) of scale speed. I've seen equations (on t'internet) to mimic scale speed that someone obviously likes, yet which are plainly wrong. So I'm very reluctant to get drawn into much of a discussion on scale speed.

Significantly, very few traders ever quote anything meaningful about scale speed and the only one I can recall seeing is a police radar speed figure and a rescaling calculation on the Mick Reeves scale Spitfire. I've checked his figures and, as far as I can tell, his Spitfire does very accurately fly at what appears to be (on paper) a true scale speed of precisely 25% for a 25% length scale. Significantly (to my mind) the Mick Reeves Spifire is both big and very light with a cubic wing loading of just 8.3 OPCF compared with some Spitfires that are smaller with much higher cubic wingloadings (well over 15 OPCF in one case).  Also significantly, the Mick Reeves Spitfire has a great reputation and has won prizes.  So if you wonder why I'm banging on about getting your cubic wing loading correct at the very start of the design phase, then Mick Reeves's Spit is a shining example (even if he didn't do the calcs first!).

My understanding is that most scale planes fly somewhat faster than scale and that is usually down to weight. Reducing the weight (reducing the cubic wing loading) will allow the plane to fly more slowly if that is the objective - this is an area where Ray's approach pay's dividends because his Depron beauties should appear to fly nice and scale like.

 

 

Edited By David Mellor on 21/02/2018 17:37:08

The Wright Stuff21/02/2018 17:37:11
avatar
1160 forum posts
225 photos

Not sure I entirely agree with your comments about 'scale speed', David. Surely the true linear scale speed is pretty well defined (and proportional to the time it takes any given aircraft to move forward a distance equal to its own length).

Whether it 'looks scale' or not is indeed a human perception matter.

What is clear, that if you want a model to look to scale in all details, fly at a scale speed, AND have scale response to control input (i.e. fly in a scale-like way) then you are very much restricted to 1:1. cheeky

Edited By The Wright Stuff on 21/02/2018 17:38:51

David Mellor21/02/2018 17:51:38
avatar
673 forum posts
230 photos

I think we're saying exactly the same thing so far.

But I think I'll take it a step further now.

The worrying aspect of "perceived" scale speed for me is that......well, it is perceived. Not measured.

True scale linear speed (nice phrase, by the way) can be measured.

So in a competition situation, judges will use their collective perception to evaluate scale speed (and control, as you say), rather than rely on a measurement.  

One consequence is that (unless someone has a radar gun trained on the model) we actually don't know what speed characteristics the judges find so appealing. In simple terms we don't have scaling data for speed, despite having had countless competitions for decades, and this makes it harder for designers to use calcs to improve designs.

Whilst I accept "thats how it is" and that "experience counts" it is counter-productive in a scientific sense because it doesn't give any usable design parameters. It is a "black art" and I don't think that is very good if we want to attract new designers into scale.

 

I think your 1:1 comment is perhaps tongue in cheek, but it does make a good point - it is a tough task to achieve a pleasing scale-like flight.  I think (I know I've already said it) Depron models have some significant advantage here because they can be made structurally stiff and still remain very light.  

Interesting design issues coming up here.

 

 

Edited By David Mellor on 21/02/2018 18:02:30

Jez Saunders21/02/2018 20:24:52
avatar
110 forum posts

Pardon my ignorance but coming at this on different level ( because I can't get my head around some of the mathes) scale speed should be a fairly straight forward conversion , 360 mph would convert to 60 mph for a 1/6 scale model Spitfire for eg. Weight is completely different equation, eg Spit again had a take off weight of around 3000 kg and our 1/6 scale Spit would weigh 500 kg which is obviously wrong but I do understand this because volume is not linear eg a sphere with a radius of 5 units has 8 times the volume at a radius 10 units.

My point is in my mind scale speed should be easy to measure?

Jez Saunders21/02/2018 21:07:37
avatar
110 forum posts

So with the above doubling radius idea increases volume eight fold I arrived at 139 kg for our 1/6 scale spitfire that weighed 3022kg 1:1. This is also far to heavy in the real world so our model aircraft weighing say 12kg has a significant advantage to fly. This has probably already been said but I had to to think about it my way I was bored !☺

Edited By Jez Saunders on 21/02/2018 21:13:55

Mike Blandford21/02/2018 22:04:13
avatar
362 forum posts
14 photos

6 cubed is 216, so volume is reduced by this factor. Assuming the build is using identical materials, 3022kg reduces to 13.9kg (not 139kg).

Mike

Jez Saunders21/02/2018 22:09:18
avatar
110 forum posts

Thanks Mike, I stand corrected, was watching the brit awards while caculating ! So it does work out close to scale ?

John Stainforth22/02/2018 00:11:29
192 forum posts
38 photos

Interesting thread. For those who are not mathematically inclined I have made an empirical log-log plot of aircraft weight versus wingspan for a very wide range of aircraft from the smallest indoor models to the fastest full-scale jet aircraft. The thin lines are lines of equal cubic wing loading in metric units of 2, 4, 8 and 16 gm/cc.aircraftscaling.jpg

The small dots are monoplanes. The ones with the smallest wingspans (between 10 and 20 inches) are indoor model aircraft, and the largest are very large aircraft such as the B747 and An-124 and 125. Some of the full-size planes are highlighted with coloured symbols. The red circles about a third of the way up the plot are my own model aircraft. I was astonished to see that the smallest indoor flyers, and my own model planes, and the largest transport planes all plot around the same cwl line of about 4 gm/cc (let's call this the Boeing 747 line!). For some of the aircraft I have both laden and fully laden data and for these the fully laden cwl's are about twice those of the unladen (8 cf 4).

All these data suggest that cubic wing loading does not have to be adjusted with size of aircraft.

I also plotted data for a variety of birds (open circles) and was surprised to find that nature uses a very wide range of wing loadings. The lightest wing loaded birds are soaring birds with cwl's better than most model gliders. The heaviest wing loaded birds are those that can dive at great speeds (e.g. falcons), which retract their wings to reduce their aspect ratios. These birds have even higher cwl's than racing seaplanes (e.g. S6b), WWII fighters and the earliest jets (e.g. Meteor) . Only the fastest jets have higher cwls. The highest of all are the Starfighter, EE Lightning, B58 Hustler, F4 Phantom and SR71 Blackbird - but those are more or less flying bricks!

John Stainforth22/02/2018 00:30:46
192 forum posts
38 photos

Interesting thread. For those who are not mathematically inclined I have made an empirical log-log plot of aircraft weight versus wingspan for a very wide range of aircraft from the smallest indoor models to the fastest full-scale jet aircraft. The thin lines are lines of equal cubic wing loading in metric units of 2, 4, 8 and 16 gm/cc.aircraftscaling.jpg

The small dots are monoplanes. The ones with the smallest wingspans (between 10 and 20 inches) are indoor model aircraft, and the largest are very large aircraft such as the B747 and An-124 and 125. Some of the full-size planes are highlighted with coloured symbols. The red circles about a third of the way up the plot are my own model aircraft. I was astonished to see that the smallest indoor flyers, and my own model planes, and the largest transport planes all plot around the same cwl line of about 4 gm/cc (let's call this the Boeing 747 line!). For some of the aircraft I have both laden and fully laden data and for these the fully laden cwl's are about twice those of the unladen (8 cf 4).

All these data suggest that cubic wing loading does not have to be adjusted with size of aircraft.

I also plotted data for a variety of birds (open circles) and was surprised to find that nature uses a very wide range of wing loadings. The lightest wing loaded birds are soaring birds with cwl's better than most model gliders. The heaviest wing loaded birds are those that can dive at great speeds (e.g. falcons), which retract their wings to reduce their aspect ratios. These birds have even higher cwl's than racing seaplanes (e.g. S6b), WWII fighters and the earliest jets (e.g. Meteor) . Only the fastest jets have higher cwls. The highest of all are the Starfighter, EE Lightning, B58 Hustler, F4 Phantom and SR71 Blackbird - but those are more or less flying bricks!

David Mellor22/02/2018 10:00:25
avatar
673 forum posts
230 photos

John, what a brilliant piece of work. Great job.

There is so much information here that I'm going to take a while to study it. I'm particularly interested in your comment that cubic wing loading doesn't have to be adjusted with size of aircraft. I think your graph shows that is not actually true, but we can discuss it because it is central to designing model aircraft in particular.

For a plot that spans 9 orders of magnitude (there would be more if you included fruit flies and gnats which also plot neatly on the line) you have done a marvellous job of proving how one simple relationship - the cubic wing loading - unites everything that flies.

Incidentally, we now have a wealth of data for individual birds by species, and they fit the pattern too (as do extinct flying pterosaurs).

The key to understanding why you might want to adjust cubic wing loading when designing a model plane is simple - not everything that flies shares the same task (or mission or flight envelope if you prefer these terms).

Birds in particular show variation in cubic wing loading and the variation accurately mirrors what they "do for a living". So slow soaring birds like vultures and buzzards that feed on the ground have different cubic wing loadings to fast flying swallows and swifts that feed inn the air. The individual cubic wing loading in detail matches the requirements of the thing that is doing the flying. Thats why your central gradient line is flanked by parallel gradient lines of higher and lower values.

What a fantastic job.

When I have a moment I'll post a comparable graph (with even more data points) from published work that shows a similar relationship between weight and cruising speed. You'll see the same cubic relationship and you'll also see similar, small but significant departures from the straight line where task (or mission etc) fine tunes the actual cubic wing loading. The graph I'll show also plots the Spitfire and we have been talking about this as a scale model example - so we will be able to deduce the performance envelope for the actual Spitfire from the graph and infer why we should adjust the cubic wing loading for a real scale model (as we have proven everyone actually does).

So - it is all coming together beautifully!

Steve Dunne22/02/2018 10:31:53
avatar
43 forum posts
9 photos

Hello John,

Fascinating and very informative graph and explanation!

Just to clear my confusion, are the cubic units ranges metric units of 2, 4, 8 and 16 gm/cc, or imperial units of 2, 4, 8 and 16 oz/cu foot, as discussed on previous posts?

Thanks,

Steve.

David Mellor22/02/2018 11:18:28
avatar
673 forum posts
230 photos

It occurred to me that some people might appreciate a very simple example to illustrate why the cubic wing loading should be varied to match the task the plane is being designed to achieve.

So here it is.

I slope soar. I fly one glider more than any other. And that is because I have designed it to accommodate lead ballast. The lead ballast adds weight (that is what ballast is, of course) and this increases the cubic wing loading.

Without any ballast added, the glider will fly in light conditions (low lift, 10 mph). But I live in the Lake District and we get howling gales of 35 mph plus very regularly. So the same glider with a pound of lead in it's belly flies superbly in these strong conditions.

Physically, it is one aircraft which can vary it's cubic wing loading to match the flight task it is assigned to.

Exactly the same thing happens in migratory birds. Migratory birds eat as much as they can to store energy for the journey and their individual cubic wing loadings are maximised at take off. As they fly they gradually burn up energy and, having no opportunity to re-fuel in midair, they gradually lose weight. This means their cubic wing loading reduces as a function of time through the migratory flight. It actually makes the task easier (because it lowers the cubic wing loading in flight).

Evolution over millions of years has equipped some migratory birds to re-absorb energy from some of their own organs - they literally "burn their own engine" for fuel. The Red Knott species does this and has been extensively studied. It burns up muscle mass and shrinks it's digestive organs to supply energy to the (shrinking) flight muscles during long migration. When it finally lands, it's cubic wing loading is dramatically lower than it's take-off wing loading (days earlier). But this is what it is "designed" to do by evolution.

eflightray22/02/2018 11:28:09
avatar
435 forum posts
97 photos

Scale speed to me is what you see when comparing a scale model to the full size in videos.

Many videos of full size aircraft from air shows, (we perhaps fly a similar routine to air shows with models, by keeping them within a restricted area), will often show the full size to appear considerably slower than similar scale model flight.

People can argue all they like about what is 'scale', but again, to me, it's all about fooling the eye of the observer.

Models flying, whether scale of not, are generally quickly perceived to be models by passersby. I doubt many will assume it is actually a full size aircraft. To many modelers, 'scale' seems to be more about what the model looks like on the ground, I build to fly in a 'scale like' manner, it's what I get pleasure from.

I wont mention, 'scale sound' wink

Ray.

All Topics | Latest Posts

Please login to post a reply.

Magazine Locator

Want the latest issue of RCM&E? Use our magazine locator link to find your nearest stockist!

Find RCM&E! 

Email News - Join our newsletter

Love Model Aircraft? Sign up to our emails for the latest news and special offers!

Latest Forum Posts
Support Our Partners
electricwingman 2017
Slec
Overlander
Gliders Distribution
Wings & Wheels 2018
Expo Tools 14 July
Airtek Hobbies
CML
Advertise With Us
Sarik
Latest "For Sale" Ads
What is the main brand of transmitter you use? (2018)
Q: What is the main brand of transmitter you use?

 FrSky
 Futaba
 Graupner
 HiTec
 Jeti
 JR
 Multiplex
 Spektrum
 Other

Latest Reviews
Digital Back Issues

RCM&E Digital Back Issues

Contact us

Contact us