A silly title, you’d say, but those who know me enough won’t even notice...and for those who don't know me, I'm Stefani Duranti, creator of Profili airfoil handling software. I'd like to describe the latest product of my insanity, which is a brand new version of “Profili”, the software that some of you have probably already used to draft ribs or foam cutting templates. I’ll tell you the story of this program shortly, but first...some background.
In 1997 while I was building a small electric model, I eventually became bored of sharp pencils and graph paper, so I started writing down some code lines in order to draw the wing profile. But I soon fell prey to programming climax, so I created something more than a simple software, capable of reading coordinates and profile printing. As a matter of fact, I started looking for profiles and how to archive them, then I added modifying capabilities and other useful things for the self-made modeler. The program got a good acceptance in my country, Italy, so I rapidly added a “macaroni” English version and Profili 1.0 crossed the national borders!
Since then versions 1.1 and 1.2 followed with some added bells and whistles, but nothing special. In the last 5 years, really, many modelers have called me asking for the addition of new functions such as leading/trailing edges and spars drafting, foam cutting templates and profile retrieving capabilities, allowing for example to locate all profiles with a camber between 2 and 2.5 %, and with a 12 - 13 % thickness in between an archive of more than 1800. And what about polars? Where could one get all those profiles’ polars?
I used to invariably answer, “Yes, yes, it could be useful. Well, I’ll see, perhaps when I have some more spare time, you know...”
But the true problem wasn’t time. It was the fact that all those requests weren’t trivial at all, and after all, what I was deeply missing was the impulse: yes, the impulse this unknown. I remember my physics teacher explaining that the impulse is such that when you stand and lean just a bit forward you may find yourself running for 10 Km !
And so it was, in fact. I leaned myself a bit over the keyboard and now (end of November) I find myself with more than 50,000 code lines from the original 10,000, which could easily be a printed “booklet” of more than 1,000 pages !
Well, you’d say, but what is so interesting about this last Profili version? In a nutshell, I could say: while version 1.x was a simple drafting tool to print airfoils on paper, the new 2.1 version is really something more complete, assisting the modeler in 3 fundamental phases:
1. Research of any airfoils set with geometrically predefined characteristics in an archive of 2,200 profiles. 2. Deep aerodynamic research of the airfoils set, comparing the airfoils in various ways, in order to find the ideal nominee for our future implementation. 3. Last but not least, printing of the rib (or ribs), including, of course, all the necessary structural items such as spars, leading and trailing edges, lightening holes where necessary. Add to this, of course, the foam cutting templates for those using foam for their wings. All above have to be printed on paper or exported to a .DXF file in order to allow full compatibility with any CAD program (those used for computer technical drafting).
"WELL DONE, BRAVO!!" you say? Well, "THANK YOU!"
And now what can I say? Perhaps may I start describing one by one in alphabetical order what Profili 2.1 allows you to do? No, I’ll save you all of this. But...where are you going, folks? I never said it’s all over!
Much more sadistically, I’ll describe to you a real life case, taking as an example what I actually did during the realization of my last depron creature, an electric fun-fly. Well, I agree with you that doing this for a model less than one meter is exactly the same as discussing sex with angels. In fact, for any one meter fun-fly any profile works, but please understand me, that night it was raining and the best TV shows was the Big Brother nomination while my brand new software was ready and waiting for me. And well, I admit it, I’m not a designer of extreme-flights-DS-capable slope soarers! So, you’ll have to settle for this miserable fun-fly of mine....
Let’s start! My basic idea is a simple trapezoidal wing with a very thick symmetrical airfoil, thicker in percentage at the tip than at the root. Converting this into numbers, this means an airfoil with camber = 0 (I’m getting upset with this orthographic aid always changing it into camper!), with 15% thickness at root and 18% at tip. By camber we mean the bending of the airfoil mean line, so in other words, symmetrical profiles won’t have any camber. Now I’ll launch Profili and, ignoring all those 22 welcome messages, suggestions etc., I’ll go straight in and open the airfoils archive with “Airfoils – Airfoils management”. In this way, we’ll see what is shown in Fig. 1, and with “Filter by parameters” we’ll get only 17 of the previous 2,200 and more airfoils. A rather good creaming-off in only a few seconds!
Now I only have to choose which one of these profiles to adopt.
Two possibilities arise:
Let’s see for a while how Profili 2.1 creates and manages all those archived airfoil polars. It has to be said that my program adopts the world-known freeware application Xfoil by Mark Drela. This is a rather complex piece of software which, among many things, allows polar calculation of an airfoil at a certain Reynolds number. What Profili does is to maximally simplify your life in managing Xfoil, which would otherwise request you to open a dedicated DOS window and type in long command lines. Did I already talk about Reynolds number? May I perhaps find him in the phone directory? Who is that man? Let me play teacher for a while. Reynolds number, or for short, Re, is a non-dimensional quantity (i.e. is not measured in Kg or liters or else, but is purely a number), allowing me, in this case, to define the working point of my airfoil. In practice this means that my profile behaves equally both on a big plane flying slowly and in a smaller one going quickly. In other words, we can say that Re express an equivalence in aerodynamic terms. Again, Re very usefully defines the aerodynamic working point of my airfoil.
And now how can I compute Re for my fun-fly?
With the “Tools–Compute Re number” command, I go to the dialog shown in Fig. 3. By inserting the wing data, I get an Re of about 80,000. We can simplify that, and by ignoring the altitude, we can conclude that small Re means slow speed and/or a short wing chord with a general worsening of the profile characteristic. By the way, this is the main reason why big planes fly better. If gnats only knew it … they’d all go on foot!
Those who already know Xfoil also know how long and boring polar calculation is. Profili helps you by already containing pre-computed polar values of all the archived airfoils at 20 different Re, from 30,000 to 500,000, practically covering all modelers’ needs. New profiles or computation at new Re numbers will be calculated on request and then archived for future use.
After this long preamble it’s now time to go and see how Profili introduces these polars graphs. To see a polar, you simply have to select an airfoil and ask for its polars on the dedicated menu. You’ll get the dialog box shown in Fig.4 where we can choose each Re with its color of our choice, then by pressing “Draw” we’ll get the graphs on our video. Fig. 5 shows exactly how Profili works in this case. In practice, each generated polar run is a dedicated document, which in turn is made by 3 sheets, each one selectable by the toolbar push buttons. Fig. 6 and 7 shows the other two sheets.
Let’s talk a bit about the meaning of these polar graphs. Let’s see Fig. 5, which shows the most popular graph all around, i.e. Cl as a function of Cd. A foreword is now in order: all Profili shown polars contain 4 basic parameters:
This is a rather brief smattering but should be enough for our purposes. As usual, purists keep away … I know they could easily criticize my explanation.
Going back to Fig. 5 we’ll see on the horizontal axis (abscissae) the Drag Coefficient (Cd) and in ordinates the Lift Coefficient (Cl). Looking at the graph we’ll immediately notice how the airfoil behaves worse as soon as Re decreases. Taking as reference the zero lift line (Cl = 0), we’ll see that Cd is a tad more than 0.01 at Re = 150,000, while it is three more times at Re = 40,000. The same happens also for other Cl curves. Even the maximum Cl value is much less at low Re. We can realize that a good profile should ideally have the widest possible graph in the vertical sense (high lift) and in the meantime the nearest possible one to the Y-axis (low drag).
Let’s shift to the second graph, as shown in Fig. 6. We have here two graphs, the first one shows the Lift Coefficient as a function of Incidence Angle (Cl(alpha)), while the second shows the Drag Coefficient vs. Alpha (Cd(alpha)). With the first we can easily see how the maximum lift angle (beyond which stall begins) is less at Re = 40,000 than at higher Re, while it creates even less lift. In the other graph we can appreciate how much the airfoil resistance increases at a faster rate at lower Re.
Let’s examine now the third and last page shown in Fig. 7. The first graph shows how the Lift vs. Resistance ratio (the airfoil efficiency) varies as a function of Alpha. The highest point of each curve gives the value and the corresponding incidence angle for the airfoil to realize the top efficiency at that Re number. The second graph shows the Moment Coefficient (how much the airfoil wants to rotate around itself) vs. Alpha. In our case we have a symmetrical profile, then we notice that at zero incidence angle, Cm is zero as well, so that if this were the wing profile, in this condition the tailplane wouldn’t be requested to produce any lift at all. Or in other words, this flight condition doesn’t need any correction by the tail to be maintained.
For our analysis I believe that the most important parameters are those shown in the first three graphs.
Our program allows us to simultaneously open one file for each profile to be analyzed in order for us to correctly evaluate each case.
Let’s now say that we have at last restricted our choice to a lesser number of profiles. Surely now it is difficult to directly compare their characteristics because these are shown on different graphs. No problem.
What we need is simply to ask the program to issue type 2 polars and we’ll see the dialog box in Fig. 8, where, for example, by pressing the “Draw” push button we can try and see what happens at Re = 80,000 (the value adopted in our model). A new document will appear as shown in Fig. 9. It’s now easy to understand that a different color is assigned to each profile for the sake of direct comparison. As before, there are two more pages with the very same rationale behind them; we only have to bear in mind that we are now comparing different airfoils at the same Re number.
The first graph highlights that E475 profile is offering the least Resistance at Zero Lift angle, by losing something regarding the maximum Lift. As a matter of fact, its curve is less wide than the others.
From now on profiles evaluation becomes more and more subjective and we have to choose our profile vs. the desired performance. Let’s say that our choice is Eppler 477 for its performance compromise.
We now have to replicate exactly the same decision path for the tip profile. But let’s also imagine that we weren’t able to find a satisfying profile, even by filtering all profiles with zero camber and 18 % thickness. Yes, we are very hard to please.
Why not then take a profile we like, such as that same Eppler 477, and, by means of the Xfoil airfoils processing capability, “Modify thickness and camber” of the same up to 18%, as shown in Fig. 10. We now ask Profili to compute the new profile polars (this time they will be computed from scratch), in order to verify whether the derived profile is a “kind-hearted” one or not. In actuality, a derived profile does not necessarily maintain the positive performance of its ancestor. We can even open both profile polars at the same time in order to compare them directly. I do believe now that you fully understand how things work. I must confess that I enjoyed very much doing all these tests and comparisons.
And there’s more: let’s imagine that after our long lucubration we ended up adopting Eppler 477 at wing root and a classic NACA 0018 at tip (ah, by the way, we can also generate new NACA profiles belonging to the so named “mathematical” series).
Let’s imagine what happens to our mid-wing profile: this will be a totally new profile being something in between (50:50) the Eppler and the NACA ones. Will this profile work as well as its illustrious parents? Profili 2.1 is capable to treat this case also.
As usual, we have to open the profiles archive and this time we will click on “Generate a mix of two airfoils”. Fig. 11 will open, which shows us how to choose two airfoils and their mixing percentage, in order to get the new profile. In other words we are able to compute the real profile existing at a given station along the wing span. Now let’s baptize the new airfoil, and ask for its immediate polar drawing by clicking on “OK”. Game over. We are now able to verify the half-breed profile performance. Amusing, isn’t it?
Well, having at last decided which profiles to adopt for our wing, it’s now time to decide the building method.
We can draw at once all ribs of a tapered wing, with different root and tip airfoils, with spars, leading edge, sheeting thickness, etc.
Let’ see how: from the menu “Tools – Draw interpolated wing ribs” a window will open as shown in Fig. 12. We simply select the two profiles here, then, after insertion of the relevant chord lengths, we impose a trapezoidal plan (or the elliptical one, if needed) and so on with the number of ribs , sheeting thickness, etc.
But there’s more here. By clicking on “Ribs components management,” a new window opens (Fig. 13), which allows us to choose between different types of leading edges, the trailing edge chord and variously shaped spar types located in different positions at our choice.
Let’s now confirm and, by clicking on “Rib lightening management”, we open another window (Fig. 14), allowing us to define all lightening holes we need. We can also add ‘building tabs’ by the dialog of Fig. 15, they help us to build a well aligned wing.
By double confirmation we finally get all ribs of our wing, as shown on Fig. 16.
For the majority of us who have an A4 or US Letter printer, all larger paper ribs will be printed in two sections, with reference lines to re-construct the rib (Fig. 17). I can also view the pure outer shape of the ribs (Fig. 18) or print the template to check their noses (Fig. 19). That’s all, I only have to decide now whether to print on paper or to export the job in .dxf format in an application like QCad (Fig. 20). If I were a lucky man, I’d even have a CAM software capable to automatically import the whole job and to drive a cutting plotter, but .. OK let me start once again with my trusty jigsaw and sandpaper.
If we want an hot-wire cut foam wing, we need the templates. Let’s start looking in the “Airfoils database” for the desired profile and after choosing our airfoil we have to click on “Begin printing a rib or a template”. As shown on Fig. 21, we have to define the required parameters, after which we click on “Ribs component management”. The next window (Fig. 22) allows us to define the template characteristics, in order to get it as shown on Fig. 23.
We can now visualize and print it both right or left headed.
We will also see the numbered cutting stations, so useful to proceed in parallel with the cutting bow on both sides of the foam block.
I’d say that’s enough, but there still are other functions, such as the possibility to modify the leading edge radius or to alter an airfoil polar by turbulating or flapping it, but if I’d tell you everything right now, you wouldn’t get a kick out of it!
Oh, by the way, someone will ask how my fun fly first flight went. Here is a short description of the test flight : duly charged and topped cells, a sure launch, one tour for trimming all around, and at last it’s time to hover it … oh, no ! I didn’t check my six! A super-hot foamy dives directly on my creature, cutting her in three pieces of equal weight and dimensions. What a sad story....
I see that Profili2 Pro will do elliptical FOAM wings. With only tip/root hotwire templates as far I can see. How does this work? I have hard time picturing it in my head. Is there any bending of the foam blank to achieve this? Has anyone actually made one using this process, and if so how did it come out?
Profili Pro helps to draw special templates to do that.
But the type of airfoil is also important to have a good result.
|Category||Thread||Thread Starter||Forum||Replies||Last Post|
|Question||what news in mpx evo software V.2.0 ?||vidhawas||Radios||18||Aug 02, 2007 07:35 AM|
|Profili & TurboCad 2.0||martin johnson||Electric Plane Talk||3||Feb 14, 2002 12:00 PM|
|Simprop Intella Charger V.2||peterv||Electric Plane Talk||1||Jul 10, 2001 01:07 AM|
|Trouble running profili software on windows 98||Daniel English||Electric Plane Talk||9||May 22, 2001 05:41 PM|
|RcCad v 2.1.4||Sparky Paul||Power Systems||0||May 14, 2001 06:38 PM|