UPDATE: If you are reading this, be sure to review the subsequent blog post.
Hello there esteemed DIYD members.
I've been very quiet lately. But I have not been idle.
Based on the topic title, you might think this belongs in "Simulation," but my target audience is here, in the APM/ArduPlane forum. I'm looking for folks who fly real life planes, so this post is a little misleading.
I wasn't satisfied with using a full scale plane, or even using the RC model shipped with X-plane to predict, simulate, and train for my missions. So I set out to create a reproduction of one of my Bixler-based ArduPlanes in X-Plane 9.70. As you may be aware, there are certain challenges and limitations to doing this, but I made a reasonable degree of headway in overcoming them.
After nearly 80 hours of experimentation, three generations of planes, and uncounted revisions, I have it working realistic (enough) for an alpha release. My intent, in a future release, is to adapt this model to increasingly perform with the same PID settings as the real thing; in this version, I focused more on getting the dimensions, wings, forces, and power of the plane to match the original as much as possible.
iMovie is a dangerous thing; I only intended to patch together a few clips quickly, and instead found the Movie trailer templates. One can create and share something so quickly that your better judgement can be circumvented ... this is a little bit of over-the-top fun, so I hope you will not take the "theme" too seriously.
Here is a slightly longer, but older clip, from a second generation of the plane:
My intent is to release the model for general use, but it is not yet ready for that. It has not yet achieved enough of the goals I have for it. So to begin with, I had in mind perhaps a progressive release, starting with providing the model to anyone who is interested - and is willing to help by providing some data on your own (real) Bixler/EasyStar flights, or to anyone who wants to help tune it up better. This could be followed by a push to rework the model to more closely simulate real life, then making it available to the entire community.
I would be happy to send you a copy of the plane, as it exists today, along with some recommended X-plane settings, and the current PID/config settings, but I would like to "exchange" it for a few of your own files.... I am looking for your data from a Bixler/EasyStar ArduPlane 2.2x flight:
I appreciate any help you might be willing to offer, including feedback on simulated flight performance, how it differs (when flying in manual, in stabilize, or in auto) from your real-life Bixler/EasyStar, any PID tuning you might do so that we might work towards a general release.
PS. A few notes on the specs - Well, I say "reproduction," but it is actually a 1.7 scale (3x in terms of mass) version, which uses a virtual (and painstakingly) replicated APC 10x7E prop in place of the most-often recommended APC 6x4E to achieve proportional lift/thrust ratios. Reynolds numbers are scale-true (but might be reduced to more closely match the 1:1 scale version.) The wings are a very reasonable approximation, for my first X-plane model, considering everything was done within Plane-Maker, and I did not do much customization of the actual airfoil yet.
I am interested in this project of yours and would like to help you refine your model. I will be doing test flights and am willing to send you the data.
Looks very good. This deserves a broader audience than you'll get in a discussion forum post. May I suggest a blog post?
Thank you Chris. That is my intent, but I wanted to put it through at least one round of improvements with a more proactive set first, that being anyone who found the forum post and was motivated to get involved. It is a pretty good plane right now, but it is still a far cry from what I believe it could become, in terms of a predictive tool.
I'm looking forward to sharing data with Robert, but I think if I do not get more replies within the coming week, I might abandon my reserve and try for a (premature) blog post in parallel.
I am also interested. In most of the cases I fly my ES with EasyGlider wings. Is that kind of data of any use for you or is it a MUST to use the ES wings for testing ?
How different are the PIDs which you use, between the two sets of wings?
Irrespective of the answer, I'm very interested in any participation and data I might get, I will simply want to account for the differences since the ultimate objective is to supply a virtual plane which might predict behavior of the actual plane.
Not much of a difference. The ES just flys better, softer with the Glider wings.
If it is important for you, I could use the Original Wings for my next flight.
Ok, you will get the data
Great work! I have a question - what if I make points cloud of my axn floater plane - can it be easily imported into XPlane? Do I need to make many adjustments?
Спосеба бвам! о бворпросе... не панимаю "make points cloud of my axn floater plane".
Б обшом это трудно, б хплане, сделать аэроплане меньшее 5lbs - но может быт помогу?
Hehe! Never expected this language from you =)))
What I meant is ''point cloud' - a 3D model made from photos.
изучал на год. семнадцать лет тому назад. много забыл. ну ладна...
хорошо, понимаю "point cloud" сейчас
фото с есть? from the side, from the top, and from the front? also some measurements, length, weight w/ everything, wing chord @ fuselage, wing length, center of gravity?