Outerra forum
User mods, screenshots & videos => Aircraft => Flight Model Development => Topic started by: bomber on March 03, 2015, 11:51:13 am
-
As we all know data is king....
http://en.wikipedia.org/wiki/Cessna_172
http://en.wikipedia.org/wiki/Lycoming_O-360
http://www.swaviator.com/html/issueAM00/basicsAM00.html
http://www.flightglobal.com/airspace/media/cutawayposters/images/14481/cessna-172-cutaway-poster.jpg
http://www.172guide.com/models.htm
the Cessna 172 3 view plans from a small stash we've got squirrelled away
http://atallguy.com/Simple-Multi-View-Plans/
-
Hi Simon,
I'll configure the engine files for the Lycoming O-360.
Regards,
Uriah
-
more data
http://temporal.com.au/c172.pdf
http://m-selig.ae.illinois.edu/ads/aircraft.html
to find airfoil 2412
http://airfoiltools.com/airfoiltools/search/index?m%5Bgrp%5D=naca4d&m%5Bsort%5D=1
to determine Co-efficients of lift....
http://www.mh-aerotools.de/airfoils/javafoil.htm
-
Hey Simon,
Which variant are you going for specifically? I've noticed some fairly substantial differences between them, including powerplant, max gross take off weight, range, service ceiling, etc.
Regards,
Uriah
-
An engine that comes with a static Power/Alt curve to compare against.
-
I head of a tool that can take a 2 image graph and convert it to data, I want to see if I can find something like that so I can convert graphs we find into a spreadsheet and compare them side by side in a single graph.
Regards,
Uriah
-
Ok so the first thing we have to do is determine area's of part of the plane and the distances to these parts from the 'datum' point... this being the pilots position on the centre line of the vehicle.
I've started on this... it's not finished, the area's are not correct... I also need to tell you that I do this in centremetres, for the simple reason that it slows me down and makes me do a cross check.
I use convertpad from googleplay..
https://play.google.com/store/apps/details?id=com.mathpad.mobile.android.wt.unit&hl=en_GB
for anything that needs unit converting.
This is my Autocad12 file, you don't have to use Autocad, it's just a tool for inserting the image of the plane, scaling it, and tracing around it with a poly line to determine areas... If you have another 2d package tool that can do the job then use it, I use Autocad and I have done so since 1984 so I'm familiar with it.
This is the ball aching job, that needs to be done to gather the data to input into the flight model... we'll use the areas and size to determine volume and from a build density value we can then determine the approx weight of each section of the plane... then with the use of a spreadsheet (coming soon) we'll find it's CoG.
Non of what we do to create a flight model requires a 3d model to have even been built, so it allows for parallel working.
right mouse click 'save as'
http://target4today.com/_posted_files_/Cesna_172/Cesna_172.dwg
-
ok download this too....
It's the FDM.... the 'D' can stand for Dynamic or Damage... I like damage as it reminds me constantly of the reason I've done this.
http://target4today.com/_posted_files_/Cesna_172/Flight-Damage-Model.xml
Now it's the front end for the ASK-13 and we're going to modify it to the Cesna 172.
-
Hey Simon,
Sorry, I've been extremely busy finishing a separate project.
I can setup the C127 with my visualization toolkit if you like. From your FDM I should have enough information already to setup the BONES for each center marker and vector you'll need. The script will call the position angle from your FDM properties, and you'll be able to hide all of the visualization elements by setting a variable from 1 to 0 at the top of the script.
The FDM looks really great so far, outstanding work Simon!
I really like what you've done with the mass_balance section. If we added a pointmass for each wing section, and configured damage according to structure point collisions with the ground we could zero the pointmasses, drag, lift, etc for a broken wing tip or whole... For calculating collision damage, I think it would be best to determine the kinetic energy of the pointmass and impact velocity with the ground.
I could even do some modeling and texturing, and create broken sections. If a wing tip were torn off, the script would hide the broken section, as well as the section just below it, and replace that one with a mesh with broken zagged end, and snapped spars, struts and internal structure. I could do the same for almost any failure mode, like broken control surfaces, fuselage, windows, propeller (in sections), etc. Interested?!
Regards,
Uriah
-
Yeh I know busy, they've had me under a train.
In the past I've seen damage added on top of a flight model and it's always come short of the target for me, in feel and granularity. My flight model has damage within it's very DNA and should along with a damage probability engine (to be designed) make the feel of damage satisfactory.
I don't think you need to worry about hitting the ground and kinetic energy, if you hit the ground you break something, normally yourself... been there got the T-shirt, been laughed at in A&E.
Damage animation needs to be discussed with Cameni, because we'll need to know if they're going to give us the capability of alpha texture overlays.
regards
Simon
-
ok I've updated the autocad file a bit and now started work on determining construction density..
http://target4today.com/_posted_files_/Cesna_172/Construction_Density-Cesna_172.xlsx
At this point I've just added in the areas and widths measurements and we're using the construction density from our nearest 'relationship' plane, which is the ASk-13.... that's not going to be as accurate as we'd like but we'll find out how far out we are when we complete the Weight Ballance Spreadsheet..(which I've yet to start on).
http://target4today.com/_posted_files_/Cesna_172/Mass_Ballance-Cesna_172.xlsx
Regards
Simon
-
Updates on all 3 files now...
Dims added, CoG added
http://target4today.com/_posted_files_/Cesna_172/Cesna_172.dwg
Modifed the wing solidity to get to the data we have on wing weight
modified the build ratios between the tail and fuselage to give them the correct weights.
http://target4today.com/_posted_files_/Cesna_172/Construction_Density-Cesna_172.xlsx
Basically this has been completed with the available data and out pops the CoG.
http://target4today.com/_posted_files_/Cesna_172/Mass_Ballance-Cesna_172.xlsx
-------------------------------
http://temporal.com.au/c172.pdf
There are interesting issues thrown up here...
1) empty weight seems to vary with source... wiki, janes etc.
2) weight of a human used is 165lbs, this says no parachute to me as it's usually 201lbs
3) and it does highlight the fact that you can carry two extra passengers, but not far as you can't put much fuel in the plane.
Overall though the CoG is where you'd expect it to be, for a very 'get in and fly it' plane.
-
Updated files
http://target4today.com/_posted_files_/Cesna_172/Cesna_172.dwg
http://target4today.com/_posted_files_/Cesna_172/Construction_Density-Cesna_172.xlsx
http://target4today.com/_posted_files_/Cesna_172/Mass_Ballance-Cesna_172.xlsx
The front end is now complete apart from the forces part which just link to the other files to come..
http://target4today.com/_posted_files_/Cesna_172/Flight-Damage-Model.xml
The one thing I have noticed is the Gross Weight is 2360lbs... If I was to take a punt I'd say that if I was to bring the pilots weight to 165lbs and consider that the engine accessories are probably for an advanced engine configuration as opposed to a 1950's version we'd be on 2210lbs
A 172R's Gross Weight should be 2450lbs, so if I keep it as it is and change the prop weight to constant speed version weighing 71lbs and give the pilots parachutes we're at 2385lbs... So I'm wondering if there's also an increase in fuel tank capacity with version..
But all in all I'm pretty happy we have the variation compromises worked through.
-
Excellent work Simon! :D
Since stability will change with wing loading, and I am interested to see how it handles max weight vs. near-empty weight, or even over-max weight.
It is fairly easy to increase the tank capacity, so I don't think there is a problem with the aircraft's gross weight being slightly off, it can always adjusted. If you are comparing sim data with real-world flight data, than I believe you will need to know the weight regardless, so if it is "off" you can add fuel to bring it up to the specified weight from the real world data.
I found a number of great tools. For converting a picture of a graph into comma seperated X, Y coordinates, http://arohatgi.info/WebPlotDigitizer/ (http://arohatgi.info/WebPlotDigitizer/) is the best for general purposes, but there are others specifically for calculating thrust tables using integrals and extended equations.
Regards,
Uriah
-
Flight modelling for me isn't about just adding weight to make the numbers, if it was then we could just add the number to the weight value and miss out all the point masses.
Of course we'd then have to guess the planes inertia value and its CoG....
If we have the balls to back up our decisions, we need to state the assumptions made and accept the challenges that come from doing so.
The weight of a man, plus parachute should be an agreed known constant.
-
ok so I've used an 11stone bloke with his 7kg 'standard' parachute.
-
http://target4today.com/_posted_files_/Cesna_172/FDM-c172-A.zip
Levi this is the completed flight model, engine and prop... I've added a wing leveller autopilot upon pressing the 'v' key...
Can you add this to your published file magigy..
cheers
Simon
p.s.
Now to add the catapult to the Ask-13
-
Hi Simon,
Sorry for the delay, here is it: c172(New FDM).Levi (http://www.mediafire.com/download/belyv2akq4ekazp/c172(New_FDM).Levi.otx)
-
Outstanding work Simon!
I finally had a chance to test the new FDM and it feels really incredible to fly!!! I received a Thrustmaster Cougar flight stick and HOTAS, so I was able to test the C172 old and new FDM using that, and performed a number of great landings. One problem I noticed is the nose wheel steering seems to pull to the right during taxi/takeoff/landing, but that might be my configuration, I am not sure.
Again, awesome work, keep it up! 8)
Best regards,
Uriah George
-
Yes that'll be the P value being too high I think is at 60, a setting for a Merlin.... Oops...
-
http://target4today.com/_posted_files_/Cesna_172/FDM-c172-A.zip
Levi this is the completed flight model, engine and prop... I've added a wing leveller autopilot upon pressing the 'v' key...
Can you add this to your published file magigy..
cheers
Simon
p.s.
Now to add the catapult to the Ask-13
Isn't the "V" key already used to switch views on Levis planes.....?
-
Yes but the V key, or any key for that matter doesn't have to be exclusive to one task... So what we have here is it turning on the auto pilot wing leveller.
Imagine flying a plane with a rear gunner, you want to jump into the gunners seat take control of the guns and not control the pilots joystick and ensure that the plane doesn't go spiraling into the ground but fly level... That's what we're aiming for here.
We just need to convince cameni that simple is best.
-
Yes but the V key, or any key for that matter doesn't have to be exclusive to one task... So what we have here is it turning on the auto pilot wing leveller.
Imagine flying a plane with a rear gunner, you want to jump into the gunners seat take control of the guns and not control the pilots joystick and ensure that the plane doesn't go spiraling into the ground but fly level... That's what we're aiming for here.
We just need to convince cameni that simple is best.
Well My plane flys straight and level now, ignoring all left/right joystick input, and I have no idea how to de-activate the autopilot. =|
-
press the v key...
-
press the v key...
Which is what I thought. But for me, the plane stays locked and uncontrollable. :-/
-
keyboard or joystick flying ?
..... and you're sure the keys or joystick is configured correctly.
-
keyboard or joystick flying ?
..... and you're sure the keys or joystick is configured correctly.
Joystick. An Xbox pad. The planes control yoke correctly and visibly turns. The plane does not. The keyboard works fine.
-
Hit Esc and go to Controls --> Airplane tab and make sure the Camera view is set to V key.
-
Hit Esc and go to Controls --> Airplane tab and make sure the Camera view is set to V key.
Eh. I decided to punt, and am using the FDM with the enhanced sounds. That seems to have resolved the issue for now.
-
Hit Esc and go to Controls --> Airplane tab and make sure the Camera view is set to V key.
Eh. I decided to punt, and am using the FDM with the enhanced sounds. That seems to have resolved the issue for now.
I have disabled the wing level auto-pilot in the .js of this one (didn't noticed it was possible to set it on/off with v):
//var wing_lvl_ap;
//this.fdm["ap/autopilot-roll-on"]=wing_lvl_ap;
You can remove the // if you want to use the autopilot.