Sign in to follow this  
Guest grb

reducing high end performance without effecting AI landing characteristics

Recommended Posts

Hi all,I have a few military jets of recent vintage that fly rather well as AI, that is, they are Gmax models that do show landing gears extended/retracted, lights etc.. However, they fly a bit to fast when used as AI. I have tried to reduce the Static lbs. Thrust for the engine(s) by increments to slow the aircraft down to make them subsonic, and at the same time not cause them to drop short of the runway during landings. I realize there are limits to this single change, and hope to gain some knowledge as to what main entries in the airfile I can manipulate without screwing things up.Are there a set number of entries I can modify in the air file using AirEdit, to accomplish this goal?Thanks for any help. PS. Using FS2002 simulator.George

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

IIRC you change the cruise speed in the flightplan.txt source file for TTools. Regards,Roman(KGRB)

Share this post


Link to post
Share on other sites

Wooooo! It can't be that simple! This is to good to be true.Thanks! Will give it a shot!regards,George

Share this post


Link to post
Share on other sites

Hi Roman.Your answer to my question about cruise speed."IIRC you change the cruise speed in the flightplan.txt source file for TTools. "OK. I changed the cruise speed to 500knts from the original 740knts.Obviously a significant amount. I also edited the cruise speed number in the aircraft.cfg file to conform to what I put in the traffic tool file (aircraft.txt), but DID NOT run FSEDIT to change it. I am using the TTools that was available prior to the friendly GUI Window version was available but we must remember that Lee Swordly's compiler interprets source files the same way. So I see no difference between editing my aircraft.txt file like:AC#100,500,"name of aircraft_title" verse having the GUI Window package stuff the same value into it's buffer file that would be used for the calculations. Just wanted to make clear what TTools version I am using and how.Bottom line is I still see the aircraft zooming along over 650 knot range at mid attitudes, e.g. 20K on up to lets say 24K. So it appears the answer is not in just changing the value in the TTools files that designates the cruise speed for a given aircraft.I wonder if it is that this aircraft in question (a later well designed F15 for FS2K2 contains an airfile that is just to #### overpowered with unrealistic oswald numbers just for instance, super low drag params etc., so it gains the cruise altitude then just flys to fast for the AI controls to control it. I believe your statement regarding the TTools cruise_speed, working for you but just don't see it happen at my end.And thanks for the info I asked elsewhere concerning where to look to learn how to edit tables using AirEdit. Funny how I missed it for a few years, or forgot I read it three years back, then did not use the tool for so long.At any rate, thanks for the help.George

Share this post


Link to post
Share on other sites

Roman,I am going to take the plunge (off course I will save the original airfile), and using Ron Freimuth's notes I found at the site you indicated, notes in the AirEdit's ini file, and some other stuff to modify the Cdo Drag Coefficient, Delta Cd0(M), Turbine Inlet Area,and one of the Engine Turbine tables (mach - drag).Surely I should be able to get the aircraft to stay around .8 to .9 mach range which is desirable, without effecting the low end and landing characteristics. Obviously I can also lower the SL Static thrust as a first step. I had already done this to one aircraft with a bit of success. Thanks again for your help.George

Share this post


Link to post
Share on other sites

Thanks for the URL link in this forum to some directly related TTools issues (cruise speed). As I mentioned above, I had not only modified the cruise speed number in the aircraft.txt file TTools uses to find available aircraft, but had set the same speed in the aircraft.cfg file.However for all readers. Here is a NOTE from the TTOOL.doc dated Feb/2002:------------------------------------------------------------------AI performance You cannot alter the cruising speeds or the climb and descent profiles of the AI aircraft. The cruise speeds are defined elsewhere (not in the aircraft.cfg files) and AI manoeuvring behaviour is probably hard-coded in the FS2002 software.------------------------------------------------------------------- Clearly. The above note, I cut and pasted it from the doc file, is in clear violation of what those and other posts indicate as to the possibilities of the aircrafts's flight characteristics somehow being changed. And I cannot contrive in my mind why the TTools doc then goes on later, to indicate that one can adjust the cruise in the GUI window area that allows one to choose an aircraft, with the net result that it actually will modify the cruise speed of the aircraft, over a long flight, especially at higher altitudes.This all seems like double talk. Also, the Simulator's AI algorithms must take into account the airfile, and some entries made in the aircraft.cfg file to allow for a given type performance envelope to be actuated (performed) during the AI flight. To try to bring out what I am trying communicate, one example:If changing the cruise speed in the aircraft.txt file of TT, really did effect the cruise speed of the AI, then how would a lets say Boeing 737 AI perform if one was to set the cruise speed at 200knots then have the AI fly at 30K feet. Obviously there would be a problem with the AI trying not to stall.Further. The TT documentation clearly indicates that the cruise speed number provides the program a means to calculate distant/times for arrivals. BUT with the understanding that one must choose appropriate aircraft and times of arrival. Example: One would not choose a Cessna 180 to fly a loop (one hour each way) between two airports having a distance of lets say three hundred miles. Right? And if one did try it, by forcing the AI to fly three times it's normal speed in the TTools GUI, I am sure it would not work. So my analysis is. TTools allows one to make slight adjustments to the cruise speed to help establish a time frame for departure to arrival, with the understanding that the selected AI has the required flight characteristics (ability to cruise at a given speed) to meet the criteria for that given leg of a flight. So in effect, the faster a given AI can fly, the better chance it can meet the demands of the flight plan. That is how I have used the tool over the years to create hundreds of flights for many downloaded aircraft.In my case I am asking the reverse process. How does one slow down a AI's actual cruise speed so that it will not fly supersonic. I think the only way to do this is to modify the aircraft's performance envelope. I say these things with an open mind.best regards,George

Share this post


Link to post
Share on other sites

To anyone that may few this thread.Per my discussions and feedback on this subject of downtuning a typically supersonic AI so that it will fly slower while used in flightplans (TTools). My observations.Using AirEdit, I started to modifiy a few major elements based on notes given by Ron Freimuth, on issues such as drag and cruise speed. His notes lead me to adjust a few major (first order or main parameters).They being:Thrust Static SL (lbs). Engine thrust.Intake Throat Area ..... a key parameter in the engines power curve.And for major elements concerned with drag (parasitic):Cd0 Drag Coefficient "Zero Lift".Delta Cd0(M) Mach Drag vs mach * 2048 Table: 430 where I modified the curve through the low subsonic range to just above mach one area.OK. At first when I just tried to adjust the elements involved with fuselage and wing drag, I noticed no changed regardless of the amounts of significant change to the parameters. SO I decided to run FSEDIT just to see what the Oswald Number (typically seen as a small "e" in equations like the one for Cdi (coefficient of drag)Cdi = Cl / PI * AR * eThe Oswald Number is the Wing Efficiency Factor. And is expressed as a n integer between 0 and 1.OK as usual, when I fire up FSEDIT and load the airfile, I see a totally un-realistic number used to represent the Oswald Number. I see 3.1389999 for instance. TOTALLY INACCURATE AND IMPOSSIBLE FROM THE VIEWPOINT OF REAL WORLD AERODYNAMICS. It must between 0 and 1.0.OK. So I change it to a more realistic number ( 0.9) which would represent a typical high effecient wing design). Safe the file.Fire up the sim, take the plane up to 24KF. All of a sudden, remember I have not changed the TURBINE ENGINE parameters as of yet.The plane will barely go supersonic (Mach > 1 ). Ahaaah. So I then fire up AirEdit and modify the two Engine Turbine parameters mentioned above and also modify the Mach Drag Table a bit more to offer more drag in the subsonic range as it approaches M1.---------------------------------------------------------------Now when I fly the aircraft, it will not exceed Mach 0.95. But it is not quite stable, other characteristics of the airfile will have to be changed to hopefully make it stop porposing up and down, almost like a slow ocsillation in the vertical. But that is not the issue, it is flyable, and on autopilot will maintain it's set altitude with this slight porposing. REMEMBER I WANT TO USE IT AS AI NOT FLY IT.OK. Looks good. The aircraft certainly cannot exceed 575 TAS at this altitude range.So I sit parked at an airport and wait for a couple of these AI to take off and fly to their destinations. RESULT:The #### aircraft still fly around 680KTS when viewing them on the Radar Screen! They are flying about a 100kts faster then I could fly it in the sim. So what is going on? Truly I could not get anywhere this cruise speed while flying the modified aircraft.Something is just plain wrong with the AI algorithms in FS2K2. I have no other way of looking at it. The AI are set in the flight plane to have a quite reasonable flight window. The flights are less then 220 miles from source to destination airport. I made the point of choosing 2 hours so that there would be plenty of time for an average flight well below the airplanes designated speed in the TTools aircraft.txt file (set to 500).I am at the point of giving up on this project. Clearly if one reads what I wrote, giving the fact that I am not fine tuning and realize many parameters could be changed to better improve the stability of the aircraft (not an issue to me, only the cruise speed was), one will see I got the aircraft to fly much slower while actually flying it in the sim, then what it flies at as an AI aircraft. So I am convinced there is much that has never been written about how AI is really handled in FS2K2. If the plane cannot go faster then 560 TAS while flying it at 24KF, how can it possibly go 660 TAS plus when used as a AI at the same Altitude?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this