Need for Research - Part 2
Results 1 to 2 of 2

Thread: Need for Research - Part 2

  1. #1

    Need for Research - Part 2

    The first part of this subject was covered many years ago and can be found here:
    http://www.sim-outhouse.com/sohforum...d-for-Research

    It involved the construction of my Corsair model which was one of my first.
    In summary, it was recognizing the mistake of going immediately into the build process and spending too little time in basic research.
    It was a beginner's mistake. My concern at the time was learning Aircraft Factory 99 and my standards for shapes was pretty low.

    It was a mistake I have made a few more times since then.
    Some of the projects turned out reasonably well despite the lack of precision or could be fixed or looked pretty good despite the shape errors.
    Some could not be easily modified and have never been released.

    -------

    The reason for revisiting this subject today is because I seem to be coming across some fairly good designs in which the author clearly has spent a lot of time on the 3D Model, but has apparently not spent enough time getting to know something about the aeroplane they are trying to represent.

    Create a Data Sheet with overall dimensions, an estimate (or better yet, a calculation for the Center of Gravity), weights, armament, etc. I go to some extremes in this data sheet because I use it for a reference, but basic information is all that is necessary.
    It is quite annoying to find a very fine looking model that has a Center of Gravity located so badly that a good flight model cannot be built around it.

    If you are building for CFS, have an idea of the aircraft's armament and ammunition loads and where they are located. This is generally pretty easy information to find. Once the armament is determined, the weight of the ammunition may be a bit more difficult to find. If necessary, use other existing aircraft as reference. DON'T USE STOCK CFS AIRCRAFT! They really didn't get their numbers right.
    Determine the default ammunition load. Many times an aircraft can carry much more ammunition than it does in its "Standard" configuration. US Fighters often carried only a partial load for MGs in standard "Fighter" loadout and didn't carry full MG ammunition until "Overload Fighter". Of course in the field, pilots tended to have as much ammunition loaded as possible.
    You decide.
    Many years ago, whether bombers should carry a default bomb load was discussed. The general consensus was that they should default to NO bomb load and players should be responsible for their own bomb loads.

    Find out where the Fuel Tanks are and their volumes. Being accurate to the tenth of a Gallon isn't necessary, but getting close is necessary if the handling is to resemble the real thing. Fuel is heavy and mislocating it will affect the balance.
    * Here is where things get somewhat interesting. *
    Although the AIR file is thought to represent the offsets in Feet, I have found that that seems to produce balance issues all out of proportion if actual values are used. Stock aircraft seem to use values that are approximately 1/2 of the distance and I find that seems to work pretty well.
    This idea should be tested but I have not thought of an appropriate test that would prove anything conclusively.

    When the project is finished, Spend a little time to try it out yourself.
    Recently, I have seen a very promising looking model that was missing some of its texture files in its archive.
    It was hosted at an obscure site that has not been updated in years and the archive was probably much older.
    I can see no easy means of recovering what is missing.

    I have also seen a few aircraft that could not load into the simulator without instantly exploding.
    One has to wonder if the designer ever tried to load it himself and how he managed it.

    If your beautiful new aeroplane loads and doesn't bounce or wobble all over the place, spend a few minutes looking it over from various angles. You might find something you missed earlier. I know I have found silly mistakes when I thought everything was done. Sometimes the mistakes may just be software goofs and just need a recompile.

    If the aeroplane looks as good as you expect, test it. Check out cockpit views. Approach it as if you were a test pilot with a brand new aeroplane. Try the controls. Try a Take-Off. Try a few aerobatics if it is capable. See if there is any strange behaviour.
    Take it high and try a high speed dive. Stall it. Try a Landing. Keep in mind that most real aeroplanes don't immediately drop like rocks out of the sky when Landing Gear and Flaps are lowered though I have seen plenty of CFS planes that did.
    Try a Belly Landing with Gear Up.

    There are many more tests and evaluations. These are just the very basic ones that do not require any significant knowledge.
    If the author does not do this, it can get pretty embarrassing when the users find something really silly that has been left in place.
    Of course, users don't generally usually comment even when there is a problem though I wish that they did.

    - Ivan.

  2. #2

    The CFS "Test Pilot"

    After all the "Obvious" Checks, Don't be in a hurry to release your brand new project.

    Spend a little time examining it as if it were a new aeroplane project that came from someone else's shop.
    Presumably at this point, there is pretty good confidence that the basic straight line performance numbers are correct.
    Remember though that most of those numbers were the result of flying with the autopilot.

    At this stage, do as much testing as possible WITHOUT the autopilot.
    The autopilot often masks bad behaviour and handling issues that make the new aeroplane unpleasant or impossible to fly under fairly common circumstances.

    The kinds of situations I have personally encountered are the following:
    1. Poor ground handling such as an inability to steer around an airfield.
    2. Poor take-off or landing characteristics such as directional instability, lack of control for a flare at landing, Odd engine settings needed for the final approach. (One should not need to be running 75% throttle or more in a warbird to maintain airspeed on final.)
    3. Aeroplane may constantly bank to one side for no obvious reason. Sometimes this might be a result of engine torque effects but look for weight imbalances if this is happening.
    4. Check for stability at low and high altitude. On one of my early projects, I found that under autopilot there were no problems but without autopilot, the aeroplane was so unstable at about 30,000 feet as to be unflyable.
    5. Confirm that the new aeroplane can be trimmed for straight and level flight. Not all flying is a combat encounter even in CFS. It helps to be able to trim the aeroplane to not require constant attention while flying with your squadron to an objective. As a corollary, confirm that the aeroplane can be trimmed properly for take-off and landing.
    6. Check the agility and maneuverability to see that it meets expectations. Easiest things to check for are turn rates and roll rates. Confirm that there is proper bleed of airspeed in maneuvering.
    7. Confirm that the control effectiveness is not unreasonable for the type of aircraft. It doesn't make sense for a heavy bomber to be able to pull 8G turns.

    There are many more tests that may be reasonable. Hopefully at this stage, there are no major changes that will affect straight line performance that was tested earlier, but that is also quite possible.

    - Ivan.

Members who have read this thread: 0

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •