HoneyFox

Members
  • Content Count

    938
  • Joined

  • Last visited

Community Reputation

98 Excellent

2 Followers

About HoneyFox

  • Rank
    Toybox Manufacturer

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. 潜艇上浮至通气管深度,伸出通气管,测试通讯桅杆,测试光电桅杆,测试无线电桅杆,测试雷达桅杆。 测试完毕,充气完毕收回通气管,下潜,下潜,下潜。
  2. hi. ah... hmm... i didn't come to this forum for quite a while. Just checked your pm, this plugin is under MIT license just like my other plugins.
  3. Hey! Engine thrust controller works for 1.1.4! The only thing is... I don't know how to edit thrust curves! Please help!

  4. Hi ferram. I've been working on a modified version of FAR which allows player to attach lua scripts (which can be reloaded at runtime so it will be easier to debug) to any control surface so that certain control surfaces can be controlled via cuztomized logic if needed. I know that this might be overcomplicating things a bit but I just want to inform you of it.
  5. I remember that I've seen some code logic which will ignore the ctrl surface if: 1. it's behind some other thing (perhaps by raycast or by voxel checking? i don't remember that clearly) and 2. its deflection angle is not big enough.
  6. As far as i've observed, two things i feel need changes: 1. the drag due to lift seems to be applied twice, first time when applying normal force as lift force for wings, second time when applying drag force using the Physics.cfg's liftDragModifier. An easy example is by pulling your a/c to nearly 90 deg of AoA, you can see by activating the debug option that both the "lift" force (blue line, actually it's normal force instead of lift force i think) and the drag force (red line) contribute to slow down the a/c. That's not quite right. 2. Control surfaces' actuator's angular rates are a bit too low, they might be just ok for some large aircrafts, but definitely not responding fast enough for fighter-class aircrafts.
  7. Well... I added lots of features into my FAR build for my own convenience. and guess it's time to release it to public. ** It's just a replacement pack based on FAR 0.14.6, you should get the original FAR installed, and replace .dll files with the ones provided in my package. ** Features: Some more autopilot helpers: Pitch-Alpha-Controller, a PID controller that tries to control aircraft's AoA based on pilot's pitch input. PAC has a built-in 1G auto-trimmer based on DCA's stdAoA and stdVelocity setting (which consider these two as the parameters of take-off AoA and take-off speed), which is suitable for aircrafts that has zero Cl at zero alpha. PAC will change to Gee-command mode when at high dynamic pressure, limiting G to around -2 ~ 12G. Roll-Yaw-Converter, an Aileron-Rudder-Interconnect implementation, with purpose to minimize sideslip angle during roll and to achieve "roll-around-velocity-vector". Yaw damper changed to Yaw PD controller, which will also try to keep zero sideslip angle instead of just damping yaw rate. Brake Rudder % option for control surfaces, useful for fly-wing aircrafts (best used with Yaw PD Controller and Roll-Yaw-Converter). Local AoA (AoA change due to nearby wings' interaction is also displayed in brackets) information display for all wing parts, especially useful to check the local AoA of control surfaces. Tweaks: DCA now uses Sqrt(scale) so that aircraft will have a bit more control authority at high dynamic pressure. (it used to be too conservative before) Wing interaction of MaxAoA is greatly changed. You no longer can get an aircraft that has a stall angle at ~60 deg, even with leading-edge-flaps. These LEFs can still delay the stall for perhaps several degrees but that's about it. Wing's AC shift due to stall is adjusted a bit. it will still shift afterwards to somewhere around 50% of MAC at 90 deg of AOA. Drag at medium AoA tweaked a bit to make it feel better. (used to be too high I think) Download link: https://kerbalstuff.com/mod/626/FAR%20Modified%20%28Replacement%20Package%29 License still follows FAR's GNU GPL v3. Source code at my github: http://github.com/HoneyFox/Ferram-Aerospace-Research
  8. This is a fly-wing aircraft i designed several days earlier. With brake-rudders and modified FAR Yaw stability enhancer (actually i changed that from a pure Yaw Damper to a PD controller with an AoA-based Roll-to-Yaw command converter which works as ARI IMO), it can be flown with no big trouble unless I pull the AoA too much and stall it.
  9. Perhaps you can try offseting your split-brakes a little bit to left/right, so that it's easier for FAR to judge which side each spoiler belongs to? just my guess...
  10. this is a brilliant application of the free rotating dock-washing disc. I'm using the KerbalWind as well, but mainly for cross-wind take-off/landing challenges. So you found that the aircraft needs to generate more pitch-up torque by these reaction wheels at alpha 45 than at 90 alpha to hold its attitude, right?
  11. I get this curve graph from x-plane's website and it seems like it's a pretty generic one for Cl(green)/Cd(red)/Cm(yellow) with alpha ranging from -180 deg to +180 deg, because we can see the stall angle judging from the Cl curve's peak point. the drag curve looking normal, and Cm curve as well having a sudden change at the stalling alpha. I will post some curve graph in FAR to make some comparison later. EDIT: Here is a sample curve in FAR. There're some differences in Cl & Cd curves, but at least the shapes are still roughly the same. But for Cm curves, I just don't get it. I talked to you on irc earlier about this, and here's it is. Looks like the Cm curve in FAR has followed a complete period of a sine-ish wave from alpha 0 ~ 180 deg, while the Cm curve in the above graph only passes half period in the same alpha range.
  12. You can always watch the Cm curve with Spoiler option on/off to check if the pitch moment variation is minimized if you activate the spoilers.
  13. You can get the acceleration vector from Vessel.acceleration, then use your cockpit's part transform to convert that vector into the cockpit's local coordinate system. Finally process the result vector.