search

CobaltWolf / Bluedog-Design-Bureau

Stockalike parts pack for Kerbal Space Program
https://forum.kerbalspaceprogram.com/index.php?/topic/122020-131mostly-functional-141-bluedog-design-bureau-stockalike-saturn-apollo-and-more-v142-%D0%B0%D1%82%D0%BB%D0%B0%D1%81-1feb2018/
126 stars 144 forks source link

Request: FAR-friendly (split wing and strongback) Pegasus wings #1506

Open JacobB094 opened 3 months ago

JacobB094 commented 3 months ago

Currently, the Pegasus wing is a single part, which FAR doesn't really know what to do with. Could this be split into a two parts, working like X-15 Delta wings? Specifically, the strongback as a separate part, and both wings as node attached parts with switcher for selecting the side. This is necessary for making the Pegasus flyable in FAR.

If that's done, I'll add it to my FAR patch. I'd recommend hiding the non-FAR strongback when FAR is installed, as it can't really be made to work (a straight wing could perhaps be, but not a swept one, and the sweep is important here). I can also add that, and any COM shift that the first stage might need to actually fly, as well.

Rodg88 commented 3 months ago

I've set up a FAR patch, (before I saw your PR), but it seems to be fairly flyable. It behaves fairly close to using two procedural wings in place of the strongback, so I don't think it needs to be split up.

I did notice your patch was just using the default transformName, but the pegasus fin has a different control surface transform name, so would have been preventing them from working. I also just slightly increased the size of the reference fin slightly to boost control authority too.

LMK how it seems with these updates

JacobB094 commented 3 months ago

I'll check how Pegasus flies with that, however be aware that with your config, you're not actually flying with a delta wing that looks like a Pegasus one. What it produces, as far as FAR is concerned, is a slightly oblique, asymmetrical wing attached in the middle. It's not that such wings don't work IRL, and it might work for the Pegasus in normal use, but this sort of trickery might come back to bite us in some unforeseen edge case.

Rodg88 commented 3 months ago

Unforseen edge cases with FAR are kinda beyond scope for this/BDB imo, but since it'd require new modelling and texturing it's really up to Cobalt.

JacobB094 commented 3 months ago

Turns out it's not just edge cases, in my testing it doesn't fly well at all. Using a 11400m and M=0.87 level release (straight from the PPG), the current configuration can't bring its nose up. This is borne out by FAR analysis that shows, in short, that COL is too far aft relative to the COM in these conditions. Fiddling with offsets is probably an option, but I'd like to try a wing that's properly set up. screenshot392 screenshot393 This is the difference between the current config and how it should be. COM likely needs to be shifted a bit aft in either case. It's not a problem with fin authority, you could probably revert to my values for the fins and add a degree or two of deflection instead (I'd also trim default roll authority).

FYI, be careful with comparisons with B9 procedural wings. They enable you to create a wing that will have FAR properties (especially mid chord sweep) completely out of whack with its appearance. The wing needs to be carefully set up to actually be useful measurements. Also, do make sure to configure FAR analysis use actual release conditions and not "Cessna on takeoff" defaults. This actually does influence the COL display.

For what it's worth, when a wing's FAR parameters are set up to match its appearance, it'll act like it should, edge cases or not. If you try to hack it instead, it might work in a narrow regime, but will more likely than not break down outside of it. It might be as simple as using the Pegasus as a ground launched X-15 booster instead of air launched light LV. With proper FAR setup, neither should have problems (except for realistic ones, but these have logical solutions).