KML VS PEGAS

Ty for the response! I checked the crossfeed and both were enabled, though you are totally right that that is typically the cause - I managed to fix it by running KML and fixing some root issues that staging after docking had caused.

It's called [KML](https://github.com/my-th-os/KML/releases) persistence file editor, but it'll work on any sfs I think. Just be sure to backup what you have before you save over it. When it reads mine, it is chock full of warnings about parts not being fully attached -- I'm guessing this is the morphing/clipping thing that happens with robotic parts. I haven't bothered telling it to fix any of that. In one instance, it found some errors in the file and gave me a chance to fix it, which caused a number of bugs to magically go away (kerbals poofed back into existence and so forth).

KML (Kerbal Markup lister) has a few things, one being 'refill resources'. Handy for testing or cheating I don't judge.

For reference here’s an implementation using a real world rocket algorithm.

PEGAS deserves a mention as a good example of an algorithm used in real space missions re-created in kOS. Of course, that's not for a detailed explanation but it might get some people's interest.

A few years back some one actually did implement a form of PEG in kOS. There repository of code for that project can be found HERE with the kOS subreddit post found HERE.