anopheles

So the procedure for deploying on a planet would be: 1. Land the DIY kit for a vessel with modded parts near a GC assembly plant 2. Bring the Material Kits to assembly plant and build the DIY kit as normal 3. Deploy the vessel, with the modded parts using a "superstructure" mesh 4. Transfer resources to the vessel 5. Click "Go" on the modded parts, swapping them to a "finished" mesh So roughly analagous to using material kits to deploy the inflatable modules in MKS, except a) using different resources and b) not being reversible.

I've recently got really into playing Kerbal with the USI Modular Kolonization Systems mod, and had an idea for a small-scale add-on for that suite of mods. I have some background coding in Unity (although my grasp of C# is still a bit hesitant), and thought this would make a fun project to build my coding skills. I've been reading through a lot of the "how to get started" guides and have a rough idea of how to... well, get started, but before I start barking up the wrong tree, I thought I'd ask here for opinions on how best to approach designing and building the mod. The core idea is to add structural parts that are affixed to the ground, like the ground pylon in the MKS mod (the one that is basically a launch clamp without the decoupler, rather than the KIS ones that are fixed to the ground on EVA), but using an "in-line" model rather than the radial surface-attached model. I know this needs three parts (the surface attachment, the vessel attachment, and a stretchy bit in between) and I think I can probably figure it out by pulling apart the stock models and the cfg file, I just wondered if there was any subtlety to it that I should be aware of before I start noodling with them? For example, would it need any custom .dll files? The other arm of the project was that the structures should be initially deployed (using MKS Ground Construction, i.e. you establish a small workshop, ship out a DIY kit and build the structure in situ) as a part-built superstructure (i.e., all the metal bits), and in order to "finish" the structure, you import a large amount of local raw materials(e.g. ore, silicates, gypsum) and hit deploy. Basically, you're making the large amount of concrete needed to build the structure on site rather than abstracting it as "material kits" or the special parts int the DIY kit). I don't really envision this being extremely complicated, but I might be missing something subtle. The idea would be to have two meshes for the part; when you're designing your base in the VAB you see the "complete" model (with, I expect, a button to preview the "incomplete" model), but when it is "launched" (either via Ground Construction or if for some bizarre reason you rolled it out of the VAB on Kerbin) it defaults to the "incomplete" model. Then, when you've accumulated the materials and hit "deploy", you get a puff of smoke and the model switches to the "complete" model (preferably without any way of toggling back to the incomplete version). So my core question is, what's the best way to implement that kind of behaviour? Does any of that get baked into the .mu file, or should I be building a .dll that gets tied to the part? Thoughts, comments and suggestions appreciated (including "this idea sucks" :P)

Hmm, I posted a reply which appears to have disappeared - or is possibly still awaiting mod approval. Anyway, it's not the throttle - the effect still happens within a closely comparable amount of time, even using below 30% throttle. Also, the engines are a long way apart; there are four of them mounted radially on a 2.5m hull. It's a moot point, as I appear to have identified the culprit - it appears to be something to do with running the ISRU, since disabling the Liquid Fuel production stopped the problem. Not entirely sure why it was having that effect, but there you go.

I've tried lowering the throttle; no joy. I'm currently trying to pulse the engines to make a ~500m/s delta V maneuver; after about a minute of full throttle burn, which consistently gives me 5.3 m/s of delta V before the engines cut down to a fuel usage of 0.00065U, at which point I very briefly put up the timewarp to 1000x to reset the heat of the components, and go around again. When I use 1/3 throttle, I only get to about 7m/s before exactly the same thing happens - not a great deal of advantage, the burns are much longer, and it's actually a lot harder to spot when the engines cut out! As for the ship layout, they're not particularly close together - again, radially mounted around a 2.5m centre, with a little extra because of the shape of the mk2 fuselage. picture here: http://steamcommunity.com/sharedfiles/filedetails/?id=451596981

I'm using LV-N's to move asteroids around. I currently have four radially mounted engines directly mounted on mkII liquid fuel fuselages and some other liquid fuel storage on the vessel; each fuselage also has two gigantor solar arrays for power (and potential helping to radiate heat, I have read?) Frustratingly, as the vessel heats up, the engines abruptly throttle down. Since this seems to occur unevenly, often leaving one engine firing on full, I basically have to turn the engine off and wait for another pulse, or risk spinning the asteroid far out of control. I can't seem to make out which component overheating causes the sudden drop in engine performance (possibly the fuselage temperature?) This is also happening long before the components are reaching dangerous temperatures... Any thoughts or suggestions on how to keep the engines firing?