Minmus Mining Base Woes

[Duke Leto]

Honestly, this is my fault for expecting better from KSP....

So, I'm having problems with my over ambitious surface mining base on Minmus.  The premise is that to eliminate the tedium of having to connect components in orbit and try to land a large and complicated base, or of going through the rigmarole of assembling component parts on the ground on Minmus, I use the Kerbal Attachment System mod to connect the component drilling/ISRU "towers" with a central "command tower" with an engineer.  (There's also another tower with a Sci lab.)  I use flags and MechJeb to guide the component parts (and eventually the fuel freighters) in for a precision landing.

The first problem, which resulted in all of the mining towers being blasted intentionally into a nearby hill as defective, was that the large mining drills, when placed too close to the ground and deployed on a landed craft, cause that craft to playfully shoot up into the air when you switch to it.  Consequently the mining towers had a marked tendency to fall over, breaking their solar panels and radiators.

I addressed that by redesigning the mining towers with the drills moved higher on the chassis.  So what's my problem now?  Well, the goddamned base keeps trying to walk away when I transfer to it or use warp while focused on it.  The individual components move relative to each other and unconnected bits like flags when either happens.  I've illustrated with a red and blue dot where the latest mining tower and the flag where I intended to land the next one were relative to the rest of the base when I left it last.  It's honestly not clear whether the latest tower and flag are moving or the rest of the base is.  The local slope is about 2-3% by the way.

So...  what the hell am I supposed to do here?  I've got one tower in orbit and three more en route, but I don't want to continue the process of setup if the base components are going to randomly slide around and clip into one another during its operational lifetime.

Is there any way to mitigate this, or do I just need to admit that I'm in an abusive relationship with a fundamentally broken game engine and get on with the rest of my life?

[king of nowhere]

i had the same problem with the drill capsizing my mining operation. on mun, even.

what i learned from that is to put 2 drills, one on each side, so their respective pushes cancel each other. you can also use 3 or 4, always symmetrical.

as for the parts moving, i put my base pieces on wheels, with landing struts. with the wheels i would move the piaces from the landing zones to where i would connect them. then i extended the struts. when extended, the struts would lift the base so the wheels would not touch the ground anymore. never had any stability problems after that.

[Duke Leto]

1 hour ago, king of nowhere said:

what i learned from that is to put 2 drills, one on each side, so their respective pushes cancel each other. you can also use 3 or 4, always symmetrical.

as for the parts moving, i put my base pieces on wheels, with landing struts. with the wheels i would move the piaces from the landing zones to where i would connect them. then i extended the struts. when extended, the struts would lift the base so the wheels would not touch the ground anymore. never had any stability problems after that.

The problem that moving the drills higher solved wasn't asymmetric drill placement.  When you switch craft to the base, all the component craft with deployed drills shot directly up into the air by several meters, although it seems towers on the edge were more violently affected.  Since the height of the jumps was different but the towers were connected by fixed length KAS fuel lines, the component craft would get torque in the direction of the static fuel line, and thus fall over.

Every single one of the towers is on landing struts rather than gear.  There might be an issue with them not settling evenly because of the length of the poodle engines.

I think that part of the problem is that the individual craft, although connected by fuel lines instead of docking ports, are considered by the physics engine to be separate craft with separate "orbits", and the calculation of where they are relative to the surface of Minmus is accurate to within "a few meters".  So they tend to wander around.  Kerbal Alarm Clock might be exacerbating the problem. 

This could be interpreted to mean that I need to use docking ports instead of KAS...  But I'm honestly thinking my whole enterprise is insane.  The Unity physics engine is just not going to allow large connected craft to exist while landed.  It's fundamentally buggy.  These bugs can not or will not ever be fixed.  Continuing to try to work around them is foolishness on my part.

Edited July 3, 2020 by Duke Leto

[king of nowhere]

not sure about that. certainly i advise docking with ports. i don't know exactly how big is your mining operation. from the pictures, though, it seems one of your towers is not bigger than one of mine, and i got 3 towers connected.

the problem with bounces is that the game can have small mistakes in accuracy, and those small mistakes can cause a part to be slightly below ground level. this will make the component recoil with a lot of force. my mining operation also tended to jump whenever i selected it, but only slightly. i think being docked does indeed help, because the whole complex bounces as one single unit. but that's about as far as my expertise on the game engine goes, someone more experienced can give better advice here.

[HansAcker]

2 hours ago, Duke Leto said:

I think that part of the problem is that the individual craft, although connected by fuel lines instead of docking ports, are considered by the physics engine to be separate craft with separate "orbits",

All 6 engines show up in the stage list, so it is just one craft. And that might be the problem. KAS lines transfer strange forces between parts in "docked" mode, pushing and pulling the towers around.

I haven't tested it myself, yet, but the Simple Logistics Mod could help you with the operation.

[Entropian]

2 hours ago, Duke Leto said:

I think that part of the problem is that the individual craft, although connected by fuel lines instead of docking ports, are considered by the physics engine to be separate craft with separate "orbits", and the calculation of where they are relative to the surface of Minmus is accurate to within "a few meters".

I had this problem in RSS as well.  Whenever I entered physics range, this happened.

 

[RoninFrog]

Try retracting the landing gear if possible.  I never put anything that isn't supposed to move on anything with springs.

[jimmymcgoochie]

I suggest using Simple Logistics, it allows you to share resources between nearby landed vessels rather than having to dock them to each other. Fuel mining operations are vastly simplified by not having to actually connect everything up and you can land, transfer fuel and take off again in mere seconds. It's easy to use, has a low memory footprint (both hard drive and RAM) and does the job, what more could you want from a mod?

 

 

[magnemoe]

23 hours ago, RoninFrog said:

Try retracting the landing gear if possible.  I never put anything that isn't supposed to move on anything with springs.

This, never use landing legs on springs then you have different modules docked together with KAS pipes, same goes for docking ports or claws on outriggers if you are mining and do time warp or load the base. 
Its less dangerous if you don't time warp, say you refuel an lander, but do save before docking. 
I have stopped using landing legs on anything large, use struts. 

[Duke Leto]

20 hours ago, jimmymcgoochie said:

I suggest using Simple Logistics, it allows you to share resources between nearby landed vessels rather than having to dock them to each other. 

That doesn't solve the underlying problem of the component base parts wandering around relative to one another.

10 hours ago, magnemoe said:

This, never use landing legs on springs then you have different modules docked together with KAS pipes, same goes for docking ports or claws on outriggers if you are mining and do time warp or load the base. 
Its less dangerous if you don't time warp, say you refuel an lander, but do save before docking. 
I have stopped using landing legs on anything large, use struts. 

1. What do you mean "use struts"?  Are you talking about attaching craft to the KAS ground bases or pylons using the autostrut part (which no longer seems to exist in KAS/KIS)?  Because I tried assembling the base in orbit and holding it together with autostruts in a previous playthrough...  And they caused the craft to rip itself apart.

2. If you're trying to convince me that the physics engine is NOT buggy garbage and the game does merit my spending any more time on it, you aren't doing a good job.

[jimmymcgoochie]

Just now, Duke Leto said:

That doesn't solve the underlying problem of the component base parts wandering around relative to one another.

It doesn't solve the problem, true, but only because it bypasses the problem entirely. Unless your base components decide to wander by a couple of kilometres and get out of physics range of each other, and I've never seen that happen before.

[Duke Leto]

38 minutes ago, jimmymcgoochie said:

It doesn't solve the problem, true, but only because it bypasses the problem entirely. Unless your base components decide to wander by a couple of kilometres and get out of physics range of each other, and I've never seen that happen before.

Well, my concerns would then be:

1.) What happens if components manage to clip into one another?

2.) Can Simple Logistics handle unified hotkey presses?  For example, I want to shut down the ISRUs and retract all of the solar panels, radiators and drills when the hulking great freighter lands or takes off...

[jimmymcgoochie]

2 hours ago, Duke Leto said:

Well, my concerns would then be:

1.) What happens if components manage to clip into one another?

2.) Can Simple Logistics handle unified hotkey presses?  For example, I want to shut down the ISRUs and retract all of the solar panels, radiators and drills when the hulking great freighter lands or takes off...

1- the same thing that happens to any two ships that clip into each other: many explosions and flying parts. Park them far away from each other (50 metres or so should do it) and they won’t clip at all.

2- No. Just land far away, it has an operational range of a couple of kilometres so land a couple of hundred metres away and you’ll be fine.

Edited July 5, 2020 by jimmymcgoochie
typo fixed

[Duke Leto]

26 minutes ago, jimmymcgoochie said:

1- the same thing that happens to any two ships that clip into each other: many explosions and flying parts. Park the, far away from each other and they won’t clip at all.

2- no. Just land far away, it has an operational range of a couple of kilometres so land a couple of hundred metres away and you’ll be fine.

Hmmm.  Then the only question that remains is if there's an efficiency loss from running all 3 resource conversion recipes at once on a single ISRU converter, because I currently have the hotkeys set up so that they can switch between the recipes on all the networked converters.  That's something I can work out for myself, though...

 

Thanks.

[Duke Leto]

Yeah, SimpleLogistics is a "no go".  One of the requirements was that the level 3 engineer in the "control tower" applied their efficiency bonus to all of the ISRUs and drills in the entire facility.  I'm sorry if that wasn't clear.

So, I either need to grit my teeth and put up with the wandering mining towers, go back to the drawing board and try and design a system which can accommodate docking assembly on the surface (and which will almost certainly be spontaneously ripped apart by the kraken on being loaded eventually), or admit that KSP hates what I'm trying to do and give up.

[Hohmannson]

20 hours ago, Duke Leto said:

admit that KSP hates what I'm trying to do and give up

Don`t give up pal. Solutions for your problem do exist.

1)Landed vessels will nearly always slide or rotate slowly even on 0deg ideal flats. Thus, you have to hold them. USI Tools or ATUtils, both lightweight, will give you ground anchoring modules, ATGroundAnchor and USI_InertialDampener. This is a toggleable PAW button that still allows vessel to slowly rotate or wobble, but not to slide away. It is not that good for large bases, just a parking brake for landers. To strap the vessel permanently you`ll need a special ground pylon part. Not the KIS/KAS one, it is bugged - a launch clamp-derived one.

2)Surface docking is questionable. If all the base components have no landing legs(important), don`t have drills too low in ground, have flat bottom and are docked together by KAS, supplemented by one of the ground anchor mods above, it can be fine. But the game engine is not meant to work with non-rigid docking. Consider assembling your base into one rigid piece with KIS, in situ, or even building it with something like Extraplanetary Launchpads or Simple Construction. 

3)A bug can be considered a challenge. KIS-attach some lander cans to your rigs, ship in some engineers from Kerbin, Simple Logistics will do the rest.

Oh, about the ground tethers. Install USI_tools plugin(adds no parts) https://github.com/UmbraSpaceIndustries/UmbraSpaceIndustries/releases, create a cfg file with notepad, call it something like tethers.cfg, and put in the following

@PART[*]:HAS[@MODULE[ModuleCommand],!MODULE[USI_InertialDampener]]:FINAL
{
    MODULE
    {
        name = USI_InertialDampener
    }
}

. Then save it. A button will apear on all probe cores and if Ground Tether Attached - it will be needed to push only it once on your KAS-docked mining complex - and hopefully it will be fine.

[vossiewulf]

On 7/5/2020 at 5:58 AM, Duke Leto said:

That doesn't solve the underlying problem of the component base parts wandering around relative to one another.

Duke, I've noticed some landing legs are more prone to walking around than others. My current favorites are legs in the Modular Rocket Systems mod (can get it from CKAN, current for 1.81 at least). The legs have good shock absorbers and that seems to let the vehicles wiggle around a bit without moving their feet. OTOH, all of the SpaceX-type legs I've tried love to take long walks across the surface. Also make sure you're set to the highest level of terrain detail, that helps stabilize things too.

[Duke Leto]

14 hours ago, Hohmannson said:

Don`t give up pal. Solutions for your problem do exist.

Of your suggestions, I like 2) and 1) but not 3).

Assembly of a single rigid craft via KIS sounds the least maddening, but sending a half dozen more engineers to Minmus makes the setup difficult to replicate on other bodies (Ike, Gilly, Moho, Dres, Eeeloo, and one of Jool's smaller moons).  Just so we're clear talking about hauling launch clamps to Minmus to act as anchors for bases or landers.

[Spricigo]

1 hour ago, Duke Leto said:

Of your suggestions, I like 2) and 1) but not 3).

Assembly of a single rigid craft via KIS sounds the least maddening, but sending a half dozen more engineers to Minmus makes the setup difficult to replicate on other bodies (Ike, Gilly, Moho, Dres, Eeeloo, and one of Jool's smaller moons).  Just so we're clear talking about hauling launch clamps to Minmus to act as anchors for bases or landers.

Anyone know if those weldable ports are working? 

 

 

[Nazalassa]

Oh, I have a pretty-foolish idea: make another game like KSP, with space squids and handmade [ game engine (no Unity or that), coordinates representation and vessel landed loading manager ] ...

I have ideas ! Let's do it in C.