Space station length limit?

[Zosma Procyon]

Is there is practical limit to the length of a space station? I have this idea I want to start building for a very long space station called "Spindle", which would just be a long series of tubular components stuck together with airlocks, and more added whenever a contract for it popped up. Is there a limit to how long it could get before the Kraken strikes?

[Geonovast]

There can be no hard limit for stuff like this.  It's all going to be "It depends."

Depends on your computer, mods, parts used, part count, autostrut options, your computer's mood, other ships in range, etc etc.

 

I have always wanted to know what would happen if you built a ship longer than your physics range, though. 

[bewing]

3 hours ago, Geonovast said:

I have always wanted to know what would happen if you built a ship longer than your physics range, though.

My bet is that you won't be able to dock with it -- because no matter how you approach it, it will get packed and disappear before you get to the end docking port.

[RealKerbal3x]

2 hours ago, bewing said:

My bet is that you won't be able to dock with it -- because no matter how you approach it, it will get packed and disappear before you get to the end docking port.

Why don't we find out? We'll probably need hangar extender, tweakscale and maybe a part welding mod to stop the part count from getting out of hand.

[Gargamel]

13 hours ago, RealKerbal3x said:

Why don't we find out? We'll probably need hangar extender, tweakscale and maybe a part welding mod to stop the part count from getting out of hand.

If we're talking about breaking the physics range with a single vessel, I have a feeling part count will be one of the least of our worries. 

But I bet @bewing is correct.   AFAIK, KSP computes where a ship is based on it's COM.   So if you are 3km away from the COM, but 50m away from where the docking port should be, the station would not be loaded, and there would be nothing to dock to. 

Now, If we do some tricky pumping of fuel back and forth, we might be able to keep the COM near one end of the station, allowing us to keep docking modules onto it.  

But...

I dunno if it's still a thing, but years ago Manley showed in a video how pumping fuel across a large station can actually be used to increase your orbit.  So by pumping large quantities of fuel back and forth to move the COM, we may end up accidentally de-orbiting the dang thing. 

Also, if we get this thing up to a size that is bigger than physics range, say 4km, and then we bring another module up to dock with it, KSP will load the station when you get within 2km (2.5?) of where COM should be, the extra 2km of station are going to go somewhere.   And that somewhere may end up being right through the ship you are trying to dock. 

20 hours ago, Zosma Procyon said:

Is there a limit to how long it could get before the Kraken strikes?

But this is the real question.   To this..... I have no idea.   Probably nowhere near physics range though.  

Edited October 10, 2018 by Gargamel

[bewing]

3 hours ago, Gargamel said:

KSP computes where a ship is based on it's COM.   So if you are 3km away from the COM, but 50m away from where the docking port should be, the station would not be loaded, and there would be nothing to dock to. 

Correctamundo.

And I doubt you can play enough games with fuel shifting to move the CoM more than a third of the way from the geometric center of your station.

Edited October 11, 2018 by bewing

[Geonovast]

Why are we assuming the docking port is as far away from the CoM as possible?  We could split the thing at any point and insert more modules if it was built correctly, unless we're welding as we go.

[bewing]

17 minutes ago, Geonovast said:

Why are we assuming the docking port is as far away from the CoM as possible?  We could split the thing at any point and insert more modules if it was built correctly, unless we're welding as we go.

Once you split it, if the CoM of either of the two new halves is outside the physics bubble, that half will vanish. So that doesn't avoid the problem. Just splits it in two, as it were.

[Geonovast]

1 minute ago, bewing said:

Once you split it, if the CoM of either of the two new halves is outside the physics bubble, that half will vanish. So that doesn't avoid the problem. Just splits it in two, as it were.

It shouldn't be until after the station as a whole is significantly wider than the bubble though, which is the goal here, right?

My computer should be as good as any to try this.  I'll likely give it a go tomorrow and see what happens.  I'm guessing kraken.

[Rocket In My Pocket]

I made a 1000 meter long object once as a test. (Took forever to load due to part count and with no autostuts it was a bit more like a rope than a rod lol. Other than that though it worked fine.)

What's the physics bubble in space? 2.5km? So it'd have to be 2 and half times longer than that. Use long parts lol, just going 1km was pretty rough on the FPS. (534 parts.)

Spoiler

Btw, no I didn't launch it; just Hyper edited it up there to test it.)

 

Edited October 13, 2018 by Rocket In My Pocket

[Geonovast]

I tried two pieces that were well over 1km each, I honestly forgot to look at actual length.  They were only about 15 or so parts each.  I was using some insanely long trusses, I believe from DSEV.  But each piece was built past the camera range using Hangar Extended.

Obviously did not launch those, just F12'd them up for rendevous.

It was taking so long for them to even get in position that honestly I just got bored and quit.  Not because of slow FPS, but they were just very limited on RCS and Wheel authority.... being so long.  Plus I kept losing target lock somehow, and it didn't like to target specific parts for some reason.

[Rocket In My Pocket]

Interestingly enough I didn't need HangerExtended, I just kept sliding it out the SPH door and adding on to it. Although it was starting to get really laggy by the end.

I might try to make it longer at some point, this whole thread has piqued my interest on the subject!