build a station low or high

[steuben]

I've got a couple of rather large stations on the drawing board. The dry weights are in the  mid two digit kilotonne range, roughly 20 odd modules ranging from 750 to 35 tonnes. Getting the modules to orbit isn't an issue, either 100 km or the operational 1000km.

For actual construction I have two options:
1. build the station a low orbit, say 100 km and then haul the whole thing up to a higher orbit with a high trust tug
2. get the modules to the operational orbit, either with a direct launch or a set of smaller tugs, and assemble it there.

With option 1 I can get it built with fewer and/or heavier launches for the modules. But I'm going to be hauling around a net full of jello to get it to the destination.

With option 2 I won't need the heavy tug but I'm looking at more and smaller launches to get everything up there. Or a similar number of heavy launches and a set of small tugs to get it all to orbit.

Any suggestions/recommendations? A third option that I haven't thought of?

 

[Sir_Robert]

Unless your station is exactly symmetrical (which mine never are), you'd probably be better of assembling it and at it's destination

[Tex_NL]

I have to agree with @Sir_Robert moving a station can be hazardous. Not only will you most likely have to deal with an asymmetrical weight distribution, most stations are also a lot more flexible.
And launching to 1000km compared to 100km won't take that much extra fuel. I'd go for the direct build at 1000km.

[Snark]

Agree with the previous posters.

That said:  Moving a jello-ship isn't too bad, if it's symmetric... as long as your tug is designed to pull the ship rather than push it.  Just mount the engines off to the side on long booms or something, and you're all set for your tractor-pull.

If you're willing to use mods, even towing an asymmetric ship isn't too bad-- you can use Kerbal Attachment System to tow the station at the end of a cable.  This allows the station to find its equilibrium orientation even if it's asymmetric.  Steering will be a royal pain, though... it'll go flopping all around if you have to change directions, so you'll need patience. 

But yeah, assembly in-place is probably simplest.

(Is it wrong that somehow I can't help hearing the subject line of this thread as a square-dance call?  "Build a station low or high, / Satellites go flying by, / Grab your partner, spin 'em round, / Aerobrake, we all go down...")

Edited February 18, 2016 by Snark

[Padishar]

Another note of general agreement, assembly at 1000km is probably a lot less hassle overall but if you decide to build it in low orbit then I would definitely recommend going above 160km (I believe that's the cutoff point where rendering of kerbin gets a lot faster) as you should get significantly better framerates.  However, you should still expect horrible performance when trying to dock high part count vessels together.

If you do build and then move then pulling is definitely better for wobbly vessels though it may complicate the docking, e.g. if the port you tow by is the port you need to dock to the station.

1 hour ago, steuben said:

A third option that I haven't thought of?

I don't think there's any other sensible option except for Hyperedit (or a simple edit of the orbit in the persistence file)...

[steuben]

36 minutes ago, Snark said:

(Is it wrong that somehow I can't help hearing the subject line of this thread as a square-dance call?  "Build a station low or high, / Satellites go flying by, / Grab your partner, spin 'em round, / Aerobrake, we all go down...")

I keep hearing classic southern gospel on the first half of it.

 

The tugs were always going to a tractor design. The stations are at least 99% mass symmetrical.  The modules are designed with docking ports on both ends to allow for attachments.

I'll have to see if KVV can spit out a pic of the two of them.

Edited February 18, 2016 by steuben

[wibou7]

Am I the only one impressed by the sheer hugeness of that station project?

3 hours ago, steuben said:

I've got a couple of rather large stations on the drawing board. The dry weights are in the  mid two digit kilotonne range, roughly 20 odd modules ranging from 750 to 35 tonnes.

In the mid two digit kilotons? Even assuming you are only launching full 3.5m tanks, that place the number of parts in the thousand. My computer would probably catch fire if I'd even consider such MASSIVE construction.

...wait, are you Skynet?

[steuben]

27 minutes ago, wibou7 said:

In the mid two digit kilotons? Even assuming you are only launching full 3.5m tanks, that place the number of parts in the thousand.

The one that is further along in the design process is approximately 7kiloparts. It takes the SPH/VAB about 10-15 mins to load.

Projections for the second station are roughly the same.

[Evanitis]

33 minutes ago, steuben said:

approximately 7kiloparts

Hmm. Nice. Hope we'll see some screens.

[wibou7]

1 hour ago, Evanitis said:

Hmm. Nice. Hope we'll see some screens.

Definitely!

If your computer manage to launch those before the heat death of the universe.

...seriously though, how long does it take to launch these? 10-15mins to load VAB probably equal to, what, 18 hours from launchpad to orbit?

[Padishar]

Yeah, though at 7k parts I expect he'll only be able to take one every few seconds...

Seriously though, I've got a pretty quick machine (4770K), a 1400+ part station runs at about 3fps (at the default max delta time setting, at minimum it can get 5 fps) and KSP performance is worse than linear with part count so I would expect 7k parts to run at somewhere between 3 and 10 seconds per frame resulting in a seriously slowed game clock.  I really don't see how you're going to manage to actually dock stuff together like that, but good luck with it...

[Laie]

6 hours ago, steuben said:

For actual construction I have two options:
1. build the station a low orbit, say 100 km and then haul the whole thing up to a higher orbit with a high trust tug
2. get the modules to the operational orbit, either with a direct launch or a set of smaller tugs, and assemble it there.

Launch-and-assemble is more straightforward than launch-assemble-move. The second variant adds one mission step, and from your description a non-trivial one.

Building a bigger lifter is comparatively easy. I predict that in the course of your project, you will move to ever bigger lifters anyway in order to lift two or three modules in one go.

[steuben]

Managed to get KVV and KSP64 to work long enough to generate my blueprint.

39 modules to that thing. I know I'm going to have to wait for 1.1 before I can float it up using any means. Part count 6654, 4590 without struts, before vessels are docked to it.

[Snark]

43 minutes ago, steuben said:

39 modules to that thing. I know I'm going to have to wait for 1.1 before I can float it up using any means. Part count 6654, 4590 without struts, before vessels are docked to it.

Good grief, that's a beast!    Thanks for sharing.

Hard to tell from the diagram what parts it's made out of... are you using stock, or are there mods in there?  For example, SpaceY has some nice big parts, and SpaceY Expanded goes up to 7.5m parts.  Big tanks = fewer parts.  (And gargantuan engines to lift big ships, perhaps fewer modules to have to jockey around.)

[Wemb]

On 18/02/2016 at 4:02 PM, steuben said:

Any suggestions/recommendations? A third option that I haven't thought of?

 

You could launch into low orbit and use a dedicated tug to move the components to the assembly point - this might be easier if launching intot he high orbit is problematic.

 

Wemb

[steuben]

She's pure stock. I'll probably end up welding the modules which will bring the part count down to the mid to low triple digits.