Launch Costs

[Pthigrivi]

How are yall doin on your launch costs? I feel like Im doing pretty good, but Im sure others are doing better. My workhorse lifter is pushes 40t to LKO at 180k off the pad, and I'm recovering everything but the fairing for a net cost of 23-24k and 580 funds/ton. I feel like I could do better with my shuttle, which delivers 18 kerbals (though its often more like 12-16) to my station at about 750 funds/kerbal. 
 

Granted I don't always land right on KSC. Half the time Im out in the grasslands nearby. Im sure I could do better with proper spaceplanes but its hard to beat the versatility of a tailsitter SSTO. Its just really nice to maintain an easy AOA and then pop the chutes as you graze over KSC.

How are you guys doing on costs?

Edit: man can someone fix the damn imgur albums? that used to be nice.

Edited April 26, 2018 by Pthigrivi

[RizzoTheRat]

On my last career I was on about 550 kredits per tonne with a VTOL SSTO that could put 45 tonnes in to LKO.  I was averaging about 97% funf recovery once I got the hang of landing it near the KSC.  Not got that far yet in this career but hopefully with kOS doing the landings I should be able to get even better recovery. 

 

 

HTOL spaceplanes work out cheaper than VTOL I belive,  but I can never be bothered with how long they take. 

[Corona688]

I think the lowest cost I've achieved is about 1000 per ton.  My reusable ones so far have ended up even worse, due to fuel weight.

[blakemw]

It depends a lot what you're going for. I play hard career a fair bit so have to factor in loss-of-vessel and missing the KSC. This favors low cost recoverable launchers.

I use a simple Twin-Boar with 2x Rockomax 64 tanks as my baseline Rocket SSTO. This can deliver ~26t into LKO, the cost of the Booster is ~32,000 and the cost of the LF/Ox is 8800, so a disposable cost of $1230/t, or with 90% recovery for landing the booster somewhere near the KSC around $430/t. Naturally, this isn't as low as using more efficient engines but the Twin-Boar is an extremely cheap LF/Ox engine and it is 100% guaranteed to not burn up on reentry even without retro burns and protects other parts during splashdowns.

SRBs are pretty exceptional value for money, for example, adding SRBs bringing the cost up to 40000 would increase payload to LKO to about 42t, this reduces the disposable cost to about $950/t, and increases the w/recovery cost to about $470/t. In fact SRBs are so cheap that when used in a "launchpad kick" role they are darn near as cheap per t as recovered SSTO liquid engine, because the solids are cheaper per kN than LF/Ox and do a much better job of overcoming gravity losses. So typically I use single-boar for smaller payloads ~26t, triple-boar for larger payloads ~80t, and single boar with SRBs for intermediate payloads.

FWIW the lowest cost possible for a purely disposable rocket, making extensive use of solids, is about $590/t. Recoverable SSTO rockets are cheaper by a smidgen, but solids are seriously not to be underestimated.

It's also possible to build reasonable low-cost spaceplanes, generally using Whiplash+Skipper or something like that, RAPIERs also aren't too bad when used 50/50 with Whiplashes. A spaceplane SSTO does not need to be dramatically more expensive than a rocket SSTO, and can generally put payloads into LKO at around half to a third the cost per t assuming you recover them without fail. But they come with greatly heightened RUD risks and it is much harder to fit a payload to one compared with just putting a big fairing on top of a rocket. I occasionally use spaceplanes for crew or fuel transport, but I don't really use them at all for other forms of payloads because efficiency is only obtained by building the plane around the payload (this was one problem with the Spaceshuttle, it could put about 25t in orbit, but itself weighed about 70t, so most of the propellant went to putting the space shuttle in orbit rather than the payload).

Edited April 27, 2018 by blakemw

[Pthigrivi]

59 minutes ago, blakemw said:

It depends a lot what you're going for. I play hard career a fair bit so have to factor in loss-of-vessel and missing the KSC. This favors low cost recoverable launchers.

I use a simple Twin-Boar with 2x Rockomax 64 tanks as my baseline Rocket SSTO. This can deliver ~26t into LKO, the cost of the Booster is ~32,000 and the cost of the LF/Ox is 8800, so a disposable cost of $1230/t, or with 90% recovery for landing the booster somewhere near the KSC around $430/t. Naturally, this isn't as low as using more efficient engines but the Twin-Boar is an extremely cheap LF/Ox engine and it is 100% guaranteed to not burn up on reentry even without retro burns and protects other parts during splashdowns.

SRBs are pretty exceptional value for money, for example, adding SRBs bringing the cost up to 40000 would increase payload to LKO to about 42t, this reduces the disposable cost to about $950/t, and increases the w/recovery cost to about $470/t. In fact SRBs are so cheap that when used in a "launchpad kick" role they are darn near as cheap per t as recovered SSTO liquid engine, because the solids are cheaper per kN than LF/Ox and do a much better job of overcoming gravity losses. So typically I use single-boar for smaller payloads ~26t, triple-boar for larger payloads ~80t, and single boar with SRBs for intermediate payloads.

FWIW the lowest cost possible for a purely disposable rocket, making extensive use of solids, is about $590/t. Recoverable SSTO rockets are cheaper by a smidgen, but solids are seriously not to be underestimated.

It's also possible to build reasonable low-cost spaceplanes, generally using Whiplash+Skipper or something like that, RAPIERs also aren't too bad when used 50/50 with Whiplashes. A spaceplane SSTO does not need to be dramatically more expensive than a rocket SSTO, and can generally put payloads into LKO at around half to a third the cost per t assuming you recover them without fail. But they come with greatly heightened RUD risks and it is much harder to fit a payload to one compared with just putting a big fairing on top of a rocket. I occasionally use spaceplanes for crew or fuel transport, but I don't really use them at all for other forms of payloads because efficiency is only obtained by building the plane around the payload (this was one problem with the Spaceshuttle, it could put about 25t in orbit, but itself weighed about 70t, so most of the propellant went to putting the space shuttle in orbit rather than the payload).

Thats a great breakdown. Makes me think I could probably shave off a bit by doing a 2 stage with SRBs.

[wumpus]

There were two contests for low cost launchers without recovery.  Here's a link to the second:

The winner appears to be a nearly single-stage Rhino with a pair of asparagused twin-boars.  646 funds/ton.  Maybe there was a set of SRBs in there I didn't see them and the files are no longer available (the video is).  Getting under 1000 funds/ton isn't that hard (in general, finding an ideal launch for a specific payload is another story).

One strategy I used was TTSO, with both stages eventually making orbit (the first stage could easily make orbit without the load of the second).  Of course, this was quite hectic without mods (you have to avoid losing the stage due to exiting the physics bubble) and takes at least four times as long as a normal launch.  It mostly burned me out of KSP before making it to other planets (I got better).  Human time is more important than game funds: use SRBs unless low cost really makes you happy enough to take that much longer to get to orbit.

[KerikBalm]

I play with a scaled up system, so my launch costs are significantly higher (3x radius and SMA, 1.25x atmospheric height, 1.5x terrain feature height)

Orbital velocity at LKO is nearly 4,100 m/s, and I need about 5,500 m/s of dV to get to orbit with a rocket. SSTO HTOL planes have to supply a lot more dV from closed cycle/rocket engines, and lose a lot more dV to drag given the higher speeds

With this, I can get about 140 tons to orbit:

So... costs... just fuel and the fairing (not needed for some payloads, and fairings are broken right now, so its often better not to use it)

Adding up the fuel:

Quote

11x jumbo 64 tanks + 4x x32 tanks = 14 jumbo 64 tanks = 37440 LF and  45760 Oxidizer. 

7x 25. to 1.25m adaptors: 2520 LF and 3080 Ox

3.75m to mk3 adaptor: 1125 LF  1375 ox

mk3 to 2.5m adaptor: 1125 LF  1375 ox

mk LF: 10,000 LF

4x mk1 LF fusalages: 1600 LF

46x big wings at 300 LF each: 13800 LF

Grand total: 66,010 LF 51590 oxidizer

0.8 funds per LF, 0.18 funds per Oxidizer....

So launching with full fuel, and coming back empty would cost: 62,094 funds

So, I seem to recall I often empty some of the tanks a bit, but anyway, I'm going to assume it comes out to 62,000 funds to get that thing to orbit and recover it, so 443 funds per ton for my SSTO in a 3x game.

More pics:

Spoiler

The leftover fuel works out to 107.825 tons. Note that the payload was sent up mostly empty, because the craft simply cannot structurally handle all that weight there (IIRC, one or two of of the 3.75m tanks were full, and the 4 jumbo 64s were only partially filled and emptied first)

The 2x 3.75 tanks add another 18 tons to the payload, for 125.825 tons. The jumbo 64s add another 16 tons of dry mass (but since some of the fuel they contained was used, I'm not sure we can count them entirely as payload), There's a KR-2L there, which adds another 9 tons, and some other miscelaneous stuff to enable the payload to function as a surface to moon orbit fuel shuttle.

an earlier version, which had less wing area and a slightly different engine arrangement:

 

 

Edited April 27, 2018 by KerikBalm

[blakemw]

4 hours ago, wumpus said:

There were two contests for low cost launchers without recovery.  Here's a link to the second:

The winner appears to be a nearly single-stage Rhino with a pair of asparagused twin-boars.  646 funds/ton.  Maybe there was a set of SRBs in there I didn't see them and the files are no longer available (the video is).  Getting under 1000 funds/ton isn't that hard (in general, finding an ideal launch for a specific payload is another story).

3

That's somewhat old (altough the general principles still hold). There's a more recent challenge, "The Lowest Bidder" on reddit, that challenge was to launch a full orange tank as cheaply as possible, and occurred at version 1.2 which was the last version to make a significant change to rocket performance (specifically, supersonic aerodynamics was changed in 1.2, I think part costs and performance were also normalized between 1.05 and 1.2). The best disposable entry was this: https://imgur.com/a/qWjcG, delivering a full orange tank (36t) for 28140.

Another interesting reddit challenge was "Twin-boar efficiency", which didn't have cheapness as a goal per-se, but rather was seeing just how much a Twin-Boar could lift, that is how much use you can get out of a single engine. The rules for the challenge were somewhat weak (to encourage creativity in entries), but there are some interesting entries.

Edited April 27, 2018 by blakemw

[Snark]

I tend to end up at a bit over 1000 per ton to 90 km LKO, for totally throwaway rockets where I don't bother recovering anything.  For example, here's a rocket that I slapped together with minimal effort that uses 38,220 of throwaway hardware to put just shy of 39 tons into 90 km LKO.  Simple, quick, easy to design, zero time spent trying to optimize cost.

(Total cost on the launchpad is 51,700.  That's 38,220 of throwaway hardware, plus a payload that costs 13,480.)

While I do try to build reasonably efficiently, and therefore avoid being egregiously wasteful, I don't particularly bother trying to min/max it to optimize for funds-per-ton.  That's because in my experience, actual launch costs aren't all that important, over the course of a KSP career.  It's the same reason why I never bother recovering anything for the funds.

For me, the important thing is to maximize my fun-per-play-time.  Different people like different things; personally, I don't find recovering hardware for the funds to be at all interesting or fun.  (It was interesting and fun the first two or three times I tried it, then it got boring for me and I never bothered with it again.)  And as long as I design reasonably efficiently, it's simply not worth it to me to spend a bunch of extra time to squeeze that last few percent of cost out of it.  Far simpler and easier to just dash off a quick contract that pays far more than I would have saved by trying to recover hardware, and/or min-max my funds-per-ton.  It gives me a lot more funds-per-minute-of-my-time, and I can then use that to build and launch rockets the way I like.

Launch costs just aren't all that big, for me.  They pale in comparison to the cost of upgrading the KSC buildings.  Compared to that, everything else is pocket change for me.