Ways to lift a lot of fuel to orbit?

[Galane]

I'd like to see how people get lots of fuel into orbit. Here's how I've done it, rather poorly so far.

My two Rockomax 64 fuel depot, with four ships docked. http://www.flickr.com/photos/27748767@N08/9656845375/in/set-72157635319720861

How I got it (nearly) to orbit. http://www.flickr.com/photos/27748767@N08/9657123873/in/set-72157635319720861

To get to 70KM it requires an in-flight fuel transfer to be able to finish the circularization burn.

The previous version http://www.flickr.com/photos/27748767@N08/9643707068/in/set-72157635319720861 could make it unaided, even laden with nose cones on top of everything. I added two more docking ports and the RCS tanks and thrusters. (When KSP nosecones are fixed to *reduce* drag, it's gonna be a real game changer...)

Before I went with the asparagus monstrosity I tried (and tried and tried and tried) using the ThunderMax parts from ReStock and a lot of struts and eventually wrapping a lot more boosters around the big Thundermax tank but never could get to orbit. Could get it to break apart or spin out of control...

One thing I did find was that piping the fuel from the upper side mounted 64 tanks to the *second* dropping ones helped out a lot. Keeping the last two Mainsails around for another tankfull achieved much more altitude. I'm thinking of moving the pipes over to the first pair of Mainsails to keep all the big engines around longer than the first two drop. Experiments showed me that Skippers are more efficient in the upper atmo, thus the final push configuration has three plus the two small ones still hanging on.

Another item of import with this vehicle is to lock the gimbal on the center Skipper before launch. It makes the wobble go away. Won't make it to orbit otherwise because of extra fuel burn by MechJeb bouncing the throttle. I dunno how just that one engine makes the launch wibble so badly, or why locking just it fixes it, when the other two Skippers and the four Mainsails are left unlocked.

You can download the craft files for Alan Aerospace Recycling & Packaging's ongoing Mun attempt here or get them with everything else I've done (that I've kept) in KSP over here I build something, modify until it does what I set out to do, then rename for a major change.

Edited September 12, 2013 by Galane

[blizzy78]

How I got it (nearly) to orbit. http://www.flickr.com/photos/27748767@N08/9657123873/in/set-72157635319720861

To get to 70KM it requires an in-flight fuel transfer to be able to finish the circularization burn.

It shouldn't need to refuel or transfer fuel between tanks during lifting when using a decent lifter. Personally I like to remove the tedium and guesswork when trying to come up with a lifter for a given payload. That's why I created an online engine layout calculator (see signature.) For a given payload, it can give you the exact layout you need to use, how many booster stacks, and the like.

[thescientist]

If you're looking for a very heavy lifter, look that: http://forum.kerbalspaceprogram.com/showthread.php/31923-0-20-Mammoth-Rocket-Family?highlight=mammoth

there are some lifters that can bring into orbit nearly everything, even a whole space station.

[bsalis]

Craft shown below gets over 600 tons to LKO. Made it recently, out of interest to see how such a design would work with the Skippers added recently-ish. It's not asparagus staged. The staging is somewhat haphazard. For something this scale, structural integrity is more important.

It works because the central core lifts itself by using a ring of Mainsails (since there are LT-Ns under there which are weak). The outer boosters all lift themselves with a Skipper. The most shear force between tanks can only be the force of one Skipper pushing an empty Jumbo tank. Fuel is all fed inward. By keeping the structural forces low between the tanks, it stays together.

Edited September 3, 2013 by bsalis

[Johnno]

It really depends what you mean by 'a lot'. New players usually feel that a full orange tank is a lot, more seasoned players will routinely be lifting multiples and then there's people like bsalis who take things to a whole new level. You can do a simple forum search for 'heavy lifter' or similar, there's also plenty of the spacecraft exchange subforum.

[capi3101]

What Johnno said.

Myself, the heaviest payload I've ever lifted was about 130 tonnes with the Thunderbolt Heavy 7 - five X200-32 tanks, five FL-R1s, five LV-Ns plus RCS blocks, fuel lines, docking ports and lights. That was in the days before I started using asparagus, so the lifter was a ridiculously inefficient onion design (had to fart around with the throttle a lot on that one to get them into orbit - succeeded three times, though).

I should get some Heavy 7s back in space...they were a good design all told.

[Jackissimus]

Craft shown below gets over 600 tons to LKO. Made it recently, out of interest to see how such a design would work with the Skippers added recently-ish. It's not asparagus staged. The staging is somewhat haphazard. For something this scale, structural integrity is more important.

It works because the central core lifts itself by using a ring of Mainsails (since there are LT-Ns under there which are weak). The outer boosters all lift themselves with a Skipper. The most shear force between tanks can only be the force of one Skipper pushing an empty Jumbo tank. Fuel is all fed inward. By keeping the structural forces low between the tanks, it stays together.

Holy mother of God...

What's the part count?

[allmhuran]

750 tons - 20 Jumbos plus assorted odds and ends. All stock of course! Launch video in sig, tutorial video on "thrust plate" technique also on channel. Tried to launch a megaton the other day but ran into part count lag issues. Pretty sure 1000 tons could be done using a thrust plate, just need to figure out a more part-efficient design.

*edit... this should of course be kiloton. Thet day I put a megaton into orbit is the day I have thoroughly beaten KSP.

Edited September 4, 2013 by allmhuran

[Moar Boosters]

I can recommend Allmhuran's design. ^^

Build one of these, put enough struts on it, and it'll be ugly as hell, make your computer run at less than 1fps, but will lift pretty much anything. If you just want fuel you don't even need to put anything on top. Just kit the central section(s) out with RCS and a probe core, and use the outer sections to get the middle tanks into orbit.

[sweetpezak]

Built this lifter for bringing large amounts of fuel into LKO, will put a 303 ton payload into anywhere from 70-120km orbit depending on the efficiency of the pilot, has 730 parts including the payload part, the lifter itself is probably around 600 or so parts... link: http://kerbalspaceprogram.com/7-up-mk-vi-303-ton-lifter/

[Motokid600]

Well those lifters are nice and all, but if you want to lift that much and maintain frames per second instead of per minute look at NovaPunch and the Soviet/American pack. I couldn't imagine heavy lifting without them. I'm working on docking 420 tons worth of fuel to my station. Lifting it will be the easy part lol. ill try and post pics later on.

[Galane]

Those are all some amazing designs for launching a lot of fuel.

Depot Marko 1-0 NC RS (No Cones, Re-Staged). Stretched the small drop tanks a bit more, stacked a FL-T100 on top of each of the first staging, moved a couple of fuel pipes, added more RCS thrusters to the core and plopped a Clampotron Sr. on top.

Put two in orbit and docked end to end just for the heck of it. Docked with Smart A.S.S. using TGT+ so the ports would automatically stay looking at each other. Had them aligned just right but once it got close the one *not* doing the docking decided it had to roll in spite of kill rot being enabled. Took some frantic tapping on Q & E then quickly switching ships and turning off docking autopilot to get it to click together, but I did get it how I wanted.

Have not tested yet to see how high an orbit it can reach now, nor done any attempts at optimizing ascent path. Just using the MechJeb default. Might fold up if the gravity turn is done after the upper 64's fly away, the small drop tanks now last until just after those go. No struts between the two core tanks.

Added to the AARP ship collection zip files.

P.S. How do I get this tagged as a Station instead of a Probe?

Edited September 4, 2013 by Galane

[Specialist290]

P.S. How do I get this tagged as a Station instead of a Probe?

Right-click on the command pod / probe body. You should see an option to rename the craft. If you click on that, it'll take you to a screen that will let you edit the name and vehicle class.

[Starwaster]

My current favorite method for lifting large loads.

The four outer rockets are KW 3.75m, the longest ones. With Griffon XX.

In the center is a Procedural Fairings fairing. The actual lifters are mounted to the fairing sides. Two of them feed into the other two and stage when they're empty. I also use Modular Fuels so usually I'll load them with LH2 / LOX to lighten the load. I can get over 100 tons up this way. Not sure how far it's expandable. If I want to lift more I'll add a third tank to each lifter but if the load exceeds the capacity of the Griffons then I'll have to add more Griffons. These four could lift a lot more if it weren't for how badly they overheat. They have to throttle back to something like 60%-70%

Edited September 4, 2013 by Starwaster

[-Velocity-]

This was my first attempt at a heavy tanker:

It was a very successful design that could lift about 200 tons to LKO. It can handle bigger payloads, though, but the payloads have to share fuel with the 3rd stage. It mainly uses KW rocketry and stock parts.

I upscaled this design into a 3000 ton heavy lifter (both asparagus staged and non-asparagus versions) that can lift about 600 tons to LKO (1000+ ton payloads are possible if the payload lends a little of its fuel to the 4th stage). The upscaled version is powered by 192 Novapunch 1.25m 285 kN aerospike engines and 18 of the 1400 kN 2.5m engines for an initial thrust of 79920 kN (it consumes over 20 tons of fuel a second during the first ~40 seconds of flight). The big drawback of this rocket is the part count is about 1500- with no payload. The game crashes a lot when I use it, mainly because for some insane reason, the devs of KSP went with a 32 bit executable and the game runs out of memory all the time, but its the only way for me to get my biggest designs into orbit. I don't have any screenshots here, but I can post them later when I get home. Anyway, there were four concepts I utilized in both these designs:

- Very high thrusts by using A LOT of 1.25m engines. I find that get higher thrusts under a stage if I pack a bunch of 1.25m engines instead of using full size 2.5m or 3.75m parts. Unfortunately, this leads to a high part count (and part clipping needs to be enabled usually).

-High efficiency by using 1.25m engines. The ISPs for a cluster of 1.25m (and smaller 2.5m engines) is MUCH better than a single, big engine. It is MUCH easier to get to orbit!

-Lateral staging instead of stacked staging- KSP rockets are too structurally weak to support huge, long rockets. So both my big lifter designs utilize a central 3.75m section surrounded by six 3.75m sections.

-Engine nacelles- the bottom of my 3.75m stalks have engine nacelles to hold additional fuel and provide a place to put more engines.

Anyway, I would have used bigger fuel tanks (Novapunch has 5 m diameter tanks, but they are only 12m long), but I have A LOT of problems with Novapunch fuel tanks (and to a lesser extent, engines)- sometimes they just randomly fall off, do not stay attached well, etc. Still, despite these problems, Novapunch is useful for other parts that they add (just, IMO, stay away from the fuel tanks!). Anyway, the KW rocket 3.75 meter fuel tanks look better and are structurally sound.

[superm18]

This is my Grand Tour Tug:351 t to orbit 150 km x 150 km (2348 t on the pad)

[Galane]

Y'all are making my little dual Rockomax 64 lifter look real puny. ;-) I need a better PC. AMD Phenom II X2 555 3.21Ghz, 2 gig DDR2 RAM, Radeon HD2600 Pro PCIe x16 video with 512 meg.

Remember when 12 megabytes RAM in an 80286 was hot stuff? Ran Windows 3.11 (not Windows for Workgroups 3.11) real nicely.

I agree on the 64bit. Why in this era anyone would even contemplate, let alone start work on, a program this data intensive in a 32bit environment? That's why I have kept the KSP mods I have installed to just a few. That's why I like ReStock, it adds almost nothing the the memory requirement while providing many new parts.

Edited September 4, 2013 by Galane

[Snarfster]

Kind of hard to answer, what is "a lot"?

I use this to get full orange tanks in to a 125k orbit for my station / fuel depot.

[HeadHunter67]

Snarfster, that looks pretty similar to the test rocket I cobbled together this morning in an attempt to answer this same question. As stock goes, that's a pretty good way of doing it. Basically, it's a payload fraction of about 12% to a 125km orbit, and the part count and mass are reasonable for the task.

I don't understand why some people overdo it - if all you're doing is adding more fuel and mass to loft the unnecessary fuel and mass you already have, what's the point? I prefer the most minimal approach that can do the job, regardless of an unlimited budget.

[-Velocity-]

I don't understand why some people overdo it - if all you're doing is adding more fuel and mass to loft the unnecessary fuel and mass you already have, what's the point? I prefer the most minimal approach that can do the job, regardless of an unlimited budget.

But that's not the case. In some cases, you DO get diminishing returns after a point, but bigger is still better, at least, once it's in orbit. Bigger fuel tank = more delta V. And besides, I probably don't share the mission profile you do. I think that flying all the way to Duna or Jool and only making one or two landings is a waste. The giant ships I'm sending AREN'T for making a SINGLE landing. I carry enough fuel for MULTIPLE landings, maybe dozens of landings, because my giant interplanetary ship has like 500 tons of spare fuel on board to fuel the landers (my landers are generally fully reusable). So basically, I fly to a planet, and keep landing until the mothership gets down to bingo fuel, then go home.

And besides, it's more fun to build gigantic stuff

Anyway, just understand that right now, in sandbox mode, people just like to play their own way and you shouldn't expect to understand why some folks do certain things

Edited September 5, 2013 by |Velocity|

[-Velocity-]

OK, here's the bigger version of my heavy lift rocket.

In that last screenshot, that's the "4th" and last stage being jettisoned by the payload, though I don't really know how to properly refer to stages in a rocket that is part asparagus staged and part vertically staged.

I actually haven't gotten the fuel flow completely correct for the asparagus staged version due to the frustrating and awful way that KSP does fuel flow priorities, but it's close enough. (If only there was a fuel duct you could connect between two tanks that made those two tanks of equal fuel feed priority!!! The current system has no way of doing this, even if you connect a fuel duct going one way and another one going the opposite way.)

I think I MIGHT be able to scale up the design one more time, but I feel like I'm approaching a limit where the structural strength of the parts in KSP just can't handle the sheer weight of an even bigger design. As an example, I have a hard time packing enough launch clamps on this thing to hold it up.

Also, my computer would probably melt down if I went much bigger and kept using the more efficient 1.25 m engines (the above rocket uses a total of 255 engines, most of them highly efficient 1.25m engines, and takes like 40 minutes to launch), so I'll have to examine actually using 2.5 meter engines rather than 1.25- which might just kill the performance of the rocket enough that it becomes useless.

Edited September 5, 2013 by |Velocity|

[Galane]

I actually haven't gotten the fuel flow completely correct for the asparagus staged version due to the frustrating and awful way that KSP does fuel flow priorities, but it's close enough. (If only there was a fuel duct you could connect between two tanks that made those two tanks of equal fuel feed priority!!! The current system has no way of doing this, even if you connect a fuel duct going one way and another one going the opposite way.)

That got me with the small drop tanks on mine. I figured it'd suck them down last as they're right next to the center engine and there's fuel coming in from asparagus around it, but nope, they got drained early. Then when I lengthened them and moved some pipes around they drained slower, but still got tapped right from the start. Running pipes both ways to make different size tanks/engines drain at the same rate was a thing I figured out soon after I first used the pipes.

What'd really be nice is a way to explicitly set fuel use order, along with either bent pipes that can jump over decouplers and separators, or have all those with a fuel crossfeed toggle that is off by default.

Looks like the question has been Answered, a whole bunch.

[-Velocity-]

That got me with the small drop tanks on mine. I figured it'd suck them down last as they're right next to the center engine and there's fuel coming in from asparagus around it, but nope, they got drained early. Then when I lengthened them and moved some pipes around they drained slower, but still got tapped right from the start. Running pipes both ways to make different size tanks/engines drain at the same rate was a thing I figured out soon after I first used the pipes.

What'd really be nice is a way to explicitly set fuel use order, along with either bent pipes that can jump over decouplers and separators, or have all those with a fuel crossfeed toggle that is off by default.

Looks like the question has been Answered, a whole bunch.

Yea, it's stupid, but I figure that the only way I am going to fully solve my fuel flow problems is to simply feed all engines in an asparagus stage from a single tank. Basically, each asparagus stage has a central, "main" tank, and all engines in that asparagus stage are either attached to the central tank, or fed by a fuel line coming off the central tank. Any additional fuel tanks in that asparagus stage have one-way fuel connections going from them to the central tank. It sucks, but as far as I can see it, it's the only way to make the fuel tanks drain in the right order. It will mean that I will to waste a lot of mass on engine mounts.

Edited September 5, 2013 by |Velocity|

[Stone Blue]

What about trying TAC Fuel Balancer?

I'm not sure if it will compensate for any issues with fuel piping, but I've used it to help with fuel balancing a couple times with added drop tanks on my 1st stage, that werent transfering correctly.

What I did was to set all the tanks I wanted to drain evenly in the 1st stage, to "Transfer Out". This tells them to transfer all fuel out to all other tanks...But with the upper stages all ready full, the fuel has no where to go...Until you light up the engines, and that becomes the only available output for the tanks you've set to transfer out...This will also force the transfering tanks to do so evenly between all the other tanks in the 1st stage.

If I have fuel tanks that are part of the payload, and not intended to be used during lifting, I can "Lock" those tanks, so the rocket will use all available fuel but these, keeping them full for post-lifting use, as intended.

I've found another use for the mod: You can use it to change the center of gravity, and even get a long, unbalanced structure to swing, if the RCS runs out or its not balanced placement wise, or to help if there isnt quite enough SAS or ASAS to get something lined up for docking in time.

[pluto101]

That got me with the small drop tanks on mine. I figured it'd suck them down last as they're right next to the center engine and there's fuel coming in from asparagus around it, but nope, they got drained early. Then when I lengthened them and moved some pipes around they drained slower, but still got tapped right from the start. Running pipes both ways to make different size tanks/engines drain at the same rate was a thing I figured out soon after I first used the pipes.

What'd really be nice is a way to explicitly set fuel use order, along with either bent pipes that can jump over decouplers and separators, or have all those with a fuel crossfeed toggle that is off by default.

Looks like the question has been Answered, a whole bunch.

While I don't think I've figured out how to ensure tanks are evenly drained, I do know a way to ensure multiple stacks all run out of fuel at the same time.

Lets say you have three tanks (it can be any number of tanks), and you want any engines attached to those tanks (lets say one engine per tank, attached at the bottom) to drain from those three tanks evenly. What you need is a choke point. Most likely, you have one central tank. This tank generally serves as the best point for your choke point.

In order for your three tanks to drain evenly, attach a fuel line from your two side tanks to the central tank. Then, attach a fuel line from your central fuel tank to your side engines. The engines will use fuel from the fuel line before they use fuel from the stack. So, all three engines are drawing from the central tank, which is in turn drawing first from the side tanks. During flight, your side tanks drain first, then your central tank. However, all three engines will tap out at the same time.

I use a modified version of this in the final stage of my heavy lifter (650 tons to LKO w/ about 350 delta V to spare; pics forthcoming, once I can get the strutting right to ensure stability during launch). In that setup, I have a central tank w/ a Mainsail and six side tanks w/ Mainsails. My asparagus boosters surrounding this final stage drain into the central tank. The central tank then has fuel lines radially coming out into the side tanks. The side tanks then have six fuel lines leading to the central mainsail. The central tank drains first. Once it is finished, the side tanks drain. The central Mainsail continues to draw fuel from the side tanks once the central tank is finished, and all seven engines run out of fuel at the same time. Because all fuel lines are symmetrical, the center of mass never wavers from the center of the ship.