Jump to content

SSTO Tylo No Refueling Or Docking


DAL59

Recommended Posts

52 minutes ago, sdj64 said:

o, I did see one error in my calculation: When I assumed the SSTO payload fraction of 35%, it didn't include any engines or wings used for Kerbin ascent (because the payload fraction challenge specified an inert payload).  My test craft had a mass of about 45 tons full and 30 in orbit, for 66%, which puts chemical engines back in the realm of possibility for the Tylo landing and ascent.

You can also use your efficient (ion or nuclear) engine in conjunction with your more powerful engines for Tylo, which could maybe save you some more.

35% has been surpassed,  I did 43.7%  on a liquid fuel only design and that's with the drag overhead of a mk3 cargo bay in a liquid fuel only lifter

https://kerbalx.com/AeroGav/partridge   ....  someone after me posted an even higher number, but the OP had stopped updating the thread long since.

However, i still think the OP is asking the impossible.    Has the OP even tried using cheats to put a lander in low tylo orbit, landing, then getting back to space again in one stage ?  Even that is really really hard.    It took me half a day to make a two stage lander that worked (three nervs, and a discardable poodle booster).    The delta v requirement combined with TWR need is brutal.    Chemical engines struggle with the fuel requirement, nukes have a hard time getting enough TWR unless the fuel fraction is quite small.     And when you have TWR that is only just good enough (nuke, or chemical stage with a huge amount of fuel) then gravity losses mean you need much more delta V than a delta V map claims.

If you succeed in all of that, then you got to try do it with the mass penalty of whatever method you need to go back to Kerbin,  and the dead weight of Jet engines/ wings etc it took to get the thing into orbit in the first place. 

 

People seem to be approaching this by designing a space plane first, which is good as everyone is learning how to make better SSTOs, but in terms of the challenge it's like designing a rocket by designing the lower stage first.     

What people should do IMHO is design a lander (test with cheats) that can do low Tylo orbit to surface and back again, then try to turn that into an SSTO.

Link to comment
Share on other sites

1 hour ago, AeroGav said:

However, i still think the OP is asking the impossible. 

I know.  This is supposed to be the ultimate ultimate challenge, because it has never been done.  

You can use command seats and use a second vessel to transfer a kerbal.  Or take command.  

Link to comment
Share on other sites

I think we would be best off getting a probe that can do this; crewed refinements can wait for version 2.0.

My current line of thinking goes along with @sdj64's approach:  The nuke can carry the load for most of the trip (an uninhibited nuke's theoretical dV is over 17,000 m/s--you can go anywhere with that ... twice) and a creative approach to the landing and take-off can pull a trick from the Kerbin ascent repertoire:  the second stage engine does not need a TWR over 1.  The second stage engine can instead do with less and add horizontal velocity while gravity bleeds off the vertical velocity from the first stage:  by the time the vertical velocity reaches zero the horizontal velocity has you in orbit.  If we apply that principle backwards, then the nuke can shed horizontal velocity while slowly pitching down until the gravity overpowers the feeble TWR of the engine, at which point we give it one good blast from an LFO engine (closed-cycle rapiers may be inefficient, but they would work well here) to keep from crashing hard enough to break things.  Nothing says that we have to use a given engine for the entire burn; we can treat it as though it is a staged rocket but without actually staging.  I have not even begun to run numbers for this scenario but I think there may be some promise to it.

For take-off, we can employ other tricks:  Tylo's highest point is quite high up,;if we can land there, then it takes less to take off (and we have more time to use the nuke to push us up before we crash into the valleys below).  It may not be much (SSTO from Eve is possible only from over 6800 m--but that's a function of pressure, not gravity), but with the kinds of margins we have, it may be necessary.

Edited by Zhetaan
Link to comment
Share on other sites

58 minutes ago, Zhetaan said:

The nuke can carry the load for most of the trip

I personally think an ion would be better.  It would be tedious, but its much lighter and more efficient.

Link to comment
Share on other sites

19 hours ago, Zhetaan said:

I think we would be best off getting a probe that can do this; crewed refinements can wait for version 2.0.@sdj64

While I normally like to go with the flow on challenges, I'm going to have to second this. The challenge is tantalizingly just barely impossible without the Kerbal. I think Kerbal dry mass may have to wait for more development first.

 

18 hours ago, DAL59 said:

I personally think an ion would be better.  It would be tedious, but its much lighter and more efficient.

Unfortunately tedium isn't the only aspect. The numbers aren't quite as in favor of ions as they might seem. Ions create tons of drymass, enough to heavily counteract their awesome Isp on an SSTO. Bizzarely, on an SSTO the order you burn your fuels matters, and you always want to burn the highest Isp fuel last. It happens to be that the high Isp of ions is wasted if it's spent pushing around low Isp fuel for a Tylo landing. Let's put some bounds on things and look at the maths. As we've been saying, it's worth it, but not the easiest justification.

(masses from one of my SSTOs in progress)

Ions LKO to LTO 2km/s = ~ 40000*ln(85/81)

Ions LTO to LKO 2km/s = ~ 40000*ln(30/28.5)

Drymass from ions = 4.3t tanks +1t dawns + 1t electric +.4t extra Tylo landing engines +.8t extra plane parts (draaaaag) = 7.5t

lost deltaV from nukes of same launch mass= 8000*(ln(85/81)+ln(30/28.5)+ln(28.5/22)) = ~2.9km/s

So we have a net gain of 1.1km/s at the expense of a bunch of extra mission complexity and 3 hours of burn time IRL with physwarp in 100 separate burns. I mean, it's worth it, but it's much more tempting to find another way to scrap 1km/s into the mission.

Still poking at this, and making little incremental improvements. Not there yet, but we'll see where it goes!

just_after.jpg

This photo was taken just after making it to LKO, after (I wasn't thinking) cheating to high orbit to test out deltaVs. I'm slowly getting the percent fuel in orbit higher, which is how I'm working on increasing mission deltaV at the moment. I've also just made and balanced an ion version, but I'm yet to make the long trip up. Those long lines of mk0 parts are the most mass and drag efficient ways to carry lots of fuel, despite the silly look.

Edited by Cunjo Carl
Link to comment
Share on other sites

1 minute ago, Cunjo Carl said:

Ions create tons of drymass,

why?  Just put on a single rtg.  

2 minutes ago, Cunjo Carl said:

This photo was taken just after making it to LKO, after (I wasn't thinking) cheating to high orbit to test out deltaVs.

Does it have enough TWR to land on Tylo though?

Link to comment
Share on other sites

Just now, DAL59 said:

why?  Just put on a single rtg.  

RTGs only make up about 1/30 of the ion drymass. The real drymass comes from those heavy xenon tanks, heavy batteries, and heavy dawn engines... Blarghlbarg and the extra wings + oxidizer to push past the extra drag in mach breakthrough and the 1700m/s to 2100m/s no man's land pushing up to LKO.

Link to comment
Share on other sites

The problem with ion drive is that you sacrifice a lot in terms of dead weight for not a lot of benefit.  Xenon tanks don't have good wet/dry mass ratio (the radial tank is best with 2.29; the inline tanks are 2.27) and lacking the ability to stage away the dead weight of empty tanks, you end up carrying that mass, which then takes away from mass you can use for power systems and other such.  Given the best xenon tank, the theoretical best dV for the ion engine is only 34,000 m/s:  only in this case because it is only twice the dV of the nuke using standard liquid fuel tanks.  Add to that the required power systems needed to run ions near Jool, and it becomes too much.  You'll need either a lot of batteries, a lot of panels, or a lot of RTGs, and none of that is a good solution.  Batteries are heavy, solar panels have low power generation near Jool (4% of Kerbin's efficiency), and RTGs have both of those disadvantages but with none of the benefits.  Fuel cells in lieu of solar panels might work a bit, but that adds additional parasitic mass in the form of extra tankage--although it may be possible to make use of space in slightly-oversized fuel tanks and avoid the extra dead weight (in fact, by making less of the existing weight 'dead').

The ion engine requires 8.74 EC/s to operate at full throttle.  Gigantor solar panels produce 24.4 EC/s, but at Jool (mean semi-major axis) they produce 3.91% of that, or .954 EC/s.  You'll need ten of them if you use no batteries, which adds three tonnes just in the panels.

Batteries all store charge at 20,000 EC/tonne.  To determine the needed numbers, we'll need to figure the length of burn for a given test vessel.  I'll use Tylo capture because it cannot be done with gravity assists and because it doesn't involve landing, which we already know cannot be done with ions.  A Tylo capture burn is (again going from the dV map) 1100 m/s--the real value will be more because you will, assuming a gravity assist, be entering the sphere of influence with decidedly more than the minimum orbital speed necessary to reach Tylo from low Jool orbit.  The rocket equation will give us a general answer here:

dV = Isp * g0 * ln (mw / md)

where mw and md refer not to true wet and dry mass, but rather mass before and after the burn.

1100 = 4200 * 9.80665 * ln (mw / md)

1100 = 41187.93 * ln (mw / md)

.02670685 = ln (mw / md)

1.027067 * md = mw

We can call that a comfortable 2.71% of the vessel's mass must be expelled in order to close and reduce the altitude to low Tylo orbit, or 271 kg of xenon per tonne of vessel (vessel tonnage includes unused xenon left in the tank).  The radial tanks hold 40 kg, though perhaps you would prefer fewer parts and less drag:  the PB-X750 holds 525 kg, which supplies enough that you may even be able to break orbit again.  The ion engine has a thrust of 2 kN, which can be used to get mass flow rate:

Mass Flow (kg/s) = Thrust / (Isp * g0)

= 2 kN / (4200 s * 9.80665 m/s2)

= 2,000 N / 41187.93 m/s

One newton is one kg*m/s2:

= .04856 kg/s

This is kilogrammes per second, per engine as well.  Xenon masses at .1 kg/unit, so the total fuel expenditure is .4856 units/s/engine.

Anyway, 271 kg-Xe/tonne, for an engine that expels propellant mass at .04856 kg-Xe/s, means that the total burn time must be just under 5581 s/tonne of vessel for one ion engine.  That's a 93-minute burn, per tonne.  Burn time shortens by quite a lot when the vessel is less than one tonne in mass, but that is why ion engines are best suited to probes.  Add in the 8.74 EC/s to power the engine, and we require 48777.94 EC, or more than 12 Z-4k batteries.  We can take the edge off with additional power generation, but that adds even more mass.  Remember, batteries store charge at 20,000 EC/tonne, and we've just added nearly 2.5 tonnes of them.  Immediately, we need to multiply our fuel usage and burn time by 2.5 because that is now the minimum size of the vessel.

RTGs produce .75 EC/s, so to power an ion, you need 12 of them. They mass less than the batteries, but they also create drag.  There may be an optimum balance of trade-offs, but that's before we've even considered how to store the xenon for all of this.

And none of this covers the fact that xenon cannot be used for anything else:  LFO for fuel cells can be stored in the same tanks as LF for the nuke, LF for the jet, and LFO for the landing rocket.  There's a truly great potential for mass savings just in using the same fuel for everything wherever possible.

Edited by Zhetaan
Link to comment
Share on other sites

Throttling an ion engine to match an RTG's output puts the engine at about 8.5% its rated output and extends the burn time by a factor of 11.65.  The 5581-second burn I mentioned before now becomes a 65,037-second monstrosity of a burn--that's nearly 1084 minutes, or a bit over eighteen hours of burn, per tonne of vessel.

However, that's not the entire picture.  First, that does not include any other electrical devices that may be needed, such as reaction wheels.  Second, it's probable that such a burn takes longer than the time you spend in the sphere of influence during the Tylo encounter--I'd have to run a few tests to verify the boundary conditions--but the more important thing is that such a burn loses all semblance of efficiency.  If you begin the burn nine hours ahead of the node on a hyperbolic entry trajectory, you end up with losses on your losses:  you spend so little of your time thrusting retrograde at the periapsis that it takes more time, and far more fuel, to close and circularise the orbit; and because you start so far out, all of your initial retrograde thrust would warp the trajectory to collide with Tylo's surface at orbital speed, so to avoid that you have to thrust in a different direction ... which wastes even more fuel, so you have to bring even more.  The theoretical gains in delta-V that you get from ions evaporate in the face of the burn time:  all of the rocket equation calculations presuppose instantaneous burns, not burns that take place over days.  There is something called a periapsis kick that can help with long departure burns, but it doesn't help with capture.

Ions are great for low-mass probes doing orbital operations.  They can also be nice for low-gravity biome hoppers and the like, though you'll probably get better response from something designed with monopropellant in mind.  It's also possible that someone will figure out a way to make them work for a Tylo land-and-return mission, but I genuinely doubt that it can be done along with the other constraints placed upon this vessel.  Admittedly, I'm not looking very hard to find an ion-powered answer, but that's mostly because I'm looking for any answer at all and nukes (and their fuel tanks) show a lot more promise.

For example, the Mk.-0 fuel tank has a wet/dry mass ratio of 10, not 9.  That is a gigantic advantage for this particular mission; it's almost as good as getting fuel for free.  Another example is the Big-S wing parts that are also fuel tanks:  wings are essential, but most spaceplane SSTO wings become dead weight once they leave the atmosphere.  Big-S parts change that.  But there are no similar solutions for xenon or batteries, so any vessel built around an ion drive will have to incorporate more parts that do fewer jobs.

Link to comment
Share on other sites

On 11/11/2017 at 12:55 PM, Zhetaan said:

  Fuel cells in lieu of solar panels might work a bit, but that adds additional parasitic mass in the form of extra tankage--although it may be possible to make use of space in slightly-oversized fuel tanks and avoid the extra dead weight (in fact, by making less of the existing weight 'dead').

I like this idea, and used it for the fun of the concept once. You wind up burning way more fuel than xenon, which lowers the effective Isp of the ion down to roughly that of the nuke. Though not the most practical, I used it to make a huge floating platform on Minmus, which was actually really fun to fly around the flats.

I like your analysis, and would just like to add one wrinkle, which is on crazy-efficiency craft ions generally aren't used with enough power generation to keep up with their enormous consumption. They're instead used in 'short' little burns each time the craft crosses its periapsis, often 10-20 in total to lift the Ap enough for escape. This doesn't work well for capture of course, so it needs to be used in conjunction with other techniques, but eccentrically efficient ion craft will slowly 'spin up' their orbits in this way to save a little bit of dry mass. Not that it helps too much, mind! Most of the drymass is in those danged xenon tanks.

 

On 11/10/2017 at 10:35 AM, AeroGav said:

Nice!

On 11/10/2017 at 10:35 AM, AeroGav said:

What people should do IMHO is design a lander (test with cheats) that can do low Tylo orbit to surface and back again, then try to turn that into an SSTO.

Yep, I've been avoiding this because Tylo's the nasty part. I think the rest of my craft is about as good as I can get it though so let's see...

...

Ha! I just went and attempted a landing. It has enough TWR and deltaV to land (I full-stopped midair a little ways above the ground), but it was about 1km/s short on the good stuff for making it back up again. Once up, it'd have plenty of LF to spare for the trip home, but it's definitely missing a big enough chunk of LF&O that I'm thinking the proverbial mountain may be too steep for me to climb.

...

I've also now tried some other combinations of engines, but nothing's helping by a significant margin. I think it's time for me to throw in the towel! I went ahead and put the things I found were working well in the spoiler. Hopefully someone can use them with more future breakthroughs to find success. But for now, looks like you're off the hook, @JacobJHC! All in all I'm very happy; this is the best SSTO I've built so far and I learned quite a bit in the course of making it.
 

Spoiler


Big things:
Pre-tilted wings are great. The most aerodynamically efficient angle is ~3 degrees, but on SSTOs, 5 degrees (a single snap) works nicely
mk0 liquid fuel tanks for low drag and lower-than-normal drymass
OscarB rocket fuel tanks for low drag
BigS strakes for best extra fuel
Checkerboarding grandparent autostruts for added structural strength (less wobbles)

Little things:
With long booms for the tail, you can get away with tiny control surfaces for large planes. Just like on a glider IRL.
Small landing gear is exceptionally light for the strength, increasing spring helps so only 3 of them can hold a 100t craft
Aerospikes have a great combination of reasonable TWR + good Isp, with low drag at a 1.25m size
Placing fairings around nukes will lower their drag
Sepratrons may be very handy for final landing on Tylo. Most of the other engines point in an inconvenient direction for landing on wheels.

Engines:
Ions are worth it by a small margin and should definitely be tried at some point. It happens to be that I don't think they'd provide enough benefit to make this particular craft into a success. Still, this combination of engines was working very well for me:
1 Aerospike
2 Rapier
2 LV-N

Flight path:
Taxi out to the berm to get extra distance for takeoff
6-8km is nice for going transsonic with either a dive or a kick from the LV-Ns
17km is nice for pushing up to 1400m/s
24km is nice for going the airbreather top speed
32km is nice for nukes

You can bleed off a lot of velocity at Tylo by starting in a higher orbit (~100-150km) and using the nukes. This uses more efficient engines, and lowers the amount of oxidizer needed (important because LF is lower drymass per ton), and I believe is ultimately more efficient despite losing to the oberth effect. Also it's smart to bring 1.5x to twice the local gravity in TWR if possible, as it saves a lot of burn time = fuel = drymass.

 

Best of luck to everyone else, and thanks for the challenge, @DAL59!

Edited by Cunjo Carl
Link to comment
Share on other sites

15 minutes ago, DAL59 said:

Where is the highest mountain on Tylo?

Near the equator, it's nearly 10 kilometers tall if I remember correctly. I tried to land here several times in one of my youtube videos once. I can check the video and tell you if you want.

Link to comment
Share on other sites

Temporary Kerbal Maps has the highest point at about 50 degrees south latitude, at an elevation of 11,280-ish metres.  The better part is that from the map, it appears that everything in the immediate area slopes at five degrees or less.  However, there are some cliffs to the northeast, which may provide a jumping-off point for a rover-turned-rocket, if we're still entertaining that idea.  I've never visited this area, so the map is all I have right now.  In theory, Tylo's orbital speed is, where we assume circular orbits to start, GM is the standard gravitational parameter for Tylo (2.82528x10^12), Tylo's surface radius is 600,000 m, and ^(1/2) is understood as taking the square root:

At 0 m:

v = (GM / a)^(1/2)

v = (2.82528x10^12 / 6x10^5)^(1/2)

v = 2170 m/s

At 11,280 m:

v = (GM / a)^(1/2)

v = (2.82528x10^12 / 6.1128x10^5)^(1/2)

v = 2150 m/s

Which saves twenty metres per second on landing and another twenty on takeoff.  That does not sound like much but remember that we're at the absolute edge of capability:  we still don't know whether this is technically possible.  For example, one thing I intend to explore is the touch-and-go landing where the landing gear contact the ground but the vessel does not stop before taking off again.

Link to comment
Share on other sites

Going in the opposite direction...

How much does gear mass, exactly?

I'm just thinking: gear is really only needed on takeoff. You don't actually even need it on landing; you can ditch in the ocean at survivable speeds if you're careful enough and you do a little design work. For the rest of the trip, it's dead weight.

What if you take off the landing gear, make it a tailsitter VTOL, and add a single Dart? Might be a slight increase in mass, but that Dart will give you a HUGE advantage for your landing burn over RAPIERs alone, both in terms of isp and in terms of terminal TWR.

A tailsitter VTOL can have a TWR of 1.001; you just punch all the engines and then slowly tilt forward until you build up enough forward speed for lift to take over. Then you'd have a TWR of closer to 3-4 by the time you get to Tylo.

Link to comment
Share on other sites

Unless some sort of out of the box trick or physics glitch is used, this is not possible.  I tried =(  Someone already linked my mission that used mining on Tylo (https://youtu.be/_OnGs7L6Wx8), and I think that suitably demonstrates how this is not possible without mining.

 

Edited by EvermoreAlpaca
Link to comment
Share on other sites

It would be cool if this was possible ofcourse, and I know that this is the discussion at hand, I'll see if I can add advice of my own.

But does it really need to be a SSTTSAB (Single stage to tylo surface and back)
I also find the term SSTO vague. Single stage to orbit, hmm, okay then! So why does the same term apply when it goes single stage to orbit, then goes to Tylo to get back in one piece?

Why is it still called a SSTO when it goes beyond orbit to land elsewhere and get back home again?
Why not SSTELB (single spaceship to Extraterrestrial Lithospheres and back)

Remember: It is single stage to orbit. It already did that when it entered LKO. There's nothing more to prove to qualify for the term SSTO, regardless of what the vessel does next.
SSTO is getting to orbit from a surface without shedding any hardware.
A SSTO is still a SSTO when it stages after getting to orbit. Whether that is when achieving LKO or entering Tylo orbit.

Hereby a idea you might allow if the original manner/method of this challenge proves unreachable.


Try making a two stage SSTO. So that's a SSTO that can get to LKO or LTO but can stage into two separate vessels.
One stage is the ship itself and all the raw engine power, landing gear, fuel tanks to get to tylo surface and back.
The other stage is what holds your interplanetary ION engines (assuming you'll go the ION route), solar panels and all the stuff you don't require on Tylo's surface.

Land, get back to orbit again, dock with the module you left in LTO and thrust your ass back to Kerbin.

Land as you took off, with everything you brought with you, minus the fuel of course.
Result: Same as a SSTO in that you bring everything home. You just stage at the destination to make it easier for you.

Link to comment
Share on other sites

8 hours ago, HeroBrian_333 said:

is making a detachable lander that redocks to the SSTTO (single stage to Typo orbit) count?

No.

 

10 hours ago, Helmetman said:

It would be cool if this was possible ofcourse, and I know that this is the discussion at hand, I'll see if I can add advice of my own.

But does it really need to be a SSTTSAB (Single stage to tylo surface and back)
I also find the term SSTO vague. Single stage to orbit, hmm, okay then! So why does the same term apply when it goes single stage to orbit, then goes to Tylo to get back in one piece?

Why is it still called a SSTO when it goes beyond orbit to land elsewhere and get back home again?
Why not SSTELB (single spaceship to Extraterrestrial Lithospheres and back)

Remember: It is single stage to orbit. It already did that when it entered LKO. There's nothing more to prove to qualify for the term SSTO, regardless of what the vessel does next.
SSTO is getting to orbit from a surface without shedding any hardware.
A SSTO is still a SSTO when it stages after getting to orbit. Whether that is when achieving LKO or entering Tylo orbit.

Hereby a idea you might allow if the original manner/method of this challenge proves unreachable.


Try making a two stage SSTO. So that's a SSTO that can get to LKO or LTO but can stage into two separate vessels.
One stage is the ship itself and all the raw engine power, landing gear, fuel tanks to get to tylo surface and back.
The other stage is what holds your interplanetary ION engines (assuming you'll go the ION route), solar panels and all the stuff you don't require on Tylo's surface.

Land, get back to orbit again, dock with the module you left in LTO and thrust your ass back to Kerbin.

Land as you took off, with everything you brought with you, minus the fuel of course.
Result: Same as a SSTO in that you bring everything home. You just stage at the destination to make it easier for you.

No.  That has been done before by @Matt Lowne.  That's what the no docking rule means.  Every part must land on Tylo.

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...