Sail ho! Sunjammers -- feasible or not?

[Guest]

Star Theory has said they intend to support torchships and brachistochrone trajectory planning in KSP2 -- pretty much required for some of the propulsion technologies used for interstellar missions. This got me thinking about sunjammers -- spaceships using solar sails for propulsion. These are very much a realistic technology, they're already up there in fact, and with advances in materials and manufacturing it would be entirely feasible to have even fairly substantial sunjammers plying the spaceways in the inner Solar System at least.

The obvious advantage of a sunjammer is infinite dV -- there is no expenditure of reaction mass or other consumable resources, the energy is all stolen from the sun.

Sunjammers would introduce a whole bunch of complications that reaction mass propulsion doesn't have, however. They're obviously an nonstarter in KSP1 because the TWR would make the Dawn look like a Funny Car, but with that hurdle cleared in KSP2, what else is there?

The main issue is that the thrust vector of a solar sail depends on multiple factors:

1. Sail area -- this is simple enough
2. Distance from the Sun -- also simple enough
3. Sail angle relative to the sun -- not a problem for the physics calculations, but how would this work in the UI, the game, and for trajectory planning?

Solar sails produce thrust by reflecting photons.* If they face the sun exactly, they produce the maximum thrust vector which points directly away from the sun. If you rotate them, the thrust vector gets shorter and will point towards the apex of the angle between the incoming and the reflected photons, until it's edge-on at which point it produces no thrust at all.

So in practice your possible thrust vectors are in a cone pointing away from the Sun. Since there's no keel, there's no way to tack into the wind like with a sailboat; however you can reduce your angular velocity relative to the Sun even all the way to zero, which means it would be possible to dive all the way into the sun.

I don't think the physics of a sunjammer would be particularly hard to simulate in KSP (even KSP 1), but how would you make them usable in a game? The trajectory planning is going to be quite tricky, and to do it at all you'd have to be able to nail down some parameters about the sail when you're doing it, and then have automation to keep it there for the duration of the trip. Would it be enough to be able to define a manoeuvre node that specifies the orientation of the sail relative to the sun, and then have the planner compute and display the resulting trajectory, taking into account the changing distance from the sun? The sunjammer would then maintain that orientation according to the planner. 

Things would be seriously funky if you're orbiting a body which is orbiting the sun -- i.e., your available thrust for any given vector relative to your local orbit varies between zero and maximum for the sail depending on your phase angle.

Are there other complications that I've missed? Would this be feasible, in the base game or as a mod? Would it be interesting or fun?

*n.b. I am aware that there are also plans for solar sails that use charged particles rather than photons for propulsion, but that's a tangent I don't want to go into right here

[The Aziz]

I think the main problem here would still be acceleration. The Dawn engine in already pretty much magic, no ion engine has this much twr.

Yes, we're going to have timewarped acceleration, but if we're going interstellar, I'd still use other means of transport ("conventional" isn't the best word here). It's about getting up to a percentage of lightspeed, so we're talking about years of difference between using solar sail and let's say Orion Drive.

I bet the mechanics regarding sail alignment would be done right.

[Guest]

Star Theory also said games can run for thousands of years. With painless warp slow speeds wouldn’t be a problem.

I was thinking of sunjammers as intra-system craft anyway although of course they could be used in interstellar missions as well.

(Also there’s no reason sail thrust couldn’t be fudged the same way as Dawn thrust.)

Edited February 2, 2020 by Guest

[Incarnation of Chaos]

Also interstellar Solar Sails would have to be propelled either by DEW's or Bombs, so you either have a more complex orion for no reason (Sail could potentially be used in reverse to slow down, but with the mass of Orion we're talking something huge here) Or you have to handle firing massive arrays of lasers across multiple SOI's, or even star systems; while being able to calculate how much the beams diffused at whatever distance the sail is and factor that in.

So while i could see them being used for planet hopping in the Kerbin system; interstellar usage would be a bit tricky. You'd almost be forced to make them procedural also; since area dictates the thrust so much. And even "Light" probes would need massive sails to be useful.

[Guest]

From what I've read, a non-assisted solar sail craft would take about 150 years to get to Proxima Centauri. If the timescale of the game is thousands of years, that would be doable.

(Also, the Kerbal universe appears to be scaled down by a factor of 10. Fifteen years to get there doesn't sound bad at all.)

Edited February 2, 2020 by Guest

[Incarnation of Chaos]

20 minutes ago, Brikoleur said:

From what I've read, a non-assisted solar sail craft would take about 150 years to get to Proxima Centauri. If the timescale of the game is thousands of years, that would be doable.

(Also, the Kerbal universe appears to be scaled down by a factor of 10. Fifteen years to get there doesn't sound bad at all.)

I'm highly suspicious of that; either the sail would have to made of some impossible material or there's something i'm missing. The inverse square law would mean by the time you got about out to Pluto the thing would be mostly dead weight, and once you crossed the heliopause there wouldn't be a solar "Wind" to catch at all.

[Guest]

1 hour ago, Incarnation of Chaos said:

I'm highly suspicious of that;

source: https://arxiv.org/abs/1701.08803

[The Aziz]

7 hours ago, Brikoleur said:

 

(Also, the Kerbal universe appears to be scaled down by a factor of 10. Fifteen years to get there doesn't sound bad at all.)

It's four years to Jool. I somehow doubt that the closest solar system would be four times the distance from Kerbin to Jool.

Keep in mind that speeds are also scaled. You must go about 9km/s to orbit Earth, while you need only 2.4 to orbit Kerbin.

[DStaal]

Solar sails within a system aren't likely to actually be all that interesting or useful - for the amount of dV needed, you can easily carry the fuel for a higher TWR.  It's only really interesting for interstellar (and boosted) - sure the acceleration isn't much, but you can keep it up quite a while.

There is a mod that puts solar sails in KSP1 - I've played around with it, but the hard part is planning the trajectory.  Timewarp under thrust is fine, but you aren't flying Hohmann transfers or anything similar anymore.

[Guest]

22 minutes ago, DStaal said:

There is a mod that puts solar sails in KSP1 - I've played around with it, but the hard part is planning the trajectory.  Timewarp under thrust is fine, but you aren't flying Hohmann transfers or anything similar anymore.

Yep that's my thinking too. For them to work, there would have to be tools both to plan and execute trajectories under solar thrust, and I'm not sure how those would look like.

[mattinoz]

2 hours ago, DStaal said:

Solar sails within a system aren't likely to actually be all that interesting or useful - for the amount of dV needed, you can easily carry the fuel for a higher TWR.  It's only really interesting for interstellar (and boosted) - sure the acceleration isn't much, but you can keep it up quite a while.

There is a mod that puts solar sails in KSP1 - I've played around with it, but the hard part is planning the trajectory.  Timewarp under thrust is fine, but you aren't flying Hohmann transfers or anything similar anymore.

Doesn't that depend on how various other mechanics work in the game. Could be fun* to have an automated fleet of them taking containers of parts out to resupply bases with no fuel load.. Slow sure but it sounds like game will hopefully push more of the planning logistics aspects of space travel compared to KSP1 "Just fly" attitude.

[Skorj]

For interstellar travel, a remarkably large laser to push the solar sail just sounds so very Kerbal, especially if there's a way for it to end in fiery death.  And it's the fastest mode of interstellar travel of the near future techs, by far, as you're freed from the tyranny of the rocket equation.  Real world, they're limited by drag and the ability to focus a laser at extreme distance, but ignoring that stuff would be fine by me for KSP2.

 

[Guest]

 

13 hours ago, DStaal said:

Solar sails within a system aren't likely to actually be all that interesting or useful - for the amount of dV needed, you can easily carry the fuel for a higher TWR.

Useful, compared to nuclear/chemical/ion propulsion, in the kerbal context where funds are no object? Probably not. 

Interesting? Here I differ with you. The reason I brought them up was because I think they would be an interesting design and operating challenge in their own right, regardless of how they compare to reaction-mass alternatives. The fact that we actually have them up there right now is an added bonus.

[Rejected Spawn]

Gonna have to man up and admit this topic has grown on me over the past 24 hours. Though interstellar use immediately struck me as foolish and seems like an even worse idea the more I look at all the factors.

The way I'd personally use solar sails would be for very small probes (comms and orbital scanning and stuff) but not as a way to get to a destination; it would make far more sense to get a probe where it's needed when it's needed than sit and wait forever just because the "fuel" is free - I'd get my probes to where I want them with more traditional rockets and have the sails packed on board so that when the situation changes and I need to alter the orbit of a probe I'd unfold the sail and be able to adjust it to my hearts content. Of course the option to use it for bigger moves still exists, for those who like sitting around waiting, even if I'd personally never bother. At any rate packing a solar sail on every small probe would be the sensible thing to do, no probe would need to become obsolete just because it's out of fuel in slightly the wrong orbit and this just seems a little romantic to me. I feel a sense of shame in terminating probes when they can no longer fulfill their role just because I didn't get them to the perfect long term spot...

This brings up other problems though, "sailing" in an orbit around something other than the star that is providing the thrust is a lot more bothersome because of two factors... One: Part of the orbit is likely in the shadow of whatever planet or moon it's located around. Two: The gravity pulls in a different direction than opposite the strongest potential thrust and it constantly changes through the orbit. I actually watched a video about this stuff some time ago laying it out, it's not that it's in any way impossible to "sail away" from a planet, it is however complicated to do it correctly without dipping any part of the orbit lower and the closer the probe is to the planet the more difficult everything about it becomes. The probe would have to keep flipping around in specific directions multiple times per orbit and for numerous orbits to start making a real difference, this would surely appeal to some more hardcore players to do manually but for me and other more casual players it's something that would lose novelty quickly. Having a powerful enough probe core on board that the probe can make its own adjustments automatically would help enormously, if implemented into a tech tree this could unlock once the player personally actually puts a sailor in some specific situations and succeeds in orbit manipulation; the KSC geniuses then use the flight data from the players movements to write a program that they automatically upload to all high tier probe cores already in use and of course install in all new ones launched later.

Of course this is some pretty high level stuff to ask of Star Theory while they're already working themselves half to death but keeping it on the radar and making any necessary preparations to the core of the game ahead of time might be good, that way it might be a little less painful to add this kind of stuff a few patches later when things have calmed down a bit. It would not only require all the work in having the game pointing a sail in the right directions depending on multiple factors and then ensuring that the sail is inactive in the shadow of a planet, it would also involve adding an ability for a craft to calculate on its own how to reach a target orbit... and it would require a novice friendly UI capable of letting the player graphically set the target orbit. Since the Star Theory fellows have been quiet for a long time we don't know if some of it is already in there but I'm leaning more to "probably not", so the workload of adding all the necessary components would probably be a big ouchie.

[Guest]

1 hour ago, Rejected Spawn said:

I actually watched a video about this stuff some time ago laying it out, it's not that it's in any way impossible to "sail away" from a planet, it is however complicated to do it correctly without dipping any part of the orbit lower and the closer the probe is to the planet the more difficult everything about it becomes. The probe would have to keep flipping around in specific directions multiple times per orbit and for numerous orbits to start making a real difference, this would surely appeal to some more hardcore players to do manually but for me and other more casual players it's something that would lose novelty quickly. Having a powerful enough probe core on board that the probe can make its own adjustments automatically would help enormously, if implemented into a tech tree this could unlock once the player personally actually puts a sailor in some specific situations and succeeds in orbit manipulation; the KSC geniuses then use the flight data from the players movements to write a program that they automatically upload to all high tier probe cores already in use and of course install in all new ones launched later.

Yep this is the biggest difficulty that I can see. Managing sunjammer trajectories manually would be unbelievably boring and frustrating; there would have to be a fairly high degree of automation in the trajectory planner. It would also be quite tricky to find the right amount of automation for maximum fun -- "plot me the fastest course to Duna and execute it" is boring, but having to lay down a gorillion manoeuvre nodes and then execute them manually would be even more so.

[Dirkidirk]

18 hours ago, The Aziz said:

It's four years to Jool. I somehow doubt that the closest solar system would be four times the distance from Kerbin to Jool.

he means 15 earth years. one kerbin year is 107 earth days, so it takes about 1 earth year to get to jool, but it still sounds too close.

Edited February 3, 2020 by Dirkidirk
no regrets

[Guest]

51 minutes ago, Dirkidirk said:

he means 15 earth years. one kerbin year is 107 earth days, so it takes about 1 earth year to get to jool, but it still sounds too close.

@The Aziz is thinking of Hohmann transfers. Four years to Jool is a good approximation for that. These aren't Hohmann transfers however, they're brachistochrone trajectories with continuous acceleration. They would be a lot faster.

(I don't know exactly how long it would take to get to Jool, but to put things into perspective, if you accelerate at 0.001g for one Earth year, your terminal velocity will be over 300 km/s -- that's in the neighbourhood of 30 times faster than your top speed on a Hohmann transfer from Kerbin to Jool.)

 

Edited February 3, 2020 by Guest