Most active commenters
  • FredPret(3)
  • trhway(3)

←back to thread

164 points pseudolus | 29 comments | | HN request time: 0.813s | source | bottom
1. FredPret ◴[] No.42475004[source]
If KSP is to be believed, this is shockingly difficult to do
replies(6): >>42475159 #>>42475235 #>>42475594 #>>42475756 #>>42476164 #>>42480023 #
2. lizzas ◴[] No.42475159[source]
The same gravity that wants to pull you in keeps you in orbit. And any object we launch starts off in the suns orbit.
replies(2): >>42475172 #>>42475242 #
3. FredPret ◴[] No.42475172[source]
It's only the v0 you start off with from being in orbit already that makes it hard. Earth (and Kerbin) orbit at a very high speed.
4. beAbU ◴[] No.42475235[source]
In order to 'land' on the sun, or any celestial body, you need to get rid of your orbital speed. Higher orbital speed means higher orbit altitude. Landing on earth is comparatively easy, because you can use the atmospheric drag to slow down. It is so difficult to land on Mars because of it's thin atmosphere. Alternatively you need a shitload of fuel to burn to kill that velocity.

Earth's orbital velcity is ~30km/s. So by extension, anything that comes from Earth will at least have that speed. So the probe needs to find 30km/s delta v in order to actually get close to the sun.

replies(3): >>42475468 #>>42475867 #>>42476089 #
5. feoren ◴[] No.42475242[source]
> The same gravity that wants to pull you in keeps you in orbit

I would say it's your velocity that keeps you in orbit. Without the velocity, you fall into the star. Without the star's gravity, you keep going away in a straight line. Any object we launch starts off with Earth's velocity.

replies(1): >>42476457 #
6. FredPret ◴[] No.42475468[source]
I wonder if you can use atmospheric drag to pull you into the Sun / a different star / Kerbol.

Long ago, playing Elite if I remember correctly, you could fly close to a star and scoop up a load of hydrogen for later resale. I'd be interested to see a graph of gas density vs tendency to melt spacecraft compared to distance from the core for a typical star.

replies(2): >>42476652 #>>42476846 #
7. aziaziazi ◴[] No.42475594[source]
In SpaceFlight Simulator it’s quite easy IIRC (it’s been a while):

1. Orbit yourself around low earth

2. When entering the transfer window (opposite side of the sun-facing earth, i.e. above midnight longitude) booooost

3. For orbit, aim for tangent with your target. For sun discovery, aim for sun center. Choose but don’t change.

The game is in 2D and you got nice auto-calculated transfert windows and trajectories. Is it one of those game simplification that makes it easy or there’s more difficulties?

replies(1): >>42475907 #
8. kccqzy ◴[] No.42475756[source]
There's a difference between flying directly into the sun as if landing there, and orbiting the sun but so close that the spacecraft is inside the solar atmosphere.

This is doing the latter.

9. nostromo ◴[] No.42475867[source]
You could use a solar sail to project a satellite towards the sun.
replies(1): >>42476123 #
10. LorenPechtel ◴[] No.42475907[source]
The math is simple enough, it's just the delta-v requirements are brutal. Or you take the slingshot approach at which point the math requirements are brutal.

And I'm not aware of any KSP mod that helps you plan slingshots. And even if there was a slingshot maneuver requires a lot of precision because your ejection angle is highly sensitive to exactly how close you came.

The Parker probe was sent outward to Jupiter and used it to slingshot away much of it's energy. (We normally think of using a planetary encounter to gain energy but it works both ways. Ejection velocity from a slingshot at Jupiter can be anywhere from hitting the sun to solar escape. It's just most probes are heading out, not in.)

replies(2): >>42476411 #>>42476533 #
11. trhway ◴[] No.42476089[source]
>So the probe needs to find 30km/s delta v

I don't understand why we aren't doing solar + ion drive everywhere (except obviously launch), and instead we settle for slow multi-year multi-grav-boosts trajectories. Current ion drives (by NASA and on Starlink) have 2500-3500 Isp. Which means that even 100+ km/s is easy doable with just 2 stages.

replies(1): >>42476382 #
12. Jeff_Brown ◴[] No.42476123{3}[source]
How do you sail into the solar wind?
replies(3): >>42476201 #>>42476336 #>>42476429 #
13. ostacke ◴[] No.42476164[source]
It requires seven Venus flybys, and 24 orbits around the sun[1]. Wikipedia has a nice animation of the trajectory[1], and NASA has another one here[2].

[1]https://en.wikipedia.org/wiki/Parker_Solar_Probe?wprov=sfti1... [2]https://svs.gsfc.nasa.gov/3966/

replies(1): >>42476852 #
14. tedunangst ◴[] No.42476201{4}[source]
Just need a triangular sail and some zigzagging.
replies(1): >>42476290 #
15. mtreis86 ◴[] No.42476290{5}[source]
Tacking doesn't work without water and a keel
replies(1): >>42476850 #
16. gpm ◴[] No.42476336{4}[source]
You sail perpendicular to the wind, cancelling your horizontal velocity relative to the sun.

Then gravity crashes you into the sun

17. margalabargala ◴[] No.42476382{3}[source]
I assumed that if I did out the math on this, it would be clear why we don't, but then I did and I now share your confusion.

The Parker Solar Probe mass is 555kg. An achievable amount of ion thrust is around 0.5N. Thus, running that thruster would accelerate the craft at 0.0009m/s2.

Getting such a craft to 30km/s of delta-v would therefore take about 33.3 million seconds of thruster time, or about 13 months.

I don't know what the duty cycle is on ion thrusters. Maybe they aren't robust enough to fire for over a year straight?

replies(1): >>42476468 #
18. lmm ◴[] No.42476411{3}[source]
> I'm not aware of any KSP mod that helps you plan slingshots.

As someone who played before they added patched conics, I'd consider patched conics such a thing.

19. Terr_ ◴[] No.42476429{4}[source]
Perhaps a sacrificial solar mirror that detaches from the main ship.

While the unfolded mirror is pushed moving away from the Sun, it reflects enough light for the smaller main body to accelerate sunward.

20. Terr_ ◴[] No.42476457{3}[source]
It's like a joke I recall from Hitchhiker's Guide to the Galaxy: You are indeed falling towards the ground... but the trick is to miss.

Your lateral velocity is what keeps you missing, whether you want to or not.

21. trhway ◴[] No.42476468{4}[source]
>The Parker Solar Probe mass is 555kg. An achievable amount of ion thrust is around 0.5N. Thus, running that thruster would accelerate the craft at 0.0009m/s2.

To be precise for 555kg probe you'd need additional 600-800kg of propellant mass and thus run the thruster(s) at about 1.5N thrust using 40-60KW - 250m2 of solar panels - everything is available at the current state of tech.

https://en.wikipedia.org/wiki/Ion_thruster

"A test of the NASA Solar Technology Application Readiness (NSTAR) electrostatic ion thruster resulted in 30,472 hours (roughly 3.5 years) of continuous thrust at maximum power. Post-test examination indicated the engine was not approaching failure.[75][3][4] NSTAR operated for years on Dawn."

replies(1): >>42480894 #
22. aw1621107 ◴[] No.42476533{3}[source]
> And I'm not aware of any KSP mod that helps you plan slingshots

If you're willing to go for the full n-body package, Principia [0] has a pretty nice flight planner that is quite usable for planning more complicated missions.

KSP Trajectory Optimization Tool [1] is a non-mod alternative with some additional capabilities beyond flight planning as well. I think this one is designed for stock gravity so it should be usable in an otherwise vanilla install.

[0]: https://github.com/mockingbirdnest/Principia

[1]: https://github.com/Arrowstar/ksptot

23. JumpCrisscross ◴[] No.42476652{3}[source]
> wonder if you can use atmospheric drag to pull you into the Sun

You can, and I believe this probe will. The Sun’s atmosphere is just much nastier than our own, which means your aerobraking destroys your spacecraft quicker.

24. anigbrowl ◴[] No.42476846{3}[source]
It was Elite. Once you had a fuel scoop there was no requirement to dock and refuel any more so it was much easier to be a pirate or privateer.

There was a bug (or was it?) in the very PC version where by if you had fuel scoops installed, set your view to looking out the rear of the ship, flew toward a star, and ignored all the warnings on your dashboard, you could fly right through the star. If you were being chased at the time you had the additional satisfaction of watching your pursuers' ships explode as they tried to follow you in.

25. dan_linder ◴[] No.42476850{6}[source]
Won't the crafts relative motion (relative to desired travel) provide the same effective force as the water+keel?
26. DiggyJohnson ◴[] No.42476852[source]
Second link is beautiful and really shows the whiplash of the last set of solar encounters. Recommend clicking through to anyone with a passing interest.
27. zamadatix ◴[] No.42480023[source]
It's actually not too brutal so long as you don't try to do it directly. Launch a craft out past Jool then when you start falling back burn retrograde. It's like the opposite of what you do to efficiently burn when leaving Kerbin.

In real life they used a set of Venus gravity assists instead. This has allowed them to slowly get closer while observing over time instead of a long wait and then one big bang close up before being shot ridiculously far back out.

28. margalabargala ◴[] No.42480894{5}[source]
Ah, fair point.

In that case I have a theory. The extra propellant mass, extra solar panel mass, etc, are all more mass against a small amount of thrust. Every bit of additional mass extends that 1-year timeline, and all the extra stuff is extra things that might go wrong.

So instead of a 7 year mission being reduced to 18 months, we have a 7 year mission reduced to maybe 4 years, but then there's possibly a higher chance of failure.

Balancing three years against failure risk, I could see that falling on one side for some missions, and the other side for others. I'm not surprised that they pick the extra time for some missions. I am surprised that they don't pick the faster option more frequently.

replies(1): >>42481722 #
29. trhway ◴[] No.42481722{6}[source]
The numbers i mentioned give 30km/s in less than 1 year. As far as i see the ion thruster makes sense even for the trips to/from Mars.