The Martian orbital mechanics

[matthiasduyck]

Hi!

Hope this is in the right place...

So I recently read The Martian, and watched a lot of the talks of the author. I get most of how the mission works but I am still unsure of some parts. I have a very basic understanding of orbital mechanics and how ion engine based ships differ from traditional rocket engines.

It is unclear to me if, when and how you can leave or get into an orbit accurately with a lage ion engine based ship with the thrust being so low. For example, to leave earth's orbit and to get into a solar orbit do you first have to make your orbit really really large? Once near Mars how do get into a proper orbit there with so little thrust?

What am I missing here?

Also how would you operate a craft like this in KSP? Are there tools that help in finding the right thrust vectors and help steering over a long period of time?

Thanks

[Red Iron Crown]

Welcome aboard!

The key is using more sophisticated planning tools than the simple maneuver nodes we currently use. Nodes assume an instantaneous burn, which for most cases in KSP is a reasonable approximation. But if you plan to apply thrust for weeks or months then that assumption becomes highly inaccurate.

Basically, it stops being a classic Hohmann transfer and becomes something closer to a brachistochrone trajectory. (Very roughly: Accelerate toward the target for half the trip time, then braking burn for the other half.)

[Chaos_Klaus]

Also, Hermes uses aerobraking to capture into an orbit.

This is referenced in the book, when they blow the inner hatch of the airlock instead of the outer one. The reason is that they want to keep the outer shape of the craft intact for aerobraking.

Edited October 7, 2015 by Chaos_Klaus

[Pax Kerbana]

Well think about it in terms of KSP, you can leave Kerbin orbit from either: accelerating to a high velocity, creating an elliptical orbit with large apopsis, creating a large circular orbit and increasing velocity, or any in between. It boils down to exceeding escape velocity and what distance you are away from the planet.

My understanding is that the Hermes is basically a big ship for traditional transfers. It had enough delta V to enter orbit at destination. The burn time was continuous because they had very low TWR to work with but massive ISP. The Mars flyby they pull is more akin to the Aldrin "Cycler" concept. Basically the Hermes being a very inefficient method to get to Mars, comfort and ion propulsion notwithstanding.

Edited October 7, 2015 by Pax Kerbana

[Kelderek]

The part that confuses me is how they capture at Mars. The book implies roughly a month to decelerate. My limited knowledge of trajectories in space from playing KSP has taught me that you can't just speed up or slow down in space and hope to keep the same trajectory. The simple act of firing your engines in any direction will always alter your path, this would be true even with very low thrust.

In KSP terms would they be starting a long slow burn from outside the SOI of Mars? What would their trajectory look like before starting that burn, in other words, what kind of periapsis would they aim for? I'm guessing that they have to start with an insanely high altitude periapsis and hope that it gets closer as a result of the burn.

It's hard for me to picture a 1 month burn in the context of planet where the final orbit would likely have a period of 1.5 hours give or take. I'm guessing that it takes a series of long burns. One burn to just get a capture, followed by multiple burns over multiple orbits to shrink it down to a low enough Mars orbit to do the mission there.

I suspect that digging deep into NASA's Dawn mission would reveal a better explanation.

[cantab]

Yeah. I've run my share of ultra-low-TWR ships. After they're launched to LKO I set them to hold prograde and just spiral outwards, before I get into a high orbit from where I can execute a more normal manoeuvre. With realistic ions you'd be doing the same spiralling outwards approach in solar orbit too.

[Teilnehmer]

(Very roughly: Accelerate toward the target for half the trip time, then braking burn for the other half.)

Wouldn’t it be closer if we say “accelerate all the time�

[Red Iron Crown]

Wouldn’t it be closer if we say “accelerate all the time�

Technically, yes. I used the word in the colloquial manner of accelerate = add speed rather than the more correct accelerate = change speed.

[Felsmak]

My limited knowledge of trajectories in space from playing KSP has taught me that you can't just speed up or slow down in space and hope to keep the same trajectory. The simple act of firing your engines in any direction will always alter your path, this would be true even with very low thrust.

My guess is that they take this into account from the beginning and plan the trajectory so that it will be passing Mars when the deceleration is complete.

[manaiaK]

The dumb answer to the question ""How does NASA do it?" is: by numerical solution of the orbit on their Pleiades supercomputer.

(NASA also has to contend with real gravitation, not patched conics, so the problem is a lot harder for them than it would be in KSP. Something labelled "Pleiades" makes a cameo appearance in the movie, but it looks a bit different to other photos of Pleiades.)

[mellojoe]

If you look up the Dawn probe, it currently is using ION engines and has performed these types of maneuvers.

This is my favorite picture from this mission, where it shows the trajectory and path of the craft:

Also, the Dawn Blog is well written and updated often, and explains much of not only the science parts of the mission, but also the hardware and craft.

Dawn.JPL.NASA.gov

PS: My favorite fact about the Dawn probe? Its TIE Craft, similar to the TIE fighters of the Star Wars universe, except it is a TRIPLE-ION ENGINE, not just twin-engine like the Empire uses.

Edited October 8, 2015 by mellojoe

[Sampak]

A game currently in development called Rogue System (referred to as "DCS in space") right now will let your ship calculate a trajectory for a constant G transfer (accelerate half way then decelerate) to other planetary bodies. The game does use very realistic N-body physics and not "patched conics" (I had no idea what this was until recently and still don't know how it works, but I'm ok with that). So I believe with the right data, IRL it may be very possible, if very complex on the computational level, to do what above posters may be asking regarding a constant acceleration/deceleration maneuver to Mars.

[matthiasduyck]

Thanks to everyone for answering, I think I get it!

[Teilnehmer]

Technically, yes. I used the word in the colloquial manner of accelerate = add speed rather than the more correct accelerate = change speed.

I mean, the vessel actually have to increase its speed all the way to follow a spiral-like trajectory. If it starts to brake (decrease its speed) in a solar orbit between Earth and Mars, it will not reach Mars. Am I wrong?

When you do a Hohmann transfer, you also increase your vessel’s speed (in the Solar frame of reference) both times, aren’t you?

Edited October 9, 2015 by Teilnehmer

[cantab]

Hmmm. If we think about the trip from Earth to Mars, our spacecraft needs to be going slower relative to the Sun when it reaches Mars than when it leaves Earth. It follows that during its journey an ion ship will be thrusting prograde but slowing down on average, but slowing down less than if it just coasted.

[Teilnehmer]

True.

[Kryxal]

Then again, at that point in the journey thrusting solar-prograde WILL be slowing down relative to Mars. It's when you get into interstellar travel that you actually start braking as such...