Flight Question Hayabusa (ion engines, slingshots)

[Pipcard]

I want to fly the Hayabusa from BrianJ's M-V mission pack. But there are some issues:

- Ion engines. They are very efficient, but very weak (i.e. you have to thrust continuously for several months). I've read that [ame="http://www.orbithangar.com/searchid.php?ID=3815"]IEAT[/ame] MFD can be used for this.

- How would you use IEAT + IMFD to transfer back to Earth, arriving on May 19, 2004 in order to perform a slingshot that would have me arrive at 25143 Itokawa on September 12, 2005? The IEAT description says to use it in conjunction with the Delta Velocity program in IMFD, but that involves manually setting the burn date and velocity vectors, which is impractical for trying to be accurate with the actual date when performing the Earth fly-by.

 

[boogabooga]

Piper has a low thrust trajectory planner, but it has issues.

I think low thrust is still a bit of a frontier in Orbiter.

 

[dgatsoulis]

I have not flown this specific mission but from my past experience with IEAT and IMFD, I remember that it was a very long and somewhat tedious process.

In effect, you are adjusting 5 variables from two different MFDs (3 burn vectors and the time from IMFD's Delta Vel program, plus the engine thrust from IEAT MFD), while you are trying to match 4 values at IMFD's Map program (PeT, PeV, PeA and EqI) with those of actual mission.
The most important one to get right is the PeT; the rest will "fall in place" rather easily after that.

It would really help if you have all the data you can get your hands on about the Hayabusa trajectory. Specifically the dates/duration of the ion engine burns, whether those burns were at 100% thrust.
Also the date, periapsis altitude, velocity and eq inclination of the slingshot.

For timing the arrival date with the map program, you can convert it to MJD, subtract the "now" MJD and multiply with 86.4e3 to get the number of seconds, which is what the map program uses when it shows you the PeT until your next encounter with your target.

IMFD's map program uses a 4 digit format for the time. For this mission, the time interval between "waypoints" is measured in tens of millions of seconds, so you have two digits after the decimal point.
That means that you have a margin of error of ~5000 seconds due to round-off which is not that big a deal -about 1 hour 23 minutes.

In general, try to break the mission into segments and replicate the actual trajectory of the first segment as accurately as you can. This way the next segment will require a lot less tweaking than the previous one.
For the slingshot, make sure you have "Period Limit" and "Hyper Limit" set to "No" in the map program's configuration page.
This will allow you to view the trajectory on the map program farther out after the encounter with the planet.

Hope someone with more experience on ion engine missions can help you a bit more.

 

[Pipcard]

Can we assume it's always at 100% thrust?

BrianJ's Hayabusa comes with a slingshot/fly-by scenario.

Unfortunately, I couldn't find any info regarding launch heading or profile. The heading and guidance file for the launch scenario do work, but plenty of room for improvement! My main source of info was the two orbit diagrams included in this folder and the following items:

Launch- Kagoshima 9th May 2003 04:29:25 UT. Reported final velocity 6.9 miles per sec.

Earth slingshot- 19th May 2004 at 06:23 UT, min. alt. 3725km (the slingshot scenario included is too high and too early, but does give a usable trajectory for rendezvous).

 

[boogabooga]

NASA has a free trajectory/mission analysis planner tool:

http://gmat.gsfc.nasa.gov/

If one gets good, perhaps they can plan low thrust trajectories in it. However, the optimization tools that you would need are not included; you would also need either MATLAB or to contact some institution in the UK to get permission.

 

[dgatsoulis]

A burn that takes months at 70% thrust will have a very different result from a burn of the same duration at 100% thrust. That's why you need to find the actual trajectory data. So you at least have some starting point to plan your burns.

If you can't, you are starting "blind" trying out different variables in IMFD's Delta Vel. program until you manage to get an encounter with the first target (in this case Earth), with the encounter properties as close as you can to the real mission.
Since you want to sling the Earth and arrive on target on very specific dates, you'll need to replicate the actual trajectory as accurately as you can. Otherwise the date of the first encounter won't match and that will throw off completely the encounter with the asteroid.

Basically you have two ways to do this.
One is to accurately replicate the trajectory, using whatever data you can find as your starting point and the other is to start blind with a lot of trial and error until you get something that resembles the actual trajectory.

I haven't used Piper's trajectory planner that boogabooga mentioned, so I can't tell you how much help it can be. It looks like it's worth a try.

---------- Post added at 23:29 ---------- Previous post was at 23:11 ----------

I just downloaded the Hayabusa. Looks like I underestimated the spacecraft's capabilities. I was thinking of thrust in the mN range but this one has a full 22 Newtons which give it a burn time of "just" 1.25 days. Which scenario are you using as a starting point?

 

[Pipcard]

You're talking about the main engines, which have RCS thrust. The ion engines are actually the hover engines. I don't understand why.

Hayabusa's Ion Engine is mounted "underneath" the craft and is represented by the Hover Engine, a Main engine (rated the same as the RCS thrusters) is implemented for compatability with IMFD, there is no Retro Engine.

 

[dgatsoulis]

Probably because BrianJ thought that most people would fly this just with IMFD, using the 22N engines instead of the real thing. It's a lot easier this way.

Anyway, I'll have another look during the weekend when I'll have time to research the actual flight. Until then, you can follow the general outline I mentioned earlier and also have a look at Piper's trejectory planner. (I'll have a look at it too.)

Question: do you know how to fly this mission with the default Deltaglider and a 4.8 km/s dV budget?

 

[BrianJ]

Probably because BrianJ thought that most people would fly this just with IMFD, using the 22N engines instead of the real thing. It's a lot easier this way.

Actually, the Ion engine is implemented as the hover engine simply due to the way multistage2.dll orients the payload on the MV5 launcher - but I must admit, I've never flown this one without cheating and using the RCS ;-)

I've managed the Dawn mission using IMFD "Delta Velocity", but there are, as you rightly observe, a LOT of parameters to tweak to set up a rendezvous.

Haven't tried the Hayabusa mission for years, so I'll be interested to see if you can come up with a solution.

All the best,
Brian

 

[Pipcard]

Why is the "Delta Velocity" program (and not the other IMFD functions) needed , anyway?

 

[BrianJ]

Why is the "Delta Velocity" program (and not the other IMFD functions) needed , anyway?

IMFD "Delta Velocity" together with IMFD "Map (Plan View)" is the only function that treats the burn as having a finite time rather than an instantaneous impulse. This is important if the burn duration is large compared to your orbit period.

 

[dgatsoulis]

Because IMFD is not designed for ion engine spacecraft. It cannot handle burns that take months to perform. The only program that works (kind of) is the Delta velocity used in tandem with the map program.
But you also need IEAT MFD to orient the spacecraft and perform the burns at x100000 time-warp, otherwise you'll have to perform the burns at a much slower time-warp making the whole thing completely impractical.
For example, with Hayabusa's ion engines a "small" 100 m/s correction takes about a month of burntime. Even at x100 time-warp you'll need roughly a third of a day in realtime to perform it.

Still waiting for an answer on the question I asked in my previous post.

 

[Furet]

I'm using a combination of these 3 tools:

• Low Thrust Trajectory by Piper: http://www.orbiter-forum.com/showthread.php?t=15036&referrerid=7609
• Orbiter Navigator by Mindblast: http://orbiter-forum.com/group.php?do=discuss&discussionid=113&pp=20
• IEAT MFD by BrianJ: [ame="http://www.orbithangar.com/searchid.php?ID=3815"]IEAT MFD[/ame]