The Dunatian, Playable Save File (Like a Scenario)

[Majorjim!]

Looking good mate!

Still two rovers though.. Maybe have one busted near the base? I'm happy to see you went for two jet engines. It's much more stable that way. I'm glad to see it's all coming together.

[JAFO]

Still two rovers though.. Maybe have one busted near the base?

Speaking personally, I'd much rather see both rovers in use, as per the book.. I was quite disappointed with the screenplay's decision to scrap one of the rovers.

- - - Updated - - -

Report on Hermes close-approach free-return trajectory testing - Round 2

I loaded up the mid-course correction named quicksave, and found that if I went with a 74m/s radial burn, instead of Round 1's 74m/s prograde burn, I got what appeared to be an 1800km close-approach behind Duna, instead of out in front. Once I actually got in close to Duna, it turned out that my trajectory didn't so much take me behind Duna, as beneath Duna.

After passing beyond Duna's SOI, I discovered that a correction of only another 80m/s or so was enough to get an encounter with Kerbin. So I arrived at Kerbin's SOI with over 1900m/s of dV remaining. Less than 80m/s was needed to get my periapse down to 100km, then came the long, 10 minute circularisation burn. By the time it was done, my apoapse was up to 158km, and the periapse down to just 48km! Fortunately, the 89m/s left in the tanks was just enough to pull the periapse back up.. I ended up in a perfect 158km circular orbit with just 3m/s remaining! And my orbital inclination was just 0.8°. A far cry from Round 1, where I ended up in a hugely elliptical polar orbit...

So in my books, Round 2 testing was a resounding, awesomely excellent, mind-blowingly fantastic success. I'm loving this trajectory so much I totally don't want to mess with it at all. In Round 3, I'm going to load up this trajectory again, stick Mark in the DAV at the official location, then try to get him to Hermes. It'll be rather like trying to shoot a bullet around 2 corners of a house, in an attempt to hit another bullet travelling past the house on the opposite side. Piece of cake!

[selfish_meme]

Looking good mate!

Still two rovers though.. Maybe have one busted near the base? I'm happy to see you went for two jet engines. It's much more stable that way. I'm glad to see it's all coming together.

Thanks Majorjim, I too like the books two rovers, he rolls in the book and has to recover, very much like Duna. It is just better with two rovers.

- - - Updated - - -

Speaking personally, I'd much rather see both rovers in use, as per the book.. I was quite disappointed with the screenplay's decision to scrap one of the rovers.

- - - Updated - - -

Report on Hermes close-approach free-return trajectory testing - Round 2

I loaded up the mid-course correction named quicksave, and found that if I went with a 74m/s radial burn, instead of Round 1's 74m/s prograde burn, I got what appeared to be an 1800km close-approach behind Duna, instead of out in front. Once I actually got in close to Duna, it turned out that my trajectory didn't so much take me behind Duna, as beneath Duna.

http://i407.photobucket.com/albums/pp159/ElYoda69/screenshot_2015-10-10--01-05-51_zps5lrysgbj.png~original

After passing beyond Duna's SOI, I discovered that a correction of only another 80m/s or so was enough to get an encounter with Kerbin. So I arrived at Kerbin's SOI with over 1900m/s of dV remaining. Less than 80m/s was needed to get my periapse down to 100km, then came the long, 10 minute circularisation burn. By the time it was done, my apoapse was up to 158km, and the periapse down to just 48km! Fortunately, the 89m/s left in the tanks was just enough to pull the periapse back up.. I ended up in a perfect 158km circular orbit with just 3m/s remaining! And my orbital inclination was just 0.8°. A far cry from Round 1, where I ended up in a hugely elliptical polar orbit...

So in my books, Round 2 testing was a resounding, awesomely excellent, mind-blowingly fantastic success. I'm loving this trajectory so much I totally don't want to mess with it at all. In Round 3, I'm going to load up this trajectory again, stick Mark in the DAV at the official location, then try to get him to Hermes. It'll be rather like trying to shoot a bullet around 2 corners of a house, in an attempt to hit another bullet travelling past the house on the opposite side. Piece of cake!

Outstanding stuff, so I'm guessing you are getting ready for a beta of the quicksave that we will be able to time forward to several hours before the game starts? I am starting to pull all the assets together for another quick save of all the new stuff and I must must do the last leg of the rover journey.

[JAFO]

Outstanding stuff, so I'm guessing you are getting ready for a beta of the quicksave that we will be able to time forward to several hours before the game starts?

Pretty much, yes.. though at this point I'd say more "planning for" than "getting ready for".. I suspect that the DAV/Hermes rendezvous is going to be trickier than anything yet seen on these forums. In trying to figure out just how to go about it, I've been comparing it with various difficult KSP manoeuvring scenarios that I'm aware of..

Consider..

Is it like a launch-direct-to-orbital-rendezvous scenario? No, not really. In those, the targeted vessel is orbiting the same planet as the launcher, the relative orbital planes have to at least be close, etc.. and the target is passing more or less overhead at time of launch. In the case of Hermes, it's not orbiting Duna, it's on a very different different orbit/plane around another body, and the DAV has to launch blind, and essentially turn corners before intercept.

Ok, maybe it's more like intercepting an asteroid on a wild trajectory? Closer, yes, but again, there are fundamental differences. On an asteroid intercept, you're simply attempting to match orbits once you're already in open space. This is more like trying to intercept an asteroid on a direct trajectory within minutes of launching off a planet.

In short, this may prove to be the most difficult single manoeuvre ever attempted in KSP. You (and Andy Weir, of course) may well have come up with something that is going to give even Scott Manley a hard time..

[selfish_meme]

Pretty much, yes.. though at this point I'd say more "planning for" than "getting ready for".. I suspect that the DAV/Hermes rendezvous is going to be trickier than anything yet seen on these forums. In trying to figure out just how to go about it, I've been comparing it with various difficult KSP manoeuvring scenarios that I'm aware of..

Consider..

Is it like a launch-direct-to-orbital-rendezvous scenario? No, not really. In those, the targeted vessel is orbiting the same planet as the launcher, the relative orbital planes have to at least be close, etc.. and the target is passing more or less overhead at time of launch. In the case of Hermes, it's not orbiting Duna, it's on a very different different orbit/plane around another body, and the DAV has to launch blind, and essentially turn corners before intercept.

Ok, maybe it's more like intercepting an asteroid on a wild trajectory? Closer, yes, but again, there are fundamental differences. On an asteroid intercept, you're simply attempting to match orbits once you're already in open space. This is more like trying to intercept an asteroid on a direct trajectory within minutes of launching off a planet.

In short, this may prove to be the most difficult single manoeuvre ever attempted in KSP. You (and Andy Weir, of course) may well have come up with something that is going to give even Scott Manley a hard time..

I suspected the book and the movie made this seem a lot easier than it is in real life, when I first read it, I said well thats some ......... The plane is the launch angle at least, then it's a matter of getting high and fast.

[JAFO]

The plane is the launch angle at least

That may be the killer, at least so far as my present best Hermes trajectory goes.. the orbital plane passes 1800km beneath Duna.. whereas my first test trajectory's plane was more or less a match for Duna's, so far as I could tell.

I may have to tweak the Hermes orbit quite a bit more than I was anticipating.

[selfish_meme]

That may be the killer, at least so far as my present best Hermes trajectory goes.. the orbital plane passes 1800km beneath Duna.. whereas my first test trajectory's plane was more or less a match for Duna's, so far as I could tell.

I may have to tweak the Hermes orbit quite a bit more than I was anticipating.

Is there a problem launching on a polar trajectory?

[JAFO]

I may have to tweak the Hermes orbit quite a bit more than I was anticipating.

Right.. a couple of minutes of thinking led to the logical solution.

In retrospect, the fact that I came in high on my first return to Kerbin, and almost bang-on the second time, after passing under Duna, meaning I got a slight kick 'upwards' from Duna's gravity, should have alerted me to the problem. I confess to only eyeballing the plane match with Duna.. at the time, I was more focused on getting a close-approach at all, than thinking about things I should have considered. The eyeball test indicated I was pretty damn close.. but as seen, by the time Hermes reaches Duna, "pretty damn close" isn't good enough.

So I dumped all my intermediate and testing quicksaves, and loaded up the quicksave made immediately after exiting Kerbin's SOI. I told MechJeb to set up a plane-matching node with Duna at the upcoming AN, and PreciseNode told me this was going to involve lots of decimal places. So I set MJ to burn with an accuracy of 0.005, and locked the engine outputs at 6% thrust. That increased the burn time to over 20 seconds, which I felt would give me some precision control, and I was right.. using Flight Engineer to watch the numbers, beginning from an inclination of 0.01937, the burn ended on 0.06000. Absolutely perfect, to the five decimal place limit of Flight Engineer's readout!

Next up is the mid-course correction, but awkward plane-changes should now be a thing of the past..

There's still the matter of the DAV launch site being situated well north of Duna's equator (interestingly enough, Schiaparelli Crater is pretty much bang on the Martian equator - I suspect Weir chose it for that reason), but that's a whole lot less problematic to deal with than Hermes passing beneath the Dunan South Pole.

So for now, I'm back to testing for passes ahead of and behind Duna, to see how they impact the return trip to Kerbin.

- - - Updated - - -

Is there a problem launching on a polar trajectory?

Yes.. think about when you exit the Mun's SOI from a polar trajectory.. the resulting orbital plane for returning to Kerbin ends up at all sorts of weird (and largely unpredictable) angles. But whatever angle you exit Duna's SOI from a polar trajectory, it's going to be at a large angle to Hermes' orbital plane. Which means there will need to be a massively expensive (in terms of dV) plane-change before intercept can take place, probably taking many days/months to set up, as well, since we're talking about a Kerbolar (is that a word?) orbit at this point.

And the DAV doesn't have the dV to make that kind of correction. Which is why I've decided to fix the Hermes orbital plane first. One less problem to deal with.

- - - Updated - - -

think about when you exit the Mun's SOI from a polar trajectory.. the resulting orbital plane for returning to Kerbin ends up at all sorts of weird (and largely unpredictable) angles. But whatever angle you exit Duna's SOI from a polar trajectory, it's going to be at a large angle to Hermes' orbital plane. Which means there will need to be a massively expensive (in terms of dV) plane-change before intercept can take place, probably taking many days/months to set up, as well, since we're talking about a Kerbolar (is that a word?) orbit at this point.

Re-reading the above, I realise that the intercept will take place whilst within Duna's SOI, or it won't happen at all, as the DAV doesn't have the dV to keep up with Hermes by the time SOI exit arrives, so the above doesn't exactly apply. So ignore the above.. the following is a better description of the problem.

In the case of Hermes passing under the south pole, you have to keep in mind that "under the south pole" is just an approximate description. Hermes isn't passing directly under the pole, but is offset some unknown distance to one side or the other, and at the same time is also crossing from outside of Duna's orbital vector to inside the orbital vector, at an angle of some 5-10 degrees, at a rough estimate. All these factors combine to make it extremely difficult to predict the precise trajectory that will lead to an intercept.

If Hermes' orbital plane is a precise match for Duna's, then we have at least one predictable point to start working from. My earlier trajectories lacked anything predictable, so making an intercept would have to be purely trial and error.

[PLAD]

Rendezvous with a ship flying by the planet you are on is not that hard, as long as you can launch into the plane of the ship flying by. For instance, if the flyby hyperbola is inclined 20 degrees to the planet's equator but the lander is sitting at 45 degrees north then it will be very expensive to get into the flyby plane. But if you don't have that problem, below is my procedure. I used this procedure 5 times in my Jool-5 entry, since I never put the carrier vessel in orbit around any moon, but rather dropped off the lander during a hyperbolic flyby and then picked it up at a later hyperbolic flyby. (it's a lot of pictures, the Tylo album shows the timing trick best, and the Bop album shows the launching into the flyby plane best.)

1) You might have to wait up to one planetary day (in this case 1 Sol) for the ascent vehicle (DAV) to move (on the planet's surface) into the plane of the flyby vessel, so start watching for the right launch time at least that far ahead. Plus you will need to be able to orbit the planet at least twice after you reach orbit before the flyby vessel reaches periapsis. Once you see the DAV in the right spot, launch into the flyby plane. You'll have to figure out the launch azimuth. Put the DAV into a circular-ish orbit that is tangential to the flyby perapsis. (so for instance if the flyby periapsis will be 800 km then put the DAV into an 800x800km orbit.)

2) Observe the first time that the DAV will be at the tangent point to the periapsis of the flyby path. Observe the time that the flyby vehicle will be at that point. If you took off early enough, the flyby vehicle will not arrive there until more than one DAV orbit period later.

3) Now place a maneuver node in the DAV's flight path that is right at the tangent point to the flyby vessel's path. Make that maneuver adjust the period of the DAV so that it will do one orbit of the planet and be back at the tangent point at the moment the flyby vessel will be there. It's nice if this 'timing' orbit requires roughly half the prograde thrust that catching up to the flyby vessel will take, so you can split one big maneuver into two half-sized ones.

4) I find a couple of small (~1m/s) tweaks are often needed about a half-orbit and quarter orbit before the rendezvous. Now just accelerate up to the speed of the flyby vessel as it comes screaming at you.

The key is that the timing orbit only has to be entirely within the SOI of the planet, and as long as the rendezvous point is not too high there will be a very large range of possible periods for the timing orbit, so you can compensate for any timing errors that the launch time forced upon you. Note that the lower the flyby will be, the better for the launch vehicle since it will get more benefit from the Oberth effect.

[JAFO]

Thanks for that, PLAD.. I'll check out your links, and work through your description until I'm sure I understand you correctly. But believe me, my original Hermes trajectory was truly weird.. the offset to Duna's orbital plane introduced all kinds of complications. At least now the planes are perfectly matched (to 5 decimal places.. yay!) we will have some kind of predictable foundation to work from.

[PLAD]

Thanks for that, PLAD.. I'll check out your links, and work through your description until I'm sure I understand you correctly. But believe me, my original Hermes trajectory was truly weird.. the offset to Duna's orbital plane introduced all kinds of complications. At least now the planes are perfectly matched (to 5 decimal places.. yay!) we will have some kind of predictable foundation to work from.

Oh wow, does this mean that the Hermes will flyby Duna precisely in it's equatorial plane? That would make the return to Kerbin easier, but unless the DAV is right on the equator it won't be able to get into the Hermes' flyby plane without a dogleg, and step 1 of my procedure will be much harder.

[JAFO]

Oh wow, does this mean that the Hermes will flyby Duna precisely in it's equatorial plane?

Correct. In the beginning, my Hermes free-return trajectory was offset by 0.04063° from Duna's orbital plane, putting Hermes roughly 1,800km below and to the side of Duna at flyby periapse. When you factor in the flyby crossing over Duna's orbital vector and the DAV launch site location in the northern hemisphere as well, working out an intercept became a complete nightmare.

So now, Hermes is in Duna's equatorial plane.

That would make the return to Kerbin easier, but unless the DAV is right on the equator it won't be able to get into the Hermes' flyby plane without a dogleg, and step 1 of my procedure will be much harder.

The DAV site is presently situated 17° North of the equator. At this point, I'm not sure just how much of a problem that will be.. but at least it's a smaller problem than it was before I corrected the Hermes' orbital plane.

I've looked over your links and descriptions, and I'm impressed.. so now I have a question for you..

In the book, the MAV is flown directly to an intercept with Hermes, without establishing an orbit first, as you suggest to do. Certainly, this would have been made easier by the MAV launch site being on the Martian equator, as presumably was the Hermes' orbit. How feasible do you think it would be to recreate such a manoeuvre?

[selfish_meme]

Remember a plane change at a high altitude takes very little dV

- - - Updated - - -

A very hyperbolic orbit with a plane change and then swing round and catch up?

[JAFO]

A very hyperbolic orbit with a plane change and then swing round and catch up?

It will probably be something like that, yes..

In other news, the plane correction has made things considerably better/easier from a close-approach perspective. I'm feeling much more confident that a rendezvous can be pulled off, now. More after the next lot of testing.

[selfish_meme]

Take us past Ike!

[JAFO]

Take us past Ike!

One of my test close encounters ran into Ike.. well, it would have, if I'd been silly enough to leave it set that way.

[selfish_meme]

Progress was slow, I made a video of a Pathfinder Launch, then I started to place some assets for another runthrough, ran into an issue with the hab so I set about fixing it and cleaning it up. Have placed the hab and a rover with mark in it, now onto the rest of the assets. It's annoying how one small change can waterfall through.

[JAFO]

Progress was slow ... It's annoying how one small change can waterfall through.

Yep.. I can relate to that..

After my earlier optimistic post, things started to go wrong.. the nice tight approaches I thought I had set up, once in close to Duna, turned out to be nothing like the numbers I'd gotten earlier. I could re-load one particular quicksave (my "best" result!), and every time I loaded it, it gave me one of three different Duna periapses.. with a total variation of several hundred kms! I was tearing my hair out, and every new attempt I made to create a new intercept ran into similar problems.

After a few hours of this kind of frustration, I decided to scrap all my quicksaves, and start over from scratch at the mid-course correction point. More frustration ensued. No matter what I did, I couldn't get an intercept I was happy with.. "happy" being defined as a Duna periapse of 200km or less. Heck.. half the time I couldn't get a working intercept inside Duna's SOI! I tried not making any course corrections until about 45 days out from Duna, regardless of the fact that applying corrections so late would be more costly in fuel. Still no joy.

Muttering lots of bad words, I decided to ignore the recommendations I'd read about course changes.. prograde and retrograde just weren't making me happy.. I was going to try radial adjustments again. That had worked for me once before, and the dV costs were a lot lower, too. I started fiddling with the numbers, and found I was getting encouraging results, but boy was the damn thing sensitive to tiny alterations! I started working to 3 decimal places, locked the engines down to 6% output, and cranked MechJeb up to a tolerance factor of 0.0001m/s to try and get the fine control I needed. And it was working. Results started to improve.

However, I noticed MJ was still a bit inconsistent around the final few seconds of burn. I decided to shut MJ off with 0.2m/s to go, and finish the burn by hand. Better.. but still not great. And then, two things happened that, combined, gave me a real breakthrough.

The first thing was, I began seeing a pattern in the results I was getting. If the map said I had a periapse of 203km behind Duna, when I actually flew the route, I'd get a periapse of 323km. If I was expecting 307km ahead of Duna, I got 187km, and so on. Every single result was a consistent 120km 'out', and the direction of the offset was consistent also.. Ok.. this I could work with! I've no idea why it's like this, but so long as it's predictable, I can adapt!

And then, I made one of those serendipitous discoveries that began, like so many scientific breakthroughs, with the words, "hmm.. that's strange..." For some of you guys, this will probably be old hat, but I've never seen this specifically explained anywhere.. for me, it was a revelation.

As I followed ships in, from the course correction point about 400 days before Duna, to the SOI change, to see how well my results stood up, a couple of times I'd double-clicked on Duna to shift the map focus, so I could keep the action centred on screen. And whenever I did, my nice little periapse and SOI markers would vanish, to be replaced with a plain ol' intercept marker. Most annoying. But then, when I zoomed in some more, I noticed something odd. It wasn't easy to spot, because the colour was a fair match for the highlighted Duna, but there was a periapse marker on Duna! Zooming WAY in, I could see both SOI markers, with a neat line between them, with a periapse marker on it, alongside Duna! Even though Hermes was still a good two weeks from SOI. Kewl!

But it gets better.. It turns out that if you have focus on the target planet, those markers are there right from the moment you set up the manoeuvre node. This gave me an idea.. I could set up the burn, halt MechJeb at 0.2m/s to go, zoom WAAAAYY in on Duna, even though its' position on screen was still 400 days away from intercept, and watch the course lines and periapse marker in real time, while I carefully worked the throttle on dead slow! And it worked! This was how I discovered that there was one limitation to that 120km offset.. if I tried setting the periapse 70km inside Duna, to get a 50km behind position, even though it subsequently told me I had a 50km position, KSP didn't like that, and would give me a final result of about 180km anyway, when I actually got to SOI.. but other than that, I was golden. A periapse of just 10km behind got me a true result of 130km.

Now that I could super-fine-tune my periapsis, I discovered that the difference between 50km ahead of Duna, and 130km behind Duna, was something on the order of 5 mm/s of dV. No wonder I'd been having problems with prograde/retrograde earlier. I'd been trying to drive pins in, with a sledgehammer. Using this technique, I'd probably get good results with prograde burns as well.

Here's Duna, 400 days before it will get to meet Hermes, as I gently nudge the vector towards me, through the planet.. it will emerge out and get halted well before that purple 224km marker, at just 10km periapsis, which will give me a Duna flyby at 130km, later on.

So now, I have two nice neat Duna flybys set up, one 47km ahead of Duna, and one 130km behind Duna, ready for rendezvous testing with the DAV.

Feeling very happy right now.

Edited October 10, 2015 by JAFO

[PLAD]

Correct. In the beginning, my Hermes free-return trajectory was offset by 0.04063° from Duna's orbital plane, putting Hermes roughly 1,800km below and to the side of Duna at flyby periapse. When you factor in the flyby crossing over Duna's orbital vector and the DAV launch site location in the northern hemisphere as well, working out an intercept became a complete nightmare.

So now, Hermes is in Duna's equatorial plane.

The DAV site is presently situated 17° North of the equator. At this point, I'm not sure just how much of a problem that will be.. but at least it's a smaller problem than it was before I corrected the Hermes' orbital plane.

I've looked over your links and descriptions, and I'm impressed.. so now I have a question for you..

In the book, the MAV is flown directly to an intercept with Hermes, without establishing an orbit first, as you suggest to do. Certainly, this would have been made easier by the MAV launch site being on the Martian equator, as presumably was the Hermes' orbit. How feasible do you think it would be to recreate such a manoeuvre?

If you launch from the equator and the flyby plane is equatorial then it is just a matter of timing, but that doesn't mean it will be easy. You want to arrive at the periapsis point of the Hermes at the same time as the Hermes does. If you are not at exactly the right place on Duna's surface though, you will still have to coast for part of a circular orbit to the intercept time/place. From the sound of things you've probably figured this out already, but here's what I'd determine:

1) How long it takes you to get from launch into a circular orbit at the altitude of the flyby periapsis, both in time and angular distance around Duna. (For instance my last launcher took about 4 minutes and traveled about 12 degrees from the launch site to orbit circularization.)

2) How far around the planet you have to coast from circulization to intercept. This will depend on where you are at launch, for instance if you have to launch from the other side of the planet you will then have to coast almost 180 degrees before intercept. From the coasting orbit's period you can then determine how long the coast will take.

3) Now you have to iterate to find the proper launch time, since the earlier you launch the longer you have to coast, which means you have to launch earlier... since the coasting orbit altitude is set by the flyby periapsis a simple spreadsheet should be sufficient.

I want to stress that it would only be possible to go directly from launch to rendezvous in one burn if the Hermes flyby is timed such that you are sitting on Duna's surface just 12 degrees before the periapsis point exactly 4 minutes before the Hermes gets there (to use the numbers I mention above). Then instead of doing a circularizing burn you just keep burning to catch up with Hermes as it flies past you. If there is some really good reason why this is the only allowed launch profile (for instance does the launcher only get one engine ignition?) then the flyby of the Hermes will have to be adjusted to time it right. Let's see, Duna rotates once every 18 hours or so so you will have to be able to adjust your flyby time by +/-9 hours and be accurate to within a minute or so... oh boy.

Let's hear from the judges...does direct to intercept mean 'less than one orbit from launch to intercept' or does it mean 'one continuous burn from launch to rendezvous'? I will note that NASA called the ascent method the LEM used in Apollo 14-17 "direct ascent rendezvous", and that involved the LEM ascent stage going once around the Moon between launch and rendezvous, coasting most of the way.

[DMSP]

I really like this thread! So many ideas working together.

I applaud everyone who works on this actively.

I was just wondering, do we have any idea what the Duna Descent vehicle will look like?

A always thought it would look like a Red Dragon/Dragon Rider spacecraft. Does anyone know?

[selfish_meme]

I really like this thread! So many ideas working together.

I applaud everyone who works on this actively.

I was just wondering, do we have any idea what the Duna Descent vehicle will look like?

A always thought it would look like a Red Dragon/Dragon Rider spacecraft. Does anyone know?

I'll get you a screenshot of the current one, I don't think there are any presently, it does indeed just look like an extended capsule. I have tested a few landings on it. I added tilted back seats like the movie in both the DAV and DDV, but it means you need to remember to fly from the probe core. The MAV also comes in a version with a hole in the front of the capsule so you can use hullcam to watch your ascent from the eyes of the Kerbal.

[onewatt]

I build a more detailed, assemble yourself version of the Hermes, needs some work but check it out here: http://www.curse.com/shareables/kerbal/236897-kermes

[selfish_meme]

I build a more detailed, assemble yourself version of the Hermes, needs some work but check it out here: http://www.curse.com/shareables/kerbal/236897-kermes

That looks incredible, whats the partcount like? Also I noticed you probably have rigidity issues because of the docking ports. You may want to steal the ones from my Hermes they will help with floppiness.

[JAFO]

I build a more detailed, assemble yourself version of the Hermes

That looks just fantastic.. but it has so many greebles that it would make my laptop run like molasses in Antarctica.

And I think selfish_meme is right.. I suspect she'd fly like overcooked pasta.

Edit:

whats the partcount like?

I count almost 50 docking ports, and roughly 90 small tanks alone..

- - - Updated - - -

Downloading now, for the sheer fun of seeing how much of her I can get assembled before my computer chokes..

Edited October 11, 2015 by JAFO

[JAFO]

If you launch from the equator and the flyby plane is equatorial then it is just a matter of timing, but that doesn't mean it will be easy. You want to arrive at the periapsis point of the Hermes at the same time as the Hermes does. If you are not at exactly the right place on Duna's surface though, you will still have to coast for part of a circular orbit to the intercept time/place. From the sound of things you've probably figured this out already, but here's what I'd determine:

1) How long it takes you to get from launch into a circular orbit at the altitude of the flyby periapsis, both in time and angular distance around Duna. (For instance my last launcher took about 4 minutes and traveled about 12 degrees from the launch site to orbit circularization.)

2) How far around the planet you have to coast from circulization to intercept. This will depend on where you are at launch, for instance if you have to launch from the other side of the planet you will then have to coast almost 180 degrees before intercept. From the coasting orbit's period you can then determine how long the coast will take.

3) Now you have to iterate to find the proper launch time, since the earlier you launch the longer you have to coast, which means you have to launch earlier... since the coasting orbit altitude is set by the flyby periapsis a simple spreadsheet should be sufficient.

I want to stress that it would only be possible to go directly from launch to rendezvous in one burn if the Hermes flyby is timed such that you are sitting on Duna's surface just 12 degrees before the periapsis point exactly 4 minutes before the Hermes gets there (to use the numbers I mention above). Then instead of doing a circularizing burn you just keep burning to catch up with Hermes as it flies past you. If there is some really good reason why this is the only allowed launch profile (for instance does the launcher only get one engine ignition?) then the flyby of the Hermes will have to be adjusted to time it right. Let's see, Duna rotates once every 18 hours or so so you will have to be able to adjust your flyby time by +/-9 hours and be accurate to within a minute or so... oh boy.

Let's hear from the judges...does direct to intercept mean 'less than one orbit from launch to intercept' or does it mean 'one continuous burn from launch to rendezvous'? I will note that NASA called the ascent method the LEM used in Apollo 14-17 "direct ascent rendezvous", and that involved the LEM ascent stage going once around the Moon between launch and rendezvous, coasting most of the way.

Many, many thanks for the detailed analysis, PLAD. Combined with your earlier post, that gives us plenty to work with. Out of curiosity, have you read the book yet?

does direct to intercept mean 'less than one orbit from launch to intercept' or does it mean 'one continuous burn from launch to rendezvous'?

So far as the original story the scenario is based upon goes, it means 'one continuous burn from launch, with a coast to rendezvous after engine burn-out'. Launch takes place a few hours at most before rendezvous. While we'd love to replicate that, it obviously isn't going to be possible with the current launch site. So far as "adjusting the flyby timing" goes, I wholeheartedly echo your sentiments.. oh boy.

The plan is that this is a scenario where, so far as possible, everything has been pre-set for the player.. a fun challenge for an afternoon's play, rather than something of brain-bending difficulty. The player's mission is simply to drive several hundred kilometres to the Ascent Vehicle site, in time to then be able to attempt a rendezvous with Hermes as it comes hurtling by.. in other words, Hermes is on rails.. it's their job to catch her, if they can. Even though in the story the Hermes crew make some emergency course adjustments to facilitate rendezvous, we'd like to avoid having the player need to adjust the Hermes trajectory as well as fly the Ascent Vehicle. Certainly, after a successful rendezvous, they're going to have to make some small adjustments to Hermes' course to intercept Kerbin, but that's fairly trivial.

I'm probably only a day or so away from releasing a beta gamesave, the purpose of which is simply to playtest the launch and try to work out the timings and techniques needed, in part so we can create some Kerbal Alarm Clock alarms tailored to the scenario. If you feel inclined, we'd love to have you try it out and see what you think. The same invitation is extended to anyone else who'd like to help playtest the scenario.

Edited October 11, 2015 by JAFO