Different windows for different ships?

[annallia]

Trying to set up a station in orbit of Duna, I have three ships like this in orbit of Kerbin, all 278-281km

Problem is their window times are all different.

The first ship I launched has a 58 day and change window, second has 148 days, third has 92 days. Now I am all good with a bit of staggering but how is this changing so much over just a minute or so?

Using MechJeb to plot the course and they all look to be similar but for some reason I have over almost 100 days between the first and second craft.

Edited August 14, 2013 by annallia

[Johnno]

Using MechJeb to plot the course

There's your problem.

It's probably something like Mechjeb is expecting to fly them one at a time, back to back.

I'd say do them manually. Once the window opens for the first one send it off, then do the second one without timewarping the first one very far. Then third in a similar manner. You can adjust their arrivals in solar orbit.

[Thomas Cook]

I don't know if this answers your question, but what I saw in other people's tutorials:

You can just place something on the launchpad or the runway and then accelarate time to 100.000x

Since a day has 86.400 seconds, you should be able to burn more than a day in a second.

[annallia]

Mechjeb shouldn't be operational on the other two since I am going to the space center to switch vessels (same orbit, close to eachother but far enough apart [] doesn't work, easier than trying to pick out the right craft ont he orbit map)

I will give plotting it manually another go, though I do suck at that which is why I use mechjeb for it...

Edit- Thomas that isn't the issue, the issue is the gaps between launches which there shouldn't be. If you have 3 ships in the same orbit with the same destination there shouldn't be a significant gap in their launch windows.

[blizzy78]

I will give plotting it manually another go, though I do suck at that which is why I use mechjeb for it...

Time to learn something new, right? We're here to help

[annallia]

Time to learn something new, right? We're here to help

More like time to practice, it isn't that I don't know how to do it it is that I suck at doing it. I am fine going from planet to moon or the other way around but planet to planet isn't so easy for me to plot.

That said I plotted them all myself and I got anywhere from 61 days to 67 days... Certainly better. 6 days between launch windows can easily be accounted for by my crappy interplanetary navigation.

Although the odd part, after I got that settled I launched a new tug that will go along with them to Duna (forgot to put a docking port on the other tug oop!). When I calculated its transfer to Duna I gave Mechjeb another try, got 58 days again, I decided to try the other crafts and they all have 58 days and are +/- a few hours of eachother... Very weird.

[Rhomphaia]

I haven't used mechJeb to transfer to another planet for a while now, most of the reason I stopped using it was I found it too fussy with it's launch windows.

I think that what is happening here is that mechJeb has decided that your second and third ships will be too far from the burn point at the first window so is plotting for later windows, even though all you would probably need is a mid course correction burn to get the encounter

[annallia]

That could be... Though again from my understanding of how it works it should be independently plotting each course if I go to the tracking station to switch..

At any rate calling this answered, they are all on their way to Duna, think I will work on some other stuff rather than time warping right now.

[blizzy78]

Though again from my understanding of how it works it should be independently plotting each course if I go to the tracking station to switch..

It does, it just is too picky about the launch windows.

[Geschosskopf]

It does, it just is too picky about the launch windows.

Agreed. If you tell MJ to transfer to another planet, it will only plot nodes at times when the ship's position in its current orbit lines up with the "ideal" time to burn for a given window. It would rather wait months for this to happen than do what it considers a non-optimum burn in the current window by missing its timing by an orbit or 2. This is pretty ridiculous because the planets don't move by each other fast enough for a few orbits around Kerbin to make any real difference to delta-V.

So, to launch a flotilla to another planet, you have to create all your own transfer nodes manually, 1 at a time. That is, on the day of the window, pick a ship to be the 1st to leave, set up its node, and burn it. Then switch to the next, set up its node, and burn it, repeat. I usually plot the node for a ship that's just passed the approximate point to do the burn, so I have most of its orbit to set the node up, in case I have trouble tweaking the node.

Thus, they all leave Kerbin over the course of a couple hours, depending on how many ships you have, which doesn't have that much effect on the delta-V required. However, this much difference in departure times gets amplified into days or a few weeks at the destination. Which is fine because you don't want them all arriving at once .

MJ comes in handy when doing the burns, with the "execute next node" thing. If using nuke engines, the burn times are LLLLLOOOOONNNNNGGGGG and you've got a whole flotilla to get on the road, so MJ is a blessing here. BUT BE SURE TO BLIP THE THROTTLE when switching to the next ship. Otherwise the burn time calculator will be way off and neither you nor MJ will do an accurate burn. So, switch to the next ship, lightly tap LSHF and then X immediately after to wake up the burn calculator, plot your node, then have MJ execute it.

[annallia]

Agreed. If you tell MJ to transfer to another planet, it will only plot nodes at times when the ship's position in its current orbit lines up with the "ideal" time to burn for a given window. It would rather wait months for this to happen than do what it considers a non-optimum burn in the current window by missing its timing by an orbit or 2. This is pretty ridiculous because the planets don't move by each other fast enough for a few orbits around Kerbin to make any real difference to delta-V.

So, to launch a flotilla to another planet, you have to create all your own transfer nodes manually, 1 at a time. That is, on the day of the window, pick a ship to be the 1st to leave, set up its node, and burn it. Then switch to the next, set up its node, and burn it, repeat. I usually plot the node for a ship that's just passed the approximate point to do the burn, so I have most of its orbit to set the node up, in case I have trouble tweaking the node.

Thus, they all leave Kerbin over the course of a couple hours, depending on how many ships you have, which doesn't have that much effect on the delta-V required. However, this much difference in departure times gets amplified into days or a few weeks at the destination. Which is fine because you don't want them all arriving at once .

MJ comes in handy when doing the burns, with the "execute next node" thing. If using nuke engines, the burn times are LLLLLOOOOONNNNNGGGGG and you've got a whole flotilla to get on the road, so MJ is a blessing here. BUT BE SURE TO BLIP THE THROTTLE when switching to the next ship. Otherwise the burn time calculator will be way off and neither you nor MJ will do an accurate burn. So, switch to the next ship, lightly tap LSHF and then X immediately after to wake up the burn calculator, plot your node, then have MJ execute it.

I don't use mechjeb for the burns, if I were going to let mechjeb do it all for me I may as well use hyper edit and just place my station in orbit in all one piece rather than several smaller sections. Instead MJ plots the course, I fine tune it and take over when I get into Duna's SOI. As I have said before, I use it as a nav computer, and occasionally to pilot probes though given that this is not a satellite that I don't care if it crashes but rather a station... I will be at the helm myself.

As far as the burns go, I know they can be long, specially with nuke/ion engines but they won't be on this. The only reason these things come in at 50 tons is the weight of their interplanetary stage (where they are now). Once I get to Duna the two that are in fairings only weigh about 4 tons each, the core is a bit heavier.

[Geschosskopf]

I don't use mechjeb for the burns, if I were going to let mechjeb do it all for me I may as well use hyper edit and just place my station in orbit in all one piece rather than several smaller sections. Instead MJ plots the course, I fine tune it and take over when I get into Duna's SOI. As I have said before, I use it as a nav computer, and occasionally to pilot probes though given that this is not a satellite that I don't care if it crashes but rather a station... I will be at the helm myself.

Calm down, dude. I'm not here to get into some argument over the relative manliness of using MJ for this, that, or the other. You asked a question and I'm trying to give you an answer.

Like it or not, in the specific case of interplanetary transfers, MJ is a very poor "nav computer" because it's too anal about getting a perfect synch between your position in orbit around Kerbin and the timing of the ejection burn. So if you want your flotilla to arrive more or less at the same time at Duna, you CANNOT use MJ to plot the transfers because it will launch the ships weeks or months apart. So there's nothing for it but to do the maneuver nodes manually for each ship. End of story.

Now, once the maneuver nodes are set, you're probably looking at ejection burns of 6-10 minutes for each ship in the flotilla. If you feel honor-bound to sit there and do all this manually, that's your business. Have fun with it.

[annallia]

Sorry about that, I see rabid people on both sides of the MJ fence so I get a bit jumpy about it.

That said I can, and did use MJ to plot the transfers, and get them all within a few hours which is how it should be at their orbit (moved up to higher orbit) they just required a little fine tuning to make sure they entered Duna's SOI without a course correction mid way (one needed to be changed so it would enter on the right side of Duna...) Just as I said earlier it plotted them for me just fine after a restart.

FWIW ejection nodes are ranging from 3 minutes 15 seconds to 4 minutes 12 seconds (core). As I had expected, the core stage is requiring a longer burn as it weighs more and has slightly less thrust due to only having 3 LVN's instead of 4 like the rest.