Launch Windows, Kerbal Alarm Clock - What am I doing wrong?

[BoilingCold]

I'm feeling very frustrated with interplanetary transfers and am wondering if anyone can point out what I'm doing wrong.

I'm using things like the alexmoon-github Launch Window Planner, Kerbal Alarm Clock, Precise Node & MechJeb, however the information I get from these tools doesn't seem to match up with what happens in-game at all.

Here's an example: I started a new sandbox game and built a simple ship to get to Duna. Added an alarm in Kerbal Alarm Clock for the next Duna transfer and then headed to http://alexmoon.github.io/ksp/ and played around a bit to get a slightly higher dV transfer that would give me an 80 day flight time. Here's the chosen transfer:

And when I create a maneuver node with those details here's what I get:

As you can see that burn takes me nowhere near Duna's orbit. My ship is in a circular, equatorial orbit at 100km around Kerbin. The node details are as precise as Precise Node allows me to set. And yet the ejection angle, the phase angle and the final positions don't match the Launch Window Planner at all.

Similarly if I go by the transfer alarms in Kerbal Alarm Clock and try to do either manual maneuver nodes or use MechJeb to create them they just don't match up with the information. The Kerbal Alarm Clock windows are often 100s of days different from MechJeb's calculated nodes!

Does anyone know what I'm doing wrong? Am I completely misunderstanding and/or misusing these tools?

Thanks for any advice!

Edited September 16, 2014 by BoilingCold

[Signo]

Hi, Im using almost the same set of mods apart from the precise node thing and yeah, they behave in a pretty unconsistent way. I have to manually correct all my nodes.

However in this case, it seems that the ejection angle to Duna (132 to prograde) is not correct in the precise node window (84.59 to retrograde).

[mhoram]

As you can see that burn takes me nowhere near Duna's orbit. My ship is in a circular, equatorial orbit at 100km around Kerbin. The node details are as precise as Precise Node allows me to set. And yet the ejection angle, the phase angle and the final positions don't match the Launch Window Planner at all.

Does anyone know what I'm doing wrong? Am I completely misunderstanding and/or misusing these tools?

Try to adjust the departure node to the other side of Kerbin.

You can use the Precise-Node "+/-"-Button in the UT: line to do this.

The Time of the node should be around 15 minutes later.

While doing this you can also see the transfer orbit change to get nearer to Duna.

[r_rolo1]

My guess is that you are burning in the wrong direction. Hoffman tranfer burns to Duna and beyond should be done in the dark side of the Kerbin if you are in parking retrograde orbit ( you most likely are ) :

From Kosmo-not excelent guide

Other thing to consider is that Alex tool and KAC might not be using the same coordinate system ( they might not be measuring the angles starting on the same place and in the same direction ) ... as I do not use KAC I can't tell. Be careful and read the fine print

Edited September 15, 2014 by r_rolo1

[Red Iron Crown]

TriggerAU (responsible for KAC) has collaborated with Alexmoon to produce Transfer Window Planner, an in-game tool similar to the webpage that permits copying of the transfer info directly into KAC. Not sure about PreciseNode.

This mod has made it so I need no external tools for interplanetary transfers.

[mhoram]

This mod has made it so I need no external tools for interplanetary transfers.

For me the combination KAC + Precise-Node made external tools unnecessary.

From KAC I get an estimation when the next Transfer window is available.

Then I setup a transfer node around that time (+/- a few days makes usually no big difference).

Using the Time- and Node- "+/-" buttons of Precise-Node I search for the departure burn (within a single orbit) with the lowest dV that ends up within the target's SOI and has the lowest Periapsis (Centering the view around the target body hepls with this).

For target bodies that are on an inclined orbit the procedure gets a little bit more complex since it mostly involves a mid-transfer-course-correction node, but the basics are the same.

If I want the transfer time to be shorter, then I just perform the departure burn a few days earlier and if I want to make a transfer later, it costs more dV and lasts longer.

And the cool thing about this procedure is that I do not have to know anything about ejection and phase angles.

The ejection angle comes automatically from playing with the "+/-"-buttons and the phase angle from the KAC departure-time estimation.

The KSP-Node-GUI is unfortunately not exact and predictable enough to use this procedure reliably, so trying Precise-Node for the first time was really an epiphany for me.

[Superfluous J]

For clarity:

You don't want to burn at EXACTLY the time that Alexmoon's calculator tells you to burn. You want to burn at the CLOSEST time to it you can to have an ideal ejection. This will usually be within 20 minutes or so of the time Alexmoon's page (or the wonderful new mod that incorporates it into the game) tells you to burn at, and for all intents and purposes that is perfectly fine.

[-ELF-]

I have not found the description of what causes everything to be so incorrect, so i will just leave it here.

You have done a correct calculation with Alex Moon's calculator.

What you did wrong is you relied on the exact departure minute.

Your craft must be in very exact point of orbit to start the burn. This point is described by ejection angle.

but, your craft is not going to be in that very point. It is going to be in some another point.

That is the source of your mistake.

What you need to do, is, as mentioned, use + and - buttons of Precise Node, to move your ejection angle to the closest possible value as in Alex Moons calculator.

Ignore that your departure time will be slightly off. Believe me, 2 days off the departure window will not cost you much, but 5 degrees off the ejection angle - will bring you to totally different place. And, of course, you will need to do mid-course correction burn, but that's a totally different story.

A-and, ninja'd!

Just as planned.

[capi3101]

Long and the short of it, you didn't launch into the exact position you needed to be in in order to leave at the exact moment the AlexMoon tool tells you. The good news is that very, very, very little in KSP is about being 100% precise all the time. I use the same tools you do (PreciseNode and KAC), and from experience I can tell you that everybody else that has posted to this thread has given you the answer already; the angle of the ejection is the crucial factor at play, and if you don't hit it at the exact moment it'll cost you a trivial amount of delta-V at the worst. You just go as close to that departure time as you can manage and you'll turn out fine.

[Kryxal]

Also, if the window is small (such as going to Moho), you can execute one burn at the proper phase angle a few hours before, and return to that point at the EXACT time you're looking to do so by just paying attention to the orbital period ... and it gets some of the burn out of the way too!

[BoilingCold]

Aha, OK thank you all, it makes more sense now that I understand to use these tools as a starting point rather than an exact guide!

I actually managed a Moho landing last night, although a miscalculation means that intended launch and return didn't happen, so now my goal is a rescue mission

[capi3101]

Yeah, Moho is a very tricky target. I've only managed to put a probe around it myself. If your target is Moho, I suggest you go maccollo. Game hasn't changed that much in the past nine months...

[BoilingCold]

Oh that's very cool. I'm still 180 days away from a launch window to rescue Bob but I'll definitely give that a go, thanks