[Tutorial] Interplanetary How-To Guide

[SergeantFritz232]

This was pretty useful as it stands but I haven't unlocked the nuclear engines in career mode yet so what engine would you recommend for a dres transfer stage. I currently have a Skipper attached with drop tanks.

[Pecan]

...so what engine would you recommend...

1. Fuel mass for the required deltaV is the first consideration. Choose the engine with the best vacuum ISP (Ion, if you have it, then LV-N, aerospike and LV-909, ...) to save as much as possible.

2. Thrust for the desired TWR is less important. Unless you're launching or landing a higher TWR just means you can do ejection, adjustment and injection burns faster - the actual transfer will take the same amount of time in any case. TWR can therefore be as low as your boredom threshold; I often use 0.5 as my minimum, giving burn-times of 10-20 minutes. Look at how many of the engines chosen in the first step it would take to give you a useful TWR. Move on to the next-best engine if you can't reasonably fit that many on your ship.

[JarvisTbone5150]

And people say video games aren't educational Excellent stuff!

[Black Talon]

Does anyone else have a problem loading this awesome Tutorial at the moment? I don't have any issues with any other web pages. The Tutorial just hang mid-load and goes nowhere.

[Black Talon]

Seems to be working again now.

[windows_x_seven]

I have a question.

What radii and gravitational parameters are used?

Kerbol gravity? Planets' orbits?

[cwalrus]

I believe since you're transferring from planet one to planet two, the gravity is planet one's, and r1 is planet one's, and r2 is planet two's.

What I'd like to know is how to find the SOI exit velocity. Is that the same as escape velocity?

[maltesh]

I have a question.

What radii and gravitational parameters are used?

Kerbol gravity? Planets' orbits?

In the above equation,

μ is the Standard Gravitational Parameter of the sun.

r₁ is the radial distance from the center of the sun of your origin planet.

r₂ is the radial distance from the center of the sun for your destination planet.

[cwalrus]

Yeah! Exactly what I said! Except much neater.

What I'd like to know is how to find the "SOI exit velocity". Is that the same as the escape velocity?

[maltesh]

Kosmo-Not kind of copied a bunch of different equation images from several sources when he made the initial post. Since a lot of them were images, the reuse of different variables for the same quantity can get a little confusing.

The ÃŽâ€v₁ in the equation in windows_x_seven's post is the /is/ the necessary SOI exit velocity to produce the desired Hohmann transfer.

[Schinxz]

I'm trying to get a better understanding of these formulae as I like understanding what I'm doing rather than just plug numbers into some formula. I'm having a bit of trouble with this v₁ = SQRT((r₁*(r₂*v₂²-2*μ)+2*r₂*μ)/(r₁*r₂)) Can someone shed some light on what/which formulae has been used to arrive at this? Would be greatly appreciated

[obeythelord]

by Kosmo-not

v₁ = SQRT((r₁*(r₂*v₂²-2*μ)+2*r₂*μ)/(r₁*r₂))

where:

r₁ = parking orbit radius

r₂ = SOI radius

v₁ = ejection velocity

v₂ = SOI exit velocity (absolute value)

µ = gravitational parameter of origin planet

so i tried to calculate a trip to duna and here's what i filled in

r1=150 km

r2=84159.286 km

v1=

v2=3431.03 m3s-2

u=3.5316*10^12

and what i got whas 216831.3755

this doesn't look like an accurate amount of delta V does it...?

help me please

[maltesh]

by Kosmo-not

v₁ = SQRT((r₁*(r₂*v₂²-2*μ)+2*r₂*μ)/(r₁*r₂))

where:

r₁ = parking orbit radius

r₂ = SOI radius

v₁ = ejection velocity

v₂ = SOI exit velocity (absolute value)

µ = gravitational parameter of origin planet

so i tried to calculate a trip to duna and here's what i filled in

r1=150 km

r2=84159.286 km

v1=

v2=3431.03 m3s-2

u=3.5316*10^12

and what i got whas 216831.3755

this doesn't look like an accurate amount of delta V does it...?

help me please

There's one definite obvious problem: You mistook Parking orbit altitude for parking orbit radius. If you're orbiting Kerbin at an altitude of 150km, the radius of your orbit is 750 km.

I don't currently have the time to thoroughly check your math but the other potential problem is that you must use compatible units.

Assuming Kerbin again, it /looks/ like you're using Standard Gravitational Parameter in m³/s² and velocity in m/s, in which case, for the calcualtion to work as written, both parking orbit radius and sphere of influence radius must be in meters, not kilometers.

[windows_x_seven]

One more question, how does one predict the next launch window to a body, like in Kerbal Alarm Clock, where it shows the next window's date?

This whole thing makes rocket science look hard...

[windows_x_seven]

Any idea?

[mhoram]

@windows_x_seven:

This is called the Lambert's problem.

The wikipedia site contains the formulas for the calculation.

[maltesh]

Kerbal Alarm Clock has two options to predict the date of the next transfer window.

There's the Formula Option. This is the one selected by default when Kerbal Alarm Clock is installed.

In the Formula option, Kerbal Alarm clock assumes that all the planets are in equatorial, circular orbits, and moving at constant speeds along them. It works out the rate of change of that angle based on the synodic period between the periods of the two worlds in question, and then uses the aforementioned assumptions to generate a transfer window time.

And then, since the worlds are /not/ in Equatorial, Circular Orbits, it constantly readjusts the alarm it set as the time of the window gets closer and closer, and the error on the Equatorial/Circular assumption gets smaller. However, you probably won't notice that Kerbal Alarm Clock does this unless you have other, fixed-time alarms to compare the transfer window alarm against.

Kerbal Alarm Clock's Mode Option uses a list of transfer windows someone generated years ago by computer modeling the first 100 earth years of a new save game, and noting whenever the worlds were in the tranfer position. In that case, Kerbal Alarm Clock looks up the universal time of the calculated model window, and generates an alarm based on that.

[spioner]

Quick question:

If I were in a parking orbit around a planet, and I would calculate my orbital velocity using

v = SQRT(μ/r)

And I would calculate required velocity (called v1 in this guide) using

v₁ = SQRT((r₁*(r₂*v₂²-2*μ)+2*r₂*μ)/(r₁*r₂))

Could I then calculate the required transfer delta-v by getting the difference between the two? As in the total amount of delta-v required to get from a parking orbit to an intercept with another planet?

Sorry if I'm being vague, I just can't explain it very well.

[troyfawkes]

Hullo! I remember seeing this guide previously and it being super useful. I'm going through the math again and everything is broken here! All of the other resources online assume that I know some pretty intense math and don't exactly answer my question. I'd love to see this guide updated to work again

[egoego]

On 9.10.2015 at 8:23 PM, spioner said:

v₁ = SQRT((r₁*(r₂*v₂²-2*μ)+2*r₂*μ)/(r₁*r₂))

It's not broken, evey KSP player should know that a ÃŽÂ¼ is a quarter pounder with cheese. So this formula just describes how many burgers you have to bring to convince Jeb to make the rocket work by sheer will power. 

[BashGordon33]

Am I the only one who can only see strange As with weird lines on top when loading the math in this page?

[Vanamonde]

This thread is very old, and some of the text has been garbled by transitions to new forum softwares, BashGordon33. It is one of the better guides, though. Perhaps @Kosmo-not could be nagged/wheedled/flattered into updating it? 

[Kosmo-not]

I decide to log into the KSP forums after many many months of inactivity, and it's @Vanamonde who wants to pull me out of retirement!

I'll go ahead and try to clean up the original post the best I can.

 

Assuming you can create maneuver nodes, there is a much MUCH easier way to accomplish an interplanetary trajectory:

1.)  While orbiting Kerbin (I use a probe for this), create a maneuver which will take you juste barely out of Kerbin's sphere of influence (theoretically, you're exiting the SOI with zero velocity, achieving the same orbit as Kerbin around the sun).

2.) From the orbit around the sun, create a maneuver for a transfer orbit to whatever you want to intercept. Manipulate it until you get an intersection.

3.) Timewarp until Kerbin is at the maneuver node created in step 2.

4.) With the craft you are going to be doing the stuff with the other planet you want to go to, create a maneuver node for an ejection from Kerbin's orbit (in either the planet's prograde or retrograde direction) and increase the velocity until an intersection is made with the target planet. You will have to adjust in the normal or anti-normal direction for planetary orbits which do not lie on the origin planet's orbital plane.

5.) Execute maneuver and later do whatever slight corrections are required.

 

^{Any such undertakings are at your own risk. Upon attempting the above steps, you agree to not hold Kosmo-not liable for any damages resulting in the loss or destruction of spacecraft, kerbals, or ego.}

 

 

*update* I think all the equations in the OP are fixed now.

Edited August 26, 2016 by Kosmo-not

[TheKraken2000]

It seems like the bottom half of the tutorial is broken again. At least for me.

A few formulas are completly gone because the image containing them could not be found and we have this magicle variable ( Ejection Angle = 180° - ÃŽÂ) used to get the Ejection Angle.

Other than that thank you for this great guide. It helped me a lot.

[bundook]

Please, I'm very interested in this, and appreciate the work that went into producing it for us. Is there any chance that the OP could be corrected so we can see the images? The imageshack links are dead. A fix for the corrupted text (  Π) would be great too.