How to preform orbital insertion????

[John Doe]

44 minutes ago, 5thHorseman said:

Except you can't be orbiting something in a perfect circle unless that thing is exactly at the center (disregarding it moving due to your gravitational influence on it) of the circle. And there is no way to orbit in an ellipse without the thing you're orbiting being at one of the foci.

only seems to hold true to a certain extent- What then, do you suppose happens if we offset a circular orbit, but not enough to lower the Pe below the threshold for re-entry?  We haven't changed the shape, we have just lowered Pe and raised Ap.

[Superfluous J]

13 minutes ago, John Doe said:

only seems to hold true to a certain extent- What then, do you suppose happens if we offset a circular orbit, but not enough to lower the Pe below the threshold for re-entry?  We haven't changed the shape, we have just lowered Pe and raised Ap.

No, that's an absolute.

If you are in orbit around a planet and not otherwise accelerating, you are in an ellipse with one focus the planet. Period. If you are in a circular orbit, the planet is at the center. Period. If you are going around in a circle and the thing you are going around is NOT at the center, you are NOT orbiting. Period.

There are no exceptions to this.

[Wemb]

13 hours ago, 5thHorseman said:

There are no exceptions to this.

What he said.  The only way you could have something that looks like an exception would be if the planet had it's centre of gravity somewhere other than at it's geometric centre and while that happens in a tiny degree IRL, I don't believe it happens in KSP.

Wemb

[N_Molson]

21 minutes ago, Wemb said:

What he said.  The only way you could have something that looks like an exception would be if the planet had it's centre of gravity somewhere other than at it's geometric centre and while that happens in a tiny degree IRL, I don't believe it happens in KSP.

Wemb

KSP doesn't simulate non-spherical gravity sources. And it doesn't takes the Mun tidal effect into account. If you are curious about those matters, try Orbiter

So yeah, in KSP the data is simple : you have a sphere (Kerbin) with a given density and a given mass. From there, Newtonian physics apply straight to the letter.

[Stoney3K]

15 hours ago, John Doe said:

only seems to hold true to a certain extent- What then, do you suppose happens if we offset a circular orbit, but not enough to lower the Pe below the threshold for re-entry?  We haven't changed the shape, we have just lowered Pe and raised Ap.

Which makes the orbit no longer circular but an elongated circle: An ellipse.

It would be easy to understand if orbits were rectangular: The long side being your Apoapsis and the short side being your Periapsis.

You can only have a 'square' (circular) orbit with Ap = Pe (resulting in an eccentricity of zero). Changing your Ap or Pe will stretch or shrink one side of the rectangle, making it not a square anymore.

[John Doe]

1 hour ago, Stoney3K said:

Which makes the orbit no longer circular but an elongated circle: An ellipse.

It would be easy to understand if orbits were rectangular: The long side being your Apoapsis and the short side being your Periapsis.

You can only have a 'square' (circular) orbit with Ap = Pe (resulting in an eccentricity of zero). Changing your Ap or Pe will stretch or shrink one side of the rectangle, making it not a square anymore.

Which makes the shape of the orbit an offset CIRCLE, NOT an Ellipse, by defintion. An perfect ellipse  is twice as long as it is wide,  with Ap and Pe at, or near, the length of the ellipse. Think about it like this. Take a coffee cup and turn it upside down. the perimeter of the cup represents a perfect circular orbit. Now, place a bottle cap in the middle. This represents the object orbited, such as Kerbin. Now, if we move the bottle cap, we see that we modify the Ap and Pe nodes, WITHOUT modifying the shape. (i.e The coffee cup  DIDN'T  change it's shape and turn into an ellipse just because we moved the bottle cap.)  So long as we keep the Pe node high enough, (i.e. we don't place the bottle cap closer than x to the perimiter of the coffee cup, where x is the maximum Pe allowed that will allow a re-entry )  the laws of physics do not allow re-entry, and because we haven't accelerated towards an escape node, (which changes the shape and path of the orbit) we will not rise higher than Ap. Thus, we cannot travel outside the path of the orbit, and we cannot re-enter unless we drop the Pe enough.  There's nothing I've been able to locate within newton's laws on gravity that state this theory to be incorrect, assuming that our velocity never changes. (which can be accomplished with a burn in the right direction at the right time.)  I've seen nothing in newtonian physics that states that this cannot happen, though it is extremely difficult to make it happen.

[Superfluous J]

1 hour ago, John Doe said:

The coffee cup  DIDN'T  change it's shape and turn into an ellipse just because we moved the bottle cap.

But if you move an ORBIT around like that, it WILL change shape.

I really by now think that we (you and then everybody else) are just talking across each other. Nobody's going to agree. I'm going to recuse myself from this point on but I will end it with a vehement statement: What you are talking about (moving a circular orbit so the planet is off-center yet the orbit remains a circle) is IMPOSSIBLE. Not by Newton. Not by Kepler. It's not a theory. It's a result of mathematics.

/done

[Plusck]

An orbit is not an *object*. You can't move it around and say - hey, it's a circle!

The orbit is a mathematical description of what happens when gravity pulls and you try to move at right angles to where gravity is pulling you. If you change the angle so that it is no longer a right-angle, then you stop going round in circles and you start going round in egg-shaped things. And whatever you want to call that egg-shaped thing, it is not a circle.

[John Doe]

3 minutes ago, 5thHorseman said:

But if you move an ORBIT around like that, it WILL change shape.

I really by now think that we (you and then everybody else) are just talking across each other. Nobody's going to agree. I'm going to recuse myself from this point on but I will end it with a vehement statement: What you are talking about (moving a circular orbit so the planet is off-center yet the orbit remains a circle) is IMPOSSIBLE. Not by Newton. Not by Kepler. It's not a theory. It's a result of mathematics.

/done

 So, then basically what your saying is that it's pretty much impossible to raise Ap And Pe at the same rate? ( you actually have me confused. trying to really get the hang of orbits as there was this point in the basic demo about establishing a lunar orbit, wherein I am at the stage of being on a collision course with the moon, in which the only hint was "remember what you learned about orbits in the previous tutorial"  which was basically nothing more than how to raise /lower Ap / Pe. So My instinct was to fire Prograde to slow down my velocity to avoid impact, and perhaps establish a Pe.  (not a good idea) So basically Yeah, I'm trying to learn all the crap that the tutorial isn't saying.  Also bear in mind that I'm used to flying STS missions in orbiter, which always involve the pretty much SAME type of orbit, and the orbiter is doing practically ALL of the hard parts- the burns are automatic once you program the computer to tell it what you want to accomplish. So really,  outside of really basic concepts, I have no idea of how newtonian physics might somehow apply differently than what would be plainly obvious.

[Plusck]

One (maybe) useful thing to consider: the biggest effect on your orbit will happen one quarter of an orbit before and after where you are now. If you burn radially, you make a big change to the height of the orbit a quarter of an orbit later and three-quarters of an orbit later. If you burn prograde/retrograde, you change the angle of your flightpath one quarter later. If you burn normal, you move the orbit up or down one quarter orbit later.

However, each effect has its exact opposite effect at every other point in the orbit, with the greatest effects at the 1/4 and 3/4 marks. You can certainly raise Ap and Pe at the same time, by burning pro/retrograde with a radial component sometime between the two, but this is never going to be an efficient maneuvre because it comes at a cost of the other effects that will apply at the 1/4 and 3/4 marks.

[Superfluous J]

1 hour ago, John Doe said:

 So, then basically what your saying is that it's pretty much impossible to raise Ap And Pe at the same rate?

Not at all. Just by doing so, your orbit will no longer be a circle. And if the Pe and Ap are at all different (which, in all but idealized situations, they will be) then your orbit is an ellipse.

Perhaps this page will help you: https://en.wikipedia.org/wiki/Kepler's_laws_of_planetary_motion

In particular, the first law about ellipses.

Edited April 27, 2016 by 5thHorseman

[John Doe]

Got it-  so the elipical shape of the orbit is really only elpitical by such small fractions that its not observable by looking at the geometric shape....(i.e. the shape is so marginally eliptical that it still appears to be circular. I was making the assumption that  hyperbolic was synonymous with elliptic. I was also understanding the orbit to be an complete escape of gravity combined with the effect of inertia.

Edited April 27, 2016 by John Doe

[Harry Rhodan]

Yeah, a circle is just an ellipsis with an eccentricity of 0.

[Stoney3K]

To be fair, you CAN move around in circles around an object that is not in the center of that circle. However, you would need to maintain your flight path (with engines) to counteract the force of gravity that wants to change your flight path from a circle into an ellipse.

You'd be hovering around the object with your engines in a circular path, so it would hardly qualify as an "orbit". Only if you're the USS Enterprise, with magic gravity-defying thrusters that allow it to move straight up and stay there forever.

[LameLefty]

5 hours ago, Harry Rhodan said:

Yeah, a circle is just an ellipsis with an eccentricity of 0.

THIS!

Gah, this thread makes my brain hurt. It's like no one studies conic sections in algebra anymore.

 

[mrclucks]

A page and a half of argueing and it seems like the OP hasn't gotten an answer...

@OP while most people will say something about getting a circular orbit, the apoapsis and periapsis can defer by some number of meters or kilometers but is for gameplay purposes circular (IE periapsis is 80257km and apoapsis is 80263).  There's not really much difference between them and realistically in this game it isn't needed.  There are other orbits that people will say are elliptical and that's when the differences between apoapsis and periapsis counts.  The further difference will give you a more profound oberth effect at periapsis and will require less DV when changing inclination at apoapsis.  

Now as your original question on raising periapsis/getting into an orbit, you want to perform a gravity turn.  An orbit is basically your vehicle going sideways around an object with significant gravity and while that object is still bringing you down, you've gone sideways enough to counteract the gravity enough that the object is basically in a different position.  Now a gravity turn is nessasary since you start off going straight away  from the object( kerbin in this case ) to escape the atmosphere and ground, and you slowly start turning sideways so you can build the sideways momentum to eventually start missing the ground.  

An efficient gravity turn to 100km or so will end with you pointed towards the horizon at 60km or so and only needing a few hundred DV or less to 'circulize' ( more like bring your periapsis pretty close to your apoapsis).  If you search the forum then you should be able to find many threads explain in detail on how to perform a gravity turn.  

You should also note that when you set a manuver node, it assumes a burn that is performed instantaneously and precisely at that time, so a way we perform burns is to take the burn time, split it in half and start thrusting that many seconds before the actual node so that your orbit closly matches what you were trying to get.  You also can probably take off the fins from an upper stage from your picture since that adds mass and drag, which can flip your rocket and will give you more TWR and Dv

[Plusck]

1 hour ago, mrclucks said:

A page and a half of argueing and it seems like the OP hasn't gotten an answer...

He got his answer quite quickly.

On 23/04/2016 at 1:05 AM, Plusck said:

To get the Kerbal 5 into orbit, the simplest approach seems to be to head up at 75° until you get an Ap of about 100km, then head 45° or so until you have an Ap of 250km, then coast and circularize

Even the old paradigm of heading 45° at 10km doesn't work that well - you're better off going much higher before trying to slurp through the atmosphere.

 

A gravity turn will not work for the OP. He is playing the 0.18 demo. With a gravity turn, he'd be lucky to escape the atmosphere.

[Reactordrone]

When testing this it took me a little while to remember that in the old demo you have to turn the SAS off in order to manoeuvre. No wonder I was having such a hard time turning the ship  

[John Doe]

The thing is that I'm trying to learn as much of the basic theory as I can, as to prepare myself for the lunar mission. I've figured out how to establish a nice, fairly ciruclar orbit. The thing is that my instinct is to then establish a parabolic orbit that places the moon on the inside of the orbit, as the transfer orbit, and then to do a burn from there to do a decent stage toward the moon, and finally to to counter my decent at a certain point by firing on the prograde, in order to establish a lunar orbit. Problem is that when trying to do this, my initial burn is a good 13 second burn, and I establish the parbolic orbit as planned. The second one usually ends up on a kerbin escape path, and I end up at an astronomical altitude (like 3K to 4K) orbiting the sun, by which point my rocket says no more, and I have to drop it, leaving poor Mr. Kerman in orbit around the sun forever.

[Plusck]

5 hours ago, John Doe said:

The thing is that I'm trying to learn as much of the basic theory as I can, as to prepare myself for the lunar mission. I've figured out how to establish a nice, fairly ciruclar orbit. The thing is that my instinct is to then establish a parabolic orbit that places the moon on the inside of the orbit, as the transfer orbit, and then to do a burn from there to do a decent stage toward the moon, and finally to to counter my decent at a certain point by firing on the prograde, in order to establish a lunar orbit. Problem is that when trying to do this, my initial burn is a good 13 second burn, and I establish the parbolic orbit as planned. The second one usually ends up on a kerbin escape path, and I end up at an astronomical altitude (like 3K to 4K) orbiting the sun, by which point my rocket says no more, and I have to drop it, leaving poor Mr. Kerman in orbit around the sun forever.

It's a bit difficult to understand what you mean here.

Firstly, I'm really not wanting to sound pedantic but you've got to be careful with your terms, and I'm not sure you mean "parabolic" here. A parabolic orbit stretches to infinity, with velocity tending to zero at infinity. If you enter the Mun's sphere of influence on a trajectory that exceeds escape velocity, that's a hyperbolic orbit. However, I think that here you're talking about an elliptical orbit around Kerbin, with Ap just the other side of the Mun's orbit.

Similarly, when you say "firing on the prograde" I think you mean "retrograde" - since you are descending and you want to slow down to turn a fly-by into an orbit. If you are actually firing prograde then that'll be why it isn't working, but I don't think that's the problem.

I think that the second burn you're talking about is the one to descend to the Mun from the top of your elliptical orbit. You shouldn't have to make any burn here at all, because you should already be getting pulled in by the Mun.

Your transfer orbit from Kerbin should end up with the apoapsis being where the Mun will be at that time. This should be visible on the map when you finish that first burn. If you have to make any corrections to fine-tune the transfer, they have to be done much earlier (halfway there or thereabouts). If the map does not show any kind of intercept, then you've missed the transfer and will need a huge amount of fuel to correct this when you get to apoapsis and find the Mun isn't there.

Finally, you don't want to place the Mun on the inside of an elliptical orbit as such. When you reach apoapsis, you are going very slowly and the Mun (which is on a circular orbit) is going much faster. You don't run into the Mun, the Mun runs into you. That relative difference in velocity puts you on a hyperbolic trajectory around the Mun, hopefully with a periapsis as close to the surface as possible. At Pe, you burn retrograde to slow into an orbit around the Mun. So basically, you want the top of your orbit to be within the Mun's SOI, not beyond it.

If you made a mistake and started burning from Kerbin too late, your elliptical orbit will have its Ap at a point that is ahead of the Mun - so you do indeed have to make a longer first burn to make your orbit longer to allow the Mun to catch up, and it will catch you as you slowly start to fall back towards Kerbin. However, this is less efficient because it means you enter the Mun's SOI will a larger difference in relative velocity, and so you'll have to spend more fuel to capture into orbit.

If you made a mistake and started burning too early, your Ap will be behind the Mun. There is simply no way (other than spending ridiculous amounts of fuel) to correct this by trying to catch up with the Mun when you get near to it. Any kind of correction for this situation is going to be very expensive and/or complicated.

Edited April 29, 2016 by Plusck

[John Doe]

11 hours ago, Plusck said:

It's a bit difficult to understand what you mean here.

Firstly, I'm really not wanting to sound pedantic but you've got to be careful with your terms, and I'm not sure you mean "parabolic" here. A parabolic orbit stretches to infinity, with velocity tending to zero at infinity. If you enter the Mun's sphere of influence on a trajectory that exceeds escape velocity, that's a hyperbolic orbit. However, I think that here you're talking about an elliptical orbit around Kerbin, with Ap just the other side of the Mun's orbit.

Similarly, when you say "firing on the prograde" I think you mean "retrograde" - since you are descending and you want to slow down to turn a fly-by into an orbit. If you are actually firing prograde then that'll be why it isn't working, but I don't think that's the problem.

I think that the second burn you're talking about is the one to descend to the Mun from the top of your elliptical orbit. You shouldn't have to make any burn here at all, because you should already be getting pulled in by the Mun.

Your transfer orbit from Kerbin should end up with the apoapsis being where the Mun will be at that time. This should be visible on the map when you finish that first burn. If you have to make any corrections to fine-tune the transfer, they have to be done much earlier (halfway there or thereabouts). If the map does not show any kind of intercept, then you've missed the transfer and will need a huge amount of fuel to correct this when you get to apoapsis and find the Mun isn't there.

Finally, you don't want to place the Mun on the inside of an elliptical orbit as such. When you reach apoapsis, you are going very slowly and the Mun (which is on a circular orbit) is going much faster. You don't run into the Mun, the Mun runs into you. That relative difference in velocity puts you on a hyperbolic trajectory around the Mun, hopefully with a periapsis as close to the surface as possible. At Pe, you burn retrograde to slow into an orbit around the Mun. So basically, you want the top of your orbit to be within the Mun's SOI, not beyond it.

If you made a mistake and started burning from Kerbin too late, your elliptical orbit will have its Ap at a point that is ahead of the Mun - so you do indeed have to make a longer first burn to make your orbit longer to allow the Mun to catch up, and it will catch you as you slowly start to fall back towards Kerbin. However, this is less efficient because it means you enter the Mun's SOI will a larger difference in relative velocity, and so you'll have to spend more fuel to capture into orbit.

If you made a mistake and started burning too early, your Ap will be behind the Mun. There is simply no way (other than spending ridiculous amounts of fuel) to correct this by trying to catch up with the Mun when you get near to it. Any kind of correction for this situation is going to be very expensive and/or complicated.

Ok to clarify things a bit, when I said prograde, I meant the retrograde marker, which is a prograde burn in orientation to the the moon. (i.e. firing in the opposite direction of velocity, in order to resist and slow the velocity.) Second, by parabolic orbit, i mean an quite elliptical orbit with around a LEO Pe to Kerbin (around 228,000 m or so ) and a really high Ap (around 3K m or so) which extends slightly beyond the moon, while keeping the moon inside the orbital path at the time I'm at or arround Ap, to where essentially I'm on the backside of the moon. (this seems to be, from what I've read in the NASA archives, what Apollo 11 did, when they did most of the dirty work of insertion into lunar orbit and the LM decent on the back side of the moon.)  What I was also finding was that, within the training tutorial, that the retrograde burn was causing me to essentially change directions and end up on what the map calls an escape  path back toward Kerbin.

[Plusck]

40 minutes ago, John Doe said:

Ok to clarify things a bit, when I said prograde, I meant the retrograde marker, which is a prograde burn in orientation to the the moon. (i.e. firing in the opposite direction of velocity, in order to resist and slow the velocity.) Second, by parabolic orbit, i mean an quite elliptical orbit with around a LEO Pe to Kerbin (around 228,000 m or so ) and a really high Ap (around 3K m or so) which extends slightly beyond the moon, while keeping the moon inside the orbital path at the time I'm at or arround Ap, to where essentially I'm on the backside of the moon. (this seems to be, from what I've read in the NASA archives, what Apollo 11 did, when they did most of the dirty work of insertion into lunar orbit and the LM decent on the back side of the moon.)  What I was also finding was that, within the training tutorial, that the retrograde burn was causing me to essentially change directions and end up on what the map calls an escape  path back toward Kerbin.

It isn't really prograde because you should be firing at Pe around the Mun, and you are no longer facing the same way. When you are in the Mun's SOI, the speed of the Mun's orbit around Kerbin is irrelevant - all that matters is how fast you are going around the Mun. However, it is obviously true that some of that velocity comes from the difference in speed between the Mun and your ship when you entered the SOI.

The Apollo missions did use a "free return" transfer, which had the benefit of ensuring that if something went wrong and the circularising burn failed, the ship would automatically find itself geting sent back towards Earth on a re-entry orbit. However, it's a retrograde orbit around the Mun (so a bit more costly to land) and its also a bit less efficient to get there (because you have to climb higher than the Mun's orbiting altitude to get into a retrograde orbit to start with).

So it would be easier just to get the most efficient transfer for a low pass by the Mun: use the maneuvre node to set your Ap altitude at the Mun's altitude about a sixth of an orbit ahead of it, and you should get a good encounter. Drag the node around until Pe is closest to the shadow Mun, then add some more until Pe is at about 20km. You may need to correct inclination halfway there to get an equatorial fly-by (you can tell because the escape path out of the Mun's SOI is in line with its orbital plane). When you reach the Mun's Pe, you burn retrograde for about 260 m/s to get into a low circular orbit.

If you burn the right way at the Mun's Pe (retrograde) you should not be 'changing directions'.

 

[John Doe]

1 hour ago, Plusck said:

It isn't really prograde because you should be firing at Pe around the Mun, and you are no longer facing the same way. When you are in the Mun's SOI, the speed of the Mun's orbit around Kerbin is irrelevant - all that matters is how fast you are going around the Mun. However, it is obviously true that some of that velocity comes from the difference in speed between the Mun and your ship when you entered the SOI.

The Apollo missions did use a "free return" transfer, which had the benefit of ensuring that if something went wrong and the circularising burn failed, the ship would automatically find itself geting sent back towards Earth on a re-entry orbit. However, it's a retrograde orbit around the Mun (so a bit more costly to land) and its also a bit less efficient to get there (because you have to climb higher than the Mun's orbiting altitude to get into a retrograde orbit to start with).

So it would be easier just to get the most efficient transfer for a low pass by the Mun: use the maneuvre node to set your Ap altitude at the Mun's altitude about a sixth of an orbit ahead of it, and you should get a good encounter. Drag the node around until Pe is closest to the shadow Mun, then add some more until Pe is at about 20km. You may need to correct inclination halfway there to get an equatorial fly-by (you can tell because the escape path out of the Mun's SOI is in line with its orbital plane). When you reach the Mun's Pe, you burn retrograde for about 260 m/s to get into a low circular orbit.

If you burn the right way at the Mun's Pe (retrograde) you should not be 'changing directions'.

 

So then from what I'm reading here, is that upon reaching the Pa node of the tranfer orbit, near the moon, the moon should then, by operation of it's field of gravity, start causing a descent in line with it, since the moon's gravitational field should have more influence than kerbin's ? and what I'm seeing on the lunar insertion burn is that I end up changing directions after about 30 into the burn..

[Plusck]

 

57 minutes ago, John Doe said:

So then from what I'm reading here, is that upon reaching the Pa node of the tranfer orbit, near the moon, the moon should then, by operation of it's field of gravity, start causing a descent in line with it, since the moon's gravitational field should have more influence than kerbin's ? and what I'm seeing on the lunar insertion burn is that I end up changing directions after about 30 into the burn..

Yes, at your Ap around Kerbin, the Mun should catch up with you and you will fall towards it, under its gravitational influence, and the closest point to the Mun is that "Pe" you see in map view.

This is the 0.18.3 demo, right?

On map view, your orbit should change colour when it enters the Mun's Sphere of Influence (SOI). Since three-body gravitational equations (or n-body, if you include the sun and other planets) would be horrendously complex to model, KSP simplifies these into Spheres of Influence. When you are within the Mun's SOI (i.e. you can see your orbit in blue making a graceful hyperbolic curve around the Mun) then Kerbin's gravity is no longer having any effect on you whatsoever.

In real life, the planet's gravity would have a very slight effect, and more so toward the edges of the moon's SOI, but close to the moon it's gravitational influence completely dominates.

Anyway, what you should be seeing in the map view is this:

 

Edited April 30, 2016 by Plusck

[John Doe]

OK. Got it, and successfully pulled it off. I was just ending up with the moon right in the line of my initial escape trajectory towards it, so that my Pe to the moon was zero, with either the option of travelling the other direction, OR the option of smashing into the moon at a relatively high velocity.