LV-N atomic engine question

[creator1629]

cool thanks guys this has been really helpful info. i was wondering, do you guys have a rough rule of thumb on how to many periapsis kicks you do for a certain amount of delta v needed? like say for example 1 kick per 1000 delta v or something like that. not an absolute rule but just as a general reference guide

[tavert]

It's more time-based than anything. So the amount of delta-V you'll get in a given time depends on your TWR. I'd say pick an angle, lower if you're more patient and want to do more kicks, and do kick burns for as long as your prograde vector is within that angle of your velocity vector at periapsis. If a kick will eject you from Kerbin's SoI towards the end of a burn, then you probably want to stop that kick early and come around one last time to do the final interplanetary ejection.

[RoboRay]

cool thanks guys this has been really helpful info. i was wondering, do you guys have a rough rule of thumb on how to many periapsis kicks you do for a certain amount of delta v needed? like say for example 1 kick per 1000 delta v or something like that. not an absolute rule but just as a general reference guide

It varies with your thrust to weight ratio, so no. The higher the thrust you have, the fewer kicks you'll need.

Of course, if you have a high TWR on an interplanetary craft, efficiency isn't your goal to begin with so you're probably not going to be doing Pe kicks.

[capi3101]

We ever get an answer on how long a periapsis kick burn should last? One minute? Two?

[RoboRay]

We ever get an answer on how long a periapsis kick burn should last? One minute? Two?

It's up to you.

The shorter the burn the more efficient it will be, but that also requires more kicks (and more time to perform them), so you'd need to plan ahead and start your kicks earlier so that you're at the right phase angle for the last one.

I usually burn for about two minutes on each side of the node (or around 20 degrees on each side of the ejection angle, which works out to about the same length of time).

[creator1629]

i have a question regarding periapsis kicks. i have been using the website ksp.olex.biz to determine the ideal phase angle and ejection angle to rendezvous with a planet. but the ejection angle seems to be based on a rather circular orbit, how is this factor affected by my orbit being very elliptical as a result of multiple periapsis kicks?

[M5000]

i have a question regarding periapsis kicks. i have been using the website ksp.olex.biz to determine the ideal phase angle and ejection angle to rendezvous with a planet. but the ejection angle seems to be based on a rather circular orbit, how is this factor affected by my orbit being very elliptical as a result of multiple periapsis kicks?

I'm not 100% sure, but I don't think it would matter in relation to that aspect. The circular orbit accuracy factor refers to the fact that finding an angle on an ellipse is harder than just a perfect circle, and that the math they did was based on circles. If you're at that singular point in orbit and it's the same phase angle as you're supposed to be at, it shouldn't matter as long as you burn at basically the same point every time.

[RoboRay]

As long as you burn evenly on each side of the ejection angle, the Pe of your orbit will stay at the ejection angle, so the eccentricity doesn't matter.

[CalculusWarrior]

ksp.olex.biz will calculate the phase angle that you need assuming an instantaneous velocity change. If you are using the periapsis kick method, use the ejection angle for your initial burn, but be sure to have the phase angle be a little bigger/smaller than is required. Ideally, you want to escape Kerbin's SoI at the exact instant the phase angle is the number calculated.

[RoboRay]

The reasonable phase angle window (within 1% of minimum-energy requirements) is days wide, so it only really becomes a factor if you maximize the benefit of Pe kicks by pushing the last one out to near Minmus (since a flight to the Mun's orbit and back is measured in hours).

It's a good thing to keep in mind, though. If you are starting the kicks late, at or just past the ideal phase angle, you can still depart within the window by skipping the final kick and setting up your node for the departure burn while your Ap is still near the Mun's altitude. Energy-wise, you should only lose a few meters per second of your bonus and your final burn will just be a few minutes longer.

Edited May 4, 2013 by RoboRay

[creator1629]

awesome this all makes a lot more sense to me now, thanks for the info guys. regarding an extended burn of several minutes using a maneuver node, a portion of it is done prior to the actual node and a portion is done after it has passed. my question is regarding which way you point your craft during the burn, say for example the maneuver requires a prograde burn, before you hit the node, the maneuver node marker is a little ways off the actual prograde direction, then when you hit the node, the marker and the prograde direction are one and the same, then after the node they start to separate again. so during the whole burn process, do you follow the marker as it moves along, or keep the direction fixed on the prograde marker knowing its a prograde burn?

[duncan1297]

this is what i usually do.

[RoboRay]

awesome this all makes a lot more sense to me now, thanks for the info guys. regarding an extended burn of several minutes using a maneuver node, a portion of it is done prior to the actual node and a portion is done after it has passed. my question is regarding which way you point your craft during the burn, say for example the maneuver requires a prograde burn, before you hit the node, the maneuver node marker is a little ways off the actual prograde direction, then when you hit the node, the marker and the prograde direction are one and the same, then after the node they start to separate again. so during the whole burn process, do you follow the marker as it moves along, or keep the direction fixed on the prograde marker knowing its a prograde burn?

If there is no normal or radial component to your injection, it's more efficient to stay on prograde through the burn. If you're incorporating an inclination change or swinging your line of apses around, though, you'll have to point at the maneuver node at some point.

I normally stay on prograde through all my kicks and then aim at the maneuver node for the final departure burn.

Edited May 4, 2013 by RoboRay

[tavert]

If you burn in the direction of the node, you will change around your orbit a bit more (if your periapsis is close to the top of the atmosphere, burning slightly down before you get to the node can drop your periapsis and you might lose a tiny bit to drag) and be less efficient than burning prograde, but your end orbit will more closely match the prediction from the instantaneous impulse that the maneuver node showed you. You probably care more about efficiency than accuracy for the kick burns, but more about accuracy for the final ejection burn.

[tssn1611]

Question regarding the periapsis kicks: that works fine when your orbit stays inside Kerbin's SOI, but what about the last one (the one that takes your periapsis outside Kerbin's SOI)? For transfers to Jool or Eeloo, nearly half your delta-V needs to be expended on the last kick, and for some of my interplanetary ships, this is a 10-minute burn (or longer). Any thoughts on how to do that efficiently?

Two days old doesn't count as necro-ing does it?

[RoboRay]

Your last burn will be much longer, as you note. You can minimize it by pushing your Ap out to Minmus on your last kick, but that makes your orbital period several days long, which may be less efficient in the long-run if it pushes your final burn past your optimal phase angle.

However, by that time, your orbit is going to be so elliptical that the curvature on each side of your Pe is much shallower than if you were in a circular orbit. This helps efficiency, as more of your thrust is going to be aligned with your prograde vector even if there is a normal or radial component to your final injection burn.

[creator1629]

yeah i had that same problem too, once the orbit becomes too elliptical the time to get back to the periapsis can take days, as a result you may miss your launch window, so i try to estimate and start my kicks several degrees before the actual launch window, though this can have a large margin of error.

hey i have a question, i have been using mechjeb to determine the planetary phase angles but as to the ejection angle i have been generally eyeballing it. which mod do you guys use and is the best/has the most complete information when determining planetary intercepts?

[wilt57]

I understand that it is possible to use 1 NERVA behind multiple Jumbo-64s and get an escape trajectory from Kerbin using PE kicks. It may be tough, but possible. My question regarding the NERVA per fuel weight ratio involves capturing orbit around the other planets. I can use all the time I want, in theory, making PE kicks to displace my D-v into multiple burns, but I need one long burn to turn my encounter into an orbit, right? Otherwise I would enter a Kerbol orbit. So, if I understand correctly, TWR may not be all that important in the step to escape Kerbin's SoI, but it is gonna be important when I get to my destination. Assuming no atmospheres for aerobraking and taking best advantage of negative gravity assists, won't it still take a long burn to close my orbit?

[tavert]

Depends where you're going, but especially in the cases of Moho, Eeloo, and Dres, then yes. You may need to be burning retrograde at very high altitude in order to successfully capture around these bodies in time if your TWR is low. But at least they're relatively small so the Oberth loss of having to burn at high altitudes will not be quite as bad as it would be for Kerbin.

[RoboRay]

I understand that it is possible to use 1 NERVA behind multiple Jumbo-64s and get an escape trajectory from Kerbin using PE kicks. It may be tough, but possible. My question regarding the NERVA per fuel weight ratio involves capturing orbit around the other planets. I can use all the time I want, in theory, making PE kicks to displace my D-v into multiple burns, but I need one long burn to turn my encounter into an orbit, right? Otherwise I would enter a Kerbol orbit. So, if I understand correctly, TWR may not be all that important in the step to escape Kerbin's SoI, but it is gonna be important when I get to my destination. Assuming no atmospheres for aerobraking and taking best advantage of negative gravity assists, won't it still take a long burn to close my orbit?

You can do kicks in reverse, on the other end...

You need one long burn at the closest point of approach, much like your last burn on the transfer injection, to get captured into a highly elliptical orbit. Once you're captured, you can make repeated retrograde burns near Pe to bring your Ap down more efficiently than with one long burn (if the delta-v requirements are high enough to even bother).

The challenge is making sure you burn enough on the first pass to actually get captured to begin with.

Personally, I usually just aerocapture if there's an atmosphere to take advantage of. I've only been to Moho and Eeloo one time each, so burning to capture is a rare event in my space program.

yeah i had that same problem too, once the orbit becomes too elliptical the time to get back to the periapsis can take days, as a result you may miss your launch window, so i try to estimate and start my kicks several degrees before the actual launch window, though this can have a large margin of error.

hey i have a question, i have been using mechjeb to determine the planetary phase angles but as to the ejection angle i have been generally eyeballing it. which mod do you guys use and is the best/has the most complete information when determining planetary intercepts?

I use the Protractor mod for my phase angle and ejection angle info.

Edited May 7, 2013 by RoboRay

[wilt57]

You can do kicks in reverse, on the other end...

You need one long burn at the closest point of approach, much like your last burn on the transfer injection, to get captured into a highly elliptical orbit. Once you're captured, you can make repeated retrograde burns near Pe to bring your Ap down more efficiently than with one long burn (if the delta-v requirements are high enough to even bother).

The challenge is making sure you burn enough on the first pass to actually get captured to begin with.

Personally, I usually just aerocapture if there's an atmosphere to take advantage of. I've only been to Moho and Eeloo one time each, so burning to capture is a rare event in my space program.

I think I'm overestimating how much D-v it will take to form the orbit. I understand that once I get an elliptical orbit, I can slowly work my AP down. I was just thinking I'd need a higher TWR to form the elliptical orbit in the first place. That and I keep forgetting, for whatever reason, that Duna has an atmosphere for aerobraking. That combined with tavert's point of Oberth not being very crucial on some planets/moons means that there should be more time to close the orbit than I was thinking there'd be.

In a most extreme case, I suppose there is a theoretical TWR where one would not be able to close the orbit, but that ratio may have many zeros to the right of the decimal point! Like I said, I think I was over-thinking it. Thanks to both of you for setting me straight!

[RoboRay]

You can always capture, even with a ridiculously low TWR, if you plan ahead. On my one ion-probe mission to map Moho, I determined that I would need a two hour burn to barely capture just into a highly elliptical orbit, and at my interplanetary velocity would only be in the planet's SOI for one hour. Compounding that, the several minutes right after Pe would be in shadow, over the planet's night-side, so I'd have no power for the engine. So, a little over an hour before reaching the SOI transition, I aimed my dinky little blue lightbulb at Moho's limb, just above the south pole, and started the burn.

I made it. It's still the only time I started a capture burn before even entering the destination SOI.

[creator1629]

hey RoboRay, i installed the protractor mod, and it works great. but i do have a question, when going to moho which has a somewhat elliptical orbit i notice that the protractor doesn't seem to compensate for it and my maneuver seems way off the mark. how do you get around this, do you wait for the next launch window when moho isn't in a particularly "elliptical" point in its orbit?

[RoboRay]

None of the current mods or calculators for KSP transfer injections account for orbital eccentricity or inclination, AFAIK.

What I usually do is use Protractor to identify roughly where my maneuver node should be (I start with it right on the ejection angle), and adjust the maneuver to get the encounter.

[Francesco]

moho can be quite painful to reach, depending on whether it is at its periapsis or apoapsis around the Sun: be prepared to spend more delta-v than planned if you get an encounter at its PE.

anyway, you can trust protractor only to a certain point after that, maneuver nodes become your best bet.

they make life so much easier than it was pre-0.18.