Maneuver nodes don't seem to be very useful for extremely long burn times (using a nuclear engine)

[Thorazine]

Hello everyone.

I'd like to take a big rocket to Duna using the nuclear engine.

When the maneuver node calculates a maneuver, I believe it assumes the engine will fire for only a few seconds, or maybe a minute at most.

However, my engine must fire continuously for 60 minutes to get to Duna, and the resultant path is quite different from the maneuver-node plotted path. In fact, it misses Duna entirely.

Those of you who use nuclear engines - how do you plot your course to other planets when you must have a very long engine burn time?

Thank you to everyone in advance.

[Red Iron Crown]

I generally use more engines to get the TWR high enough that burns don't take that long. It's less efficient but more enjoyable play.

If that is not an option (using ion engines or trying to squeak every bit of delta-V out of a nuclear or chemical design), it is better to split the big burn up into several smaller burns, all performed at the same spot in the orbit. I combine this with a higher parking orbit, around 150km or so, so that the vessel doesn't move as much while burning.

[Scarecrow88]

The manoeuvre node system generally works on you being able to change your velocity instantaneously. As we all know, this isn't feasibly possible, so the accepted method of working is to split your burn time around the node point. i.e. if it is predicting a 2 minute burn, you should start your burn one minute before reaching the node and continue burning for one minute after. Having said that, I have never built a ship that requires a 60 minute burn as you are required to do, so am not sure what inaccuracies creep in for a burn of that duration.

[OhioBob]

I think the longest burn that I ever had to do was somewhere in the 5 to 10 minutes range. If longer than that, I do what Red Iron Crown suggested and use more engines. If you have to go with a 60 minute burn, then perhaps you can break it up into a series of shorter burns. Make a maneuver node, burn for 10 minutes and then stop. Create a new maneuver node and burn for another 10 minutes, and so on. You might have to perform your first several burns through successive periapsis passages. Each time you raise your apoapsis higher and higher until the last burn places you on your escape trajectory. It sounds like a real pain to me but it might work. You'd likely have to start the process days before your intended launch window because of the time it will take to cycle through several periapsis passages.

Edited February 1, 2015 by OhioBob

[eddiew]

Your problem is that the orbit is continuing while you burn... during that 60 minutes, you've probably looped Kerbin, which the node never accounted for.

A very high starting orbit (past Mun) should help, since you won't be fighting your angular momentum I believe you will use slightly more delta-v however, due to the oft befuddling Oberth Effect.

More thrust-to-weight is another answer. I wouldn't care to hit a transfer burn from LKO with less than 0.4 TWR.

[juanml82]

My interplanetary transfer stages usually have four nuclear engines (or the MRS quad nuclear engine). That typically provides decent thrust while not lowering too much delta-v.

Here's my transfer stage

Otherwise, as others said, you should split your burn into several burns. However, if your burns put your Ap at around Minmus orbit, you'll find that your orbital period will be of several days, and by the time you hit the Pe, the transfer window might be more expensive, because the best window had already gone through. If I need to split burns, I usually only split twice: one prograde burn of some 800 dV (or, if I have a leftover ascend stage with several hundred dV left, enough to burn that stage so I don't need to calculate burn times with different TWR). Then the second burn puts me in a rough intercept course.

In fewer words: moar engines!!

[cantab]

A 60 minute departure burn is seriously long. As you've found, the manoeuvre nodes might not be entirely useful there.

[GoSlash27]

When setting up an interplanetary burn with long burn times, I'll park in a highly eccentric orbit while waiting for the window.

This gives me a much shorter burn during the actual transfer while keeping the Oberth effect in play.

Although I've never had to perform a 60 minute burn...

Best,

-Slashy

[Pecan]

60 minute burn to Duna! Hell, that would mean more than 41 minutes just for a Mun burn!

Either you're seriously under-powered (heavy craft, single engine) or you're trying to burn all the way until you get there.

@all: Even splitting that up over multiple orbits there's no way that can be practical, surely?

Pictures, .craft file or at least the mass of your ship please, so we can give a better recommendation.

[Laie]

The manoeuvre node system generally works on you being able to change your velocity instantaneously.

(nitpick) relatively instantaneously. Kerbin being such a small world and orbital periods short makes this worse than it is IRL. When returning from a Tylo-like orbit around Jool, a 20-minute node&burn will be as precise as a ten-second burn in low Kerbin orbit.

@all: Even splitting that up over multiple orbits there's no way that can be practical, surely?

I just assumed that the OP was firmly in the "more than one engine is a waste of fuel" camp. But yes, 60min to Duna is quite extreme. I wonder if the OP knows about launch windows? If yes, he could do the first 45min in small increments, raising his apoapsis past Minmus. That would be *many* small increments... I don't consider this practical, either. But it can be done and would kinda-sorta work.

[Empiro]

How many m/s is the maneuver? While it's possible that you need more engines, I think it's more likely that your transfer is not efficient as it could be. It can take less than 1100 m/s to transfer to Duna if you do it right.

[Dispatcher]

For interplanetary burns, you might try doing shorter burns; setting up new burns with the maneuver nodes in between those burns. Also, as others have pointed out, using multiple engines helps. I usually use 3 LV-Ns in tandem.

[Venusgate]

From what I have seen, the Maneuvering nodes don't factor in ISP or stages, they just work off of your current TWR.

Let's say you start a 30 second burn at 10.0tons at 15 seconds to node. You're probably done burning at +14seconds with 9.9tons (.1 tones fuel spent) and your TWR has gone up a little.

Magnify this to 60 minutes, and yes, the discrepancy will become much bigger.

This also happens during "suicide" burns for landing.

My advice until SQUAD develops a more complicated "Burn Time" equation: start taking note of the leftover burn time for each class of ship you are using, and subtract the split from the burn time. I.E. if you're space stations are ending up with 20 seconds left over on a 3 minute burn, start burning at 1 min 20 seconds to node, instead of 1 min 30 seconds.

Even this method is not perfect, but it is more precise.

[OhioBob]

If you want a more accurate answer, the burn time can be calculated using the following equation:

t = ( m_o * I_sp * g_o / F ) * ( 1 - 1 / EXP[ ÃŽâ€v / ( I_sp * g_o ) ] )

where,

t = time (s)

m_o = initial mass (kg)

I_sp = specific impulse (s)

g_o = standard gravity (9.81 m/s²)

F = thrust (N)

ÃŽâ€v = change in velocity (m/s)

[juanml82]

Or use Mechjeb, which has an accurate maneuver node calculator.

BTW, could it be that the OP had a very long estimated time because the last time he had fired the engines, he did it at low throttle and the game used that to calculate?

OP, where are you?

[KerikBalm]

Perapsis kicking... learn it, love it

[KerBlam]

Perapsis kicking... learn it, love it

What is it, and where do you learn it?

[Red Iron Crown]

What is it, and where do you learn it?

It's splitting the burn into many smaller burns, all done at the same periapsis. This pic (not mine) is worth 1000 words, shows each successive orbit nicely:

[maltesh]

I usually don't set a starting orbit higher than 150km, because 150 km will take a 20-minute slavish node-burn without dropping your spacecraft in the atmosphere, and any spacecraft that /needs/ a 20-minute burn is probably sufficiently stocked with delta-V that it can eat the losses to get it to that altitude. That said, 7-10 minute burns are more the norm for my travels.

I generally use the Protractor mod, as I've not yet found a better mod to constantly show my closest-approach to the target.

If I'm going to do a twenty-minute burn, it's generally Mechjeb to hold the course (and to give me a true reading on how long the burn is going to last, the in-game UI sill lies about it fairly often), and Kerbal Alarm Clock to let me know 30-60 seconds before it is scheduled to end.

I then watch Protractor until my closest-approach stops getting closer, and kill the burn. After which, I plot a mid-course correction.

Edited February 3, 2015 by maltesh

[WuphonsReach]

For long burns, you will need to know the burn time using a tool like KER (Kerbal Engineer Redux):

http://forum.kerbalspaceprogram.com/threads/18230-0-25-0-90-Kerbal-Engineer-Redux-v0-6-2-12%280-25%29-and-v1-0-14-1%280-90%29

In the Orbital window, it will tell you how long the burn will take at 100% thrust as well as how soon before the maneuver node you need to start the burn.

Alternately - let MechJeb execute the burn.

For ejection burns (leaving Kerbin's SoI) over 5 minutes in duration, I prefer to park the vessels in a 3000km orbit well prior to the transfer window. With such a long orbital period, a 10 minute burn is only a small fraction of the total orbital period at 3000km compared to a 100km orbit.

Orbital period at 100km is about 30 minutes, orbital period at 3000km is a bit over 6 hours. Much easier to long-duration precision burns when you don't have to time the burn down to the nearest second.

Edited February 3, 2015 by WuphonsReach

[Brainlord Mesomorph]

I'm writing a tutorial on what someone here just called "perigee kicking" or IRL the Mangalyaan Maneuver (named after an Indian Mars probe, with apparently a very very low TWR).

All stock, no mods. No (hard) math. The key is knowing when to start and which direction to go off in. And I have now figured out "one weird trick" for doing just that.

OP, pls PM me, and I'll explain more (not quite ready to go public with this yet)

PS: did you use this page to calc your departure window? (http://alexmoon.github.io/ksp/ )

Edited February 5, 2015 by Brainlord Mesomorph

[Brainlord Mesomorph]

Here's my transfer stage

the RC module and the SAS module both have torque. if you move one to the other end of the ship it'll handle better.

[OhioBob]

OP, pls PM me

The OP appears to be AWOL.

[MarvinKitFox]

I'd like to take a big rocket to Duna using the nuclear engine.

However, my engine must fire continuously for 60 minutes to get to Duna, and the resultant path is quite different from the maneuver-node plotted path. In fact, it misses Duna entirely.

60 minutes, to achieve 1080 m/s???

so, your TWR is under 0.03?

You could be efficient, and execute the ejection burn as a sequence of short burns at perigee, thus maximising Oberth.

But realistically, just pump your initial orbit up to 3000 km

Your ejection burn will now be under 400m/s, thus taking 25 minutes .

25 minutes at this alt is less than 1/20 of an orbit, thus your burn's distortion is negligible.

Frankly, that sort of TWR ratio is what I expect from an ION ship, not a nuke.

[Red Iron Crown]

I'm writing a tutorial on what someone here just called "perigee kicking" or IRL the Mangalyaan Maneuver (named after an Indian Mars probe, with apparently a very very low TWR).

All stock, no mods. No (hard) math. The key is knowing when to start and which direction to go off in. And I have now figured out "one weird trick" for doing just that.

Ugh, I didn't know someone else was working on this, I just today posted a precomputed periapsis-kick-style escape in Tutorials (see link in sig). It's pretty basic, though, if you'd like to use some of the information from there in yours or use the spreadsheet I made to do some computation of your own by all means feel free. Or PM me if you'd like to kick some ideas around.