Transfers - again

[technion]

Hey guys,

I know this topic is somewhat done to death but hopefully I'm asking a slightly newer question.

I've never been overly precise with transfers and often didn't mind just circling the system once or twice, but a new game with the combination of TAC, and not all of career mode opened, is killing me.

I'm trying to get a rocket to Eeloo. I've done it three times now but TAC has killed me on the way back each time

One of the difficulties to come up each time has been how to do the transfer. By the time my rocket is in orbit, there is ~800dv left on its (Edit: Skipper powered ) lifter stage, and planning a manoeuvre node tells me it's going to take about two minutes. Only one minute in, I drop that stage, and now I'm powered by one nuclear engine and have 24 minutes left on the burn. So question number one, how do you time this?

In several of my efforts, I've found that, by the time I'm 15 minutes in, I'm pointing directly at the blue manoeuvre marker and yet my remaining dv on the node starts going up. My orbit doesn't look like it's moving. I can turn and point prograde and watch the orbit get a lot larger a lot quicker - but that doesn't address the fact that, based on the Alexmoon calculation, a large amount of the Eeloo burn is supposed to be on the normal plane.

How do you plan a move like this?

Edited February 24, 2014 by technion

[brooklyn666]

Your biggest problem is that you're starting your burn at the wrong time. The maneuver node is telling you your burn time based on your current engine, but it also assumes your burn will be one instantaneous impulse. What you want to to is time it so that you are halfway through your burn AT the node. If your burn time is 24 minutes, start your burn 12 minutes before the node.

Also, you can try doing multiple periapsis burns. This will make your final ejection burn shorter. And if you're feeling adventurous, try a Jool slingshot. You can save a ton of dV that way.

My biggest pieces of advice would be to time your burns, and use only one stage for the ejection burn if you can.

[Kasuha]

I've found that, by the time I'm 15 minutes in, I'm pointing directly at the blue manoeuvre marker and yet my remaining dv on the node starts going up. My orbit doesn't look like it's moving. I can turn and point prograde and watch the orbit get a lot larger a lot quicker - but that doesn't address the fact that, based on the Alexmoon calculation, a large amount of the Eeloo burn is supposed to be on the normal plane.

Yes, this typically happens when you're doing transfer burns with very little thrust. After a while you happen to be in a very different place and the blue marker is misleading - it doesn't show you right direction or dv anymore.

I have two tricks for that.

Trick 1: Notice where the blue marker is pointing at T+0 and keep that direction even if the marker runs away.

Trick 2: Go to map view and check your trajectory and your planned trajectory. The planned trajectory has an escape marker on it (a circle) and when you put your mouse over it, you can see the time in which you are supposed to be leaving the SOI. Rotate the map and zoom in to see them both at maximum possible zoom (you need to rotate it so the planned trajectory is running "towards you"). To get roughly correct escape trajectory, you need to place your real trajectory's escape circle over this planned circle with matching timestamps. Forget about blue marker - continue burning in original direction while the two markers get closer. If the real one appears to run by the planned one, add radial/normal deviation to make them converge again. Also watch the time, don't let the real time go below the planned time or you'll need to correct retrograde.

You will spend more dv than the maneuver told you. Not only you lost on Oberth effect, you also spend some dv on not burning in the right direction. But in the end, only small correction should be needed after leaving the SOI to get the intercept you need.

Edited February 24, 2014 by Kasuha

[technion]

. If your burn time is 24 minutes, start your burn 12 minutes before the node.

.

But how do you align that with the multiple stages I referred to? My burn time is 2-3 minutes until I'm halfway through it.

[blizzy78]

By the time my rocket is in orbit, there is ~800dv left on its (Edit: Skipper powered ) lifter stage, and planning a manoeuvre node tells me it's going to take about two minutes. Only one minute in, I drop that stage, and now I'm powered by one nuclear engine and have 24 minutes left on the burn. So question number one, how do you time this?

My advice would be: You don't. Drop unnecessary delta-v from the lifter stage, and instead increase the delta-v of your cruise stage. This way you can stay on the same engine for the complete burn.

[Kasuha]

My advice would be: You don't. Drop unnecessary delta-v from the lifter stage, and instead increase the delta-v of your cruise stage. This way you can stay on the same engine for the complete burn.

It makes things simpler but does not solve the problem. If the burn is 20 minutes long, starting 10 minutes ahead does not get rid of the problem with navball maneuver marker becoming unreliable towards the end of the burn.

[irongut]

One of the difficulties to come up each time has been how to do the transfer. By the time my rocket is in orbit, there is ~800dv left on its (Edit: Skipper powered ) lifter stage, and planning a manoeuvre node tells me it's going to take about two minutes. Only one minute in, I drop that stage, and now I'm powered by one nuclear engine and have 24 minutes left on the burn. So question number one, how do you time this?

Drop the launch stage and do the transfer burn purely using your transfer stage. I'd also add more engines to the transfer stage to increase your TWR and reduce the burn time. Then time your burn so you do half of it before the node and half after - so a 10 minute burn starts 5 minutes before the node.

[brooklyn666]

It makes things simpler but does not solve the problem. If the burn is 20 minutes long, starting 10 minutes ahead does not get rid of the problem with navball maneuver marker becoming unreliable towards the end of the burn.

1) Like I said before, the navball is inaccurate because it assumes your entire burn takes place as one instantaneous impulse. There's no way around it, so you have to learn some tricks to cope with it. Like I said before, you cab do multiple Pe burns. Shorter burns are more accurate, so that will help. Also, towards the end of the burn, keep pointing towards your burn marker until it starts to move, then stay where it was when it starts to drift, cutting off as close to 0 as possible. Then, if necessary, use RCS to trim.

2) If you want, you multiple periapsis burns to reduce the length of each burn

[Kasuha]

1) Like I said before, the navball is inaccurate because it assumes your entire burn takes place as one instantaneous impulse. There's no way around it, so you have to learn some tricks to cope with it.

I know. See my post above. I was not disputing that.

Like I said before, you cab do multiple Pe burns.

That helps only to a degree. You can't afford the orbit to be too long or you may miss your transfer window (and we don't have very good tools to plan that in advance). And with transfers far away (Moho/Jool), it might still be only small part of the total burn (been there, tried that).

Plus you got to set up the maneuver again after each pass.

[Kryxal]

I'd probably do two burns, first burn is everything you can SAFELY get out of the skipper and jettison, plus about 5 seconds of thrust from the nuke. Second burn breaks Kerbin orbit and is going to be pretty long, but you'll have a better idea HOW long after the 5 seconds you burned earlier. Just be careful you're not running into the Mun ... or if you do, use it!