Optimum ascent path for EVE

[cicatrix]

I've finally managed to construct a craft having about 12 km/s of dV which is theoretically capable of taking off from Eve's sea level. To test that I hyper-edited it to Eve and tried the ascent. Well, in theory I should have entered the low Eve orbit but the ascent path was far from being optimal and thus my craft never reached the orbit.

Since I know that 12 km/s should be enough, the only reason I could think of about my failure was about wrong ascent path.

I reasoned, that since I make a gravity turn at about 9-10 km on Kerbin, I should make the turn at about 15 km on Eve. Anyway, either I'm too sloppy or I am missing something.

So, the question is - is there any rule of thumb I should follow when making an ascent from a planet with atmosphere in general and on Eve in particular? I even tried MJ's ascent guidance and used it 'edit ascent path' option, but it gave me no clues about where should I really make a G-turn and how sharp it should be.

Edited December 1, 2014 by cicatrix
change prefix

[LABHOUSE]

You need to get a pressure thing and see the pressure at 10,00 on kerbin and wait for that pressure on eve.

[Jouni]

It's been a while since I last visited Eve, so I might remember some things wrong.

Eve's atmosphere is 5x thicker than Kerbin's, and the scale height is 40% higher, so you need to climb much higher than on Kerbin before starting the turn. As a rule of thumb, you might want to climb to 30 km without exceeding the terminal velocity (see the wiki) before using any significant fraction of thrust horizontally. At this point, your climb rate should be around 500 m/s, which you should aim to sustain, until the apoapsis hits 100 km. After that, you can let the climb rate fall, while still accelerating horizontally.

So: Climb vertically until 30 km at close to terminal velocity. Then sustain the climb rate, while using any remaining thrust for horizontal acceleration. After the apoapsis hits 100 km, concentrate on horizontal acceleration, while keeping the apoapsis at 100-110 km. This is not optimal, but it's close enough that it usually doesn't matter.

[Red Iron Crown]

Straight vertical climb until 30km, then gravity turn as appropriate for the vessel's TWR. If using mechjeb, set turn start to 30km and turn end to 95km, the curve shape will again depend on TWR.

[MarvinKitFox]

12km is enough, but only if you fly a very neat ascent.

As per the guys above, your ascent profile is pure vertical to about 30, then start a gradual turn.

Don't even think of being near horizontal until something above 75km.

I would advise you up your delta-v quota to something like 14.5, it will give you the elbow room to ascend without forcing the need for a perfect curve.

[mhoram]

What is the TWR of your ship? If it is too low, you will lose a lot of Delta-V to gravity.

Regarding the ascent path for Eve, fibonatic made some calculations about them:

http://forum.kerbalspaceprogram.com/threads/46194?p=1407898&viewfull=1#post1407898

But be aware: the ideal ascent path depends a lot on the rocket configuration itself.

Edited November 29, 2014 by mhoram

[OhioBob]

A few weeks ago I was planning my first return mission from Eve. To try to get my Eve launch vehicle right I created a computer simulation to test various designs. I adjusted many different factors, such as TWR, to try to optimize the design the best I could. My actual final design configuration ended up a little different than what I simulated, but the simulation provided a good basis to start from. I haven't actually flown my mission yet, so I can't vouch for how well it will work out in actual game conditions.

The simulation is based on the stock game, so if you're using any mods the results will likely be different. My simulated design used a three-stage central core with three pairs of strap-on boosters. The strap-ons and the first stage were arranged to utilize asparagus staging. The two upper stages of the core were ignited in series after the strap-ons and first stage were jettisoned. The simulation assumed launching from am altitude of 3,000 m, and I was shooting for an orbital altitude of 105,000 m.

I found that TWR is critical to a good design, thus I spent quite a bit of time going through multiple iterations until I hit on what appeared to be the optimum. Below is a graph of my simulated launch vehicle's TWR versus time. Each peak represents burnout of one of the stages. The first three peaks are the pairs of strap-ons and the final three peaks are the stages of the central core. The TWR is based on Eve gravity, where g_o = 16.677 m/s². If your launch vehicle's TWR is significantly outside the range represented by the graph below, I'd recommend modifying it. I found that being significantly outside the optimum range yielded disastrous results in most cases.

Controlling the rocket's attitude is also an important factor, though I didn't find the altitude of initial pitch-over to be all that critical. I started a slow gradual pitch-over immediately after jettisoning the last of the strap-ons stages, which for the simulation was at an altitude of about 18,000 m. I think it is important that we don't put too much emphasis on what the pitch needs to be, but rather on what the flight path angle should be. The flight path angle is the angle that the velocity vector makes with the local horizon. We alter the pitch simply to maintain the correct flight path angle. Below is a graph of the simulation's surface velocity flight path angle versus altitude.

At some point during ascent the NAV ball will switch from surface velocity to orbital velocity, but by the time that happens the two vectors should be pretty close to each other. The result shown here was obtained by keeping the pitch vector just barely ahead of the velocity vector throughout the gravity turn (the pitch led the velocity vector by a constant 0.23^o). In practice it's unlikely that a proper separation between the pitch and velocity vectors can be constantly maintained, thus frequent small adjustments will undoubtedly be necessary. Furthermore, your rocket configuration may require a different ascent profile than what I simulated. You're going to have to find out what works for you. I would definitely recommend trying to make a slow smooth transition from a vertical velocity vector at about 20 km to a horizontal velocity vector at about 100 km.

For the simulation, engine cutoff occurred at an altitude of about 99.5 km with a flight path angle of +1 degree. This was followed by a coast up to 105 km and a small circularization burn.

Edited December 1, 2014 by OhioBob
spelling

[Signo]

I think that (like OhioBob said) TWR is the key while designing a succesful Eve lander. If I may give a hint based on my experience, I made it just when I abandoned the idea of using aerospikes.

[EdFred]

If using mechjeb and departing from sea level, 12k is enough. Set the turn start at 30km, the turn end at 95-97km and a turn profile of 70%.

I have under 12k on my Eve ascent vehicles from sea level.

Click on my Eve ascents in my signature, and you can see the dV and TWR on each stage.

[cicatrix]

Well, finally. As it appeared, I missed the fact that 2 out of 14 stages had in fact TWR 0.98 and 0.92. Now I'm facing a hard choice - either to re-design the craft or tweak fuel loadout and hope it will be enough. 14 asparagus stages is not what you'd normally want to rebuild.

[EdFred]

Well, finally. As it appeared, I missed the fact that 2 out of 14 stages had in fact TWR 0.98 and 0.92. Now I'm facing a hard choice - either to re-design the craft or tweak fuel loadout and hope it will be enough. 14 asparagus stages is not what you'd normally want to rebuild.

You can get away with the final stage being that little. But prior to that you're in trouble.

[numerobis]

Well, finally. As it appeared, I missed the fact that 2 out of 14 stages had in fact TWR 0.98 and 0.92. Now I'm facing a hard choice - either to re-design the craft or tweak fuel loadout and hope it will be enough. 14 asparagus stages is not what you'd normally want to rebuild.

Add some 24-77s to the second of those stages. Turn them on at liftoff. Turn them off when you reach terminal velocity or close to it. Toggle them on/off for whenever your TWR is too low. They have great TWR, and the extra little squirts can help you maintain a better ascent speed.

Low TWR, even 0.5, works fine on a high stage when you're mostly going sideways at high speed and altitude. It works OK on a lower stage, but you lose a lot of efficiency.

[OhioBob]

Click on my Eve ascents in my signature, and you can see the dV and TWR on each stage.

Wow, those are huge. I employed a different philosophy and designed the smallest, most stripped-down lander/launcher that I could come up with. Of course my mission is to simply land one Kerbonaut, plant a flag, take a sample, collect some science, and take off again. The following is what I came up with. With only two pairs of asparagus strap-ons I was able to get a dV of about 11,100 m/s*. The total mass as shown is 35.6 t. According to my simulations, it shouldn't have any problem getting to Eve orbit from a ground elevation of 3,000 m with, if everything goes perfectly, about 1000 m/s dV in reserve.

(* This is actual simulated performance. Calculating dV based on vacuum ISP yields about 11,450 m/s.)

Edited December 1, 2014 by OhioBob

[Laie]

Wow, those are huge.

In part because the Eve Rocks Challenge demands that Kerbals sit in a pod during ascent, and encourages landing at sea level.

I'm late to the party. When I give advice, I tell people that their vessel must be capable of as sustained TWR=2 for three minutes, or TWR=1.5 for five minutes, or anything in between, just to get out of the worst atmosphere (which I assume to be ~25km). These figures are glimpsed from successful launches, and the beauty of this approach is that it is possible to test the vessel on Kerbin just by keeping it tied down on the launchpad and see how long it will work under atmospheric conditions.

However, the part about "anything in between" is a bit unsatisfactory. Could your software create a few plots, assuming constant TWR?

[OhioBob]

In part because the Eve Rocks Challenge demands that Kerbals sit in a pod during ascent, and encourages landing at sea level.

Sounds interesting, I'll have to try that.

However, the part about "anything in between" is a bit unsatisfactory. Could your software create a few plots, assuming constant TWR?

Is that comment directed at me? If so, can you explain further because I'm not sure I understand?