the vertical launch approach

[farmerben]

I was playing KSP again recently and I remark how easy it is to leave Kerbin's SOI going straight vertical the whole way.  Launching at dawn ( for which there is a convenient warp) has a near optimal trajectory for increasing the solar apoapsis, conversely dusk launches go closer to the sun.  To optimize the angle you just have to go a little west to offset the initial planetary spin.  Generally speaking, I think it works out to close to the same dV to escape Kerbin's SOI with or without orbiting first.  Maybe somebody has the calculations.  The vertical approach is easier than orbiting, not that the latter is difficult, but the former lets you get away with much sloppier designs like using lots of SRBs or having more aerodynamic drag in the nose.

Do you think the vertical launch approach is a good one to take in reality? 

I was thinking the shuttle and SLS style SRB's are a really good value, then use 8 of them and take the vertical approach.  

Or maybe the problem is wanting everything to land in the ocean...?

 

 

[darthgently]

It might make landing reusable boosters simpler.  Though they'd be coming mostly straight down from a higher apoapsis with a more compressed and intense aero experience I imagine, so not sure about if more or less fuel for booster return would be required.   

I'm already hearing in my mind the argument that going straight up means you are "fighting directly against gravity longer", but if your goal is solar orbit I'm not sure fuel spent going orbital around the launch body isn't just as wasteful. 

Good question!  Looking forward to the responses

[AckSed]

As I've heard, in a two-stage reusable rocket, going up on a flatter trajectory is a lofted trajectory, and it's indeed an advantage for reusable boosters if you can push the work of getting to orbit on to the second stage. Because if the fuel burn rate is constant, the booster uses less energy to punch through the thick lower atmosphere, and to boost back to cancel its sideways momentum so it can return to the launch site.

In the first instance, though, boosting till you leave the SOI is a tad trickier to do from Earth.

If this cheat-sheet is correct, the amount of delta-V to escape Kerbin and into an elliptical orbit (3400 + 930 = 4330 m/s) is slightly less than that of escaping the SOI of Mars (3800 + 1440 = 5240 m/s):

Spoiler

An equivalent manoeuvre to leave Earth and reach its escape velocity requires 9400 + 3210 =  12,610 m/s. So the bar is set nearly three times lower for Kerbin.

[sevenperforce]

5 hours ago, darthgently said:

I'm already hearing in my mind the argument that going straight up means you are "fighting directly against gravity longer", but if your goal is solar orbit I'm not sure fuel spent going orbital around the launch body isn't just as wasteful. 

It's not wasted, though. Your orbital speed contributes to your ejection burn -- that's why you can't just burn for solar orbit anywhere; you have to burn at the point where your orbital velocity vector lines up with the ejection burn direction.

Burning straight up on Kerbin is doable in part because of Kerbin's high density. Kerbin has the same gravity as Earth but is physically much smaller, meaning that F_g = GMm/r² (that's the equation for gravity) falls off rapidly as you climb. If you were to burn straight up off of Earth's surface, the gravitational attraction only would be reduced by a few percent by the time you got to space; on Kerbin it is substantially lower even while you're still in the atmosphere.

There's virtually no delta-v wasted by going into a stable orbit first before you start your ejection burn.

[kerbiloid]

First Luna crafts were sent (in)to the Moon without intermediate LEO.
https://en.wikipedia.org/wiki/Luna_1

Luna-16 return rocket has targeted the Earth vertically.

LK-700 had two return options, and the main one was direct vertical trajectory, without LLO.

The intermediate orbit just help to soften the possible ballistic errors, or simplify the transfer, or land on the invisible hemisphere.

Edited March 22 by kerbiloid

[farmerben]

Orbit first is the base case.  The vertical approach probably does not save delta-v.  The amount wasted is a function of "fighting gravity longer".  If you have TWR of 1.5 or so, then the vertical approach is way less efficient.  With TWR greater than 3 the amount wasted is very small.  

[kerbiloid]

The near-parabolic delta-V is almost absolutely same, regardless of direction.

The full energy reaches the infinity distance limit in any case.

Only the acceleration time is important. Not for total delta-V but for additional losses.

Edited March 22 by kerbiloid

[tomf]

Just to confirm my intuition I just did a test in ksp. I built a simple two stage rocket with about 5000 m/s delta v with fairly low twr.

First launch was directly upwards and it arrived at the edge of kerbin's soi with a speed of 856m/s.

The second launch did a conventional gravity turn before burning parallel to the surface. That one arrived at the soi with 2000m/s

So it seems clear that the gravity losses for burning directly up are pretty substantial.

[PakledHostage]

37 minutes ago, tomf said:

So it seems clear that the gravity losses for burning directly up are pretty substantial.

That's the Oberth effect. In both cases, you're ending up in a hyperbolic escape trajectory,  but one has it's pariapsis near Kerbin's center (i.e. below the planet's surface) while the other has its periapsis near the orbital altitude that you're ejecting from.

In the burn straight up case, you're doing your burn well past the periapsis point, where the vehicle's speed is much lower than at periapsis. The other case has you doing your escape burn from very near the resulting hyperbolic orbit's periapsis point, where you're moving much faster, and therefore achieve a more efficient burn.

[farmerben]

1 hour ago, tomf said:

Just to confirm my intuition I just did a test in ksp. I built a simple two stage rocket with about 5000 m/s delta v with fairly low twr.

First launch was directly upwards and it arrived at the edge of kerbin's soi with a speed of 856m/s.

The second launch did a conventional gravity turn before burning parallel to the surface. That one arrived at the soi with 2000m/s

So it seems clear that the gravity losses for burning directly up are pretty substantial.

I did my tests a little differently.  burning until the escape bubble shows up and then seeing how much dV I have left.  I just tried it with a simple two stage rocket with 7000 m/s, I had about 1700 left in both cases.  If anyone else wants to test this please do.

[PakledHostage]

Because Kerbin's atmosphere is so thick, and it ends at ~70 km altitude,  it stands to reason that maximum efficiency will be achieved by doing the escape burn from a 70 km apoapsis. Probably after following a launch profile similar to those in this challenge (minimum delta-V to LKO)

Edited March 23 by PakledHostage
Fixed link

[darthgently]

I think overall TWR and ascent profile for the non-straight ascent would make a big difference.  Not to say the straight to escape is better overall, but to spend less time fighting the gravity well the higher your TWR must be.

Another thought experiment is to consider that the most efficient apoapsis raising burn is done at periapsis.  A straight out to escape burn is going to put some portion of your burn higher than your effective periapsis. 

Maybe some rough rule of thumb can  apply.  Like if your TWR is high enough to reach escape velocity before your altitude is greater than the lowest possible orbital periapsis (70.xx km kerbin) then straight out to escape could have a case.  I'm guessing, so I better not see this posted as a "fact" someone "read somewhere" next year, lol.  Does that happen to anyone else?

You better have ablator on the nose cone.

Edited March 23 by darthgently

[PakledHostage]

High TWR has the effect of mitigating gravity drag, but intuitively I still think doing the escape burn from a 70 km periapsis is going to be more efficient than going straight up (even with high TWR) for the reason I gave before about Oberth effect.

[kerbiloid]

https://en.wikipedia.org/wiki/Vis-viva_equation

There is no angle in the formula.

[darthgently]

11 hours ago, PakledHostage said:

High TWR has the effect of mitigating gravity drag, but intuitively I still think doing the escape burn from a 70 km periapsis is going to be more efficient than going straight up (even with high TWR) for the reason I gave before about Oberth effect.

The more TWR early on the more you are burning at your initial "periapsis" re vis viva.  The moment you leave the ground, vis viva applies and you are very close to your "periapsis".  That is the aspect I'm looking at it from as a thought experiment.

So if you can reach escape velocity going straight up in a short enough time going to orbit first would be more wasteful.  Going to orbit first is purely limited by TWR for the same reason that very low TWR requires multi-pass Oberth burns to raise AP.   

Having trouble not seeing it this way.  Convince me.   The aero phase is problematic as drag goes up non-linearly with early TWR 

Edited March 23 by darthgently
*Raise AP, not PE

[jimmymcgoochie]

Vertical launch to escape velocity is a bad idea, for several reasons:

1. Gravity. Every second you’re ascending vertically, you lose 9.81m/s of delta-V just not-falling. Real rockets pitch over as soon as they can to minimise this gravity loss and gain horizontal velocity towards orbit.
2. Drag. A high-TWR rocket will encounter significant drag very quickly in the atmosphere, losing even more delta-V in the process. The structural and thermal stresses would also be much greater, requiring a sturdier rocket that can handle those stresses, which adds more weight.
3. Mass. Higher TWR requires more thrust, which adds more weight, which reduces delta-V, which requires more fuel, which adds more weight, etc. etc. It’s a vicious cycle and not easily broken.
4. Scale. Earth is ten times the radius of Kerbin, its atmosphere twice the height, escape velocity more than triple and density substantially lower meaning gravity decreases much more slowly. Launching straight up on Earth would be hideously wasteful.
5. Sol system is inclined, no two planets are in the same orbital plane. Most launch sites are nowhere near the equator either, plus Earth has a significant axial tilt. It’s incredibly unlikely that shooting straight up would actually result in a trajectory that went anywhere near another planet, whereas launching into a parking orbit can allow a departure burn from any launch site, and once per orbit (1.5-2 hours) rather than once per day.

[darthgently]

9 minutes ago, jimmymcgoochie said:

Vertical launch to escape velocity is a bad idea, for several reasons:

1. Gravity. Every second you’re ascending vertically, you lose 9.81m/s of delta-V just not-falling. Real rockets pitch over as soon as they can to minimise this gravity loss and gain horizontal velocity towards orbit.
2. Drag. A high-TWR rocket will encounter significant drag very quickly in the atmosphere, losing even more delta-V in the process. The structural and thermal stresses would also be much greater, requiring a sturdier rocket that can handle those stresses, which adds more weight.
3. Mass. Higher TWR requires more thrust, which adds more weight, which reduces delta-V, which requires more fuel, which adds more weight, etc. etc. It’s a vicious cycle and not easily broken.
4. Scale. Earth is ten times the radius of Kerbin, its atmosphere twice the height, escape velocity more than triple and density substantially lower meaning gravity decreases much more slowly. Launching straight up on Earth would be hideously wasteful.
5. Sol system is inclined, no two planets are in the same orbital plane. Most launch sites are nowhere near the equator either, plus Earth has a significant axial tilt. It’s incredibly unlikely that shooting straight up would actually result in a trajectory that went anywhere near another planet, whereas launching into a parking orbit can allow a departure burn from any launch site, and once per orbit (1.5-2 hours) rather than once per day.

1. Still seems to only be relevant if your goal is local orbit as the DV req'd for that horizontal velocity isn't free either and keeps you in the atmo longer.  Not sure the lower velocities make it worth it.  Would depend a lot on design.

But I'm persuaded strongly to your view by 2 through 5.  I'm convinced.  Thank you

[farmerben]

Staying in the atmosphere longer means more loss due to drag, no doubt about it.  The inclination difference is no big deal.  It's easy to adjust inclination after leaving the atmosphere.  Nobody claimed the vertical approach would save dV.  But if the difference is not significant, what advantages does it have?

One of the big advantages in game is you can have oversized payloads without a fairing.  The aerodynamics hurt you less.  

[Kerbart]

1 hour ago, darthgently said:

1. Still seems to only be relevant if your goal is local orbit as the DV req'd for that horizontal velocity isn't free either and keeps you in the atmo longer.  Not sure the lower velocities make it worth it.  Would depend a lot on design.

Keep in mind that if your TWR is 1.5, ⅔ of your thrust—and thus ⅔ of the fuel— is spent battling gravity. That fraction goes down as the TWR increases, with a TWR of 3.0 it's “only” ⅓ but we all know what the issues are with launching at such a high TWR.

The flatter the trajectory, those fractions go down and if you're able to go fully horizontal, all of your thrust goes into velocity. Yes, there's atmospheric drag and you do have to get out of the thicker atmosphere, but at a 45° angle you'll gain velocity faster to do that.

[PakledHostage]

13 hours ago, kerbiloid said:

There is no angle in the [vis-viva] formula.

The vis-viva formula only applies when the only force acting on the object is gravity. It's irrelevant when the vehicle is under acceleration due to rocket thrust or atmosphericdrag, such as is being discussed here.

[darthgently]

3 minutes ago, PakledHostage said:

The vis-viva formula only applies when the only force acting on the object is gravity. It's irrelevant when the vehicle is under acceleration due to rocket thrust or atmosphericdrag, such as is being discussed here.

I disagree.  The vis viva solutions merely change continuously given the thrust vector.  They are still there

[PakledHostage]

16 minutes ago, darthgently said:

I disagree.  The vis viva solutions merely change continuously given the thrust vector.  They are still there

Sure, but how are you going to use them in some useful way? That's like saying your orbit changes continuously given the thrust vector.  We know that to be true. But how is it relevant to answering what's more efficient: launching straight up,  or using a gravity turn to reach an altitude from which we can either burn directly into an escape trajectory,  or where we can park temporarily before doing so?

[kerbiloid]

Launching straight up allows you increase distance from the gravity center faster, but spiralling to LEO assists you with centrifugal force.

Total amount of energy to reach the infinity is equal for the cases of throwing a stone up, or horizontally. In both cases you have to add same energy.

The air drag losses also should not differ very much, as most part of the atmosphere is concentrated at first several kilometers of altitude, which you are anyway passing almost vertically.

[darthgently]

33 minutes ago, PakledHostage said:

Sure, but how are you going to use them in some useful way? That's like saying your orbit changes continuously given the thrust vector.  We know that to be true. But how is it relevant to answering what's more efficient: launching straight up,  or using a gravity turn to reach an altitude from which we can either burn directly into an escape trajectory,  or where we can park temporarily before doing so?

Given your constraints the Oberth effect is pointless because your orbit is "changing"

[PakledHostage]

Sorry if I have ruffled feathers. Maybe I will just "get my own coat", so to speak. Bye.