Vertical Ascent vs. To LXO First

[diomedea]

Gents, while I can read very interesting and valuable opinions and ideas from all who posted lately on this thread, the discussion seems also to be descending to a rather hostile level at times.

Please, keep it in the most civil way possible. Really, there is no need to show a lack of respect for the opinions of others.

[arkie87]

Yessir, I've got that... but all these other factors are critical to the comparison between the two modes. I understand you think you've yourself a favor by eliminating the characteristics of the vehicle itself from the analysis, but in doing so you have eliminated all *practical* vehicles from consideration.

Your analysis, mathematically rigorous as it is, boils down to this:

"If stranded on a desert island, there may be some case where there's a more efficient way to open a coconut than banging it on a rock. You could, for instance, bring an elephant along with you. The math shows that the larger the elephant, the more efficiently it can open a coconut."

I think the reason this gives so many people so much consternation (myself included) is that you speak of metaphorically "bringing along an elephant" and efficiency as if they are not mutually- exclusive concepts.

Stepping back out of the analogy...

Your analysis completely stacks the deck in favor of vertical mode by assuming a vehicle optimized for it while rejecting a vehicle optimized for horizontal launch and erroneously concludes that "there's not much difference between the two modes".

You cannot compare the relative merits of the two modes fairly without comparing *all* of the merits. Vertical launch is easier, while horizontal mode saves DV, weight, cost, and fuel. The difference between the two modes *in total* is not minor in any of these categories.

Vertical launch from the munar surface is not the preferred option *under any circumstances* from a mission planning, engineering, or fiscal standpoint.

As for LD's analysis, the holes in his thinking are threefold:

1)

^This

2) The t/w of the vehicle increases over time as fuel is consumed.

and

3) The pilot is going to vary the pitch of the vehicle over the duration of the launch, not keep it at a fixed angle.

Best,

-Slashy

Your point about it being impractical is conventional wisdom; however, this thread will show that it is possible to have a high TWR craft that is also quite cheap/competitive (that is sort of the point of the linked thread): http://forum.kerbalspaceprogram.com/threads/103449-A-New-Thread-to-Discuss-5thHorseman-s-Challenge?p=1606814&viewfull=1#post1606814 and http://forum.kerbalspaceprogram.com/threads/103449-A-New-Thread-to-Discuss-5thHorseman-s-Challenge?p=1607366&viewfull=1#post1607366

Second, I am not trying to stack the deck in favor of anything. I merely want to understand under what conditions conventional wisdom is correct (i.e. horizontal burn) and under what conditions vertical is better...

Furthermore, even if for an ideal launch conventional wisdom is better, it is important to know how badly one needs to mess it up to offset its gains. I was surprised to learn that the values were so low for the Mun case-- for TWR 3, if you get your craft up to 7.5 km, it is more efficient to have just launched vertically... i dont know about anyone else, but when i launch from the moon, i first climb up to a parking orbit at least 10km high... i dont skim across the surface like a daredevil... it is possible that climbing up to this height actually makes the gains less. Finally, if you are in a crater, it might easily be better to launch vertical...

And my last point--for heavy landers, the difference in mass between a rockomax 48-7s and the lv-909 is large, but the different mass is negligible on the scale of the large craft. So the ISP gains from switch to LV-909 actually favor using the LV-909, since the LV-909 will give more deltaV... so it's entirely possible that a higher TWR craft will be both cheaper and have more deltaV. For the example of a craft where this is the case, see:

And FYI: for the last stage before the lander/payload, i had the same design using that smaller engine (rockomax 48-7s), but i noticed i actually get more deltaV from the LV-909 since its ISP is larger; the increase in mass of the LV-909 to the total package isnt significant enough to offset its increased ISP.

Edited December 17, 2014 by arkie87

[arkie87]

See spreadsheet below for a case study of where a higher TWR also happens to be more fuel efficient i.e. larger deltaV

[GoSlash27]

Your point about it being impractical is conventional wisdom; however, this thread will show that it is possible to have a high TWR craft that is also quite cheap/competitive (that is sort of the point of the linked thread): http://forum.kerbalspaceprogram.com/threads/103449-A-New-Thread-to-Discuss-5thHorseman-s-Challenge?p=1606814&viewfull=1#post1606814 and http://forum.kerbalspaceprogram.com/threads/103449-A-New-Thread-to-Discuss-5thHorseman-s-Challenge?p=1607366&viewfull=1#post1607366

Sorry, but that thread has no bearing on this discussion. If you believe that you can make a vertical lifter that is "competitive" in any way, shape, or form to what I have already presented here ($1688, 45 units of fuel burned, total mass under 1.9 tonnes), then by all means, have a go.

Second, I am not trying to stack the deck in favor of anything. I merely want to understand under what conditions conventional wisdom is correct (i.e. horizontal burn) and under what conditions vertical is better...

I think you've over-analyzed it pretty badly if that's the case. Any DV chart will tell you that 850M/sec is plenty for this job. You also know that escape velocity for the Mun is 807 M/sec.

Knowing that your goal is to achieve escape velocity and knowing that the gravity is robbing you of 1.63 m/sec during the duration of the launch (1G), you're asking at what point a vertical launch can achieve 850M/sec total expenditure or less.

This is actually a very simple problem to solve.

(x+1)/x*807=850

Simplifying this to solve for X is x= 807/43, which is 18.8:1 t/w. At that t/w, it will cost you 850 m/sec to achieve escape velocity. As t/w approaches infinity, DV approaches 807.

You're welcome

Furthermore, even if for an ideal launch conventional wisdom is better, it is important to know how badly one needs to mess it up to offset its gains. I was surprised to learn that the values were so low for the Mun case-- for TWR 3, if you get your craft up to 7.5 km, it is more efficient to have just launched vertically... i dont know about anyone else, but when i launch from the moon, i first climb up to a parking orbit at least 10km high... i dont skim across the surface like a daredevil... it is possible that climbing up to this height actually makes the gains less. Finally, if you are in a crater, it might easily be better to launch vertical...

I swear I don't mean to be a bastage when I say this, but yeah, you're doing it wrong. Climbing straight vertical and then circularizing is incredibly inefficient.

And my last point--for heavy landers, the difference in mass between a rockomax 48-7s and the lv-909 is large, but the different mass is negligible on the scale of the large craft. So the ISP gains from switch to LV-909 actually favor using the LV-909, since the LV-909 will give more deltaV... so it's entirely possible that a higher TWR craft will be both cheaper and have more deltaV. For the example of a craft where this is the case, see:

This is actually the same point as the first, and already answered.

Whatever payload you choose, you can't make a booster that will launch it at 18.8G that can compete with the mass, cost, and fuel burned by the conventional method. Below this acceleration, you're still heavier, more expensive, and wasteful, but in addition you're wasting DV.

Best,

-Slashy

[arkie87]

Sorry, but that thread has no bearing on this discussion. If you believe that you can make a vertical lifter that is "competitive" in any way, shape, or form to what I have already presented here ($1688, 45 units of fuel burned, total mass under 1.9 tonnes), then by all means, have a go.

I think you've over-analyzed it pretty badly if that's the case. Any DV chart will tell you that 850M/sec is plenty for this job. You also know that escape velocity for the Mun is 807 M/sec.

Knowing that your goal is to achieve escape velocity and knowing that the gravity is robbing you of 1.63 m/sec during the duration of the launch (1G), you're asking at what point a vertical launch can achieve 850M/sec total expenditure or less.

This is actually a very simple problem to solve.

(x+1)/x*807=850

Simplifying this to solve for X is x= 807/43, which is 18.8:1 t/w. At that t/w, it will cost you 850 m/sec to achieve escape velocity. As t/w approaches infinity, DV approaches 807.

You're welcome

I swear I don't mean to be a bastage when I say this, but yeah, you're doing it wrong. Climbing straight vertical and then circularizing is incredibly inefficient.

This is actually the same point as the first, and already answered.

Whatever payload you choose, you can't make a booster that will launch it at 18.8G that can compete with the mass, cost, and fuel burned by the conventional method. Below this acceleration, you're still heavier, more expensive, and wasteful, but in addition you're wasting DV.

Best,

-Slashy

I dont mean to be rude, but i dont think our discussion is going anywhere. None of your comments actually address the points i've made or "refute" an imaginary point i never made... from my perspective, it seems like you arent actually reading what i am writing, but rather, just typing your response to imaginary arguments.

I'm not sure you should continue responding to my threads, as we dont seem to communicate well.

[GoSlash27]

arkie,

I absolutely agree that we aren't communicating well, but I'm going to continue to participate in your threads until I finally give up in exasperation.

*somebody's* got to counter this notion that vertical launch ain't so bad.

Now... you've had a lot of people come in here and attempt to explain to you why this whole approach is borked, argue around in circles, and eventually give up in disgust. Are you really certain the problem isn't on *your* end?

I mean... just as you get the impression that I'm not reading your posts, I get the impression that you simply don't want to hear any counterarguments. It's not like I have these miscommunications often, but you have had them ongoing in these threads with a lot of people besides me.

Isn't at least theoretically plausible that maybe you're not conveying your meaning very well?

Sorry

-Slashy

[Fyrem]

For god sake people it has been 19 pages on this thread alone..

If only we had at our disposal a computer program that could run though the examples given

if only someone could ACTUALLY provide a SHIP FILE, and you know.. USE it .. to collect data....

key words in that sentence : Provide a SHIP FILE ....

really, all of you .. need to just stop, build a ship, LAUNCH it, record the data, and provide the ship file. REALLY.

Just stop. do that. come back. end.

Arkie, you believe there could be situations/ships/payload sizes, etc, where method X is "nearly as good as" ... ok, THEN MAKE A SHIP FILE and SHARE IT, test it, and show us the data.

Arkie, 5thHorsman has his $10k ship, (but no file yet, poke poke) .. do you have yours?

Slash, SHIP FILE ... DATA. done.

TDLR, it is time to put the pencils down, and DO it.

People would LOVE to come here and see this question actually WORKED out, ... so.. DO the work.

ok, I will shut up now.

[arkie87]

For god sake people it has been 19 pages on this thread alone..

If only we had at our disposal a computer program that could run though the examples given

if only someone could ACTUALLY provide a SHIP FILE, and you know.. USE it .. to collect data....

key words in that sentence : Provide a SHIP FILE ....

really, all of you .. need to just stop, build a ship, LAUNCH it, record the data, and provide the ship file. REALLY.

Just stop. do that. come back. end.

Arkie, you believe there could be situations/ships/payload sizes, etc, where method X is "nearly as good as" ... ok, THEN MAKE A SHIP FILE and SHARE IT, test it, and show us the data.

Arkie, 5thHorsman has his $10k ship, (but no file yet, poke poke) .. do you have yours?

Slash, SHIP FILE ... DATA. done.

TDLR, it is time to put the pencils down, and DO it.

People would LOVE to come here and see this question actually WORKED out, ... so.. DO the work.

ok, I will shut up now.

Yes, i built an 11k$ ship with nearly the same performance.

http://forum.kerbalspaceprogram.com/threads/103449-A-New-Thread-to-Discuss-5thHorseman-s-Challenge

[Fyrem]

Thank you Arkie and 5thhorseman. Ideas + Ships * Data = Learning.

[GoSlash27]

Fyrem,

Sure thing. I'll get it posted here this evening, along with pics from a mission and all figures.

Best,

-Slashy

[arkie87]

arkie,

I absolutely agree that we aren't communicating well, but I'm going to continue to participate in your threads until I finally give up in exasperation.

*somebody's* got to counter this notion that vertical launch ain't so bad.

Now... you've had a lot of people come in here and attempt to explain to you why this whole approach is borked, argue around in circles, and eventually give up in disgust. Are you really certain the problem isn't on *your* end?

I mean... just as you get the impression that I'm not reading your posts, I get the impression that you simply don't want to hear any counterarguments. It's not like I have these miscommunications often, but you have had them ongoing in these threads with a lot of people besides me.

Isn't at least theoretically plausible that maybe you're not conveying your meaning very well?

Sorry

-Slashy

I appreciate your desire to help and willingness to do so until you give up in exasperation. I recommend you have a look at http://forum.kerbalspaceprogram.com/threads/103449-A-New-Thread-to-Discuss-5thHorseman-s-Challenge as well where i have provided a 11 k$ ship which can do the same task as 5thHorseman's 9k$ ship (which is only 20% increase in price...), so while you might believe vertical ascent is much worse, that thread provides information that says otherwise.

I do appreciate you actually quoting my position more correctly (i.e. saying that vertical launch might not be so bad vs. saying that it is always better). If you read the OP, you will see that I do not actually have an opinion. I am looking to find more information, not expressing an opinion.

I will readily admit that the problem can potentially be on my end. But i read, and re-read your posts before i respond. When i respond, i cite what i'm responding to so there is no confusion and try to be clear as possible.

When i get my responses (from you, and plenty other), i see things that immediately indicate to me that the responder did not accurately read my statements (like the whole confusion over the mun lander vs the horseman challenge which i repeatedly tried to clarify...) or that discredit my arguments without giving any reasoning. In general, I've noticed you dont provide counter arguments, just counter-opinions. Even if these opinions are correct, you seldom provide proof to back them up. If the logic i present is wrong, explain what is wrong about it, not merely *that* it is wrong... i dont really understand why this is so hard to understand or accept....

[arkie87]

Thank you Arkie and 5thhorseman. Ideas + Ships * Data = Learning.

Are you sure its linear with Data? Shouldnt it be Data^2?

[arkie87]

Sorry, but that thread has no bearing on this discussion. If you believe that you can make a vertical lifter that is "competitive" in any way, shape, or form to what I have already presented here ($1688, 45 units of fuel burned, total mass under 1.9 tonnes), then by all means, have a go.

Best,

-Slashy

Can you please repost your lifter design? I want to design my own (on paper). Just to clarify, we are referring to your craft taking off from the Mun, leaving Mun's SoI such that it gets an aerobreaking encounter with Kerbin, correct?

Edit: nevermind. I found it.

Edited December 17, 2014 by arkie87

[LethalDose]

In general, I've noticed you dont provide counter arguments, just counter-opinions. Even if these opinions are correct, you seldom provide proof to back them up. If the logic i present is wrong, explain what is wrong about it, not merely *that* it is wrong... i dont really understand why this is so hard to understand or accept....

Nail.

Head.

You can only hear 0:35 here so many times before it's clear it's a waste of everyone's time and stops being funny.

Edited December 17, 2014 by LethalDose

[arkie87]

Nail.

Head.

You can only hear 0:35 here so many times before it's clear it's a waste of everyone's time and stops being funny.

I dont think that link worked the way you intended... but i'm glad you understand my sentiment.

Edited December 17, 2014 by arkie87

[GoSlash27]

Can you please repost your lifter design? I want to design my own (on paper). Just to clarify, we are referring to your craft taking off from the Mun, leaving Mun's SoI such that it gets an aerobreaking encounter with Kerbin, correct?

Yes, you're correct. My payload will be a Mk.1 lander can, and I will launch it from the location you described on the munar surface.

The equipment to place it on the surface and the solar panels will not count towards the cost of the vehicle (unless you just want to include that in the mission profile).

Right now, the design is simply a mk.1 pod, t-100 fuel tank, and LV-1 engine, but I may revise the final design to de-pork it a bit.

The only requirement for the lifter itself is that it must have a manned Mk.1 command pod and the only stipulation for the test is that debug menu options and kraken drives are illegal.

I think that covers it...

Best,

-Slashy

[arkie87]

Fyrem,

Sure thing. I'll get it posted here this evening, along with pics from a mission and all figures.

Best,

-Slashy

Against my better judgement, I am going to post these results. The top part is your results, the bottom is mine:

My design is cheaper, has more deltaV, and higher TWR... (high TWR enough, in fact, that you would kill your kerbals with the acceleration), and I can even forgo some of the fuel mass since i dont need all that deltaV, which would make my design cheaper still...

Now granted, it is a solid booster, so you cannot turn it off mid burn, so you would have to land it on the Mun with something else that you would jettison; however, that wasnt part of the problem description.

For a heavier lander, this type of vehicle might be more practical...

[arkie87]

As another example, if you have a vehicle landed on the Mun made out of Mk1-2 command pod and a rockomax X200-16 fuel tank, what engine should you strap on it:

The two most obvious are Rockomax 48-7S and the LV-909. In this case, the lighter engine does not result in more deltaV due to it's reduced ISP. Furthermore, in this instance, choosing the LV-909 results in both more deltaV and more TWR, which is why it is a good example...

[LethalDose]

Actually, if you replaced the LV1 in *someones* design with a 48-7s, it gives the craft:

• 1379.4 m/s dV
  
• 1.5125 t mass
  
• 1150 funds cost
  
• Munar sea level TWR of 12.2
  

The engine swap results in a gain of 150 m/s dV, a price reduction of 50 funds, and straight up baller TWR at the expense of 70kg...

I can't see how some could reasonably argue that additional 70 kgs makes this an "impractical" Munar lander. IMO, choosing the LV1 instead of the 48-7s given that comparison is simply idiotic.

Further it very easily pushes the vessel into the range were vertical ascents are reasonable give the data I provided.

[Superfluous J]

Yes, i built an 11k$ ship with nearly the same performance.

http://forum.kerbalspaceprogram.com/threads/103449-A-New-Thread-to-Discuss-5thHorseman-s-Challenge

Thank you Arkie and 5thhorseman. Ideas + Ships * Data = Learning.

On that thread that arkie linked to, I'm posting the craft file. There's already a video and pictures of my craft on there though and it's 18 parts so it's not like you couldn't recreate it.

[arkie87]

Actually, if you replaced the LV1 in *someones* design with a 48-7s, it gives the craft:

• 1379.4 m/s dV
  
• 1.5125 t mass
  
• 1150 funds cost
  
• Munar sea level TWR of 12.2
  

The engine swap results in a gain of 150 m/s dV, a price reduction of 50 funds, and straight up baller TWR at the expense of 70kg...

I can't see how some could reasonably argue that additional 70 kgs makes this an "impractical" Munar lander. IMO, choosing the LV1 instead of the 48-7s given that comparison is simply idiotic.

Further it very easily pushes the vessel into the range were vertical ascents are reasonable give the data I provided.

Very good point. I dont know why i didnt see that, but thanks for pointing it out. besides, as far as i know (i could be wrong) you cannot place just one LV-1 since it is a surface-attach engine...

TWR of 12 is not enough! Let's go to 100

Either way, i think LethalDose and I have provided enough examples that show that it is possible to obtain a higher TWR, deltaV and reduced cost all at the same time. The three are not always mutually exclusive, (since they are arbitrary functions of the part specs in KSP...)

On the plus side, this thread seems to be back on track... (not to jinx it or anything)

Edited December 17, 2014 by arkie87

[Starwaster]

Pffff do it from Kerbin.

In Real Solar System

[Superfluous J]

Either way, i think LethalDose and I have provided enough examples that show that it is possible to obtain a higher TWR, deltaV and reduced cost all at the same time. The three are not always mutually exclusive, (since they are arbitrary functions of the part specs in KSP...)

I hate to throw a wrench in your calculations, but unless you're starting on Krebin (or use Extraplanetary Launchpads) then weight of the craft is also important. SRBs are so good on the launch pad because you don't have to pay for the fuel to lift them off of some other planet and transfer them to Kerbin.

Kind of like in that other thread we just talked in, how the LV-909 was better on the top stage than the 48-7s and got more thrust and better dV. However the lower weight of the 48-7s allowed all the other stages to carry the rocket much further, to the point that I'd not have been able to get it to its destination without adding more fuel and/or boosters to it.

[arkie87]

I hate to throw a wrench in your calculations, but unless you're starting on Krebin (or use Extraplanetary Launchpads) then weight of the craft is also important. SRBs are so good on the launch pad because you don't have to pay for the fuel to lift them off of some other planet and transfer them to Kerbin.

Kind of like in that other thread we just talked in, how the LV-909 was better on the top stage than the 48-7s and got more thrust and better dV. However the lower weight of the 48-7s allowed all the other stages to carry the rocket much further, to the point that I'd not have been able to get it to its destination without adding more fuel and/or boosters to it.

I readily admitted that my RT10 example is impractical, but that wasn't the challenge posed by Slashy. Nevertheless, that is just one example of many I've given. Furthermore, LethalDoses is both practical and advantageous in every way.

Second, are you saying that the 487s is better than lv909 as the final stage lifter? Isp difference is 4/35 while mass is only 4/60.? Please explain....

[Superfluous J]

Second, are you saying that the 487s is better than lv909 as the final stage lifter? Isp difference is 4/35 while mass is only 4/60.? Please explain....

As I said in that other thread we're not to name:

Adding an LV909 cut my total dV from 6972 to 6110, though it did raise that one stage from 1186 to 1235. So while it's better for that stage, it's worse for the rocket as a whole.

I don't care about Isp or mass or TWR. I care about how much further my rocket can go. I guess I'm more of a Bill than a Bob in that way