Jump to content

Most less DV to get into LKO


Recommended Posts

Whats the actual most less DV to get into LKO 70KM ?

I onced was trying to get into LKO with a One-Burn Ascent, was trying to make it so much like in Real Life like possible, that means fall naturaly into Gravity Turn and with extremly low Thrust after about 10KM.

So i was trying to stay max 30-40 Second behind AP all the time. From about 20 attempts i onced got to 3080 Delta-V into LKO, but i dont have clue how I did it, I couldnt repeat it afterwards ( Most attempts were about 3200 to 3300 DV )
With the same Vessel Mechjeb was using 3300 to 3400 DV.

Did someone managed to get into LKO with unter 3K DV ? I think it should be possible with very dragless vessel and very good Ascent Skills.
 

Link to comment
Share on other sites

I've heard rumors that it's been done for under 3000 m/s, though I've never tried it myself.  Trying to see how little Δv you can get to orbit for may be a fun challenge to chase after, but as a practical payload delivery system, it's a bad design.  The lowest that I've ever done it for was somewhere around 3350-3400 m/s (with the current version).  I just don't design for minimum Δv.  If minimizing Δv if what you really want, what I've heard is that you want a high TWR, need to pitchover quickly, and peg the throttle to punch your way through the atmosphere with minimum gravity loses.

 

Edited by OhioBob
Link to comment
Share on other sites

How do you guys calculate the Δv of your atmospheric flights?

The classical Tsiolkovsky rocket equation ln(mfull/ mdry) = Δv / Isp applies for constant Isp only.

Do you solve the general-form impulse conservation equation with Isp as a function of pressure/height?
Or I’m nuts, and there is a simple way to measure the Δv?

Link to comment
Share on other sites

56 minutes ago, Teilnehmer said:

How do you guys calculate the Δv of your atmospheric flights?

The classical Tsiolkovsky rocket equation ln(mfull/ mdry) = Δv / Isp applies for constant Isp only.

Do you solve the general-form impulse conservation equation with Isp as a function of pressure/height?
Or I’m nuts, and there is a simple way to measure the Δv?

It's actually pretty simple-- the math doesn't get too hard.  Two things to bear in mind:

  • Most engines don't actually change Isp all that much, in terms of percentage, as long as you don't do something dumb like use a vacuum engine (Terrier, Rhino) at sea level.  For example, the Mainsail's Isp is only about 8% worse at sea level than in vacuum.
  • The atmosphere thins out really fast with altitude.

That second one is the kicker.  By the time you're at 10 km, the atmosphere is already only a small fraction of what it is at sea level.  And typically, your ship has used only a very small fraction of its total dV by that time.

So if you just do like this:

  • For your first stage (e.g. typically your SRBs), just use their sea-level Isp to figure out that stage's dV
  • Assume vacuum Isp for everything else

...you'll get a pretty darn accurate measurement of your dV.  It might be a percent or two off, but not much more than that.  Might not be up to NASA standards, but plenty accurate enough for KSP.  :)

 

Edited by Snark
Link to comment
Share on other sites

3 hours ago, Snark said:
  • For your first stage (e.g. typically your SRBs), just use their sea-level Isp to figure out that stage's dV

I've found that I get a better answer by averaging sea level and vaccum Isp for the first stage.

Link to comment
Share on other sites

1 hour ago, Teilnehmer said:

Thanks. I actually do the same way. I just thought that a more accurate method could be needed for the ‘least Δv to LKO’ research.

Most of the time when people around here discuss the Δv of their launchers, the numbers are based on vacuum Isp.  Although this practice overstates the actual Δv, it's just easier to do it that way.  I doubt if the difference is any more than 100 m/s.

Link to comment
Share on other sites

33 minutes ago, OhioBob said:

Most of the time when people around here discuss the Δv of their launchers, the numbers are based on vacuum Isp.  Although this practice overstates the actual Δv, it's just easier to do it that way.  I doubt if the difference is any more than 100 m/s.

This is true. ISP rises very fast on Kerbin. At 5000m, you're already at near vacum value. I always calculate my dV as vacuum value. I go to space with 3200 to 3300m/s.

My standard Cygnus recoverable SSTO rockets are fueled with 3400m/s, but they're designed to have 100 to 200m/s before detaching payload so they can deorbit, do some minor slowdown burns (since 1.0.5) and do a powered landing (helped with chutes).

I think I did a 3150m/s (vac) in 1.0.4 once. My usual flight path was to turn at 60m/s and cross 45° at 8000m. But in 1.0.5, heating is now an issue, so I cross 45° at 10km, even 12km now. That costs a bit more dV, but it's safer.

Link to comment
Share on other sites

We discussed about that too on the dedicated topic in the tutorial forum. We acknowledge that VAC value is better than ASL value.

Beware that dV is dependent on engine ISP, but also with ship aerodynamics and flight plan. Ascent path is very very important. I discovered that when practicing landing/lifting from Eve.

In 1.0.4, I designed a Eve ship with 9400m/s. I hyperedited it at 100km, landed (on 1000m Midlands) and returned to 140km. I had 2900m/s left. So I used 6500m/s (VAC). As the ship was really heavy, I tried to sized it down. The new ship had 7000m/s (VAC). and I failed to return to orbit multiple times...

Link to comment
Share on other sites

Well, only the first two replys do have something to do with my question hehe.

You can use for example Mechjebs Readouts for exact Delta-V spended.

Just make a Costum windows ( or use Vessel Stats ) and add the Delta-V expended Information. You dont have to use the Mechjebs Autopilot functions, Im talking only about the readouts.

Then you can go to Orbit and have a exact measurement, how much DV you spended to go to LKO.
Would love to see a pic and even more important the description about the ascent way from someone who gets to LKO with under 3000DV.

 

Edited by Rucki
Link to comment
Share on other sites

The optimal transfer from stationary on the surface to orbit at 70km is 2480 m/s. Anything extra is gravity, drag, and steering losses. (Or just miscounting deltaV by approximating on vacuum Isp).

Edited by numerobis
Link to comment
Share on other sites

3 hours ago, numerobis said:

The optimal transfer from stationary on the surface to orbit at 70km is 2480 m/s. Anything extra is gravity, drag, and steering losses. (Or just miscounting deltaV by approximating on vacuum Isp).

That sounds good, but something about minimum 800dv drag loss looks uninevitable. I think 70dv drag loss is possible.

Link to comment
Share on other sites

2 hours ago, Rucki said:

That sounds good, but something about minimum 800dv drag loss looks uninevitable. I think 70dv drag loss is possible.

I few months ago I performed a bunch of launch vehicle optimization simulations.  If you're interested, you can read the results here:

Launch Vehicle Optimization Test Results

One of the interesting parts of those simulations was that I was able to compute the losses due to gravity and drag.  For the three different launcher configurations that I tested, here is what I got:

Liftoff TWR = 1.90 ---> gravity loss = 669 m/s, drag loss = 210 m/s, total = 879 m/s

Liftoff TWR = 1.46 ---> gravity loss = 765 m/s, drag loss = 152 m/s, total = 917 m/s

Liftoff TWR = 1.21 ---> gravity loss = 968 m/s, drag loss = 118 m/s, total = 1086 m/s

Of course these numbers will vary depending on the aerodynamic characteristics of the vehicle and the ascent profile.

Link to comment
Share on other sites

18 hours ago, OhioBob said:

I few months ago I performed a bunch of launch vehicle optimization simulations.  If you're interested, you can read the results here:

 

 

I read your optimization simulations and its very interesting, thank you for sharing :). Though, one Question, did you tried that also in practice or only on formulas ?
It would be interesting to know that formulas, if they are 100% correct, than i could know, why i fail to get to these values.

I am testing since 2 or 3 days whenever i have the time, to get to the most effective Design&Ascent-Way which is only possible to minimize dV to LKO.
In my last attempt i had a TW for about 1.90 and got to LKO with 3072 dV ( Im sort of proud after so many testing flights :D )
BUT: you showed a gravity loss of 669, i got to 835, you showed a Drag loss of 210, i got to 119 = total 945m/s ( well plus 25 m/s steering loss, so total-totaly = 970m/s loss )
And that is after dozens of dozens of tests.

A few things got in my head after a few test:

1.) I thought all the time that in new KSP Atmosphere we should stick to a max-Q ( and I just took the advice from MJ for about 20.000pa)
That is not true when the most important thing for you is most less dV to Orbit. You should stick with max-Q if you dont have a very well drag-optimized design to minimize stress on the rocket, thats also the way like they do it in Real Life, because an exploding rocket is fun in KSP, but it isn´t in Real Life.

2.) Further on beeing under Terminal Velocity is very important, you should not go over or you are waisting loss to drag. though in 1.05 its very hard for me to go over the TV limit.

3.) It seems that if you have a very well aero designed rocket, that the most important factor is not drag loss anymore, the most important factor is gravity loss, if you stick with max q limit to 20.000pa you will always have a too high gravity loss and if your rocket is stable you are not gaining anything. ( in a gamesim like KSP )

4.) If you are trying to get the best optimized ascent, you can´t use Mechjeb Autopilot Ascent, you have to do it manually !
Though i was using MJ functions like maxQ, Terminal Velocity limit ect. for testing purposes. Also to get a perfect circularizing maneuver, the MJ Smatblabla function is useful to stay in prograde at Apoapsis.
I found out that you can safe dV if you make the circularization manually and without a maneuver node, just stay 10 seconds behind time-to-ap and then burn exactly so that you dont go over or under the 10 seconds to ap, you will make perfect circularization maneuver.

5.) I cant make a totally handsoff Gravity Turn, the gravity turn is always starting to early or if starting at a good point, will end to early and I have to correct the pitch by myselve, thats a thing which grind my bears, because I wanted to make a gravity turn with so less drag and steering loss as only possible. And also I read that in Real life they are also trying to make a One-Steer only Gravity Turn ( the beginning pitch over ) though i also read that sometimes even in real life they have to use gimballed help to correct steer but at a minimal level )

I tried to give my rockets more weight to the top through tanks which i dont really need/use, so that I thought more weight on the top will result in faster gravity turn, but it looks like that is not true, further testing from my site is needed though.

6.) Im not sure about that thing, but it seems that a rocket with more width has also less drag in KSP and i think it should be the other way. Im testing this further on. I can only say that with similiar Ascent behavior, my first rocket with 1.25m did always had much higher drag loss than the rockomax parts rocket.

7.) Round about you should stick with that Degrees to Ascent:
Pitch Over at : 100 m/s ( which is with a good TWR Vessel always something between 600m high and 1400m high, you should look only at m/s, because under 100m/s will return in too fast gravity turn and too much over 100m/s will return in a too straight gravity turn with much steering corrections and loss )

You are at 10Km = ca. 45° degrees
Your AP is at ca. 40KM = turn down to 20° degrees
Your AP is ca. 55-60KM= turn down to something about 0° degrees, you need velocity now

It would be nice to go to 0° at 30Km but the heat will melt the cone and wings and may even melt more of the rocket.


3072 dV to LKO Rocket:

Start

Ascent at 28KM

Ascent at 47KM after Burn

Final Orbit

 

If someone can show me that Im not right with some or even all facts, i would like to hear that !

 

Update
1 hour later i got to under 3000dV :=)

Pic

Edited by Rucki
Link to comment
Share on other sites

3 hours ago, Rucki said:

I read your optimization simulations and its very interesting, thank you for sharing :). Though, one Question, did you tried that also in practice or only on formulas ?
It would be interesting to know that formulas, if they are 100% correct, than i could know, why i fail to get to these values.

I did play around a bit with the designs in the game.  They performed similarly to the simulations, but I couldn't quite obtain the same payload capacity.  I'm sure this is because the simulations used a highly efficient and precise ascent profile, which I couldn't replicate in the game.  My angle of attack frequently exceeded the values used in the simulations, which resulted in greater drag losses.  (From my in-game experiments with the test vehicle, I estimated that the drag increased about 8% for every 1o increase in AoA.)

Another problem that I experienced in game was that the 1.21 TWR vehicle was extremely long and slender, making it wobbly and difficult the steer.  For this reason I found that a liftoff TWR of 1.3 to 1.4 worked much better, which is what I now target when designing new launch vehicles.

Quote

In my last attempt i had a TW for about 1.90 and got to LKO with 3072 dV ( Im sort of proud after so many testing flights :D )
BUT: you showed a gravity loss of 669, i got to 835, you showed a Drag loss of 210, i got to 119 = total 945m/s ( well plus 25 m/s steering loss, so total-totaly = 970m/s loss )
And that is after dozens of dozens of tests.

Maybe we're computing it differently.  For drag loss, I calculated the acceleration resulting from the drag force and integrated it over time to obtain the total velocity change.  I then took the total Δv needed to attain orbit, subtracted the final orbital velocity and the drag loss, and called this the gravity loss.  Note that the total Δv was an integrated value computed using the actual instantaneous Isp.  This value was 3167 m/s, compared to 3202 m/s using vacuum Isp (which is how it is commonly computed).

Quote

 

1.) I thought all the time that in new KSP Atmosphere we should stick to a max-Q ( and I just took the advice from MJ for about 20.000pa)
That is not true when the most important thing for you is most less dV to Orbit. You should stick with max-Q if you dont have a very well drag-optimized design to minimize stress on the rocket, thats also the way like they do it in Real Life, because an exploding rocket is fun in KSP, but it isn´t in Real Life.

2.) Further on beeing under Terminal Velocity is very important, you should not go over or you are waisting loss to drag. though in 1.05 its very hard for me to go over the TV limit.

 

I typically don't worry about any of that.  It might be more of an issue with a high TWR rocket, but since I'm typically ≤1.4, I just put the throttle at 100% and go. 

Quote

3.) It seems that if you have a very well aero designed rocket, that the most important factor is not drag loss anymore, the most important factor is gravity loss, if you stick with max q limit to 20.000pa you will always have a too high gravity loss and if your rocket is stable you are not gaining anything. ( in a gamesim like KSP )

To me, drag losses are over rated; they just aren't that big.  Just make a rocket that is reasonably aerodynamic and don't worry about it.  It's not a big enough problem to overthink it.

If you are designing for minimum Δv, then reducing gravity loss is more important.  Of course my design philosophy is to maximize cost efficiency, so I'm not overly concerned about gravity loss either.  I pack on the fuel to increase my payload capacity and just accept the fact that I'll incur high gravity losses.

Quote

4.) If you are trying to get the best optimized ascent, you can´t use Mechjeb Autopilot Ascent, you have to do it manually !

I don't use Mechjeb, so I always do it manually.

Quote

6.) Im not sure about that thing, but it seems that a rocket with more width has also less drag in KSP and i think it should be the other way. Im testing this further on. I can only say that with similiar Ascent behavior, my first rocket with 1.25m did always had much higher drag loss than the rockomax parts rocket.

I've also observed that launch vehicles using 1.25m parts always require more Δv to attain orbit.  I usually budget at least 3600 m/s, and as high as 3800 m/s if I have a non-aerodynamic payload.  With a 2.5m rocket, I'm usually right around 3400 m/s.

Quote

7.) Round about you should stick with that Degrees to Ascent:
Pitch Over at : 100 m/s ( which is with a good TWR Vessel always something between 600m high and 1400m high, you should look only at m/s, because under 100m/s will return in too fast gravity turn and too much over 100m/s will return in a too straight gravity turn with much steering corrections and loss )

I usually don't pay too much attention to either velocity or altitude.  I just start the turn when it feels right.  Since I try to design all my launch vehicles to a standardize set of guidelines, they usually fly similarly.  What I try to do is keep the angle of attack as small as possible throughout the turn.

Link to comment
Share on other sites

The drag force is proportional to the cross-sectional area, drag coefficient and the vehicle velocity. It is independent of vehicle mass (and therefore likely liftoff thrust). A 1.25m rocket would experience 1/4 the drag force of a 2.5m rocket using an identical boby shape and ascent profile. However, your rocket is 8 times heavier (mass scales with the cube of radius). Therefore, the drag force would only cost half as much delta-v due to your higher mass and inertia. 

Basically, 2.5m or 3.75m rockets are far better at overcoming drag due to their sheer mass - it becomes negligible for 3.75m stacks especially with decent payload shapes/fairings. 

I routinely launch groups of 4 2.5m or 3.75m stacks held together with struts and a central payload (only way I can find to get 200 tons to orbit) and drag is not an issue with a 1.25 starting TWR - the engine's control authority from thrust vectoring is larger than any drag force assuming my AoA is not crazy. 

Link to comment
Share on other sites

Mass scales with the cube of radius if you assume a spherical rocket.

For a cylindrical one, mass scales quadratically with radius, just like drag. What matters is the height: drag doesn't (much) change with height whereas mass scales linearly with height.

The maximum height you can achieve depends on construction ability.

Here, for example, is the result of my latest attempt at making a very tall rocket:

NUZ5EJY.png

Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...