How much reserve fuel (delta-V) do you take?

[paul23]

One can, considering the delta-V map created by the community, easily optimize the amount of delta-V for a specific mission. Creating the cheapest rocket for the job.

 

However the other day I made a Mun lander, which had 250 surplus delta-V from the community map (250 after escaping from the moon, to get myself into the atmosphere at periapsis). After escaping the Mun I had only 78 delta-V to spare though, about 1-2% o the initial delta-V. So that was a bit close for comfort. I had to do a lot of aerobraking approaches to get the apoapsis down.

This made me wonder: how much "extra" do you take above what the community map says?

[Zeiss Ikon]

I currently play RO, and don't have restarting engines yet -- so, none.  Ideally, I want to make my orbit with just a whisper left...

[Streetwind]

14 hours ago, paul23 said:

I had to do a lot of aerobraking approaches to get the apoapsis down.

This made me wonder: how much "extra" do you take above what the community map says?

Just a tip - you can generally do a direct reentry from the Mun, unless your lander is really temperature sensitive. Just a service bay, or a heatshield without ablator, is more than enough protection (even from Minmus). A periapsis of ~30km is ideal for this, and you can set that periapsis directly during your Mun escape burn. In fact, it is cheapest to do so at that point, unless you were in a highly inclined Mun orbit.

As for the question - I don't generally use the community map. I might resort to it when I want to go somewhere I rarely visit, but the typical destinations (Mun, Minmus, Duna, Eve) I have rule-of-the-thumb memorized. For example, when I build a Mun lander, my thought process goes like this: "3,500 to Kerbin orbit, 900 to the Mun, 300 to insert, 600 to land, 600 to take off, 300 to eject... so I'll bring 6,200m/s." These figures are already rounded up, so that usually takes care of having a safety margin. Also, I don't sweat the details. If I build something that I like, and it just so happens to come out at 6,400 m/s, I don't bother re-engineering it - I just fly as is.

 

Edited November 7, 2019 by Streetwind

[RizzoTheRat]

Too much usually

My early game (career) mun/minmus landers tend to be a similar design so I have a pretty good idea what I need to use to get there, but the return tends to be less than a tank of fuel so I usually have too much.

Where I do try and make a bit of effort to get the fuel right is on launch, where I like to drop a stage shortly before achieving orbit  limit the amount of orbital debris (I know it doesn't really matter but it does to me), however I've now got my reusable SSTO working so hopefully won't need to dump stages any more.

 

[Pecan]

1 hour ago, Streetwind said:

the typical destinations (Mun, Minmus, Duna, Eve) I have rule-of-the-thumb memorized

Good grief, well done!  I still use the map for everything beyond Kerbin orbit.

16 hours ago, paul23 said:

how much "extra" do you take

I tend to build space-infrastructure vehicles I will have, say, "transfer vehicles" that will wander between all the planets/moons, carrying various payloads and refuelling at space stations as necessary.  As such they tend to be built for maximum flexibility rather than with any specific journey in mind.  I sometimes build landers, on the other hand, with as much as +50% dV so they can plane-change for a wide range of landing sites or do multiple hops.

[evandisoft]

I aim for about 400 extra dv per stage.

[MechBFP]

Early career is pretty close because I don’t really have a choice. After that things tend to have 1000+ more than they actually need, assuming money isn’t tight.

[purpleivan]

Depends on how much attention I'm paying to my construction. I've ditched a stage meant to just circularise my Kerbin orbit, then push me towards the Mun, just a few hundred metres above the the surface, on the way to a landing, more times than I'd like to admit.

 

Edited November 7, 2019 by purpleivan

[Superfluous J]

I usually tack on 10% or so and these days rarely need it.

Except Kerbin orbit. I've done so many of those I don't bring any surplus except maybe rounding up to the nearest tank.

[Piper]

I generally, I multiply my total needed deltaV (as calculated with a deltaV road map) by 1.25x.  However if I'm doing any type of precision landing like landing at an anomaly or a base, I'll multiply the amount I need for that portion of the mission by 1.5x.  I usally also add on the max plane change amount without any multiplication.  Actually liftoff from Kerbin I usually just make an even 3500m/s, no 1.25x added on.

So as an example, let's say I was sending a craft to the surface of Dres to investigate an anomaly, I would calculate it as this:

Liftoff from Kerbin: 3500m/s

Kerbin Escape: 950m/s x 1.25 = 1187.5m/s

Transfer to Dres: 610m/s x 1.25 = 762.5m/s

Max Plane change: 1010m/s

Dres Insertion burn: 1300m/s x 1.25 = 1625m/s

Dres precision landing: 430m/s x 1.5 = 645m/s

Total mission deltaV: 8730m/s

[Francois424]

depending on missions, but I'm most comfortable with an extra 30%.

Like say going to Moho at your current phase is 7k dv, I try to put 9k on my craft.
The mission I am having a lot of problems with is a Jool exploration ship. 
Speaking of doing it without refueling capacity, ofc.
I plan ahead, but I always am short when returning home and must either abandon the mission or get out and push for an hour.
So for a Jool-5 I always DOUBLE what I think I will need.  That usually takes care of it

[Edit] If you get Kerbal Alarm Clock, it helps tremendously.  When I have the mod installed it's very rare I run out of dV as I know all the proper transfer windows with "the flick of a switch"

Edited November 7, 2019 by Francois424

[septemberWaves]

I optimise my rockets to fly with MechJeb's autopilot without manual input (at least as much as I can avoid it); I determine the maximum payload by first experimentally determining the required delta-v to launch the rocket into the desired parking orbit (which can vary somewhat depending both on the target orbit altitude and the design of the rocket, as well as the ascent trajectory used), and then using that as a basis to adjust the NRAP test weight (with smallest increments of 100kg) until the excess delta-v (whatever is left above the number I determined previously) is within 10 to 20m/s above the required minimum. I do de-orbit my upper stages, but they always have RCS which is partly for this purpose - and often there's a bit more delta-v left than what there is at maximum payload capacity anyway because I don't always launch payloads at the full capacity of the rocket.

For anything besides low Kerbin orbit, the margins can vary a bit more than that, depending on how familiar I am with the specific intended mission profile and on how versatile the spacecraft needs to be. My crew spacecraft tend to be designed before I have a specific idea in mind of the mission profile, so I typically go with an estimate of about 30 to 50% extra compared to a specific baseline - for example, if I know a spacecraft is meant to return from Munar orbit after being put there by another stage, and that this is the largest single burn it will ever have to make (as well as the only significantly large burn it will make in that particular kind of mission), I use that as the baseline and then add a decent amount extra for the sake of covering most small orbital maneuvers. Other times, I typically design modular stages that can send a certain payload to a certain place, and I like to set the margins for those kind of stages pretty low in a similar way to with the launch vehicles - by designing the vehicle first and then determining its payload capacity. Ascent vehicles for atmospheric celestial bodies other than Kerbin are generally the hardest to design with precise margins, since the payload tends to consistently be "crew capsule + docking equipment + orbital propulsion + life support and basic functional stuff" and it makes less sense to design the lower stages of a lander first - although having broader margins is generally better in these situations for safety reasons, particularly since landing locations can also cause a lot of variability in delta-v requirements.

[paul23]

13 hours ago, Streetwind said:

Just a tip - you can generally do a direct reentry from the Mun, unless your lander is really temperature sensitive. Just a service bay, or a heatshield without ablator, is more than enough protection (even from Minmus). A periapsis of ~30km is ideal for this, and you can set that periapsis directly during your Mun escape burn. In fact, it is cheapest to do so at that point, unless you were in a highly inclined Mun orbit.

As for the question - I don't generally use the community map. I might resort to it when I want to go somewhere I rarely visit, but the typical destinations (Mun, Minmus, Duna, Eve) I have rule-of-the-thumb memorized. For example, when I build a Mun lander, my thought process goes like this: "3,500 to Kerbin orbit, 900 to the Mun, 300 to insert, 600 to land, 600 to take off, 300 to eject... so I'll bring 6,200m/s." These figures are already rounded up, so that usually takes care of having a safety margin. Also, I don't sweat the details. If I build something that I like, and it just so happens to come out at 6,400 m/s, I don't bother re-engineering it - I just fly as is.

 

I did that, after burning to 50 km I got  78 delta-V to spare. Lower than 50km and the tanks are ripped apart from each other (FAR, tanks just next to each other). And I already have to very carefully aim to not destroy the science jr. part. And for center of gravity situations the science part is between the landing can and mk1 command pod: so I most certainly don't want to lose this connection XD.

[Skorj]

I take a lot of extra delta-V to Mun and Minmus, but only because I like to shorten the trip.  An extra 500 m/s comoing home from Minmus makes a huge difference.

I generally bring about 100m/s over my plan, in Kerbal's SOI, and a few hundred elsewhere.  You really pay for over-engineering in this game, so I don't like to go much beyond that.

[paul23]

Well I got my income at 20% (I noticed that at 10% a lot of contracts just don't make any money anymore, and you're left with tourist contracts and part testing in kerbin orbit).

[PositronLance001]

I always play sandbox, and over-engineer  a LOT  (typically 500+ spare m/s) because I am a TERRIBLE  pilot.

Edited November 8, 2019 by PositronLance001

[SlinkyMcman]

I think I'm taking +50% extra dv, but the reality is more like -50%.

[RizzoTheRat]

On 11/7/2019 at 11:10 PM, paul23 said:

I did that, after burning to 50 km I got  78 delta-V to spare. Lower than 50km and the tanks are ripped apart from each other (FAR, tanks just next to each other). And I already have to very carefully aim to not destroy the science jr. part. And for center of gravity situations the science part is between the landing can and mk1 command pod: so I most certainly don't want to lose this connection XD.

For my early game landers I eva to get the science from the experiments and then only land the capsule.   A 30-40km pe from the mun or minus is fine without a heat shield.  This keeps the weight, and therefore the fuel requirement, down a lot. 

Edited November 9, 2019 by RizzoTheRat

[Pecan]

4 hours ago, RizzoTheRat said:

only land the capsule

To take it to the extreme - once you can dock you only ever need one fully-equipped science module.  After launching it, position it as required with tugs/landers, extract and return the science separately, clean, refuel, reposition and reuse the module.

[esmenard]

For landing and launching in orbit, I bring about 100 to 200 m/s margin, because I never know exactly how much I will need. For other stages, either I plan the flight plan before and I bring almost no margin, or I don't have a flight plan and i just scrap a random amount of delta-v that looks good enough. 

[pandaman]

Typically either not enough or way too much.

[STeven1366]

Mostly for another 30-40% of Delta-V that allows me to consider some other extended mission after the main mission was completed, for example go to Saturn's inner moons after Manned Landing in Neptune,

[Terwin]

I must admit that I generally only use transfer windows by chance.

For a nuclear powered scan-sat probe heading to Duna, I realized after my ejection burn that when I achieve polar orbit around Duna, I will have the entirety of my 'core' stage(~7km/s) and several hundred m/s in my drop-tanks still remaining.

On the other hand, my self-fueling Munar Hopper only had ~500m/s left in LMO, and I had to send a fuel tanker out to it so it had enough fuel to land.

(my Mk2 version launched with ore in the tank and had quite a lot of fuel left after landing on Minmus.  Unfortunately the 'Viewing Cupola' in the contract description seems to be the core cupola pod, and not the 'viewing cupola' part from MKS, so it can only fulfill half of the contracts it was intended for)

 

Generally I favor self-fueling ships and I like to have a fairly large fuel/ore capacity so that they can be used for a variety of missions, with the Life-support(USI-LS) generally providing the hardest limit on the range of a given vessel.

 

[Pds314]

The community maps are not an exact science.

Take Tylo: escape velocity is listed by the wiki at 3068 m/s. So in theory, a cannonball launched at that speed from the ground should reach Tylo escape. However, the recommended delta-V for Tylo ascent and escape is 2270+1050= 3320. So 250 m/s. Delta-v is assumed necessary for gravity and turning losses.

Laythe, meanwhile, recommends a whopping 1000 or so Delta-V over the escape velocity.

 

Depending on craft design, mission profile, and pilotting, you may need significantly more or less delta-V than the chart says.

Edited November 11, 2019 by Pds314

[paul23]

18 minutes ago, Pds314 said:

The community maps are not an exact science.

Take Tylo: escape velocity is listed by the wiki at 3068 m/s. So in theory, a cannonball launched at that speed from the ground should reach Tylo escape. However, the recommended delta-V for Tylo ascent and escape is 2270+1050= 3320. So 250 m/s. Delta-v is assumed necessary for gravity and turning losses.

Laythe, meanwhile, recommends a whopping 1000 or so Delta-V over the escape velocity.

 

Depending on craft design, mission profile, and pilotting, you may need significantly more or less delta-V than the chart says.

Escape velocity is very much exact science. And in ksp if you have escape velocity you *will* escape if you instantaneous reach that speed and don't have any aerodynamic drag. Actually a bit less since you don't even need to escape to infinity, having an elliptical orbit with apoapsis outside the sphere of influence counts as having escaped.

So the actual rocket equation is:

delta_v = v_exhaust * ln(m0/mf) - g * time_burn - integral(dragForce * time) ds