• 0
Sign in to follow this  
NASI Director

Which Is More Efficient?

Question

If we use the same engine, and the same amount of starting fuel, then is it more efficient to burn slow,* or is it better to burn fast?

My theory is that it is the same, because fuel consumptions goes down if you throttle down, and goes up if you throttle up. However since you're burning slow you spend more time burning and more fuel is used in that longer time. But if you burn fast, the fuel consumption goes up and in that short time to complete a burn you used up more fuel.

What is the truth here? Is there some sort of mathematical formula that can be done to see if my theory is correct? If I use a Skipper that had the same Isp as a nuke (theoretically), which would be better to use? Is Isp the all-reigning dictator in fuel consumption and it doesn't what I do?

These are the real questions being asked here!

EDIT: Also, if I wanted to figure out the Isp of an engine, say throttled down to 50%, then would I just multiply the Isp by two?

EDIT EDIT: How do I calculate burn time for said engine?

Edited by NASI Director

Share this post


Link to post
Share on other sites

18 answers to this question

Recommended Posts

  • 0

Kerbal Space Program doesn't simulate chamber pressure, so burning at lower throttle doesn't generate a decrease in efficiency. What this means is that if you have a stupidly high (and inefficient) TWR, throttling back will give you more deltaV. This was proved when people started to use throttle management and they were able to reach orbit with 4200m/s deltaV, contrary to the supposedly 4450m/s minimum mechjeb set.

Share this post


Link to post
Share on other sites
  • 0

Ignoring gravity and atmospheres, the total amount a ship can accelerate is constant. If you are say, fighting gravity on your way to orbit than it does matter. Gravity is applying acceleration to your ship, so the faster you can escape gravity, the less time gravity has to act on you (the further away from the source you are, the weaker gravity becomes). So with no atmosphere on a planet, your most efficient throttle setting to get into orbit is 100% (not strait up, you should still gravity turn). If you want an easy way to see this, think about hovering. To hover a craft above the surface (in this case hover means you are not moving laterally relative to the surface either), you have to continuously apply thrust, so you are always consuming fuel but not moving.

In an atmosphere, efficiency is detected (in this game) by terminal velocity. The velocity at which an object in freefall will be in equilibrium between the force of gravity pulling it down, and air resistance pushing it up. In this game, a craft moving at terminal velocity is experiencing the smallest amount of air resistance possible at that altitude. Slower means you are fighting gravity more than you need to, and faster means you are fighting the air more than you need to.

If you check out this, they have a table that shows the terminal velocity for Kerbin (and you can find similar tables for the other planets). Follow this table during ascent as best you can, and combined with a gravity turn starting at 10km, you'll be getting into orbit with the minimum fuel required.

Edited by Randox

Share this post


Link to post
Share on other sites
  • 0

KSP engines have the same Isp at all throttle settings, and your (paper) ÃŽâ€V integrates over the entire burn time, so in that sense throttle doesn't matter.

In practice, throttling down is of some value in some designs so as to maximize ascent efficiency in atmosphere. It's tempting to call these designs over-engined. In orbit, the standard maneuvers assume essentially infinite TWR, so again throttling down only helps for control reasons. For landing, intermediate settings are useful, but that's for ease of use. A suicide burn is in theory more efficient.

Kerbal Space Program doesn't simulate chamber pressure, so burning at lower throttle doesn't generate a decrease in efficiency. What this means is that if you have a stupidly high (and inefficient) TWR, throttling back will give you more deltaV. This was proved when people started to use throttle management and they were able to reach orbit with 4200m/s deltaV, contrary to the supposedly 4450m/s minimum mechjeb set.
Link to the 4200 m/s ascent, please? The lowest I'm aware of is ~4380 - 4350 m/s.

edit to address additional questions: the wiki is a useful discussion of Isp and fuel consumption. For burn time, you'd want starting mass, final mass and mass flow of the engine.

t = (initial-final)/flow

Edited by UmbralRaptor

Share this post


Link to post
Share on other sites
  • 0
Link to the 4200 m/s ascent, please? The lowest I'm aware of is ~4380 - 4350 m/s.

Not on the forums, not on reddit, you guess where it is.

KSP engines have the same Isp at all throttle settings

Exactly, they should lose Isp when throttling down due to the combustion chamber losing pressure, not by a big number but there should be some loss. KSP doesn't simulate this and reducing the throttle to a more efficient setting while getting out of the atmosphere can give us some benefits, well, ACTUALLY, we are keeping the rocket slower than terminal velocity while maintaining it's Isp, so it's not a benefit gained, but rather a non-existent penalty.

Share this post


Link to post
Share on other sites
  • 0

Moving this to a more appropriate forum. Now you can use the thread tools edit your title to say "answered" "unanswered" as well!

Share this post


Link to post
Share on other sites
  • 0
Exactly, they should lose Isp when throttling down due to the combustion chamber losing pressure, not by a big number but there should be some loss. KSP doesn't simulate this and reducing the throttle to a more efficient setting while getting out of the atmosphere can give us some benefits, well, ACTUALLY, we are keeping the rocket slower than terminal velocity while maintaining it's Isp, so it's not a benefit gained, but rather a non-existent penalty.

Right, you aren't making your engine any more efficient at turning fuel into thrust, but you are getting to the same destination using less fuel overall.

I've gotten to low orbit in under 4300 m/s, but that was including a Kerbal sitting in an external seat and I think they have lower drag coefficient than most parts. Because of the way drag works you can reduce drag losses by using lots of low-drag-coefficient parts like aircraft cockpits or RCS blocks, but that's kind of cheaty.

For low delta-V ascents, try making a rocket with a high TWR, set MechJeb ascent to start turn at 8 km, end at 37 km, final flight path angle 5 degrees, turn shape 66% (credit to Nao for finding this trajectory). Limit to terminal velocity, and as soon as you reach terminal velocity also enable an acceleration limit of 22 m/s^2 and keep that on for the rest of the flight.

Share this post


Link to post
Share on other sites
  • 0

I would say that in an existing orbit at least, it's more efficient to burn at maximum throttle. The theoretical Hohmann transfer uses infinitely short impulses to change course; the closer you can match that, the closer your fuel consumption will be to ideal.

And as for whether to use an LV-N or an equally efficient Skipper, I'd go with the nuke. Unless I'm mistaken, it weighs less for the same Isp, so you should get better dV overall.

Share this post


Link to post
Share on other sites
  • 0
Not on the forums, not on reddit, you guess where it is.
Google has not turned up anything useful. Again, I request a link.
Exactly, they should lose Isp when throttling down due to the combustion chamber losing pressure, not by a big number but there should be some loss. KSP doesn't simulate this and reducing the throttle to a more efficient setting while getting out of the atmosphere can give us some benefits, well, ACTUALLY, we are keeping the rocket slower than terminal velocity while maintaining it's Isp, so it's not a benefit gained, but rather a non-existent penalty.
Fair enough. While we're at it, can we request altering Isp so it affects thrust instead of fuel consumption?
I've gotten to low orbit in under 4300 m/s, but that was including a Kerbal sitting in an external seat and I think they have lower drag coefficient than most parts. Because of the way drag works you can reduce drag losses by using lots of low-drag-coefficient parts like aircraft cockpits or RCS blocks, but that's kind of cheaty.
Yeah, I'm fairly sure that they have a drag coefficient of 0.1.

Share this post


Link to post
Share on other sites
  • 0

During launch to orbit, there is an ideal design that will place a given payload into orbit with less fuel used. With the limitations of engines available in stock, it can be more difficult to find that sweet spot using the Kerbal School of Hard Knocks. (trial and error experimentation).

On my fuel can stock experiment, I first started with a launcher of all Mainsails. Later I tried using all skippers but found it way underpowered. With the use of skippers with two mainsails under the last asparagus pair, I found a sweet spot in launch that was not only very stable spinwise, but gave me a better cushion of fuel left in the core booster.

The design;

0x77eoW.jpg

The results;

HfUTTmt.jpg

The rocket is designed to place a fuel can into orbit to dock with a refueling station. The core booster and fuel can are designed to return to Kerbal for recovery and reuse.

Share this post


Link to post
Share on other sites
  • 0
If we use the same engine, and the same amount of starting fuel, then is it more efficient to burn slow,* or is it better to burn fast?

My theory is that it is the same, because fuel consumptions goes down if you throttle down, and goes up if you throttle up. However since you're burning slow you spend more time burning and more fuel is used in that longer time. But if you burn fast, the fuel consumption goes up and in that short time to complete a burn you used up more fuel.

What is the truth here? Is there some sort of mathematical formula that can be done to see if my theory is correct? If I use a Skipper that had the same Isp as a nuke (theoretically), which would be better to use? Is Isp the all-reigning dictator in fuel consumption and it doesn't what I do?

These are the real questions being asked here!

EDIT: Also, if I wanted to figure out the Isp of an engine, say throttled down to 50%, then would I just multiply the Isp by two?

EDIT EDIT: How do I calculate burn time for said engine?

I think you are right in common sense. In orbit there shouldn't be difference between burning low or high. Although if you don't have plans for high burns why bring big engines?

In ascent there will be big difference as it is all about escaping gravity ASAP. The longer you burn, the longer you stay affected by down force. In casual use I don't see the need for throttle limiting.

Limiting throttle might be useful in achieving perfect ascent path in planets with atmosphere. But limiting itself already means that construction is not perfect.

Share this post


Link to post
Share on other sites
  • 0

Your question is a bit ambiguous, as you haven't defined "efficient" very well. Let me share what I understand about how it works:

Isp is a measure of how efficiently an engine converts fuel into thrust. It is *not* a measure of fuel consumption, though fuel consumption is a function of thrust and Isp. Think of Isp in terms of thrust per unit of fuel; i.e. a higher Isp engine produces more thrust for a given amount of fuel than one with a lower Isp. If you are trying to get the most dV from a given amount of propellant, a higher Isp engine is what you are looking for. For game balance reasons, engines in KSP with a higher Isp have a lower Thrust-to-Weight Ratio (TWR); this is not necessarily the case in real life.

In real life, Isp varies with throttle setting as an engine is only most efficient under a narrow set of operating conditions. In KSP's approximation of engine behavior the Isp stays the same throughout the throttle range and all other variables are linear, too. So an engine at half throttle in KSP produces half the thrust with half the fuel consumption while Isp remains the same. So it is no more or less efficient to burn longer at a lower throttle setting.

What does matter though, is how the in-game conditions affect the use of a given amount of thrust. During an atmospheric ascent, if your craft is exceeding terminal velocity for its altitude it is wasting fuel trying to punch a hole through the thick air. In that case it would be wise to throttle back; even though the thrust:fuel-consumed ratio is the same you would be using the thrust generated more efficiently.

Similarly, once in orbit and trying to complete a maneuver node the calculation assumes an instantaneous change in velocity. The longer the burn takes, the more fuel is expended away from the optimal point. Even though the Isp remains the same, the thrust is used less efficiently. So in orbit it is better, efficiency wise, to complete your burn as quickly as possible, i.e. at full throttle. This effect is large enough that sometimes a lower Isp but higher TWR engine can complete a maneuver using less dV than the more efficient engine, even though it converts fuel into thrust less efficiently, as more of the burn is completed near the ideal point.

Hope this helps.

Share this post


Link to post
Share on other sites
  • 0
Kerbal Space Program doesn't simulate chamber pressure, so burning at lower throttle doesn't generate a decrease in efficiency. What this means is that if you have a stupidly high (and inefficient) TWR, throttling back will give you more deltaV. This was proved when people started to use throttle management and they were able to reach orbit with 4200m/s deltaV, contrary to the supposedly 4450m/s minimum mechjeb set.

I mean no offense, but you are wrong here. DeltaV doesn't change with throttle setting as the engines have the same Isp over their entire throttle range. Throttle management consumes less dV on ascent in the case you describe because it avoids exceeding terminal velocity; it does *not* change the amount of dV available.

Share this post


Link to post
Share on other sites
  • 0
If we use the same engine, and the same amount of starting fuel, then is it more efficient to burn slow,* or is it better to burn fast?

My assumption: You are defining efficiency as minimizing fuel consumption.

Answer:

It depends?

In atmosphere, during ascent, you only want to burn fast enough to reach terminal velocity. If you go too fast, you waste fuel pushing against the atmosphere - too slow, and you waste fuel fighting gravity.

In space, always fast - as fast as you can. The reason is to maximize the effect of the Periapsis and Apoapsis helping you. The most efficient place to affect the Periapsis or Apoapsis is at the opposite node. Also, during ejection burns (to go interplanetary) you get the most efficiency due to the Oberth effect of high velocities in a gravity well.

My theory is that it is the same, because fuel consumptions goes down if you throttle down, and goes up if you throttle up. However since you're burning slow you spend more time burning and more fuel is used in that longer time. But if you burn fast, the fuel consumption goes up and in that short time to complete a burn you used up more fuel.

In essence, your hypothesis is correct; the rate at which you burn fuel has nothing to do with how efficient fuel consumption is - at least in space. The only factors that determine your overall efficiency are Fuel to Mass Ratio and Specific Impulse.

What is the truth here? Is there some sort of mathematical formula that can be done to see if my theory is correct? If I use a Skipper that had the same Isp as a nuke (theoretically), which would be better to use? Is Isp the all-reigning dictator in fuel consumption and it doesn't what I do?

Actually, yes, there is an equation that answers your question - it's the Tsiolkovsky Rocket Equation which describes how efficiency works. Also - here is a

that I did to help understand the math behind how this works.

These are the real questions being asked here!

EDIT: Also, if I wanted to figure out the Isp of an engine, say throttled down to 50%, then would I just multiply the Isp by two?

Isp does not change with different engine throttles - it is a measure of the energy potential per second per unit of fuel. Full throttle, no throttle, half-throttle, are all the same Isp - because the potential kenetic energy available is based on how energetic each unit of fuel is and how well the rocket motor can transfer the potential energy into momentum.

Say you burn a full tank of propellant, at 100% throttle, in 10 seconds and get 10k/s delta-V from it.

Now, say you burn a full tank of propellant at 50% throttle. You correctly guess that it would take 20 seconds to empty the tank - but what about the delta-V?

In the second case, you are delivering half as much thrust, which means it takes twice as long to reach the same velocity. Net result: no improvement in delta-V.

EDIT EDIT: How do I calculate burn time for said engine?

That's actually really easy, and you don't need any plugins or fancy tools:

1. Put the stage you want on to measure on the launch pad, by itself - make sure it's tied down.

2. Ignite your engines - go to full throttle.

3. Pull down the Resources menu in the upper right.

4. Look at the total fuel, and the consumption rate (in parenthesis).

5. Total Propellant / Rate per Second = Burn Time in Seconds!

Share this post


Link to post
Share on other sites
  • 0
Your question is a bit ambiguous, as you haven't defined "efficient" very well. Let me share what I understand about how it works:

Isp is a measure of how efficiently an engine converts fuel into thrust. It is *not* a measure of fuel consumption, though fuel consumption is a function of thrust and Isp. Think of Isp in terms of thrust per unit of fuel; i.e. a higher Isp engine produces more thrust for a given amount of fuel than one with a lower Isp. If you are trying to get the most dV from a given amount of propellant, a higher Isp engine is what you are looking for. For game balance reasons, engines in KSP with a higher Isp have a lower Thrust-to-Weight Ratio (TWR); this is not necessarily the case in real life.

In real life, Isp varies with throttle setting as an engine is only most efficient under a narrow set of operating conditions. In KSP's approximation of engine behavior the Isp stays the same throughout the throttle range and all other variables are linear, too. So an engine at half throttle in KSP produces half the thrust with half the fuel consumption while Isp remains the same. So it is no more or less efficient to burn longer at a lower throttle setting.

What does matter though, is how the in-game conditions affect the use of a given amount of thrust. During an atmospheric ascent, if your craft is exceeding terminal velocity for its altitude it is wasting fuel trying to punch a hole through the thick air. In that case it would be wise to throttle back; even though the thrust:fuel-consumed ratio is the same you would be using the thrust generated more efficiently.

Similarly, once in orbit and trying to complete a maneuver node the calculation assumes an instantaneous change in velocity. The longer the burn takes, the more fuel is expended away from the optimal point. Even though the Isp remains the same, the thrust is used less efficiently. So in orbit it is better, efficiency wise, to complete your burn as quickly as possible, i.e. at full throttle. This effect is large enough that sometimes a lower Isp but higher TWR engine can complete a maneuver using less dV than the more efficient engine, even though it converts fuel into thrust less efficiently, as more of the burn is completed near the ideal point.

Hope this helps.

If you read my second post you will see I corrected myself.

Share this post


Link to post
Share on other sites
  • 0

Isp is constant for an engine, with the exception of in atmosphere vs vacuum. (In KSP)

If you burn engine x at 50% throttle, you burn 1/2 the fuel as at full throttle, but receive only 1/2 the DV. Therefore, at 100% throttle, you gain (v) velocity at the cost of (f) fuel. At 50% throttle, you burn 1/2 (f) and get 1/2 (v). There is no way you can increase your DV by throttle management in a vacuum; you burn the same amount of fuel for a given (v), no matter what.

The values in () in the resources tab are resource consumption per second. You burn 1/2 the fuel at 50% throttle. So if you had a consumption of .5 at 100%, you'd have .25 @ 50%.

Share this post


Link to post
Share on other sites
  • 0

Exhaust velocities should remain the same for a particular engine regardless of throttled thrust. When you reduce thrust you are reducing the number of particles escaping the engine, not the speed of those particles. Specific Impulse therefore is unaffected, however, the efficiency of the launch vehicle itself is affected by the thrust of the vehicle as pointed out by others above.

Also in a vacuum I never worry about acceleration. Specific Impulse to me is far more critical. I regularly pack a low thrust, high efficiency second stage into my launches, once my gravity turn flattens out I generally cease to care about my acceleration. I will make that orbit.

Share this post


Link to post
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
Answer this question...

×   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.

Sign in to follow this