How to achieve good TWR when you need lots of fuel

[JackBush]

A few days ago, I got Snark's sweet rover to the Mun but didn't have sufficient fuel in it to get back to Kerbin. 

This has led to my adding more fuel and therefore more mass (is mass the same as weight?). And therefore a more powerful ship, which I'm currently experimenting with. But this leads to an overall question for me: how do you balance the need for more fuel while achieving a good TWR? 

[Randazzo]

I suspect you'll see a lot of references to the tyranny of the rocket equation pop up in here. Basically, you start get diminishing returns rather quickly in terms of adding more fuel/thrust. You need a little more dV so you add more fuel only to discover your TWR is too low, so you add more thrust, only to discover you need more fuel, so on and so forth. This is why rockets use stages, and specific to your situation, why the Apollo lander had two stages. You probably won't need to go to that extreme just to land on and re-ascend from the Mun, but if you want to return with the same vehicle, a delivery stage would help.

I've probably rambled a bit farther than necessary, but the answer to

25 minutes ago, JackBush said:

But this leads to an overall question for me: how do you balance the need for more fuel while achieving a good TWR? 

is: Stages

[steve_v]

Ahh, the tyranny of the rocket equation. There's no escaping it.
The only ways around this: decrease your payload mass, decrease overall non-fuel mass, increase engine efficiency, or stage more.
Or just add moar boosters, moar fuel and end up with a monster.

And no, mass is not quite the same as weight. Mass x gravitational acceleration = weight. On earth, they're often used interchangeably, but in space (microgravity) you mass the same, but weigh less.

Edited December 28, 2015 by steve_v

[OhioBob]

24 minutes ago, steve_v said:

And no, mass is not quite the same as weight. Mass x gravitational acceleration = weight. On earth, they're often used interchangeably, but in space (microgravity) you mass the same, but weigh less.

To add to this, weight is actually a force and is measured in different units than mass.  Mass is measured in kilograms (or metric tons) while weight is measured in Newtons (or kilonewtons).  Newtons is also the unit used to measure thrust (another force).  This is why thrust-to-weight ratio is dimensionless - the units cancel and we are left with a dimensionless ratio.

[Dman979]

So, mass is not the same as weight. Weight is a vector quantity, as given by the equation W=MG, where W is Weight, M is mass, and G is the local acceleration due to gravity. Weight can change if the acceleration on an object changes.

Mass is a different unit: it gives the amount of matter in a given object. You can derive mass from weight, and vice versa, but they are two different quantities.

The rocket equation is ∆V= Isp*G*Ln(M₀/M₁)

∆V is the potential change in velocity your spacecraft can achieve, measured in meters/second
Isp is the effiecny of your craft, in seconds
Ln is the natural log (you'll need to use a calculator)
M₀ is the total mass of your craft, including your LFO, measured in Kg
M₁ is the empty mass of your craft (including any propellant that you don't want to use, like monoprop), in Kg.

There's another equation, too: T/W
T is the total thrust of the rocket at a given point in time, and
W is the weight of the rocket (see above)

Here's where you make compromises:

In real life, ∆V has a higher priority than TWR. This is because most chemical fuels have a comparatively low efficiency. An orbital engine is needed which maximizes the Isp of the fuel. If this means having a low TWR, most missions are ok with this, because they are multi-year operations.

You need to find a balance between how much ∆V you need and the TWR you want in orbit. I've been running missions with 20m+ burn times and 0.03 TWR, and split them into multiple orbits. I'm OK with this, because I only needed to put a small amount of mass in orbit, but i have ~1,500 km/s of ∆V. I'm using Realism Overhaul, so maximum ∆V in minimum mass is my priority. Since you're using the stock game, you need to decide what your priorities are.

What I would recommend would be to calculate how much ∆V you will need for everything after you first get into orbit, multiply that by 1.25 to 1.5 (depending on how confident you feel in your piloting skills), and then find the engine which strikes the best compromise between TWR and Isp for your needs.

[Snark]

What the previous posters have said:  that's what staging is for.

Also, note that rovers in general are not great rocketships.  Craft that are special-built for one purpose generally work better than craft that try to do two things and end up being kind of half-assed at both. 

If I want a mission that can land on the Mun, rove around, and then come home, I usually do it with two pieces:  a lander and a rover.  The rover has just enough fuel to land itself, and that's it. The lander has land-and-return capability.  Land the lander first, then land the rover nearby.  Use the rover to rove around, then when done, go home in the lander, leaving the rover behind.

That said:  the rover pictured easily has enough fuel to go from low Mun orbit to the surface and then back to orbit again, as long as it starts with a full fuel load.  So if you're having trouble landing it and getting back to orbit, you may need to work on your landing/takeoff technique.    (Search forums/tutorials/etc. for the term "suicide burn" and learn how to do that, if you're not already.)

Note that I just siad "back to orbit," not  "back to Kerbin."  A design like this works well for the "orbiting mothership / fuel-tanker" type of mission, where you make repeated round trips to the surface from orbit, and refuel in orbit each time.

That's one strategy to consider for your Mun missions:  an Apollo-style mission, where you leave the return module in orbit, and go down to the surface with something that has only just enough fuel to go down to orbit and back.  You don't have to do that, it's perfectly possible to do a land-and-return Mun mission with no fancy orbital docking, but it's an option.

 

Edited December 28, 2015 by Snark

[FancyMouse]

I thought the answer should be MOAR SRB! no?

[Snark]

27 minutes ago, FancyMouse said:

I thought the answer should be MOAR SRB! no?

Depends.  If the problem is that there's not enough overall dV for the whole mission when it's sitting there on the launchpad, then adding SRBs could help.  However, I get the impression that that's not really the issue here, it's more a problem of dV when at the Mun.

And you really, really don't want to be sending SRBs to the Mun.  Or anywhere.  They're heavy and have crappy Isp, so they're horrible as upper-stage engines.  Don't get me wrong, I love them like a brother, but they have their place.  Great for the launchpad, terrible anywhere else.

[GoSlash27]

JackBush,

You're bound to get into trouble when you start adding stuff to an existing design. The design should ideally be built to the mission spec in the first place.

Step 1 is to figure out how much total mass one of your engines can lift at your desired t/w ratio.

Step 2 is to figure out how much of your total vehicle needs to be fuel in order to meet the DV requirement. You also need to figure in the tank mass, which is 1/8 of the fuel mass.

 Total mass minus engine,fuel, and tank mass= payload.

 Once you have that, it's a simple matter to scale the design linearly for your payload requirement.

See here for more info:

Best,

-Slashy

 

 

Edited December 28, 2015 by GoSlash27

[numerobis]

The main thing I do is target a lower TWR. You could use one engine instead of two, save 100kg, and still have TWR nearly 2 on your landing burn.

This is part of going for an overall reduction in payload mass. Engines tends to be where the big savings is. Do you need the docking ability -- you have 160kg of RCS nozzles, 60kg monopropellant, and a docking port. Do you need two goo containers? If one is enough, that's 50kg. Can you use lighter wheels, maybe build a tricycle? You've got 300kg of wheels at the moment. Do you need the command seat, 50kg? Can you reduce the number of solar panels and batteries?

All these reduce the capability of the rover, and force you to be more careful flying and driving it. But they reduce dry mass.

If you shrink the payload but keep the same fuel mass, the delta V you can achieve increases by the log of the mass ratio: if your new payload is alpha times as heavy as the old payload, your delta V increases by -Ve ln(alpha). With your engine, if you shrink your dry mass by 10% (so alpha is 0.9), your delta V will increase by 310 m/s.

[Yasmy]

5 hours ago, Dman979 said:

<...> Weight is a vector quantity, as given by the equation W=MG, where W is Weight, M is mass, and G is the local acceleration due to gravity.

<...>The rocket equation is ∆V= Isp*G*Ln(M₀/M₁)

<...> M₀ is the total mass of your craft, including your LFO, measured in Kg <...>

Overall, nice, helpful post.  A couple suggestions:  

1) kg, not Kg.  Case matters for SI unit prefixes.

2) G is typically Newton's constant, and g is the local acceleration. Again, using conventional case helps with clarity.

3) Your post suggests G (which should be g₀) in the rocket equation is the same as G in the relation between mass and weight. It's not.

[Plusck]

Another helpful little rule of thumb:

To get down to the surface of the Mun, you will generally use a bit less than half of your fuel (assuming your vessel is mostly fuel). Getting back up will generally take a bit more than half of what's left.

So for the Mun, you should have enough fuel to get down and back up again if your tanks are full in low Munar orbit and you aren't carrying too much weight mass and you land efficiently. That last bit - landing efficiently - basically means you need to do something like a suicide burn, even though it doesn't have to be a perfect one.

Edited December 29, 2015 by Plusck

[Chaos_Klaus]

1 hour ago, Plusck said:

To get down to the surface of the Mun, you will generally use a bit less than half of your fuel (assuming your vessel is mostly fuel). Getting back up will generally take a bit more than half of what's left.

Nope. You will use about half your delta v ... but due to the logarithmic nature of the rocket equation that translates to way more than half of the fuel.

[monstah]

2 hours ago, Plusck said:

Another helpful little rule of thumb:

To get down to the surface of the Mun, you will generally use a bit less than half of your fuel (assuming your vessel is mostly fuel). Getting back up will generally take a bit more than half of what's left.

 

1 hour ago, Chaos_Klaus said:

Nope. You will use about half your delta v ... but due to the logarithmic nature of the rocket equation that translates to way more than half of the fuel.

Half of what fuel or delta-v? You can surely design a lander to accomplish each of the cases

[Mastikator]

The ISRU is the best option when you need lots of fuel, it really is. Refueling is the only way to defeat the tyranny of the rocket equation.

[Snark]

2 hours ago, Mastikator said:

The ISRU is the best option when you need lots of fuel, it really is. Refueling is the only way to defeat the tyranny of the rocket equation.

Absolutely.  The key here being "when you need lots of fuel"-- adding an ISRU and assorted paraphernalia really bumps up the minimum workable vehicle size, which in turn can make the overall sitting-on-the-launchpad mass a lot bigger.

You can do just fine with a round-trip to Duna and back without using ISRU, with a surprisingly small vehicle.  If the OP just wants to do a simple-go-and-back at minimum mission cost, with the added wrinkle of mining a bit of ore while there, then ISRU may or may not be a good idea.  Doing ISRU would significantly increase the scale of the mission.

It's also the case that the OP here is someone who's new to Duna, and building an ISRU-scale ship is a significantly bigger design challenge.  Not sure if the OP feels like biting off the added complexity?

[Mastikator]

Does it though, if you're going to land and take off multiple times, perhaps on multiple bodies then the ISRU instantly makes the initial rocket smaller

That said, about mun rovers and TWR you only need a good TWR on take-of and landing. It takes about 5.6km/s d-v to land on the mun, and a little over 1.1km/s to take off from the mun and land on Kerbin.

Start off with a high TWR stage using SRBs, then go to high efficiency burn to circularize around Kerbin, then a low TWR burn to the Mun, followed by a low TWR capture, circularize, go sub-orbital and finally when you land is when you need >1 TWR

A single terrier with a bunch of detachable tanks can go really far.

If you want a ship that can go to Ike and then Duna and then potentially Ike again (or even back to Kerbin orbit) then you'll need an ISRU.

[Snark]

45 minutes ago, Mastikator said:

Does it though, if you're going to land and take off multiple times, perhaps on multiple bodies then the ISRU instantly makes the initial rocket smaller

Yes, definitely-- if you're going to be bopping around, doing multiple takeoffs and landings, etc., then ISRU wins, hands-down.

But not necessarily if it's just a simple land-once-then-return mission, which is more like what it sound the OP is trying to do here, if I'm reading it correctly.

[JackBush]

18 hours ago, Snark said:

Absolutely.  The key here being "when you need lots of fuel"-- adding an ISRU and assorted paraphernalia really bumps up the minimum workable vehicle size, which in turn can make the overall sitting-on-the-launchpad mass a lot bigger.

You can do just fine with a round-trip to Duna and back without using ISRU, with a surprisingly small vehicle.  If the OP just wants to do a simple-go-and-back at minimum mission cost, with the added wrinkle of mining a bit of ore while there, then ISRU may or may not be a good idea.  Doing ISRU would significantly increase the scale of the mission.

It's also the case that the OP here is someone who's new to Duna, and building an ISRU-scale ship is a significantly bigger design challenge.  Not sure if the OP feels like biting off the added complexity?

I'm not going to Duna YET. And yes I am doing a simple go-get-science-and-back. Hadn't thought of mining ore and don't have a clue (YET) how to do it. Or why.

[Wanderfound]

While agreeing with all of the "tyranny of the rocket equation" comments, from a pure engineering POV: radial engines.

When you find that your stage doesn't have quite enough TWR, don't start bolting on a dozen SRBs, just add some radials. Twitch, Thud or Vector, depending on the size of the rocket. They're quite handy for balancing large ungainly landers, too:

 

[Wanderfound]

33 minutes ago, JackBush said:

I'm not going to Duna YET. And yes I am doing a simple go-get-science-and-back. Hadn't thought of mining ore and don't have a clue (YET) how to do it. Or why.

Because that thing shown landing above is Grand Tour capable thanks to its ISRU mining rig; that's why.

Edited December 30, 2015 by Wanderfound

[Plusck]

On 29/12/2015 at 0:12 PM, Chaos_Klaus said:

Nope. You will use about half your delta v ... but due to the logarithmic nature of the rocket equation that translates to way more than half of the fuel.

"Nope."?

It is extremely disagreeable to contradict someone just for the sake of it, and counter-productive to do it in a thread that is supposed to be a help to other players.

So I repeat: a rough rule of thumb is that you can get down to Mun surface from a low-ish orbit using slightly less than half of your fuel, and back up to orbit with slightly more than half of what's left.

And that applies for any LfO-burning lander where most of the mass is in the fuel.

I would add that it will be very difficult to design a lander that is able to do much better than that for the Mun, and still be useful for doing any of the other things you might want a lander to do (i.e. ferry people, carry scientific experiments, dock with orbiting mothership). The only real way of doing better is to use an LV-N and a lot of fuel, but that makes a very unwieldy craft.

This is my Mun lander doing a flag-planting trip. Starting at a roughly 80kmx40km eliptical orbit and 135 fuel. Landing with 79 fuel left (slightly more than half). Redocking in orbit with 36 fuel left (slightly less than half of landed fuel). The disadvantage of this design is the lack of a pod, and therefore the possibility of storing experiments, so you're limited to one EVA report and one surface sample per Kerbal.

 

Edited January 3, 2016 by Plusck

[Kryxal]

Personally, I like the idea of a drop tank on the bottom, it will reduce your TWR but you have lots for landing and you'll be rid of the tank before you're on the ground.

[Chaos_Klaus]

On 3.1.2016 at 1:53 PM, Plusck said:

It is extremely disagreeable to contradict someone just for the sake of it, and counter-productive to do it in a thread that is supposed to be a help to other players.

So I repeat: a rough rule of thumb is that you can get down to Mun surface from a low-ish orbit using slightly less than half of your fuel, and back up to orbit with slightly more than half of what's left.

And that applies for any LfO-burning lander where most of the mass is in the fuel.

I would add that it will be very difficult to design a lander that is able to do much better than that for the Mun, and still be useful for doing any of the other things you might want a lander to do (i.e. ferry people, carry scientific experiments, dock with orbiting mothership). The only real way of doing better is to use an LV-N and a lot of fuel, but that makes a very unwieldy craft.

I'm not contradicting for the sake of it. I'm just saying that that is not true. Landing on the Mun and returning to orbit both takes the same amount of delta v. The same amount of delta v can stand for different amouts of fuel. You can see that your ascent takes less then half of your fuel. The fact that in your case it takes exacly half of the total fuel and you have 1 quater left is purely incidental.

What portion of your fuel you use, depends on the total delta v budget of the rocket/lander.

I'm sorry if I upset you. But it is how it is.

Edited January 4, 2016 by Chaos_Klaus

[Plusck]

58 minutes ago, Chaos_Klaus said:

I'm not contradicting for the sake of it. I'm just saying that that is not true. Landing on the Mun and returning to orbit both takes the same amount of delta v. The same amount of delta v can stand for different amouts of fuel. You can see that your ascent takes less then half of your fuel. The fact that in your case it takes exacly half of the total fuel and you have 1 quater left is purely incidental.

What portion of your fuel you use, depends on the total delta v budget of the rocket/lander.

I'm sorry if I upset you. But it is how it is.

No, it's merely annoying.

First, I said it is a "rule of thumb". Second I said if your craft is mostly fuel, then... blah blah blah. So you saying "nope" was just stupid and annoying, since what I said is most certainly true, as a rule of thumb, for a lander design where the mass is mostly fuel.

What might have been intellgent would have been if you added something interesting, like the average delta v budget of a single-stage lander-type vessel in vacuum that uses neither the LV-N nor ion engines. And I'm guessing it'll be about 2.5x to 3x Mun low orbit velocity.

Edited January 4, 2016 by Plusck