Highest Dv with decent TWR?

[Veldez]

I have been trying to built a nuke engine setup that can hit 15,000+ Dv and still have a TWR of 0.2 or 0.25. So far the best I have come up with is under 12,000 Dv. Can anyone give me pointers on the best way to built a massively high Dv rocket, or do I have to settle for assembling such ships in space? I hate the wobbling clamp-o-tron attachments cause lol. Using stock parts BTW.

[Armisael]

What's your payload?

[Snark]

It's certainly doable, but it will require a lot of fuel in drop-tanks, and appropriate mathematical attention to ship design.  You'll need an overall mass ratio of 6.77 to hit 15 km/s of dV with nukes, which is shaving it pretty thin given that fuel tanks themselves are only a mass ratio of 9.  So definitely need a multi-stage design.

And you're definitely not going to be able to do it with a TWR of 0.2 or higher (unless you're willing to throw away lots of equipment and make a truly gargantuan, excruciatingly expensive ship for a tiny payload).  You'll need to go lower than that.  To hit that kind of dV, you need to keep dead weight to an absolute minimum, and nukes are heavy.  For example:  Suppose you have a craft with just a single nuke on it.  To get a TWR of 0.2, your total mass can't be any higher than 60 / (0.2 * 9.8) = 30.6 tons.  To hit a dV of 15 km/s, you need a mass ratio of 6.77, meaning that the dry mass of your ship can't be any higher than 30.6 / 6.77 = 4.52 tons.  The nuke engine itself is 3 tons, leaving 1.52 tons for the entire remainder of the ship.  And just the fuel tanks alone, for 25+ tons of fuel, will be a lot heavier than 1.52 tons.

You could do it at high TWR if you're willing to throw lots of equipment away.  For example, suppose you have an asparagus design, with a lot of radial boosters, with each radial booster having an LV-N and 8 tons of liquid fuel.  That would give each booster a mass of around 12 tons with the tanks full, or a TWR of 0.51.  So you could have lots of those stages... but you'd need to throw away a lot of LV-N engines, which is not only pricey, but also you'd need to be putting that monstrous ship up into LKO in the first place, which would be a royal pain.

So, in general:

• yes, you can get 15 km/s of dV
• no, you can't do it nuke-powered with a TWR of > 0.2, unless you want a ridiculously huge/expensive ship for a tiny payload
• you'll need a multi-stage design (i.e. drop-tanks)

If you could post a screenshot of your best attempt, it would be a lot easier to offer more specific advice. 

Also, what are your parameters / priorities?  e.g. do you care how wasteful the design is?  are you willing to go with lower TWR if it makes the ship a lot cheaper/lighter?  etc.

[Veldez]

Well, it was half-thought exercise, half-real attempt. I was curious just how high a Dv we could get in KSP, given the limits of the engines and fuel tanks available. My best attempt was using about 60 or 80 of the Mk0 liquid fuel tanks meant for small planes, and a single nuke. I hit just under 12k Dv with a TWR of 0.2ish. This was with zero payload though.

The origin of this idea for me was to see if I could make a vehicle that could transport 2 kerbals to the orbit of any body in the system, dock with a pre-positioned station, and do it while (mostly) ignoring launch windows or any other optimal transfer concerns. I could either assemble it in space or refuel after launch, either way would work, as long as it is reusable once built. It would just be a personnel transporter for a pre-built refueling network.

The reason for the TWR minimum is I HATE doing super long burns, plus overheating is always a factor too. Since it is intended to be reusable, the cost isn't important, although asparagus designs that include 50 orange tanks in a cluster make me gag a little lol.

Edited March 25, 2016 by Veldez
added info

[Snark]

4 hours ago, Veldez said:

The reason for the TWR minimum is I HATE doing super long burns, plus overheating is always a factor too. Since it is intended to be reusable, the cost isn't important, although asparagus designs that include 50 orange tanks in a cluster make me gag a little lol.

"Reusable" is gonna make it hard, since that means you can't stage empty parts away, which makes hitting that 6.77 mass ratio really difficult.  Without staging, it's mathematically impossible to get to 15 km/s with nukes and have a TWR over 0.2.  If you're willing to accept a really low TWR, then it's mathematically possible, barely.  Will require a ship that is basically, 1. the nuke itself, 2. a lot of fuel tank capacity, and 3. a tiny payload.  Most of the ship's dry mass will be empty fuel tank.

(The absolute maximum possible dV to get on an LV-N-powered ship, without staging, is 17,226 m/s.  That's the unattainable ideal which you can asymtotically approach but never reach-- it's the result of fuel tanks' 9:1 mass ratio.  So 15 km/s is barely doable, but skating fairly close to the limit.)

Overheating's not a problem at all, just put some radiators on.  For a ship powered by one LV-N, four of the smallest-size foldable radiators will be plenty enough to let you burn forever without overheating.

[Veldez]

8 minutes ago, Snark said:

(The absolute maximum possible dV to get on an LV-N-powered ship, without staging, is 17,226 m/s.  That's the unattainable ideal which you can asymtotically approach but never reach-- it's the result of fuel tanks' 9:1 mass ratio.  So 15 km/s is barely doable, but skating fairly close to the limit.)

Ok, I guess that answers my question pretty well then. Thank you for doing the math lol! I guess I should aim a bit lower then, maybe 8-10k dV.

[GoSlash27]

This question can be answered mathematically.

A single NERV can generate 60kN of thrust, which can accelerate 60/9.81= 6.12 tonnes of mass at 1G acceleration, or 30.6 tonnes at .2G

If your target DV is 10 km/sec, then you can plug this into the reverse rocket equation.

Rwd= e^(DV/g0Isp)

Rwd= e^(10,000/9.81*800)

Rwd= 3.58

So the ship will weigh 3.58 times it's empty weight when loaded with fuel

Converting this to fuel fraction

Ff= (Rwd-1)/Rwd

Ff=.721

The fuel mass will be 72.1% of the total mass of the ship

The fuel tanks weigh 1/8 the mass of the fuel, so .721/8=..0901

The tanks weigh 9.01% of the total mass of the ship.

Fuel plus tanks is 81.1% of the total ship mass

.811*30.6 tonnes = 24.8 tonnes of tanks and fuel

And the engine weighs 3 tonnes.

Payload mass is total mass minus engines, fuel, and tankage.

30.6-24.8-3= 2.8 tonnes.

So a single NERV can impart a DV of 10 km/sec to a 2.8 tonne payload at .2G acceleration using 24.8 tonnes of loaded fuel tanks.

You can scale this linearly for whatever payload you need to move. 2 engines can move 5.6 tonnes of payload with 49.6 tonnes of fuel and tankage, and so on.

 

HTHs,
-Slashy

 

Edited March 25, 2016 by GoSlash27

[Plusck]

Fully reusable? That would be hard.

Getting 15k with a TWR always above 0.2 should be doable with drop tanks and asparagus staging, but the main problem is part count.

Getting 8k should be relatively easy.

I tried to see what an extension of one of my standard designs could do. Minimum TWR is 0.3 for the LV-N stages. Fully returnable Mk1-2 pod with heatshield, parachutes, science, Remote Guidance Unit. I managed to get 13.5k m/s in LKO, after losing a couple of Mk0 tanks to a poorly executed stage. The lifter stage wasn't pretty. Total dv for the whole thing about 17k m/s. Part count is horrible - but still flyable.

Spoiler

Spoiler

 

[Norcalplanner]

14 hours ago, Snark said:

You could do it at high TWR if you're willing to throw lots of equipment away.  For example, suppose you have an asparagus design, with a lot of radial boosters, with each radial booster having an LV-N and 8 tons of liquid fuel.  That would give each booster a mass of around 12 tons with the tanks full, or a TWR of 0.51.  So you could have lots of those stages... but you'd need to throw away a lot of LV-N engines, which is not only pricey, but also you'd need to be putting that monstrous ship up into LKO in the first place, which would be a royal pain.

 

I did exactly what Snark described as part of a challenge back in 0.90.  The goal was to see how fast you could get going before leaving Kerbin's SOI, so I went with a massively onion- and asparagus-staged behemoth that managed to accelerate to almost 32 km/s before it ran out of fuel (and 15 minutes before it left Kerbin's SOI).  Lowest TWR on the upper stages was around 0.55.  MechJeb is a bit confused in the VAB because of fuel lines.

 

Edited March 26, 2016 by Norcalplanner
fixed formatting