Scott Manley weighed in on the Nuke discussion - what do you think?

[sal_vager]

I don't mind using the LF only aircraft tanks that KSP already provides, and the LV-N is a great vacuum engine for large designs, that's its purpose, not a lift or lander engine or for little craft where its own mass works against it.

Like with this thing.

Scott pretty much addressed everyone's concerns though, including the one about LF only tanks, you're allowed to you aircraft tanks in space

[Brotoro]

Unless I didn't read the fuel flow-numbers right, the engine should burn through the same tank without oxidiser in the same time as it once burned through both. Thrust is still 60kN, and 60kN for the same time as before but without the mass of the oxidiser that was can only result in more delta-v. Sure, theoretically the rocket equation would state that higher wet mass with the same dry mass should give more delta-v (actually less dry mass because the engine was lighter then), but due to the way the game works (accelerating dynamically by thrust over current mass times time while mass changes) the oxidizer didn't really "count" as propellant. If one were to compare the working delta-v of nuclear craft in both .90 and 1.02, they'd find a crossover point where the new engine did better, once the lack of oxidiser mass overcame the added engine mass.

Though thinking about it...this may've screwed with mechjeb and KE's calculations.

I don't know WHAT you are looking at, but when I actually test this out in the program, a ship in 0.90 with LFO exhibits a delta-V of 3980.0 m/s. The same ship in 1.0 (with LF only, and the oxidizer removed) exhibits a delta-V of 1731.6 m/s.

The ship tested in 0.09 with an extra part added to make up for the difference in engine mass between the two versions gets 3279.1 m/s of delta-V.

- - - Updated - - -

...Scott pretty much addressed everyone's concerns though, including the one about LF only tanks, you're allowed to you aircraft tanks in space

No, that doesn't address the concern that there are no good 2.5 meter LF-only tanks. Kerbals engineers wouldn't make a tank that was designed to be used half-empty. And clipping fuel tanks together (as you did) would not be an acceptable solution for me..although that looks cool.

Edited May 25, 2015 by Brotoro

[cybersol]

What I really want is a large LF only tank similar to the big orange that fits into Mk3 cargo bays.

[klesh]

Maybe do what B9 aerospace (through the firespitter mod?) does and have multiple tweakable versions of each tank, some with a different texure/model.

After seeing how well that works in that mod, I am at a loss as to why the developers don't use a system like that.

[Wjolcz]

After seeing how well that works in that mod, I am at a loss as to why the developers don't use a system like that.

Same.

[somnambulist]

I don't know WHAT you are looking at, but when I actually test this out in the program, a ship in 0.90 with LFO exhibits a delta-V of 3980.0 m/s. The same ship in 1.0 (with LF only, and the oxidizer removed) exhibits a delta-V of 1731.6 m/s.

That's not quite a fair comparison -- the .90 and 1.0 Nerva burn through reaction mass at the same rate.

13 476 m/s -- .90 LV-N - orange tank of LfO

2 880 Liquid Fuel 14 400 kg
3 520 Oxidizer    17 600 kg
      Dry Mass     4 000 kg
	          36 000 kg

=============================================

8 767 m/s -- 1.0 LV-N - orange tank, Lf only

2 880 Liquid Fuel 14 400 kg
      Dry Mass     4 000 kg
	          18 400 kg

=============================================

8 767 m/s -- .90 LV-N - orange tank of LfO,
propellant mass equal to 1.0 Lf tank

1 296 Liquid Fuel  6 480 kg
1 584 Oxidizer     7 920 kg
      Dry Mass     4 000 kg
	          18 400 kg

(Note: Engine mass of 3 000 kg used when calculating the .90 dV values.)

Before 1.0 you were carrying around 1 ton of dry mass for every 8 tons of propellant. With 1.0 you have 1 ton of dry mass for every 3.6 tons of propellant. KSP's unrealistically heavy tanks are at fault.

[Archgeek]

I don't know WHAT you are looking at, but when I actually test this out in the program, a ship in 0.90 with LFO exhibits a delta-V of 3980.0 m/s. The same ship in 1.0 (with LF only, and the oxidizer removed) exhibits a delta-V of 1731.6 m/s.

Welp, looks like I did indeed misunderstand the fuel flow numbers. I started in on some testing, getting no less than six data points in .90, both with delta-v and burn time, as well as clocking fuel flow at .69 LF and .84 O per second each, and before I could set in on comparing KE and practical delta-v in 1.02, I saw that the fuel flow is now the sum of those, 1.53/s, more than double what I expected. So yeah, that's actually a pretty big nerf.

Edit: well dang, the sleepwalker up there beat me to it.

*Stupidly cool looking spacecraft*

Holy crap that thing looks awesome. It's like Dynasoar writ large.

Edited May 25, 2015 by Archgeek
Ninja'd, it seems.

[MathmoRichard]

Surely the problem is as simple as the ratio between full tanks with LF and empty tanks, and the difference in these numbers between the LFO tanks and LF only tanks.

Full/dry = ratio:

Different LFO tanks with oxidiser removed:

4.6/1 = 4.6

20.7/4.5 = 4.6

41.4/9 = 4.6

Avionic LF tanks:

0.9/0.2 = 4.5 (the smallest one)

4.6/0.6 = 7.666

28.6/3.6 = 7.9444

57.1/7.1 = 8.04 (the largest one)

So first thing to note is that the discrepancy between the rocket tanks and plane tanks is massive. The second thing is, why do the plane ratios change but the rocket ratios are constant???

[problemecium]

I agree the smallest LF tank is kinda lame. IMO it should either have 300 units or be made half the length.

As far as the rest, the fact that the giant LF tank has such a huge mass ratio makes me very happy with the LV-N changes:

[KerikBalm]

it turns out LH2 has such an abysmally low density that the tank's dry mass actually weight more than LH2 they store. It would be more realistic that way

Umm... no, and this is a mistake scott manley makes in the video....

With 1.0 you have 1 ton of dry mass for every 3.6 tons of propellant. KSP's unrealistically heavy tanks are at fault.

KSP's tanks are unrealistically heavy, but the dV needs of any craft are unrealistically small too... I figured this was a balance/gameplay decision.

If we had realistic Isps, tank weights, and realistic sized planets... payload fraction of anything would be abysmal.

Now... when you try to use the density of LH2, in KSP's fuel tanks (assuming 1 kerbo-meter = 1 human meter), then it gets unrealistic.

A LH2 tank can store much much more mass in the form of H2 than its own mass.

http://en.wikipedia.org/wiki/Space_Shuttle_external_tank#Liquid_hydrogen_tank

The *entire* space shuttle external tank had a dry mass of 26.5 tons.

That tank could store:

~630 tons of Oxidizer

~106 tons of LH2

Now... I don't know what the tank would mass if it just stored LH2, and nothing else, with no attachment hardware for the space shuttle, etc etc.

But we can see at a minimum, one can make a LH2 tank with a mass ratio of (106+26.5)/26.5 = 5:1

KSP's LF+O tanks, with no O in them, have a mass ratio of 4.6:1

The Isp is realistic, the LFO tanks with no O have realistic mass ratios...

The only problem is the unrealistic dVs.... it takes over 3x the dV to get to the Moon from LEO, as it takes to get to Mun from LKO.

Pretty much the same ratio for Earth-> Jupiter vs Kerbin-> Jool.

When you push a large payload, the tank mass fraction doesn't matter much anyway... (I'm not going to go through the math again, or my old posts just yet, but I showed for specific payload and dV cases, that the change in craft mass is quite small using LF+O tanks, or LF only tanks, or a hypothetical LF only tank with a 9:1 ratio

[enroger]

Umm... no, and this is a mistake scott manley makes in the video....

KSP's tanks are unrealistically heavy, but the dV needs of any craft are unrealistically small too... I figured this was a balance/gameplay decision.

If we had realistic Isps, tank weights, and realistic sized planets... payload fraction of anything would be abysmal.

Now... when you try to use the density of LH2, in KSP's fuel tanks (assuming 1 kerbo-meter = 1 human meter), then it gets unrealistic.

A LH2 tank can store much much more mass in the form of H2 than its own mass.

http://en.wikipedia.org/wiki/Space_Shuttle_external_tank#Liquid_hydrogen_tank

The *entire* space shuttle external tank had a dry mass of 26.5 tons.

That tank could store:

~630 tons of Oxidizer

~106 tons of LH2

Now... I don't know what the tank would mass if it just stored LH2, and nothing else, with no attachment hardware for the space shuttle, etc etc.

But we can see at a minimum, one can make a LH2 tank with a mass ratio of (106+26.5)/26.5 = 5:1

KSP's LF+O tanks, with no O in them, have a mass ratio of 4.6:1

The Isp is realistic, the LFO tanks with no O have realistic mass ratios...

The only problem is the unrealistic dVs.... it takes over 3x the dV to get to the Moon from LEO, as it takes to get to Mun from LKO.

Pretty much the same ratio for Earth-> Jupiter vs Kerbin-> Jool.

When you push a large payload, the tank mass fraction doesn't matter much anyway... (I'm not going to go through the math again, or my old posts just yet, but I showed for specific payload and dV cases, that the change in craft mass is quite small using LF+O tanks, or LF only tanks, or a hypothetical LF only tank with a 9:1 ratio

This page gives a detailed breakdown on individual Ox tank and LH2 tank on the external tank:

http://science.ksc.nasa.gov/shuttle/technology/sts-newsref/et.html

LH2 tank actually have a mass ratio of about 8:1.

[MathmoRichard]

The real world figures are fairly irrelevant since KSP's weights and forces are balanced differently.

I just wish the tanks all had the same ratio and we had rocket shaped LF tanks.

[boxman]

It is a pain, but I am sure the developers will add more LF only variants. If not, there's a mod for that.

Funny thing is that I thought this was one of the points of tweakables, so that you could yourself fill up those tanks with whatever fuel you wanted.

But like with most features they started on it seems like they quickly forgot about them before actually finishing it.

Was the same thig with the hidden rotor/ball bearing tweakable that was supposed to be enabled several versions ago and yet nothing...

[KerikBalm]

Scott manley's 1/10th fuel challenge is so ridiculous... the mass ratio (Full to empty) of the tanks goes from 8:1 to 1.7:1

ln 8 = 2.08

ln 1.7 = 0.53

IRL, a Lh2 tank does have about an 8:1 ratio...

And a LFO tank is more like 30:1

In KSP, its 9:1

ln 30 = 3.4

ln 9 = 2.2

So, LFO tanks are nerfed about (1-2.2/3.4)* 100% = 35.3%....

But he wants us to play with LH2 tanks nerfed by (1-0.53/2.08)* 100% = 74.5%...

ie capable of only 1/4 the dV of a real tank, whereas the LFO tanks are capable of 2/3s the dV of a real tank.

yea... no