Fuels in KSP

[KingDominoIII]

Share your theories on what liquid fuel, oxidizer, and monopropellant actually are.

 

[Foxster]

"Actually" they aren't anything (just bits in a game) or they are whatever you want them to be. 

My guess is that they are pixie dust, fairy flakes and snot. 

[wumpus]

Basically it is whatever fuel is convenient.  I *think* there was a normalization based on a rocket engine that used hypergolic fuels, but they've moved a bit beyond that.

NTR uses only fuel: which pretty much has to be hydrogen (I suspect "oxidizer" would be better than any other fuel).
Terrier screams RT-10 type hydrox [expander cycle] engine
most of the bigger engines appear to be kerolox engines

Monoprop could reasonably work as HTP, it doesn't appear to have multiple uses.  Solid is presumably aluminum perchlorate.

This being KSP, I'm going to go with an "essense of snacks [US edition]":  fuel is partially hydrogenated soybean oil, oxidizer is high fructose corn syrup, monoprop is carbon dioxide (think fizzy drinks), and solid is solidified powdered cheese (preferably artificial).  No comments that HFCS won't reduce PHSO (you can reverse them, but I don't think the problem will go away).

[suicidejunkie]

Liquid fuel is elemental explodium.  Eve has seas of it.

Oxidizer is air with a job to do.  Kerbin is covered in it.

[RealKerbal3x]

I think monoprop is probably hydrazine. I don't know, maybe LFO is liquid hydrogen and liquid oxygen. But it could be different for different engines, like what @wumpus said.

[Zeiss Ikon]

My own assumption has been that Lf is kerosene (or equivalent) or just possibly gasoline -- hence same fuel used in air-breathing jets as in rockets -- and oxidizer is HTP (high test peroxide).  These give good density, allowing small/cheap tanks that hold large fuel mass (that operate at low pressure, so can be thin walled).  Either admixture of a small percentage of a peroxide catalyst (colloidal nickel, for example) in the fuel, or presence of catalyst packs in the engines to decompose the peroxide allows for hypergolic starts.  Monoprop is a stoichiometric mix of 50% peroxide and 90% ethanol, fed through a catalyst pack (yes, this is a real world monopropellant that will not detonate, used in Armadillo Aerospace's early lander designs).  Peroxide catalysts, where used, are ceramic beads with embedded platinum/rhodium particles (yes, same stuff that's in a car's catalytic converter).  "Activating" an engine (i.e. staging with throttle down, or using the right-click menu and "activate engine") triggers a negligible trickle of peroxide to keep the catalyst, if present, hot for immediate starts.

The additional advantage here is that all three propellants are long-term storable, within temperature limits.  Kerosene will freeze, and so will peroxide and peroxide-alcohol, at an assortment of "cold to very-cold" temperatures; a little tweaking of emissivity of the tank coatings will keep temperature within range from well inside Eve's orbit to that of Dres (perhaps beyond).  Gasoline freezes at close to liquid nitrogen temperature; your oxidizer will be a solid block before that happens.

Contrary to what was said above, a nuclear-thermal engine will run on anything you can pump through the core.  It's most efficient on hydrogen -- but hydrogen isn't long-term storable without heavy/expensive refrigeration systems (your reactant will have boiled off before your nuke-powered vessel can get to Duna or Eve, never mind Jool or Eloo).  Kerosene gives lower Isp, but if you can run the core hot enough to dissociate ("crack") the stuff, not that much lower.  Waste heat from even a fully scrammed reactor core will be sufficient to keep the kerosene warm on a long voyage.

Now, with these propellants, the game Isp figures for several engines are fantasy numbers -- but this is a game and it needs to be playable more than it needs to be so accurate it's a pain to do anything.  If you like being frustrated by real world fuels, install whatever the current version of Real Fuels mod is called, and deal with ullage as well as needing to put the right fuel in the right tanks.

[OhioBob]

In KSP, liquid fuel and oxidizer have the same density, a mixture ratio of 1.22, and good specific impulse.  Probably the thing that comes closest to matching those properties in real life is liquid oxygen and some derivative of hydrazine, or a hydrazine blend.  Though that is not a combination used in any real life rockets.
 

Edited April 7, 2018 by OhioBob

[OhioBob]

Below is what we have in KSP.  Densities are not known, but we do know that oxidizer and fuel have the same density.  I computed an approximate density by estimating the volume of a Jumbo64 tank.

Propellants	Formula	Density	O/F	Isp,sl
		(g/ml)	Ratio	(s)
Oxidizer	?	≈0.98	1.22	≈280
Liquid Fuel	?	≈0.98

 

And below are the propellants that I think come closest to matching the KSP properties.  The mixture ratios and specific impulses are my own calculations (done years ago).  Specific impulse is "ideal" and based on a combustion chamber pressure of 1000 psi (68 atm).  Actual performance will be less; probably close to the 280s value used in KSP.

One of the two hydrazine blends (Aerozine 50 or UH25) is probably most likely because they come reasonably close to matching both density and mixture ratio, and they have enough thermal stability to be used as an engine coolant.  Hydrazine and MMH are not stable enough, therefore they aren't used in large regenerative cooled rocket engines. 

Propellants	Formula	Density	O/F	Isp,sl
		(g/ml)	Ratio	(s)
Liquid oxygen	LO2	1.141	0.74	303.2
Hydrazine	N2H4	1.004
Propellants	Formula	Density	O/F	Isp,sl
		(g/ml)	Ratio	(s)
Liquid oxygen	LO2	1.141	1.15	299.6
Monomethyl hydrazine (MMH)	CH3NHNH2	0.866
Propellants	Formula	Density	O/F	Isp,sl
		(g/ml)	Ratio	(s)
Liquid oxygen	LO2	1.141	1.38	297.4
Unsymmetrical dimethyl hydrazine (UDMH)	(CH3)2NNH2	0.791
Propellants	Formula	Density	O/F	Isp,sl
		(g/ml)	Ratio	(s)
Liquid oxygen	LO2	1.141	1.06	299.9
Aerozine 50	50% UDMH + 50% hydrazine	0.885
Propellants	Formula	Density	O/F	Isp,sl
		(g/ml)	Ratio	(s)
Liquid oxygen	LO2	1.141	1.22	298.6
UH25	75% UDMH + 25% hydrazine	0.835

Edited April 7, 2018 by OhioBob

[steuben]

two words. dioxygen diflouride... as the oxider and anything else as the fuel.

[Zeiss Ikon]

Just remember, FOOF isn't just the formula -- it's the sound you make if any of it gets loose.

[wumpus]

23 hours ago, Zeiss Ikon said:

Contrary to what was said above, a nuclear-thermal engine will run on anything you can pump through the core.  It's most efficient on hydrogen -- but hydrogen isn't long-term storable without heavy/expensive refrigeration systems (your reactant will have boiled off before your nuke-powered vessel can get to Duna or Eve, never mind Jool or Eloo).  Kerosene gives lower Isp, but if you can run the core hot enough to dissociate ("crack") the stuff, not that much lower.  Waste heat from even a fully scrammed reactor core will be sufficient to keep the kerosene warm on a long voyage.

A nuclear rocket can be designed to run on anything, but getting anywhere near 800s Isp will require hydrogen (or an extremely efficient (and hot enough about to meltdown) use of helium, nothing else comes close).  Using water would net you a similar Isp to hydrolox and likely worse.  Using CO2 would definitely be worse.  The only saving grace is that these are likely better than any other ISRU chemical rocket.

And I stand by my "essense of snacks".  I can't see kerbals developing rocket power if snacks aren't involved some way.

[blakemw]

18 hours ago, wumpus said:

A nuclear rocket can be designed to run on anything, but getting anywhere near 800s Isp will require hydrogen

Exactly, NTR exhaust is actually a bit cooler than that of typical chemical rockets (which is how the reactor doesn't melt) and ALL of the ISP advantage comes from using low molecular weight propellant and not needing a high molecular weight oxidizer to make it hot. Hydrogen is the only good NTR propellant. Helium, Ammonia and Methane can work poorly but still slightly better ISP than LH/Ox rocket. Pretty much anything else is going to be worse than a good chemical rocket because the exhaust is cooler, with the only advantage of a NTR being that it can actually heat up certain kinds of propellants (i.e. plain water or nitrogen), this is a potential ISRU advantage but not an ISP advantage.

There might be some theoretical NTR designs that could get better performance out of plain water than a LH/Ox engine, but the problem always is that using a high molecular weight propellant AND having high ISP means it's going to be reactor-meltingly hot. Any design that CAN get that hot would likely require extremely specialized propellant to not just explode or fizzle.

Edited April 9, 2018 by blakemw

[FlyingPete]

They're whatever is convenient, in whatever measure you like.

So the currency is simply 'funds', research is measured in 'science', material gained through mining is a generic 'ore', and liquid fuel is just 'liquid fuel'. Whatever is is exactly isn't important.

[AeroGav]

Monopropellant

Hydrazine fits the bill, unless anyone can think of a reason to look further

 

Liquid Fuel 

• It's far too dense to be Hydrogen
• It doesn't give nearly enough ISP to be Hydrogen (450 top end chemical, over 1000 NTR) 
• It can be stored indefinitely
• It does not cause soot clogging of NTRs with prolonged use (so not Methane or Kerosene)
• It can be used in jet engines
• Not particularly explosive.   No sustained ground fire when the container is destroyed

My theory is that it's Liquid Ammonia.   This stuff can burn in air in air breathing engines (though it only has half the calorific value of kerosene)  and it can be reacted with storable oxidizer in a chemical rocket.   In an NTR,  some of it dissociates to hydrogen,  giving moderately high ISP.         In a crash, you really don't want be breathing the vapours of this stuff, but then the same goes for Hydrazine.    It probably helps our little green friends wear space suits,  but then again they seem to be made of strong stuff all round.

 

Oxidizer

• Storable

Nitrogen Tetroxide?

[wumpus]

On 4/8/2018 at 10:15 AM, Zeiss Ikon said:

Just remember, FOOF isn't just the formula -- it's the sound you make if any of it gets loose.

FOOF: So powerful the chemical formula alone will crash some Intel processors.

[OhioBob]

On ‎4‎/‎8‎/‎2018 at 2:21 PM, blakemw said:

Exactly, NTR exhaust is actually a bit cooler than that of typical chemical rockets (which is how the reactor doesn't melt) and ALL of the ISP advantage comes from using low molecular weight propellant and not needing a high molecular weight oxidizer to make it hot. Hydrogen is the only good NTR propellant. Helium, Ammonia and Methane can work poorly but still slightly better ISP than LH/Ox rocket.

According to my calculations, the next best NTR fuel after hydrogen is methane or propane.  There's a considerable drop in ISP vs. hydrogen, but it's still better than any chemical propellants.  And the density is so much better than hydrogen that, when we take tank size into consideration, the overall mass of an NTR rocket using methane is about the same as a hydrogen NTR.

After that I think MMH or UDMH is next in line, though they have only marginally better performance than LOX/LH₂ chemical rockets.  These fuels have the advantage of high density and storability.  Pentaborane is another fuel I looked at that has performance just a bit below MMH and UDMH.

I found ammonia to be actually a little worse than LOX/LH₂ chemical rockets.  It ranked very low on this list of fuels I tested.

Helium sounds promising based on its low molecular weight, but I found it to be a pretty poor NTR propellant.  Its ISP is not as good as one might think because of its high specific heat ratio.  And only hydrogen has a lower density.  Also its boiling point is darn near absolute zero.
 

[AeroGav]

5 minutes ago, OhioBob said:

According to my calculations, the next best NTR fuel after hydrogen is methane or propane.  There's a considerable drop in ISP vs. hydrogen, but it's still better than any chemical propellants.  And the density is so much better than hydrogen that, when we take tank size into consideration, the overall mass of an NTR rocket using methane is about the same as a hydrogen NTR.

After that I think MMH or UDMH is next in line, though they have only marginally better performance than LOX/LH₂ chemical rockets.  These fuels have the advantage of high density and storability.  Pentaborane is another fuel I looked at that has performance just a bit below MMH and UDMH.

I found ammonia to be actually a little worse than LOX/LH₂ chemical rockets.  It ranked very low on this list of fuels I tested.

Helium sounds promising based on its low molecular weight, but I found it to be a pretty poor NTR propellant.  Its ISP is not as good as one might think because of its high specific heat ratio.  And only hydrogen has a lower density.  Also its boiling point is darn near absolute zero.
 

But Methane leads to soot deposits within the reactor doesn't it, which precludes us from using the engines the way we do in KSP (over and over with no maintenance, for decades on end, in an IRSU grand tour)

The values given for Ammonia and Hydrazine as NTR fuels vary a lot depending on what source you're quoting and on the reactor temperature (degree of dissociation).

Remember though,  we're not comparing their values with Hydrolox,  an Ammonia/Hydrazine NTR  is competing against storable chemical propellants,  that give between 310 sec (Space Shuttle OMS)  and 375 (SpaceX Raptor, as yet unbuilt,  if you treat mild cryo Methalox as "storables")

[KSK]

On 4/8/2018 at 3:15 PM, Zeiss Ikon said:

Just remember, FOOF isn't just the formula -- it's the sound you make if any of it gets loose.

No - that would be parrrrrrrrpppp, followed by the sound of rapidly retreating footsteps.

On 4/8/2018 at 2:08 AM, steuben said:

two words. dioxygen diflouride... as the oxider and anything else as the fuel.

Engine-rich exhausts for the (temporary) win!

[Greenfire32]

It's obviously liquid Schwartz.

[OhioBob]

34 minutes ago, AeroGav said:

But Methane leads to soot deposits within the reactor doesn't it, which precludes us from using the engines the way we do in KSP (over and over with no maintenance, for decades on end, in an IRSU grand tour)

Most likely, yes.  The possibility of sooting is my greatest concern with any carbon containing compound.  Particularly in an NTR where we're just breaking the compound down into constituent parts and not burning it.  In a chemical engine most of the carbon will combine with oxygen to form CO or CO₂.  In an NTR it will be free carbon.  But the person I quoted mentioned methane, and if we compare it strictly on a performance basis, I think it ranks second to hydrogen.
 

Quote

The values given for Ammonia and Hydrazine as NTR fuels vary a lot depending on what source you're quoting and on the reactor temperature (degree of dissociation).

I'm not quoting any source.  I'm the source of the information I posted.  I based my calculations on a temperature of 2500 K.

 

Quote

Remember though,  we're not comparing their values with Hydrolox,  an Ammonia/Hydrazine NTR  is competing against storable chemical propellants,  that give between 310 sec (Space Shuttle OMS)  and 375 (SpaceX Raptor, as yet unbuilt,  if you treat mild cryo Methalox as "storables")

That's not what the person I quoted was talking about.  He was making comparisons to a hydrogen NTR and LOX/LH₂.
 

Edited April 10, 2018 by OhioBob

[OhioBob]

Something else that hasn't been mentioned in the discussion of NTRs is how much delta-v is needed.  Because of the cost and complexity of an NTR, they are just not worth it unless we need to deliver a large amount of delta-v.  For instance, @AeroGav compared an ammonia NTR to storable chemical propellants (e.g. nitrogen tetroxide and MMH).  If we need 10 km/s Δv, then I think there is no question the ammonia NTR will outperform the chemical rocket.  But if we need only 3 km/s Δv, then they are about equal in terms of the total overall mass; and the chemical rocket is undoubtedly much less expensive.

[AeroGav]

3 hours ago, OhioBob said:

Something else that hasn't been mentioned in the discussion of NTRs is how much delta-v is needed.  Because of the cost and complexity of an NTR, they are just not worth it unless we need to deliver a large amount of delta-v.  For instance, @AeroGav compared an ammonia NTR to storable chemical propellants (e.g. nitrogen tetroxide and MMH).  If we need 10 km/s Δv, then I think there is no question the ammonia NTR will outperform the chemical rocket.  But if we need only 3 km/s Δv, then they are about equal in terms of the total overall mass; and the chemical rocket is undoubtedly much less expensive.

Most of the cost issue is down to people using NTR for deep space probes.  Obviously if you're not recovering the part,  the price hurts.    That said,  since the NTR will be on the upper stage, it really should be possible to recover the entire upper stage at the end of the mission , though the re-entry aerodynamics and parachute landing attitude will need careful design.

I only really use them on space planes and shuttles, so the recovery of the engine is a given (and necessary delta v would be hard to achieve otherwise).  Though some of these nuclear spaceplanes aren't SSTOs, discarding jet engines or riding up to 20km on SRBs

[OhioBob]

3 minutes ago, AeroGav said:

Most of the cost issue is down to people using NTR for deep space probes.  Obviously if you're not recovering the part,  the price hurts.    That said,  since the NTR will be on the upper stage, it really should be possible to recover the entire upper stage at the end of the mission , though the re-entry aerodynamics and parachute landing attitude will need careful design.

I only really use them on space planes and shuttles, so the recovery of the engine is a given (and necessary delta v would be hard to achieve otherwise).  Though some of these nuclear spaceplanes aren't SSTOs, discarding jet engines or riding up to 20km on SRBs

Just to be clear, I was commenting about the hypothetical use of NTRs in real life, not in KSP.  Since we were talking about the pros and cons of different fuels, I assumed we were talking about real life.  Much of our conversation has no application in KSP.

That being said, there's probably a Δv threshold in KSP as well, below which NTRs are not economical, and above which they are.  However, I've never made any attempt to figure out what it is.

[OhioBob]

21 hours ago, AeroGav said:

Liquid Fuel 

My theory is that it's Liquid Ammonia.  

Oxidizer

Nitrogen Tetroxide?

While that fits with your criteria, it fails to meet some of the other physical properties known to exist in KSP.  Liquid ammonia and nitrogen tetroxide have densities of 0.68 and 1.45 kg/l respectively.  And I estimate an ideal mixture ratio of about 1.9.  None of that is very close to the KSP values.

Let's face it, there's really no real life equivalent to liquid fuel and oxidizer in KSP.  We might get close to matching one group of properties, but then something else won't match.

And I really don't think there is any reason we must assume that "liquid fuel" is always the same stuff.  "Liquid fuel tanks" and "fuel tanks" could contain different fuels.

 

[Zeiss Ikon]

Back a few exchanges, someone pointed out that hydrazine derivatives (themselves long-term storable monopropellants and liquid fuels) give similar performance in an NTR to hydrolox in a chemical rocket.  That points to an answer: Lf is hydrazine-ish, likely MMH or UDMH, usable in both NTRs like the NERV (giving higher Isp than hypergolic chemical rockets) and as the fuel in chemical bipropellant engines (even in jets, though standing behind one would be a Bad Idea if you're human).  The oxidizer could be MON (mixed oxides of nitrogen) or red fuming nitric acid, either of which is hypergolic with most hydrazine derivatives.  Dense, long-term storable, and they'd explain why everything in the Kerbal world explodes when overstressed.  In this scheme, monopropellant would likely be straight hydrazine, catalyzed in either a pellet bed or with platinum/rhodium plated stainless or tungsten screen packs (dense elements like these ought to be common and cheap on Kerbin, given its extreme density).