nuclear rockets powered by hydrogen - What am I missing?

[Sresk]

So I've read the discussion on the merits and weaknesses of the stock nerv and I can see where it has it's place. However I've begun playing with alot of mods in particular interstellar extended, kerbal atomics and the cryogenic engines. These mods introduce hydrogen fuel which is way less dense and supposedly has a much higher isp on nuclear rockets. However I've run tons of test on various combinations in the VAB and no matter what I do I can not find a way to get any engine running on hydrogen to perform as well as any engine running on LFO.

Sure one orange tank of hydrogen is way lighter than a tank of LFO but offers no where near as much delta v. and if I throw in enough tanks to match the same mass as the LFO then the delta v is still worse and the TWR is garbage. What am I missing? Do I need some other mods to balance out the hydrogen fuel. Are the mods I'm playing with just poorly balanced? Or am I missing something else?

[Streetwind]

I'm not up to speed on how Kerbal Atomics handles things these days... and you run Interstellar, which has a history of patching other mods to its own subjective standards, so I don't know what values you are actually dealing with here. But I doubt that you would get as strongly worse performance as your wording seems to suggest.

Given a known value for Isp, your final dV depends only on one variable: your mass fraction, AKA the ratio of wet mass to dry mass. If you provide 10 tons of propellant A and 10 tons of propellant B, with the only difference between the two setups being that the two propellants come in different tanks, then the only way you can still get different dV numbers is if the tanks have different mass fractions. If the tankage of propellant A weighs 100 kg for every ton of propellant load, and the tankage of propellant B weighs 500 kg for every ton of propellant load, then the spacecraft using propellant B will be 4 tons heavier. This extra dry mass will reduce both the available dV and the effective engine TWR.

The reason I'm doubting that your results could be as bad as you describe is because you stated "any engine running on LFO" is better than any nuclear thermal rocket. You suggest that an engine with three times the Isp has worse dV output, just because of tank dry mass differences? That would have to be a huge difference. Are you sure you're using the correct tank configuration for the NTRs? As in, hydrogen only, not hydrogen + oxidizer? While one of the Kerbal Atomics engines is a LANTR that can optionally use oxidizer, the standard operation mode for all the NTRs is hydrogen-only. Like how in stock it's LF-only, and people regularly post in Gameplay Questions about how they only get terrible Isp and TWR, because they use LFO tanks instead of LF tanks. Your problem at first glance looks exactly like that.

[Sresk]

Thanks for the response, it'll be a couple days before I can put up some solid numbers as I'm traveling but working from memory what I was doing was using the terrier engine and the poodle engine as my LFO's to compare to. And I was looking at several of the nuclear engines including the nerv, LANTRN and lightbulb. I was careful to check that the tanks where set to just h2 and not LHO. and I was building dummy rockets with just a mk1 capsule and a whole buttload of fuel. Basically I was following along with this you tube video. The main difference being that my nuclear engines run on H2 and I would need about 4 times the volume of tank to reach the same mass of H2 to LFO.

When I deleted the patches that changed the fuel in those rockets to hydrogen and left them running on just LF I got results very much like the video above (even better because I could use tanks without oxidizer). But when I left the patches in and those parts run on hydrogen then the DeltaV's where garbage. This was surprising as the light bulb is supposed to have an ISP of 1500+. So I'm sure I'm getting some bad interaction between mods but I can't figure out whats breaking what. 

[Sresk]

Ok I've done some tests with a clean install on a different computer and I haven't completely pinpointed the problem but it looks like interstellar does not play well with any of the other mods that introduce nuclear engines of their own. Also it looks like interstellar has its lower tech level nuclear engines that run on hydrogen under tuned in comparison to the nerv when running on liquid fuel. So really what I need to do is move this to a balance discussion somewhere where the intersteller mod devs can see this and talk to them  

[wasml]

Are you using Community Resource Pack? If so have you looked at the density for H2 - when I looked at it some time ago it seemed way low. Maybe you don't have as much H2 as you think you do?

[Sresk]

Oh I sure am how do I fix it? Is there a value I can change?

[Mikki]

Sidenote: Nerteas Atomic rocket mod works fine without KSPIE, the ships become rather large but the TWR ( and isp) of the both big engines in this mod is way better than anything stock. Its just the the huge tanksize that is missleading to assume they are not effective, i had pretty much the same impression as you by first installing Atomic rockets.

I recently made a "medium size"  modular interplanetary transfer tug with atomic rockets and to reuse it later for larger travels i simply attach more tanks with dockingports. The tanks need lots of electricity to prevent boiloff, put a small nuclear powergenerator from Nertea NFE into. His mods work fine, never had any issues really.  

Edited August 24, 2016 by Mikki

[Streetwind]

6 hours ago, wasml said:

Are you using Community Resource Pack? If so have you looked at the density for H2 - when I looked at it some time ago it seemed way low. Maybe you don't have as much H2 as you think you do?

Pretty sure CRP is using the RL density value for liquid hydrogen there. At least, that was Nertea's intention. If the number is wrong, I'm pretty sure that showing him proof will quickly get CRP updated with the proper value.

[Sresk]

Yeah I get that h2 is way less dense so you need way more volume of tanks to get the same mass of h2 as compared to the mass of liquid fuel. But it feels like it's burning through that mass about 5 to 10 times faster than it should without giving 5-10 times the delta-v and thrust.

I'm convinced I've got some mods interfering with each other I'm just not familiar enough with ksp yet to know which values to adjust.

If I increase the mass of h2 as defined by the crm mod will that pack more h2 into each tank increasing fuel density and lsp or will I simply increase the weight of the fuel without improving performance.

Also @mikki about nerteas atomic rocket mods your right I installed that along side interstellar without anything else and your right those two did play well together. However if I installed mods which patched nerteas engines forcing them to burn h2 then they experienced a severe performance drop as well. 

So all of this leads me to believe that one of my various mods is breaking the values of h2.

 

Hmmm Since I already have so many mods installed I didn't really feel the need for anymore complexity but would something like realfuels or smurf make things better or worse?

Edited August 24, 2016 by Sresk

[wasml]

@Streetwind Going back and looking at things I think what derailed me was that (unless I'm making a really stupid math mistake) water is in Kg/L. I had assumed hydrogen was too - but looking at the other gases they all seem to be in Kilo tons per Liter - except for intakeAtm which is off by its self.

[Sresk]

@wasml so does that mean you know which mod is causing my problem and if so can you tell me which values to change or help me patch it?

[wasml]

My problem was I had the units wrong so when I filled a tank based on volume I actually only got 1/1000 as much as I wanted (Kg/L vs Mg/L). If you're filling your tanks based on volume and using something other than Kilo tons per Liter this would cause problems - other wise it's something else - sorry.

[Sresk]

Ok So I've only been at this for less than a month and while I'm comfortable diving into the .cfg files I'm still not up to speed with what all of the values equate to. (and I realize about half way through this post that I'm starting to use this as my own personal scratch pad jump to the bottom if you don't want to see the cfg files I'm pasting in :P)

So from the community resource pack here is how hydrogen is defined Which I assume is the minable refinable resource

RESOURCE_DEFINITION
{
    name = Hydrogen
    density = 0.00000008990
    unitCost = 0.000055836
    flowMode = ALL_VESSEL
    transfer = PUMP
    isTweakable = true
       isVisible = true
    volume = 1
}

And LyqHydrogen is defined as :

RESOURCE_DEFINITION
{
    name = LqdHydrogen // General propellant, used for high thrust electric engines
    density = 0.00007085000
    unitCost = 0.0367500
    hsp = 9690 // specific heat capacity (kJ/tonne-K as units) at Crygenic Storage temperature
    vsp = 448500 // heat of vapourization (KJ/tonne as units)  or 8.97 * 10^5 or 8.97E5?
    flowMode = STACK_PRIORITY_SEARCH
    transfer = PUMP
    isTweakable = true
    isVisible = true
    volume = 1
}

CryoTanks has a fuel switcher patch and I think this is the relevent part

%LH2 = #$RESOURCE[LqdHydrogen]/maxAmount$
    
    %massOffset = #$LH2$
    @massOffset *= 0.00003125 // <- EDIT HERE (dry mass per unit LH2 capacity)
    @massOffset *= -1
    
    %costOffset = #$LH2$
    @costOffset *= #$@RESOURCE_DEFINITION[LqdHydrogen]/unitCost$
    @costOffset *= -1
    
    !RESOURCE[LqdHydrogen] {}

And this 

{
        TANK_TYPE_OPTION
        {
            name = LqdHydrogen
            dryDensity = 0.1
            costMultiplier = 0.035
            // Based on observation that LH2 is 1/20 the cost of LF
            RESOURCE 
            {
                name = LqdHydrogen
                unitsPerT = 16000
                // Based on nertea's 10x total of mixed tank
            }

Interseller has a fuel switch patch which looks like this:

{
        name = InterstellarFuelSwitch
        tankSwitchNames = LFO;LiquidFuel;Oxidizer;MonoPropel;XenonGas;Hydrogen
        resourceGui = LiquidFuel+Oxidizer;LiquidFuel;Oxidizer;MonoPropellant;XenonGas;LqdHydrogen
        resourceNames = LiquidFuel,Oxidizer;LiquidFuel;Oxidizer;MonoPropellant;XenonGas;LqdHydrogen
        resourceAmounts = #$../LF$,$../OX$;$../totalCap$;$../totalCap$;$../totalCap$;$../totalCap$;$../onlyLH2$
        tankTechReq = start;start;start;advFuelSystems;ionPropulsion;nuclearPropulsion
        tankResourceMassDivider = 8;8;8;6.66666666666;1.272727272727;5.5
        orderBySwitchName = true
        hasGUI = true
    }

    MODULE:NEEDS[WarpPlugin]
    {
        name = InterstellarFuelSwitch
        tankSwitchNames = LFO;LiquidFuel;Oxidizer;MonoPropel;XenonGas;Hydrogen;Hydrolox
        resourceGui = LiquidFuel+Oxidizer;LiquidFuel;Oxidizer;MonoPropellant;XenonGas;LqdHydrogen;Hydrolox
        resourceNames = LiquidFuel,Oxidizer;LiquidFuel;Oxidizer;MonoPropellant;XenonGas;LqdHydrogen;LqdHydrogen,LqdOxygen
        resourceAmounts = #$../LF$,$../OX$;$../totalCap$;$../totalCap$;$../totalCap$;$../totalCap$;$../onlyLH2$;$../LANTRmixLH2$,$../LANTRmixOX$
        tankTechReq = start;start;start;advFuelSystems;ionPropulsion;nuclearPropulsion;improvedNuclearPropulsion
        tankResourceMassDivider = 8;8;8;6.66666666666;1.272727272727;5.5;8
        orderBySwitchName = true
        hasGUI = true

Soooo when I install kerbal atomics is brings along the Cryo fuel switcher... is that interfering with the interstellar fuel switcher? And is there some other patch that I'm missing that changes the values of LqyHydrogen?

[wasml]

Just had a thought - you can see what the game actually ended up with through the F12 menu - not sure which tab it's under. Might poke around in there and see if you can find the final definition for LqyHydrogen too - not sure if resources are in there or not.

[CaptainArchmage]

We just have Liquid Fuel and Oxidiser in KSP. The density combination is more like Kerbal equivalent of N2O4 and Aerozine for the record, but lore tends to go for Liquid Oxygen and Kerosene.

 

The following statement is 100% valid only if actual volume calculations are carried out by the mods, given the tank volume. It appears this is the case.

 

Hydrogen and then nuclear rockets are essential IRL due to the square cube rule combined with the efficiency. Starting with liquid hydrogen and oxygen, the square cube rule says, essentially if you go for a small system hydrocarbon fuel is cheaper than hydrogen to work with, and that's mostly to do with tank costs (the structure is more expensive than fuel). Going to a modern heavy-lifter the sizing gets to the point you're either spending on mass to support the dense kerosene, or on insulation for otherwise less dense hydrogen, and not only is the insulation lighter, but the volume-to-surface-area is higher due to the square cube rule. Additionally, the insulation needs will drop as the area through which heat gets, per unit volume in is smaller (so the tank stays cold for longer). This means that the cost of the tank for the hydrocarbon fuel, including insulation, will eventually outstrip the cost for the liquid hydrogen tank and the cost of hydrogen doesn't make a difference because the tank is about a hundred times the price of the fuel it holds, unless you're sitting just on the margin. The emphasis is big tanks. Think SLS and Energia scale, or bigger. This is particularly about minimising surface area and support mass.

 

Probably you should run a graph for this. I think there might be an issue of balancing.

 

The nuclear engine will provide a lot more specific impulse, but the TWR is poor, including IRL. These are used in orbit. To be honest the ISP should be something like 800-1600s for a hydrogen nuclear rocket.

[Sresk]

Still struggling with this. Yes the ISP of the engines is much higher but the deltav is much lower when the fuel is hydrogen. When the fuel is lqdfuel then the engine performs as expected but if I change the fuel to hydrogen and then adjust the mass to be the same mass then the deltav falls through the floor.

[Sresk]

OK I think I actually finally ran this to ground. SMURF had broken things. I had to uninstall SMURF through CKAN reinstall KSP through steam and reinstall the other mods I really wanted. because SMURF nerf's hydrogen fuel it was seriously breaking anything that used hydrogen fuel with the combo of mods I was useing. So just FYI for anyone useing interestellar extended and other nuclear engines be real careful with SMURF.