[1.12.x] Cryogenic Engines: Liquid Hydrogen and Methane Rockets! (August 13, 2024)

[rasta013]

It looks like fuel switching for LF/O/MP tanks was left out of the download. I don't see it in-game or in the config files.

Edit:

Also, in the config for LF/OX tanks, the line "tankMass = #$../mass$;#$../mixLH2mass$;#$../mass$;#$../mass$;#$../onlyLH2mass$" does not appear to be working as intended. In game, as you cycle through the fuel options, the "dry mass" remains constant at the part's stock mass. Contrast this with the Modular Octo-Girder, whose dry mass changes as intended. Unfortunately, I have not yet been able to identify the problem.

I'm seeing the same behavior with the DryMass being identical but I'm definitely not able to tease out the problem spot...

[Fraz86]

I'm seeing the same behavior with the DryMass being identical but I'm definitely not able to tease out the problem spot...

After further testing, I think it is most likely a bug within IFS. It appears that if you use a list of variables to define tankMass, it just copies forward the value of the first variable in the list and ignores the rest. Using fixed numbers (instead of variables) works fine, which is why the Octo-Girders function as intended, but that obviously won't help for the LF/OX config.

[Nertea]

I fixed both, but won't have time to upload a fix until tomorrow. Sorry about the two misses.

[funk]

I've recognized that the drymass changes when gambling with the tweakables of the resources.

THX Nertea for this patch.

I've made two small changes in CryogenicFuelTankSwitcher.cfg, so that MM checks for dependencies correctly, please could you change:

@PART....

,NEEDS[!modularFuelTanks],NEEDS[!RealFuels]

to

:NEEDS[!modularFuelTanks]:NEEDS[!RealFuels]

I've made a new, adjusted MFT patch to fit ISFuelswitch values as much as possible. Costs behave odd in MFT.

// by funk 06/04/2015 released under ... Public License
//------------------------------------------------------//

// Adds LqdHydrogen to default tank definition
@TANK_DEFINITION[Default]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
mass = 0.0000625
utilization = 10
}
}

// Adds LqdHydrogen to fuselage tank definition
@TANK_DEFINITION[Fuselage]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
utilization = 10
}
}

// Adds LqdHydrogen to fuselage tank definition
@TANK_DEFINITION[Structural]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
mass = 0.000375
utilization = 10
}
}

// Adds LqdHydrogen to ServiceModule tank definition
@TANK_DEFINITION[ServiceModule]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
mass = 0.0000625
utilization = 10
}
}

// Adds tank definition for CryoEngines
TANK_DEFINITION
{
name = Cryogenic
basemass = 0.000625 * volume
baseCost = 0.815 * volume
TANK
{
name = LqdHydrogen
mass = 0.0000625
amount = full
maxAmount = 50%
utilization = 10
}
TANK
{
name = Oxidizer
amount = full
maxAmount = 50%
}
}
//------------------------------------------------------//

// Adds tank type Cryogenic to all default tanks
@PART
[*]:HAS[@MODULE[ModuleFuelTanks]]:FINAL
{
@MODULE[ModuleFuelTanks]:HAS[#type[Default]]
{
%typeAvailable = Default
typeAvailable = Cryogenic
}
}

// CryoEngine MFT (Modular Fuel Tanks) Configuration File

Download here

I'm looking forward to the upcoming NFT releases. Perhaps I can provide a MFT patch, too.

regarding Boil-Off:

I've thought of boil-off, too. Perhaps anyone can help with the values we should use for temperature and loss-rate in stock KSP.

RealFuels assumes a temperature of 20.15°K at which LqdHydrogen needs to be cooled without boil-off. I think we can deal with it.

For loss-rate RF assumes 0.0432% for an insulated tank and 3.456% without insulation. Values are per earth day, per liter tank volume and per degree deviation.

If we assume the same loss-rates should apply per Kerbin day, then the loss-rates per second have to be four times higher:

loss_rate = 0.0000016 (non-insulated)

loss_rate = 0.00000002 (insulated)

According to NASA and at first Wikipedia and some other sources, this could lead to a small part mod which offers some active cooling features. Perhaps there are any in nuNFT?!?

Edited June 4, 2015 by funk

[mikegarrison]

I've recognized that the drymass changes when gambling with the tweakables of the resources.

THX Nertea for this patch.

I've made two small changes in CryogenicFuelTankSwitcher.cfg, so that MM checks for dependencies correctly, please could you change:

@PART....

,NEEDS[!modularFuelTanks],NEEDS[!RealFuels]

to

:NEEDS[!modularFuelTanks]:NEEDS[!RealFuels]

I've made a new, adjusted MFT patch to fit ISFuelswitch values as much as possible. Costs behave odd in MFT.

// by funk 06/04/2015 released under ... Public License
//------------------------------------------------------//

// Adds LqdHydrogen to default tank definition
@TANK_DEFINITION[Default]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
mass = 0.0000625
utilization = 10
}
}

// Adds LqdHydrogen to fuselage tank definition
@TANK_DEFINITION[Fuselage]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
utilization = 10
}
}

// Adds LqdHydrogen to fuselage tank definition
@TANK_DEFINITION[Structural]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
mass = 0.000375
utilization = 10
}
}

// Adds LqdHydrogen to ServiceModule tank definition
@TANK_DEFINITION[ServiceModule]
{
TANK
{
name = LqdHydrogen
amount = 0
maxAmount = 0
mass = 0.0000625
utilization = 10
}
}

// Adds tank definition for CryoEngines
TANK_DEFINITION
{
name = Cryogenic
basemass = 0.000625 * volume
baseCost = 0.815 * volume
TANK
{
name = LqdHydrogen
mass = 0.0000625
amount = full
maxAmount = 50%
utilization = 10
}
TANK
{
name = Oxidizer
amount = full
maxAmount = 50%
}
}
//------------------------------------------------------//

// Adds tank type Cryogenic to all default tanks
@PART
[*]:HAS[@MODULE[ModuleFuelTanks]]:FINAL
{
@MODULE[ModuleFuelTanks]:HAS[#type[Default]]
{
%typeAvailable = Default
typeAvailable = Cryogenic
}
}

// CryoEngine MFT (Modular Fuel Tanks) Configuration File

Download here

I'm looking forward to the upcoming NFT releases. Perhaps I can provide a MFT patch, too.

regarding Boil-Off:

I've thought of boil-off, too. Perhaps anyone can help with the values we should use for temperature and loss-rate in stock KSP.

RealFuels assumes a temperature of 20.15°K at which LqdHydrogen needs to be cooled without boil-off. I think we can deal with it.

For loss-rate RF assumes 0.0432% for an insulated tank and 3.456% without insulation. Values are per earth day, per liter tank volume and per degree deviation.

If we assume the same loss-rates should apply per Kerbin day, then the loss-rates per second have to be four times higher:

loss_rate = 0.0000016 (non-insulated)

loss_rate = 0.00000002 (insulated)

According to NASA and at first Wikipedia and some other sources, this could lead to a small part mod which offers some active cooling features. Perhaps there are any in nuNFT?!?

First of all, I'm not terribly keen on simulating boil-off, so I hope this is optional.

Secondly, if you are trying to make this realistic, there is no reason to multiply the rate of boil-off by 4. There is no reason why, just because Kerbin's day is shorter, hydrogen would cook off faster than it does here.

[iVG]

First of all, I'm not terribly keen on simulating boil-off, so I hope this is optional.

Secondly, if you are trying to make this realistic, there is no reason to multiply the rate of boil-off by 4. There is no reason why, just because Kerbin's day is shorter, hydrogen would cook off faster than it does here.

It won't boil faster, just readings would be synced with Kerbin's 6 hour day rather than 24h.

[Temeter]

It won't boil faster, just readings would be synced with Kerbin's 6 hour day rather than 24h.

That doesn't mean a hour on kerbin has more happening than an hour on earth, tho.

[Fraz86]

regarding Boil-Off:

I've thought of boil-off, too. Perhaps anyone can help with the values we should use for temperature and loss-rate in stock KSP.

RealFuels assumes a temperature of 20.15°K at which LqdHydrogen needs to be cooled without boil-off. I think we can deal with it.

For loss-rate RF assumes 0.0432% for an insulated tank and 3.456% without insulation. Values are per earth day, per liter tank volume and per degree deviation.

If we assume the same loss-rates should apply per Kerbin day, then the loss-rates per second have to be four times higher:

loss_rate = 0.0000016 (non-insulated)

loss_rate = 0.00000002 (insulated)

According to NASA and at first Wikipedia and some other sources, this could lead to a small part mod which offers some active cooling features. Perhaps there are any in nuNFT?!?

I wouldn't be fundamentally opposed to boil-off for cryogenic tanks, however, it would need to be implemented in a manner that prioritizes fun & transparent gameplay mechanics. If people want maximum realism, they can use mods like RealFuels and RO. For "stockalike" gameplay, I don't think we should concern ourselves too much with realistic numbers for boil-off temperature and rate.

Nertea's new HeatManagement mod (and the new version of NFE) will include radiators and heat pipes with active cooling capabilities. However, this active cooling has a floor of 273K. Therefore, the boil-off threshold would need to be above 273K. Something like 400K would probably be reasonable. Thematically, we could assume that cryogenic tanks have built-in internal refrigeration that prevents boil-off, as long as the temperature of the tank itself remains < 400K. It would also be important to include a means of alerting the player when boil-off is occurring; e.g., RCS effects coming off of the tank would be ideal.

Edited June 4, 2015 by Fraz86

[woot]

I tend to agree that while boil-off would be a nice addition, the rate of boil-off would need to be pretty slow. These cryo engines are a lot of fun, but I don't think their performance is spectacular enough to compensate for boil-off that would significantly impact near-Kerbin operations. Matching the real rate per hour seems like a decent starting place. 4x seems like it'd be very prohibitive. I don't know though. It would be fun for insulation and active cooling to have significant advantages too.

I do worry that adding these types of considerations to cryogenic engines will make them less attractive overall

[mikegarrison]

Yeah, I was thinking about active cooling too. Maybe you would have boil-off unless you pay the price for active cooling, which would be probably some extra weight, loss of volume in the tank, and electrical draw.

Personally I don't think the added complexity is worth it, so I hope it would remain optional.

[funk]

These numbers were just a first thought. Of course boil-off should have a stockalike gameplay feeling and should be optional for those who like to have an additional callenge. Maybe I'm wrong but is CryoEngines the only stockmod except NFP which uses cryogenic fuels? Eventually we can use a MM-Patch to buff the engines in case of boil-off.

Referring to the Nasa source from my first post active cooling is needed when the trip is longer than 30 days. For gameplay reasons in KSP the lossrate could be balanced in a way that active cooling is needed for trips outside Kerbins SOI only, perhaps already for Minmus trips?!? I haven't played with heat control, yet, but if the zero point temperature is 273K there might be some problems with stock heating - perhaps it is turned off? Will test it now...

Edited June 4, 2015 by funk

[FreeThinker]

If we assume the same loss-rates should apply per Kerbin day, then the loss-rates per second have to be four times higher:

loss_rate = 0.0000016 (non-insulated)

loss_rate = 0.00000002 (insulated)

At what quantity is this? Liter, Kg or Mt? I'm trying to establish what are realistic boil off for IFS

[funk]

At what quantity is this? Liter, Kg or Mt? I'm trying to establish what are realistic boil off for IFS

Because I'm using MFT this is from its description:

"loss_rate = controls how fast the resource is lost. The boiloff = loss_rate * maxAmount * (current_temp - desired_temp) * time elapsed."

Time in seconds. Temp in Kelvin. The loss_rates I've mentioned are taken from RF multiplied with four.

This means the numbers are per unit/liter tank volume, per degree deviation and per second.

Edited June 4, 2015 by funk

[Fraz86]

I haven't played with heat control, yet, but if the zero point temperature is 273K there might be some problems with stock heating - perhaps it is turned off? Will test it now...

It's not the stock heat system, it's the intentional design of Nertea's new active cooling functionality. Basically, radiators and heat pipes will now transfer X watts of heat from their parent part - into themselves for radiators, or into their target part for heat pipes. During early testing, this behavior resulted in the parent part getting cooled to ridiculously low temperatures (~4K) while the radiators would get very hot, with various undesirable gameplay consequences. Therefore, Nertea introduced a cooling floor, in which the heat transfer rate tapers down to zero as the parent part's temperature declines toward 273K. I assume 273K was chosen largely because this is the initial temperature on the launchpad, and so having zero heat transfer at this temperature allows a spacecraft to start out in equilibrium, avoiding unpredictable/volatile behavior.

I think the current active cooling behavior, with a lower limit of 273K, is desirable and well conceived. If boil-off is implemented, I believe the boil-off temperature should be set at a point that integrates well with the active cooling as it is currently implemented (regardless of realism), rather than trying to change active cooling to allow lower temperatures.

[Nertea]

Quick update, 0.1.7, fixes these issues:

• Fixed LF/O/MP tank patch functionality
  
• New MFT patch from funk
  
• Fixed dry mass in tank patch not adjusting
  

[rasta013]

Bringing this discussion of NTRs vs Cryo engine efficiency over from the NFT thread...

I tested with all sizes of fuel tanks, including the largest ones from NFP, and even added weight to cryo engine setups to mitigate as much as possible the effect of dry mass, and cryo engines are still much more efficient.

I thought it was that at first and repeated the tests very carefully. I'm always testing cryo engines with LH2/OX, and NTRs with hydrogen only. I usually use 5 tanks of hydrogen in nuclear-powered setups to get about the same fuel mass.

No matter what I just can't find any setup where nuclear rockets are more efficient than cryogenic engines. Test it for yourself if you don't believe me.

LH2 Only : Rockomax Jumbo 64 w/ + LV-N Nerv + 10t NRAP = 19891 kg....this comes to a total dV of 1704

LH2/Ox : Rockomax X200-16 + Rockomax X200-8 + VL1-Volcano + 10t NRAP = 20000 kg...this comes to a total dV of 1686

This is vacuum numbers of course but I'm not seeing much difference. If I could add just a bit more fuel to the NTR I'd get even better dV. I've checked a number of different configurations including using some of the Atomic Age engines, some of Ven's Cryo tanks, some stock tanks, some of the Klockheed Martin cryo tanks and I keep coming up with the same thing. I'm finding all kinds of configurations where the NTRs are more efficient. Now that said, there are a number of situations where they are very close and I listed the above example as one since that's about as close to equal efficiency that I can get but it also would depend on situations as to what I was trying to accomplish.

Some basic testing I'm doing with the NFP tanks, here's one result where the Cryo engines perform better with similar weights:

HI-M-10 + 2x HI-M-5 + LV-N + 10t NRAP (adjusted fuel to give equal weight) = 16473 kg gives dV of 1017 m/s

HI-M-5 + VL-1 Volcano + 10t NRAP = 16471 kg gives dV of 1210 m/s

With this setup the Cryo engine is more efficient but the NERV would not really be my choice in this situation anyway. Turning to the Atomic Age LANTERN engine, I actually end up with full tanks, a weight of 16357 kg and dV of 1379 m/s - far more efficient. Regardless, I'm finding that there are so many variations of configuration between tanks and engines that I can find all kinds of performance ranges provided I maintain equal weight. LH2 is so much lighter that it requires more tanks but once I get the weights the same, the performance difference becomes very obvious. Obviously depending on the situation I'd be using differing setups but I'm definitely not finding that the cryo engines are ALWAYS more efficient, nor am I finding that the tanks from NFP are heavily unbalanced in any way. Actually, quite the opposite. The NFP tanks for me tend to be on the lighter side, yes, but I pay for that in dV and rocket size, specifically, length or width one of the two.

I'm just not seeing any problem here. I'll keep checking numbers and testing of course but in my opinion my first few passes have not really shown any unbalance to me.

[toric5]

would ti be possible to add some smaller transfer engines? possibly convert the ones from NFSP?

[Vitasalato]

Hi, i love this mod and the idea of high ISP engines, but i think that it might be a problem.

I've measured with mechjeb the DeltaV of some stock engine and one Cryo, and the result is not what i was looking for. Both stock engine have a higher DeltaV with lower ISP. This is true using the same fuel tank for all the engine, and the same MASS of fuel. Take a look at the album to see some numbers.

Where is the problem?

http://imgur.com/a/Onzp8

[Crzyrndm]

Where is the problem?

Poodle/Tunguska - Engine mass: The extra 1.5t dry weight is significant in a vessel of 10-15t total mass. Try it pushing a decent payload so the 1.5t extra isn't 10% of the vessel weight

Skipper/Tunguska - Not enough fuel. For the same tank volume you get less hydrogen by mass because of it's low density (and therefore less dV). Add another tank and it will surpass it

Cryo engines need 50-100% more tankage than their LFO equivalents in my experience. They still end up notably lighter though.

Edited June 6, 2015 by Crzyrndm

[mikegarrison]

A cryo engine setup is not always going to be better than an LFO setup. For instance, with very light unmanned probes, the small LFO engines are the way to go (until you get access to electric engines). Play around and find which setup works best for each application.

[Vitasalato]

@Crzyrndm @mikegarrison

Really guys I've tried to set up a launch of 66t (same fuel tanks), and i can't find any PRO for the Cryo. Ok they have a slight better TWR ('cause is 100t lighter), but they have worst D-v (not so much) even if they have a lot better ISP, and the there is nothing that encourage so much to ignore the high cost.

I didn't do the math, and i don't have the time to do that, so i'm not sure if there is a real problem of bad equivalent fuel amount (between LF-OX and LH-OX with the same tank) but i feel that is something wrong.

I mean how its possible that with the same TANK, the one with LF-OX (heavier) and low ISP (in compare with Cryo) that have better D-v then the one with LH-OX and better ISP engine?

In my opinion that is something that as to be fixed, but if some of you have done the math and proof that what I'm saying is wrong, I'll be very happy.

[rasta013]

@Crzyrndm @mikegarrison

Really guys I've tried to set up a launch of 66t (same fuel tanks), and i can't find any PRO for the Cryo. Ok they have a slight better TWR ('cause is 100t lighter), but they have worst D-v (not so much) even if they have a lot better ISP, and the there is nothing that encourage so much to ignore the high cost.

I didn't do the math, and i don't have the time to do that, so i'm not sure if there is a real problem of bad equivalent fuel amount (between LF-OX and LH-OX with the same tank) but i feel that is something wrong.

I mean how its possible that with the same TANK, the one with LF-OX (heavier) and low ISP (in compare with Cryo) that have better D-v then the one with LH-OX and better ISP engine?

One big thing to remember, cryo engines consume LH2/Ox faster than LFO engines consume LOX and therefore require more tanks than a standard LFO engine does. If you use a Jumbo-64 w/ LFO and then just switch it to LH2/Ox and a cryo engine, you will always have worse dV. However, if you match the fuel mass between the two (which means more LH2/Ox tanks) you will see the improvement in performance and efficient. You'll also find that you will still end up conserving weight despite using more tanks because of the mass of LH2/Ox. Match the fuel mass and don't just switch the fuel in the tanks alone otherwise you aren't getting a real comparison of the two.

[mikegarrison]

@Crzyrndm @mikegarrison

Really guys I've tried to set up a launch of 66t (same fuel tanks), and i can't find any PRO for the Cryo. Ok they have a slight better TWR ('cause is 100t lighter), but they have worst D-v (not so much) even if they have a lot better ISP, and the there is nothing that encourage so much to ignore the high cost.

I didn't do the math, and i don't have the time to do that, so i'm not sure if there is a real problem of bad equivalent fuel amount (between LF-OX and LH-OX with the same tank) but i feel that is something wrong.

I mean how its possible that with the same TANK, the one with LF-OX (heavier) and low ISP (in compare with Cryo) that have better D-v then the one with LH-OX and better ISP engine?

In my opinion that is something that as to be fixed, but if some of you have done the math and proof that what I'm saying is wrong, I'll be very happy.

One of the big pros IRL tends to be that the overall weight of an upper H2/O2 stage is less than it would be for a Kero/O2 stage. So that means you can have a smaller first stage to lift it all. I think that is also true in this mod when you use useful payloads and choose your staging wisely.

Besides, they burn with a nifty colored fire that is fun to have shooting out of your rocket.

[Temeter]

One of the big pros IRL tends to be that the overall weight of an upper H2/O2 stage is less than it would be for a Kero/O2 stage. So that means you can have a smaller first stage to lift it all. I think that is also true in this mod when you use useful payloads and choose your staging wisely.

Besides, they burn with a nifty colored fire that is fun to have shooting out of your rocket.

KSP has very high engine weights, tho, so the difference makes up a higher fraction. Afaik the Soyuz rocket had something like a conventional 1kn engine weighting a single ton, the Space shuttle had ~2.2kn with 3.4 ton with a cryo. That's e.g. why the saturns huge 2nd Stage could be cryogenic.

If you try to achieve the same t/w ratio between stock and cryo, then it's often very hard to get a D/V advantage, while costs tend to be somewhat higher nontheless.

[Nhawks17]

I don't know if anyone else is having trouble getting TweakScale to work by default with these engines but I went a made a custom TweakScale config. I don't know how balanced they are when scaled as all I did was add the modules to allow them to be changed. It doesn't seem to have broken anything with IFS installed like I read a couple pages back but if anyone wants it here yah go!

CryoEngines TweakScale

Edit: I don't know why Chrome is warning that it could be dangerous, I promise you its safe

Edited June 8, 2015 by Nhawks17