Jump to content

[1.12.x] Cryogenic Engines: Liquid Hydrogen and Methane Rockets! (Jan 22, 2022)


Nertea

Recommended Posts

Just now, Idleness said:

Great screens!  My 1.8.1 game doesn't look nearly that good. /envy

Thanks, this is on 1.7.3, I still havent moved (and wont until Kopernicus/JNSQ and some other stuff). That said the visual set up I'm using is linked in my sig including the custom KS3P config. This is all in game processing :)

I really liked the old CryoTanks but the new ones with the reflective textures are another leap ahead!

Link to comment
Share on other sites

  • 2 weeks later...
  • 2 weeks later...
On 11/6/2019 at 7:38 PM, Nertea said:

CE 1.1.0

  • Rebuilt for KSP 1.8.x
  • Updated B9PartSwitch to 2.12.1
  • Updated DynamicBatteryStorage to 2.1.0
  • Updated DeployableEngines to 1.2.0
  • Updated CryoTanks to 1.4.0
  • Updated ModuleManager to 4.1.0
  • DEPRECATED OLD CRYOENGINES MODELS
  • Etna RealPlume tweak (Zorg

Hey, thanks for the awesome work! Is there any way to replace the old deprecated engines in my save file, so i can continue my game?

Link to comment
Share on other sites

21 hours ago, Cice said:

Hey, thanks for the awesome work! Is there any way to replace the old deprecated engines in my save file, so i can continue my game?

Download an old version and extract the appropriate folders (the difference should be apparent).

Link to comment
Share on other sites

  • 2 weeks later...

I must say that I really, really enjoy your mod in my JNSQ game. Dealing with boiloff on my Mun low orbit fuel depot was an interesting way to spice things a bit, and so is the larger fuel tank volume required.

I noticed that the cryo vacuum engines for 1.875 and 2.5 stacks have better thrust (180kN for pavonis and 390kN for the Ulysse) than the LF engines like the cheetah (120kN), poodle (250kN) and wolfhound (375kN), while weighting less. This is not the case for the 1.25 engine, so I wonder if this was an oversight or a deliberate choice. Especially since the 3.75 Tharsis engine sits at 400kN while weighting quite a bit more than the Ulysse (making the 3.75 vacuum engine arguably worse than the 2.5 one).

My point it seems to me that this pushes the scales a bit too much in favor of the cryo engines. The best example is the Wolfhound (3.3T, 375kN) that is completly edged out by the Ulysses with (2.25T, 390kN).

The Hecate engine OTOH is very well balanced compared to the terrier. Heavier, lower thrust and much larger engine bell. Maybe the Pavonis should have 110kN compared to the 120kN Cheetah to mirror the Hecate-Terrier difference ? And 220kN for the Pavonis for the same thing with Poodle ? This would allow the Tharsis to be the true Wolfhound counterpart.

And, oh, thank you very much for adding the Stromboli engine, it's a neat piece of hardware for landers. :)

Link to comment
Share on other sites

On 12/27/2019 at 2:47 AM, Muetdhiver said:

I must say that I really, really enjoy your mod in my JNSQ game. Dealing with boiloff on my Mun low orbit fuel depot was an interesting way to spice things a bit, and so is the larger fuel tank volume required.

I noticed that the cryo vacuum engines for 1.875 and 2.5 stacks have better thrust (180kN for pavonis and 390kN for the Ulysse) than the LF engines like the cheetah (120kN), poodle (250kN) and wolfhound (375kN), while weighting less. This is not the case for the 1.25 engine, so I wonder if this was an oversight or a deliberate choice. Especially since the 3.75 Tharsis engine sits at 400kN while weighting quite a bit more than the Ulysse (making the 3.75 vacuum engine arguably worse than the 2.5 one).

My point it seems to me that this pushes the scales a bit too much in favor of the cryo engines. The best example is the Wolfhound (3.3T, 375kN) that is completly edged out by the Ulysses with (2.25T, 390kN).

The Hecate engine OTOH is very well balanced compared to the terrier. Heavier, lower thrust and much larger engine bell. Maybe the Pavonis should have 110kN compared to the 120kN Cheetah to mirror the Hecate-Terrier difference ? And 220kN for the Pavonis for the same thing with Poodle ? This would allow the Tharsis to be the true Wolfhound counterpart.

And, oh, thank you very much for adding the Stromboli engine, it's a neat piece of hardware for landers. :)

Hey cool, it's not often that people say this mod is overpowered rather than complaining about the size of LH2 tanks!

I'll think on these points, but I do know that comparing directly to the LF engines on a pure thrust basis might be misleading. Recall the lower mass fraction of cryogenic tanks is quite a penalty and that specific impulse matters too. Specifically (hah), I'm thinking of the Ulysses, which is designed to have a higher thrust (but lower Isp) than the other engines as a bit of a niche.

It also might be worth ignoring the MH engines which don't work well in the stock balance curves, IMO.

Thanks for the input!

Link to comment
Share on other sites

  • 2 weeks later...

Since I wanted to ckeck what I said, I made a few calculations in python to compare the different engine pairs (terrier - hecate, cheetah - pavonis, poodle - ulysses, wolfhound - tharsis)

The booster mass is the tank with the correct mass fraction for LFO (0.889) and LH2O (0.728) respectively for regular tanks, and includes the engine mass too. The payload mass is just that, the stuff you put on the rocket.

Below is four plots with the DV ratio for mutliple pairings of booster mass - payload mass, giving a 2D color map.

Spoiler

Vn3fPkV.png

mRkAbZM.png

KBmChAZ.png

Tyv44J9.png

Long story short : with regular tanks, LH2O is nearly always worse in terms of DV. Not by much in quite a lot of cases, but worse. All pairs except wolfhound - tharsis, are nearly perfectly balanced with the LFO engines in terms of DV. The wolfhound - tharsis pair is the exception with the wolfhound being slightly better to a lot better in terms of DV. This is due to the very good (godlike) ISP of the wolfhound.

LFO always wins for large booster mass - small payload mass from the much better mass fraction from the tanks.

However, the TWR is massively favoring LH2O in the following pairings : pavonis has +40% thrust over the cheetah, the ulysses has +50% thrust over the poodle. terrier - hecate and wolfhound - tharsis are balanced in term of thrust.

Using dedicated cryo tanks for the LH2O engines give them +30% to +55% DV over an LFO setup. (so, not for upper stage in atmo since cryo tanks would blow up, but out of the atmosphere, it's amazing.) If one account for the powersupply and power storage as part of the mass fraction the DV edge over LFO goes down to +20% to +45%. Still very good.

I haven't tested, but it might be better to use a cryo tank with a fairing rather than a regular tank if one build an LH2O upper stage.

If you are interested I can pretty much test any engine pairing, such as comparing nuclear-thermal to LH2O engines.

Edited by Muetdhiver
Link to comment
Share on other sites

Thanks for the analysis. One thing that jumps out for me is the way you talk about different DV results for dedicated tanks and other tanks. This shouldn't be the case - all the performance ratios should be identical, the mass difference should be abstracted into the power generation with the non dedicated tanks being more electricity intensive to cool.

Can you elaborate there? We should make sure all the balance variables are on the same page before we start thinking about changes 

Link to comment
Share on other sites

If anyone else has run into an error during the past month where CryoTanks textures are rendering improperly the issue is caused by KSPIE from v1.25.2 onward. I have commented on the support thread here: 

 and filed an issue on GitHub here: https://github.com/sswelm/KSP-Interstellar-Extended/issues/551

In the meantime if you are confident in patching .cfg files yourself then the fix is to edit line 171 of GameData\WarpPlugin\Patches\B9PartSwitch\CryoTanksFuelTankSwitcher.cfg

Line 171 Before:

!MODULE[ModuleB9PartSwitch] {}

Line 171 After:

!MODULE[ModuleB9PartSwitch]:HAS[#moduleID[fuelSwitch]] {}

 

Link to comment
Share on other sites

21 hours ago, Nertea said:

Thanks for the analysis. One thing that jumps out for me is the way you talk about different DV results for dedicated tanks and other tanks. This shouldn't be the case - all the performance ratios should be identical, the mass difference should be abstracted into the power generation with the non dedicated tanks being more electricity intensive to cool.

Can you elaborate there? We should make sure all the balance variables are on the same page before we start thinking about changes 

Yes, I can elaborate here.

I did compare the cryo (dedicated) tanks without EC hardware for cooling to cryo (dedicated) tanks with EC hardware for cooling. This cooling equipement only exists for the benefit of the cryo booster, so I added it the booster dry mass, instead of counting it as payload.

I have used 0.5T of power gen + storage hardware for your dedicated cryo tank version of the jumbo (the one with 64K units of volume). This makes the tank mass fraction go down from 0.879 to 0.849. I then compared the cryo without EC hardware to cryo with it.

To be clear I have yet to check the exact required cooling hardware mass for such a tank. I wanted to have a looksee of what adding a gigantor + service bay + 100kg of batteries would do to the DV, as one would use for long duration missions like a Duna or Jool mission. It also allowed me to have an idea of the threshold in terms of mission length where EC hardware become worth the DV tax, rather than letting the H2 boil off. Turns out that since the impact is about 10% DV, the EC hardware is only worth it if the mission time exceeds ~100h (a very rough estimate). So a Mun mission would likely not need to bother with dealing with the boiloff, Minmus would be on the fence and anything further would need the EC hardware.

I did not look at regular tanks with added EC hardware as I only considered its use case as an upper LH2O stage and/or Mun injection stage, where one would forgo the cooling (given that it'll be used up way before boiloff matters). If one was to use an orbital stage with regular tanks and EC hardware for LH2O cooling, then the DV would become even worse since the the already poor mass fraction would get slapped with a rather heavy addition of dry mass from the EC hardware.

Does this clarify things ?

 

Link to comment
Share on other sites

I found a bug but do not know who I should report it to. @Nils277 or @Nertea. When I use the K&K fuel containers with LH2 and LOX it has -mass and causes the typical negative mass issues.  I have tested it with only B9part switch, cyrotanks, planetary base system and resourse pack to make sure its isolated. Since the patch is in cyrotanks I used this topic. 

Link to comment
Share on other sites

12 hours ago, dave1904 said:

I found a bug but do not know who I should report it to. @Nils277 or @Nertea. When I use the K&K fuel containers with LH2 and LOX it has -mass and causes the typical negative mass issues.  I have tested it with only B9part switch, cyrotanks, planetary base system and resourse pack to make sure its isolated. Since the patch is in cyrotanks I used this topic. 

After some research I have discovered that some fuel tanks in KPBS have very, very low dry mass values - approximately 1/5th of stock. This will create problems, as the universal patch assumes that fuel tanks obey the stock mass ratios.

Ideally KPBS would adjust this ratio to be consistent to the rest of the game, or perhaps can provide a patch that makes it so CT will correctly understand the tanks.

On 1/14/2020 at 6:52 AM, Muetdhiver said:

Yes, I can elaborate here.

I did compare the cryo (dedicated) tanks without EC hardware for cooling to cryo (dedicated) tanks with EC hardware for cooling. This cooling equipement only exists for the benefit of the cryo booster, so I added it the booster dry mass, instead of counting it as payload.

I have used 0.5T of power gen + storage hardware for your dedicated cryo tank version of the jumbo (the one with 64K units of volume). This makes the tank mass fraction go down from 0.879 to 0.849. I then compared the cryo without EC hardware to cryo with it.

To be clear I have yet to check the exact required cooling hardware mass for such a tank. I wanted to have a looksee of what adding a gigantor + service bay + 100kg of batteries would do to the DV, as one would use for long duration missions like a Duna or Jool mission. It also allowed me to have an idea of the threshold in terms of mission length where EC hardware become worth the DV tax, rather than letting the H2 boil off. Turns out that since the impact is about 10% DV, the EC hardware is only worth it if the mission time exceeds ~100h (a very rough estimate). So a Mun mission would likely not need to bother with dealing with the boiloff, Minmus would be on the fence and anything further would need the EC hardware.

I did not look at regular tanks with added EC hardware as I only considered its use case as an upper LH2O stage and/or Mun injection stage, where one would forgo the cooling (given that it'll be used up way before boiloff matters). If one was to use an orbital stage with regular tanks and EC hardware for LH2O cooling, then the DV would become even worse since the the already poor mass fraction would get slapped with a rather heavy addition of dry mass from the EC hardware.

Does this clarify things ?

 

Yes it does. Sorry for the delay in replying - I was trying to get the most current version of my balance spreadsheet off my broken work laptop so I could speak to this a bit more clearly. However all the data on the laptop was wiped, so I don't have that anymore.

Overall your conclusions seem pretty correct (previous post) - looks like the LH2O engines are winning when they should and losing when they should. I'll take a bit of a further look at the Ulysses and Pavonis, but I'd still say that overall total performance is more important than ratio performance to another engine (particularly an MH engine). There are some considerations - a Tharsis should equal or be very close to 2 Pavonis...es, for example. The Ulysses is also bit special because I wanted to pull in a little from its RL counterpart which has quite a high thrust level and takes a hit in specific impulse. 

I'll take a further look at those two engines though!

Link to comment
Share on other sites

 

5 hours ago, Nertea said:

After some research I have discovered that some fuel tanks in KPBS have very, very low dry mass values - approximately 1/5th of stock. This will create problems, as the universal patch assumes that fuel tanks obey the stock mass ratios.

Ideally KPBS would adjust this ratio to be consistent to the rest of the game, or perhaps can provide a patch that makes it so CT will correctly understand the tanks.

wow cannot believe you figured that out because it is only one part and exactly that part that is bugged/unbalanced. The T400. Cheers. 

Link to comment
Share on other sites

  • 3 weeks later...
3 hours ago, Crixomix said:

Is there a way to get procedural tanks to be able to do the cryo-tank thing where they use EC to prevent boiloff? Or is this not possible?

You can try dropping the module into the part config, and see what happens. 

Link to comment
Share on other sites

5 hours ago, Nertea said:

You can try dropping the module into the part config, and see what happens. 

Holy cow... it just... worked. Might I suggest putting this into the .cfg that already comes with your mod aimed at procedural fuel tank? In the patches folder, there's already a procedural fuel tank patch that gives the LH/OX options, so if it's possible, it seems like the boiloff should be added there too? 

In my very minor testing on the launchpad, it seems like the boiloff works properly. EC costs scale with size, is toggle-able, % of LH2 boils off, etc. 

This is the code I copied into the procedural tank cfg.

MODULE
	{
		name =  ModuleCryoTank
		// in Ec per 1000 units per second
		CoolingEnabled = True
		BOILOFFCONFIG
		{
			FuelName = LqdHydrogen
			// in % per hr
			BoiloffRate = 0.05
			CoolingCost = 0.05
		}
    BOILOFFCONFIG
		{
			FuelName = LqdMethane
			// in % per hr
			BoiloffRate = 0.005
			CoolingCost = 0.02
		}
	}

 

Link to comment
Share on other sites

  • 3 weeks later...

CE 1.1.1

  • KSP 1.9.x
  • Updated B9PartSwitch to 2.13.0
  • Updated DynamicBatteryStorage to 2.1.5
  • Updated DeployableEngines to 1.2.1
  • Updated CryoTanks to 1.4.2
  • Updated ModuleManager to 4.1.3
  • Updated Chinese localization (tinygrox)

@MuetdhiverI have completely forgotten about your issue. I created https://github.com/ChrisAdderley/CryoEngines/issues/75 to track it, hopefully addressed in the next version.

Link to comment
Share on other sites

  • 2 weeks later...

@Nertea Hi! First of all: thank you for your mods, they're amazing, I play with almost everyone. Also apologize for my English, it's not my native language.

I found the mistake of breaking the economy. When mod CryoTanks is combined with mod B9 partswitch. For example two fuel tanks: MK3(longfuselage)LF100 and MK3(longfuselage)LFO100. 
In stock game MK3(longfuselage)LF100 costs 17200 credits, MK3(longfuselage)LFO100 costs 10000 credits. This is understandable: the cost of LF is higher than the combination of LFO. But if you switch LFO100 tank to liquid fuel, its cost with equal amount of fuel becomes less than that of a specialized LF tank. Thus, purchase of some tanks with liquid fuel, with this mod loses all sense. This also happens with a shortened version of these tanks and with others. The mk2 tanks, for example, are initially the same in cost and they are fine.

I have a suggestion: it is possible to make the cost of a universal (LFO) tank more expensive than a specialized LF tank. At equal cost of fuel, it will be less profitable to store liquid fuel in a universal tank. Or disable the switch to liquid fuel where there is a specialized equivalent tank. 

If I've written incomprehensible - tell me about it, I'll try to supplement it

Link to comment
Share on other sites

On 3/4/2020 at 3:37 AM, Husky777 said:

@Nertea Hi! First of all: thank you for your mods, they're amazing, I play with almost everyone. Also apologize for my English, it's not my native language.

I found the mistake of breaking the economy. When mod CryoTanks is combined with mod B9 partswitch. For example two fuel tanks: MK3(longfuselage)LF100 and MK3(longfuselage)LFO100. 
In stock game MK3(longfuselage)LF100 costs 17200 credits, MK3(longfuselage)LFO100 costs 10000 credits. This is understandable: the cost of LF is higher than the combination of LFO. But if you switch LFO100 tank to liquid fuel, its cost with equal amount of fuel becomes less than that of a specialized LF tank. Thus, purchase of some tanks with liquid fuel, with this mod loses all sense. This also happens with a shortened version of these tanks and with others. The mk2 tanks, for example, are initially the same in cost and they are fine.

I have a suggestion: it is possible to make the cost of a universal (LFO) tank more expensive than a specialized LF tank. At equal cost of fuel, it will be less profitable to store liquid fuel in a universal tank. Or disable the switch to liquid fuel where there is a specialized equivalent tank. 

If I've written incomprehensible - tell me about it, I'll try to supplement it

I'll look into it. 

Link to comment
Share on other sites

On 3/4/2020 at 3:37 AM, Husky777 said:

@Nertea Hi! First of all: thank you for your mods, they're amazing, I play with almost everyone. Also apologize for my English, it's not my native language.

I found the mistake of breaking the economy. When mod CryoTanks is combined with mod B9 partswitch. For example two fuel tanks: MK3(longfuselage)LF100 and MK3(longfuselage)LFO100. 
In stock game MK3(longfuselage)LF100 costs 17200 credits, MK3(longfuselage)LFO100 costs 10000 credits. This is understandable: the cost of LF is higher than the combination of LFO. But if you switch LFO100 tank to liquid fuel, its cost with equal amount of fuel becomes less than that of a specialized LF tank. Thus, purchase of some tanks with liquid fuel, with this mod loses all sense. This also happens with a shortened version of these tanks and with others. The mk2 tanks, for example, are initially the same in cost and they are fine.

I have a suggestion: it is possible to make the cost of a universal (LFO) tank more expensive than a specialized LF tank. At equal cost of fuel, it will be less profitable to store liquid fuel in a universal tank. Or disable the switch to liquid fuel where there is a specialized equivalent tank. 

If I've written incomprehensible - tell me about it, I'll try to supplement it

So there's this problem:

LF Tanks Dry Cost Fuel Amount( u) Ratio ($/u)
Mk0 Liquid Fuel Fuselage 160 50 3.2
Mk1 Liquid Fuel Fuselage 230 400 0.575
Mk2 Liquid Fuel Fuselage 810 800 1.0125
Mk3 Liquid Fuel Fuselage 4600 5000 0.92

If you do the math with the stock LF tanks, there is a close to order of magnitude difference between the various size classes in terms of dry cost ratio. That's crap to balance against (doesn't matter whether a tank switches or not) and it's Squad's idea of cost balance. Please note that this is by far not the worst cost balance issue in the game. LFO tanks are fairly consistent if they're rocket or spaceplane, with spaceplane costing more. 

LFO Tanks Dry Cost Fuel Amount( u) Ratio ($/u)
Rockomax-64 2812.4 6400 0.4394375
Mk3 Rocket Fuel Fuselage 2705 5000 0.541

Cryo Tanks takes a relatively universal approach, where the base cost of the tank is unchanged between all the variants. That works well in most cases but obviously not here. I do not want to make switching tanks more expensive, that defeats the purpose of the fuel switch. I can add an additional cost to liquid fuel tanks, because by the looks of the ratios above there will be a consistent underestimation of cost.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...