[0.25] Dibnah Engineering - Water Splitting (and more to come)

[Deadpan110]

After playing with other peoples mods and wanting to give a little back to the KSP community, I have created 'Dibnah Engineering' - a part modification and tuning company headed by the great Dibnah Kerman; a mechanical engineer who can just about modify anything with his trusty lump hammer in order to get jobs done.

Currently contains only 2 parts (but I hope to add more).

WaterSplitting

A collection of parts dedicated to the separation of water.

• DE-H20 Electrolysis unit
  
  • Converts Water into Oxygen and Hydrogen (cryogenic) using rates based on Paul Kingtiger and Daishi's Universal Storage
    
  • Available in 1.25m and 2.5m variants
    
  • Part model used under licence and with permission from RoverDude (Karbonite)
    
  • Resource colour coding used with permission from AetherGoddess (ART hatches for CRP)
    

  [*]Hydrogen Container with compressor

  • Converts/Pumps Hydrogen (cryogenic) into LiquidHydrogen for use with FTT and NearFuture
    
  • Part Model used under TAC Life Support licence CC BY-NC-SA 3.0 by Taranis Elsu
    

Installing

• Download the latest release from GitHub
  
• Delete any existing DibnahEngineering directory from your GameData folder
  
• Unzip and copy/merge DibnahEngineering into the GameData Folder
  
• (Optional) Merge any required ModuleManager files from the Extras folder into GameData\DibnahEngineering
  

Frequently Asked Questions

• Why is it too hard to create a small amount of liquidHydrogen, I was hoping to land, refuel and take off with out having to wait years?
  
  • Real life Hydrogen production is not done in large quantities using electrolysis of water as it is very costly and highly time consuming (most Hydrogen from Earth is produced by the processing of fossil fuels). These parts use the same conversion ratios as the Universal Storage Elektron (which is based on the real thing operating in the International space station) but also uses larger conversion rates which will requires a lot more ElectricCharge. Please feel free to use the included ModuleManager files so as to produce liquidHydrogen at unrealistic but more playable rates.
    

  [*]The consumption of ElectricCharge is too high/too low - is that right?

  • ElectricCharge usage is based on the rates from Universal Storage which uses the value of 1EC=33J. Other mods and parts use differing values so the correctness of what the actual values are could possibly remain forever in debate until Squad publish what the actual values should be. As these parts follow the ratios from Universal Storage, the ElectricCharge usage will only be changed if Universal Storage changes and so will always remain in sync. Please feel free to use the included ModuleManager files so as to use less ElectricCharge if you feel the current settings are too high for your style of gameplay.
    

  [*]I filled up all my Oxygen tanks and the Electrolysis unit stopped working, why didn't it just carry on producing regardless?

  • When it comes to space exploration, Oxygen is an important resource and it seems a shame to waste it. More from a gameplay point of view, managing resources and visiting your outposts once in a while introduces small challenges and reasons for more smaller missions to accomplish minor goals. I recommend using the Fuel Valve from Smart Parts to bleed off the excess. Please feel free to use the included ModuleManager files so as full Oxygen storage will not halt your Hydrogen production.
    

  [*]So how exactly are you playing your game with these parts?

  • I certainly do not play with the default settings as being too realistic can destroy fun, gameplay and make these parts totally useless for most players. I do however, like to use the Smart Tools Fuel Valve so having no space for Oxygen stops Hydrogen production. Please feel free to use any or all of the included ModuleManager files to tailor these parts to the way you want to play.
    

  [*]Why should we just drop in ModuleManger files, wouldn't it be easier just to supply the parts with those values already included?

  • Having parts where values are totally unrealistic and way off is a simple approach but will not be believable. The default configs match Universal Storage with Paul Kingtiger and Daishi's research on the Electrolosys unit aboard the International Space Station. Regardless of the accuracy, the production of Hydrogen is very time consuming and requires large quantities of power. With this as a baseline, it is then possible to tweak and adjust the values to individual needs in the hope of providing the best ratios for everyones style of gameplay. For Example: If you want to use less ElectricCharge, you have a base value to work from. Feel free to use any of the provided ModuleManager configs to adjust the parts to suit your game style.
    

  [*]The ModuleManager configs don't suit my game style - what should I do?

  • Feel free to write your own - really, it is entirely up to you how hard or how easy Hydrogen production should be. Use the existing configs if you are unsure on how to write one or ask in the DibnahEngineering forum thread for help.
    

Requirements (usually provided by the modpacks you would need these parts for)

• Community Resource Pack
  
• ORSX
  
• USI-Tools
  

Optional/Recommended Mods these parts are useful with

• USI Kolonization Systems (MKS/OKS) - (Drill for water resource)
  
• FTT - Freight Transport Technologies (Uses LiquidHydrogen)
  
• Near Future Propulsion (Uses LiquidHydrogen)
  
• Universal Storage (Uses Hydrogen cryogenic)
  
• Smart Parts - Fuel Valve (Used in the above pic to bleed off excess Oxygen)
  

Links

• 
  [*=1]

Current releases on GitHub
[*=1] GitHub Project Page

[U]Changes[/U]

0.1.3.0
[LIST]
[*]Extras directory with more ModuleManager configs for players to further tweak production values 
[/LIST]
 0.1.2.0

[LIST]
[*]Hydrogen to liquidHydrogen conversion based on CRP density values 
[*]Override with 'Easy Mode' ModuleManager config (just drop anywhere into your GameData folder) 
[*]New 2.5m Electrolysis unit 
[*]3 converters to toggle on/off on each Electrolysis unit 
[*]Converter power requirements are based on Near Future Nuclear Reactor sizes (one reactor per converter) 
[/LIST]
0.1.1.0

[LIST]
[*]Nerfed Hydrogen to LiquidHydrogen conversion 
[/LIST]

0.1.0.0 - Initial downloadable release
[LIST]
[*]Added .version file for KSP-AVC users 
[/LIST]

Dibnah Engineering Licence

CC BY-NC-SA 3.0

Edited December 2, 2014 by Deadpan110
Tag release 0.1.3.0

[SlimJim89]

Sweet, will require the ability to make LH this week, nice timing!

[White Owl]

Oh hey, this could be reconfigured to Real Fuel's liquid hydrogen and liquid oxygen, right? How does the mod handle determining where you're likely to find water ice?

[Deadpan110]

Sweet, will require the ability to make LH this week, nice timing!

Me too - and after making the parts for my own use, I figured I should share them with the community. I am still unsure of the actual balancing of values though - but would like these parts to be used more for fun over realism.

Oh hey, this could be reconfigured to Real Fuel's liquid hydrogen and liquid oxygen, right? How does the mod handle determining where you're likely to find water ice?

These are just parts and should work nicely with any mod that makes use of ORSX and the Community Resource Pack (Detection can be done with SCANSat). These parts make use of the converters within USI-Tools which take values for inputs and outputs of any resource supported by the Community Resource Pack - in essence, these parts are converting the named resources 'Water' along with 'ElectricCharge' into 'Hydrogen' and 'Oxygen' and the 2nd part converts the named resource 'Hydrogen' into 'LiquidHydrogen'. You could easily modify the config file and change the resources into anything your game uses if so required.

In the spirit of the Community Resource Pack - yes you can convert 'Space Cows into Plutonium-239' if you wanted to.

[Ar_Taq]

The balance is just way wrong. 42,000 units of LiquidHydrogen from .04 units of water in less than an hour. Can this be fixed to better balance it?

Edited November 21, 2014 by Ar_Taq
Better stats

[lodestar]

This is exactly what I was looking for. Thanks.

[FreeThinker]

Lol, I made the exact same parts for myself (except for the textures). Good thing someone made the textures and released it.

Have you realised it cost a lot of power to split water into H and O? I suggest you include a basic kOS script players can use to run the watersplitter only when enough power is available. I use it myself on my space station to prevent me for running out of electricity durring shadow time.

Edited November 21, 2014 by FreeThinker

[Deadpan110]

The balance is just way wrong. 42,000 units of LiquidHydrogen from .04 units of water in less than an hour. Can this be fixed to better balance it?

Yes, the reason it was set as it is currently was because it followed Universal Storage's conversion rates to convert water into Oxygen and Hydrogen (cryogenic - as Paul Kingtiger describes it).

Paul from Universal Storage:

The Elektron wedge contains a water electrolyzer which reverses the fuel cell reaction by converting water and electricity into Oxygen and Hydrogen (and compressing these resources for storage). The Elektron is based on the real thing operating in the International space station.

I then used his ratio (mentioned elsewhere) to convert Hydrogen into LiquidHydrogen - the numbers are the wrong way around - I will fix shortly.

I would have left them as they are just for fun vs realism - but I would rather the figures be more realistic than over the top ridiculous - thanks

Edit: Fixed in commit 3f41a70 - release to follow.

Edited November 21, 2014 by Deadpan110
Added commit

[Ar_Taq]

Thanks Deadpan110! I appreciate it!

[Newnard]

First of all, thank you so much for making this mod, it is really appreciated.

Game balance is now way off in the other direction. Example:

I want to fill one of the big hydrogen tanks from future propulsion (40,000 units liquid hydrogen). Assume I have an infinite supply of water.

Each second, one of your devices produces 0.022 units of hydrogen. There are 851 units of hydrogen per unit of liquid hydrogen.

851 * 40,000 = about 34 million units of hydrogen required, and at 0.022 per second that's about 1.5 billion seconds, which is just under 430,000 hours, or 71633 kerbal days.

Of course, if I wanted to fill the tank in about six hours, I could always just use 71,633 devices but that would cost over half a million units of energy per second, which works out to 125 or so fission reactors.

I don't know the compression level of gasseous to liquid hydrogen, but this feels like it is off by a couple orders of magnitude.

http://en.wikipedia.org/wiki/Hydrogen_storage may be useful; I haven't wrapped my head around it yet.

[Deadpan110]

Thanks, I do need help with this part of the mod which is why I initially left the number the opposite away around (didn't take long to magically fill my Liquid Hydrogen tanks this way).

I have found where I got the original ratio numbers from: Community Resource Pack Thread

Both hydrogens will stick as they are different resources with different uses - also one is used by NFT, the other by US, and in general once a mod is on CRP I'll pretty much do my best not to deprecate their things because part of CRP is to encourage sharing.

Universal Storage Hydrogen is Gas (although it's stored cryogenically in our parts which is why they have large capacities).

This is a mod in itself but it would be brilliant if we could have a single entry for each resource, then a phase tag with compression/expansion information. That way you could have a Hydrogen[LIQ] storage container which would feed a Hydrogen[GAS] processor using resources at a rate of 861 - 1.

That's wishful thinking though. A good stop gap would be a modders agreement to use a standard naming format for resources to show if they were liquid, solid or gas, something similar to the above.

This is something that spins my head around. the only thing that changes (as far as KSP is concerned) in phase transitions is volume/mole, but internal volume for resources aren't tracked in KSP, so the difference between a container of 1 mole liquid ¹H⁺ and 1 mole gasious ¹H⁺ is the size of the container's model. resource mass is unchanged, since the mass of the resource is 1 g (from definition of mole), and resource amount is unchanged (from definition of mole).

Unless there is a reason to track phase changes, like, say, requiring electricty to convert gasious ¹H⁺ into liquid coolant, then it shouldn't matter what state it's in, because you have 1 mole of it, and it's in a part that doesn't care how big 1 mole is. Even for that use case, the cooling part can just directly use EC to account for that and make everyone else's life easier.

To derive a KSPish value for LH2, I assumed that LiquidFuel is Kerosene, and worked from there.

LH2 density: 70 kg/m

Kerosene density ~800 kg/m3

That's a ratio of 0.0875, stock LiquidFuel has a density of 0.005, so applying the ratio we get 0.0004375. Rounded to 0.0004 for nicer numbers. So it's correct in a relative sense only, you'll need RF to get properly scientific numbers.

I will see if I can get RoverDude to put his thoughts into this (could it be that conversion should use 861 to 1, or 1 to 861, or 1 to 1, or 851 to 1 as is currently).

The tail end is that I want some value thats Fun and not tedious but also should feel realistically possible - In the mean time, feel free to experiment with the numbers by editing the config and report your findings so perhaps I can update based on every ones views.

[Newnard]

Ok, continuing to brainstorm here and I might submit this comment just to checkpoint my current thinking, don't take it as final opinion or in any way authoritative.

H2O is roughly 2 parts hydrogen, 16 parts oxygen by mass.

So, for every 18kg of water, we should get about 2kg hydrogen out.

In the default TAC life support, it seems that water is simply 1 unit == 1 kilogram. A standard sized water tank from TAC holds 240 units/kilograms of water. Which should translate to a little over 26kg of hydrogen.

In Near Future Propulsion, it seems that liquid hydrogen is 1 unit == 0.4 kilograms.

So one unit of water should give us 1/9 kg of liquid hydrogen, or 0.04444444 units. Reversing this logic means that for each unit of liquid hydrogen, we would need 3.6 units of water.

My favorite liquid hydrogen tank is 40,000 units large, so I *should* need 144,000 units of water to convert and fill it up.

So that's my idea of how the ratios SHOULD work. Now to examine what you have implemented at the moment to see if it matches:

Input water per second = 0.0000176697 units. Weird value, but ok. Translates to 0.0000176697 kg, which contains 0.0000176697/9 == 0.0000019633 kg of actual hydrogen.

The output resource we care about is the hydrogen resource from US, which is NOT the same as the liquid hydrogen I am seeking. We get 0.0219907427 units of this hydrogen resource out.

We must assume at this point that the 0.0219907427 units of "hydrogen" equates to 0.0000019633kg. So I'll simply take the kilograms and divide by the number of units to get kg/u which is about 8.93kg per unit? That seems like a rather large value, but maybe that's why they are using such weirdly small numbers in the conversion process.

So now we need to convert this "hydrogen" resource to "liquid hydrogen".

The converter simply takes the input resource, divides by 851. I'm still murky as to where this 851 number comes from. But that means that 0.0219907427 units of "hydrogen" converts to 0.000002584106 units of "liquid hydrogen"; at 0.4kg per unit of liquid hydrogen that is only 0.000000103364243blah kg.

Going back to the start, we determined that our output should be 0.0000019633kg of liquid hydrogen.

So somewhere the numbers are off by a factor of about 19.

It seems to me that the universal storage people went to some effort to come up with a reasonable value for "hydrogen" units/kg, and that the near future propulsion people likewise chose what seemed like a good number for "liquid hydrogen" units/kg. We needn't question either of their choices, we just need to assume that they are talking about the same element with the same atomic weight and convert appropriately.

So: "hydrogen" == about 8.93kg per unit

and "liquid hydrogen" is 0.4kg per unit.

So when converting between them, we should use a factor of 8.93/0.04 == 22.325, not 851.

My point with all this is that we should try to conserve mass when doing our conversions, as mass is critical to spaceflight and having it mysteriously grow or shrink is very, very confusing.

[Deadpan110]

So when converting between them, we should use a factor of 8.93/0.04 == 22.325, not 851.

My point with all this is that we should try to conserve mass when doing our conversions, as mass is critical to spaceflight and having it mysteriously grow or shrink is very, very confusing.

Nice - I was also running backwards and forwards between the numbers and trying to tie them in with the values of the Community Resource Pack (CommonResources.cfg).

Lets see how it works

I will update the config using the ratio of 22.325

As for other questions about powering the devices, this is all open for debate but it is more likely the conversion electric charge needed will be based on Stock gigantor solar panel output (i.e. how many you should be landing on a sunny planet with in order to get some nice production).

The actual Electroliser itself, I am toying with the idea of having 3 converters on it (one for each tank side) that you can enable/disable depending on your power usage (use just one while you have a Kerbal landed and docked so it does not consume too much power that the poor Kerbal dies and then activate the 2nd and 3rd should power production permit). Again, this is all open to comments and suggestions.

Importantly - balancing conservation of mass as well as making for good game play is a very important thing and I appreciate everyone’s input!

[Ramarren]

I calculated the mass ratios with CRP numbers and that is what I got. Where did the 851 number came from, because I am seeing about 4449?

                                     water                    hydrogen  liquidhydrogen  oxygen
density (tons per unit from CRP)    0.0010000000         0.0000000899    0.0004000000  0.0000014100
units per ton                    1000.0000000000  11123470.5228031000 2500.0000000000  709219.8581560280
kg per unit                         1.0000000000         0.0000899000    0.4000000000  0.0014100000
ratio in water                      1.0000000000         0.1111111111    0.1111111111  0.8888888889
kg per 1 water unit                 1.0000000000         0.1111111111    0.1111111111  0.8888888889
units per 1 water unit              1                 1235.9411692004    0.2777777778  630.4176516942


electrolyzer                        0.0000176697         0.0219907427                  0.0109953713
normalized                          1.0000000000      1244.5453346690                  622.2726645048

compressor                                             851.0000000000    1.0000000000    
normalized                                            1235.9411692004    1.4523397993    

should be?                                            4449.3882091213    1

[Deadpan110]

I calculated the mass ratios with CRP numbers and that is what I got. Where did the 851 number came from, because I am seeing about 4449?

Yes, that was the reason I was running around - I also had about 4449.388 using the values within CRP; but when implemented, it was even more ridiculously hard to get LiquidHydrogen.

I used the 851 figure (it is actually a typo - it should have been 861) based on Paul Kingtiger's calculations for Universal Storage but implemented and also reversed - that number makes it either too difficult or too easy to get the resource.

I am going to test Newnard's value - but if anyone else has the time to tweak values, then please do and let us know of your results!

[Ramarren]

Unfortunately I'm pretty sure Newnard moved the decimal point the wrong way when calculating the mass per unit for Hydrogen, so with that the compressor will be magically getting a kilogram of LiquidHydrogen from every five grams of Hydrogen, which is probably going to strain even the most flexible suspension of disbelief.

The actual problem is not really mass ratios, but rates and electric consumption. Water electrolysis is quite power consuming, and splitting enough water (144000 units, 144 tons) to get 40000 units of liquid hydrogen (16 tons) will realistically require power in the range of terajoules (it takes more than 237kJ per 18 grams to split water).

[Deadpan110]

Quick test using just the electroliser using Universal Storage ratios:

*CAVEAT : Please note that I am posting this so that myself and others know the current status of getting a decent gameplay feel and some kind of realism sense - these notes may or may not help me or anyone else come up with an elegant solution - but I am throwing them out there for completion's sake.

    MODULE
    {
        name = USI_Converter
        converterName = Water Electrolizer
        conversionRate = 2
        inputResources = ElectricCharge, 7, Water, 0.0000176697
        outputResources = Hydrogen, 0.0219907427, True, Oxygen, 0.0109953713, False
    }

Note that ElectricCharge within Universal Storage is 14 and not 7 as I have set it.

Using just 1 TAC Life Support Water Container (Large) and after 30 earth days (120 Kerbin days), I got the following results:

Resource     used          available     storage
Water        45.9482       2154.0518     22000       

Hydrogen     57184.5866
Oxygen       28592.2932

    *approx 2 Hydrogen units and 1 Oxygen units per 0.000531766 units of water

This is the way things currently work within Universal Storage:

Universal Storage Hydrogen is Gas (although it's stored cryogenically in our parts which is why they have large capacities).

-snip-

That way you could have a Hydrogen[LIQ] storage container which would feed a Hydrogen[GAS] processor using resources at a rate of 861 - 1.

From the Universal Website we see that:

The Elektron wedge contains a water electrolyzer which reverses the fuel cell reaction by converting water and electricity into Oxygen and Hydrogen (and compressing these resources for storage). The Elektron is based on the real thing operating in the International space station and can help to recycle resources on craft with plentiful power such as large solar arrays.

From: Universal Storage Core Pack

So IF we assume that to maintain compatibility with Universal Storage, the above is correct - (except the Electric Charge I adjusted from 14 down to 7), the only real big issue is the conversion of Paul's Hydrogen Gas into Near Future and USI's LiquidHydrogen.

The above also suggests that even using 14 as the Electric Charge value is way too low in Universal Storage - but as it is a great mod with fantastic parts, I would settle somewhere in that ballpark.

The only thing remains is Universal Storage Hydrogen (cryogenically stored gas) conversion and getting it somewhere close to not being magical or even too extremely difficult to land and refuel in less than 30 real earth days (hence the reason I openly accept peoples suggestion on what the ratio should be).

Edit: To add to the above, After 30 earth days (120 Kerbin Days), Newnard would have filled his favourite Near Future tank (40,000) with only 2561.459 of Liquid Hydrogen using a ratio of 22.325 to 1 in the Hydrogen to LiquidHydrogen resource conversion.

Edited November 23, 2014 by Deadpan110
Extra info

[Newnard]

Unfortunately I'm pretty sure Newnard moved the decimal point the wrong way when calculating the mass per unit for Hydrogen, so with that the compressor will be magically getting a kilogram of LiquidHydrogen from every five grams of Hydrogen, which is probably going to strain even the most flexible suspension of disbelief.

Ramarren is correct, and I am mortified. I was using a calculator to do the math and the answer was 8.927847625628397e-5 -- I only paid attention to the beginning part and missed the e-5 at the end. I will lamely mention that it's been a while since I slept and that might have contributed.

The actual problem is not really mass ratios, but rates and electric consumption. Water electrolysis is quite power consuming, and splitting enough water (144000 units, 144 tons) to get 40000 units of liquid hydrogen (16 tons) will realistically require power in the range of terajoules (it takes more than 237kJ per 18 grams to split water).

That makes a lot of sense... On my copy I edited the converter to process water a hundred times faster than what Deadpan110 had it at, but at the same (not 100x) energy cost... and that felt a lot better, gameplay wise. But I think I'd be willing to invest in a huge power infrastructure on minmus if that's what would be required... I just don't want it to take years and years to fill my tanks.

In reference to the bolded (by me) part of your quote above, I have two questions:

1) Can you point me at a reference that explains the theoretical power needs of this process? Keeping in mind that I'm bad at chemistry.

2) Does anyone have a clue how much energy one kerbal energy unit is supposed to represent? I've seen a couple different values for kilowatts and megajoules flying around at various points since I started playing this game, and don't know if any of them had a basis.

[Ramarren]

That makes a lot of sense... On my copy I edited the converter to process water a hundred times faster than what Deadpan110 had it at, but at the same (not 100x) energy cost... and that felt a lot better, gameplay wise. But I think I'd be willing to invest in a huge power infrastructure on minmus if that's what would be required... I just don't want it to take years and years to fill my tanks.

In reference to the bolded (by me) part of your quote above, I have two questions:

1) Can you point me at a reference that explains the theoretical power needs of this process? Keeping in mind that I'm bad at chemistry.

2) Does anyone have a clue how much energy one kerbal energy unit is supposed to represent? I've seen a couple different values for kilowatts and megajoules flying around at various points since I started playing this game, and don't know if any of them had a basis.

I just looked it up on Wikipedia ( http://en.wikipedia.org/wiki/Electrolysis_of_water#Efficiency ). 237kJ per mole (for water ~18grams) is a thermodynamic lower bound, any real system would use more.

I have typically seen 1EC=1kJ, but that seems to be a very rough approximation. It more or less works out between solar panels and lights and so on. The problem with electrolysis is you cannot simply use the standard KSP "witchcraft" solution (see description for stock ion engines, which consume way less power than they should), because of the Universal Storage Fuel Cell. If you produce oxygen and hydrogen using less energy than the fuel cell produces from the same oxygen and hydrogen you can get infinite power from nowhere. If you ignore what the "real" energy cost is and use that as a bound, for the fuel cell you get 1.78e8 EC units per ton of hydrogen. So you would need 2.85e9 for 16 tons. (Unless I miscounted something)

Near Future and Freight Transport Technologies nuclear reactors have output in the several thousand range, with the Starlifter reactor giving 10000 EC per second. So with a single one you could generate 16 tons of hydrogen in eighty hours, which seems actually reasonable.

Edit: something seems wrong after all, maybe this time I lost a bunch of orders of magnitude...

Yes, actual numbers seem to be: 2.75e9 per ton, 4.4e10 for 16 tons, 1222 hours (50 earth days) at 10000EC/s which is not that terrible still.

Edited November 23, 2014 by Ramarren

[Newnard]

Yes, actual numbers seem to be: 2.75e9 per ton, 4.4e10 for 16 tons, 1222 hours (50 earth days) at 10000EC/s which is not that terrible still.

I can easily imagine deploying five 4000 EC/s fission plants to minmus, to cut the production time down to 25 earth / 100 kerbin days per huge tank. Especially if I can do something creative and subsequently take a couple of those fission plants with me on the ship using the hydrogen. Reuse is good.

So if I assume the numbers above (16 tons of hydrogen in 1222 hours) that means processing 144,000kg of water in the same time period, or 118kg per hour, which works out to 0.032733etc units of water per second. I figure if I'm running five fission plants to power the operation, there are probably multiple separate electrolysis units at work... possibly 10-ish? So if the per-electrolysis-machine water consumption were 0.0033-ish units per second, that would be sufficient for my whims. Which is only, um... 187? times faster than the current configuration.

[Ramarren]

After some testing this is not going to work, because KSP can't handle parts consuming multiple thousands of EC under high warps without having really unreasonable amounts of batteries, and it doesn't seem to work with the vessel unfocused either. Even with both generators (FFT generators) and consumers being USI_Converters. Actually looking at the source, there might be a straight up bug in USI_Converter.cs... I will log a github issue.

[Tellion]

Very interesting project!

If i did not make any mistakes (that really is not guaranteed), the conversion rates for RealFuels are 1:809.219858156 for Oxygen (density from CRP) to LqdOxygen and 1:788.097886541 for Hydrogen (density from CRP) to LqdHydrogen if anyone is interested in that.

I am not sure what the exact roadmap for this mod is, but I think it has great potential for providing realistic-ish ISRU options in this game. Besides water electrolysis and conversion to usable rocket fuel, what about gathering resources from planetary atmospheres? This shouldnt be too hard to implement with the framework of ORSX i imagine. The Karbonite Scoops/Particle Collectors already work for Argon and Oxygen, but i dont think its possible to gather Hydrogen from Jool's atmosphere for example.

[Deadpan110]

Yes, actual numbers seem to be: 2.75e9 per ton, 4.4e10 for 16 tons, 1222 hours (50 earth days) at 10000EC/s which is not that terrible still.

The ElectricCharge values I will try to keep in line with Universal Storage because if I make them too high, you may as well completely avoid using my part altogether and just use a bank of Universal Storage Elektron's to save on ElectricCharge use. It would also work out cheaper in part count too (use 8 Elektrons and some solars or use 1 DE-H2O-125 Electroliser and a bank of reactors).

One thing I didn't account for when halving the ElectricCharge rate that Universal Storage used - was that the Universal Storage Alkaline Fuel Cell could then create more ElectricCharge than what was initially put into the system by reversing the Hydrogen/Oxygen process.

I can easily imagine deploying five 4000 EC/s fission plants to minmus, to cut the production time down to 25 earth / 100 kerbin days per huge tank. Especially if I can do something creative and subsequently take a couple of those fission plants with me on the ship using the hydrogen. Reuse is good.

When I initially wanted to make some means of Hydrogen production, one of the key areas I wanted to avoid was part count but I still wanted to closely follow Paul Kingtigers Universal Storage Elektron - just make one a little bigger and a little faster without necessarily having the need to launch too many parts to get up and running.

It came as more of a surprise that people are actually interested and want to use the parts (wow - thank you to all).

The hardest part is trying to tie in Hydrogen and LiquidHydrogen resources so they balance nicely - I will do some more testing with the ratio you provided (223.25 Hydrogen = 1Liquid Hydrogen) as the conservation of mass seems a helluva lot better with those figures.

Another note that sways the balance askew is that there are pretty playable ModuleManager configs for Kethane and Karbonite conversion into fuels - both of which do not exist in reality - so it becomes quite easy to grab figures and balance them based on what feels right.

After some testing this is not going to work, because KSP can't handle parts consuming multiple thousands of EC under high warps without having really unreasonable amounts of batteries, and it doesn't seem to work with the vessel unfocused either.

I would have said that USI_Converters work during the proverbial 'on-rails' along with USI reactors - I know that Near Future and anything stock will not, but as the whole USI on rails thing is not actually a background process (it is a calculation of what would have been done during the time there was no focus up to the point of your return) and as long as there is some means of ElectricCharge storage then hopefully there would still be enough done by the USI_Converter.

I will have another tweak at the numbers and then play test to see what happens.

Very interesting project!

If i did not make any mistakes (that really is not guaranteed), the conversion rates for RealFuels are 1:809.219858156 for Oxygen (density from CRP) to LqdOxygen and 1:788.097886541 for Hydrogen (density from CRP) to LqdHydrogen if anyone is interested in that.

I am not sure what the exact roadmap for this mod is, but I think it has great potential for providing realistic-ish ISRU options in this game. Besides water electrolysis and conversion to usable rocket fuel, what about gathering resources from planetary atmospheres?

Nice - RealFuels rates will be good once some kind of balance has been established as it would be easy enough to include a ModuleManager config for it.

As for the roadmap, Dibnah Kerman wanted a way to fill up his craft when landed on Minmus and not have to wait around days for someone at KSC to send some.

Now that people are interested in the parts - I plan to do my best taking on board suggestions about realism vs gameplay - I do like to stretch reality a little in my game (what with RoverDude's torchship and K+ engines - and now the Warp Drive awesomeness) but I also like my game's story to have some sense of realism and processes for making Hydrogen are as real as you can get due to real life science!!

For now I will be tweaking the numbers about trying to find a happy medium that tries to fall in with peoples differing styles of gameplay.

Future plans... well, I wanted to feed water + substrate to bacteria and get Hydrogen too - lets see how this one plays out first

[Ramarren]

The ElectricCharge values I will try to keep in line with Universal Storage because if I make them too high, you may as well completely avoid using my part altogether and just use a bank of Universal Storage Elektron's to save on ElectricCharge use. It would also work out cheaper in part count too (use 8 Elektrons and some solars or use 1 DE-H2O-125 Electroliser and a bank of reactors).

Those numbers were in line with Universal Storage, there were fuel cell numbers inverted plus I think ten percent energy loss. The thing is that you need orders of magnitude more hydrogen for nuclear propulsion than for the fuel cell, which means that you would need hundreds of Elektron parts to make enough in reasonable time. If the new electrolyser is not at least hundred times more powerful than Elektron then it still wouldn't be very useful for producing liquid hydrogen for propulsion purposes. Which is where the thousands of ECs per second come from.

I would have said that USI_Converters work during the proverbial 'on-rails' along with USI reactors - I know that Near Future and anything stock will not, but as the whole USI on rails thing is not actually a background process (it is a calculation of what would have been done during the time there was no focus up to the point of your return) and as long as there is some means of ElectricCharge storage then hopefully there would still be enough done by the USI_Converter.

I have logged an issue on USI Github about that, it seems there was supposed to be a modifier for electricity, where no matter how long the delta is always counted as at most one second, but there is a Math.Max instead of Math.Min. So you still need enough capacity to hold as much as is produced by all reactors for a full physics step, so it breaks at high warps. It is possible something else was wrong when I tested it with the vessel not loaded. Unfortunately, I don't know enough about how KSP processes event to know how the catchup code is exactly, or enough about building mods to test it myself.

The hardest part is trying to tie in Hydrogen and LiquidHydrogen resources so they balance nicely

I still think that fully mass conserving ratio is the only proper one (~4450:1). Magic resources are all very well, but increasing mass of the vessel by pumping stuff around just seem weird.

[Newnard]

I still think that fully mass conserving ratio is the only proper one

For the record I 100% agree, magic mass is bad.

(~4450:1)

I don't have Universal Storage installed, so my only point of reference for the non-liquid "hydrogen" resource is the compressor that comes with this Dibnah pack. That compressor tank holds 10kg of hydrogen for a total of 114,068 units, or 11,406.8 units per kilogram, or 0.00008766700564575516 kg/unit. This is slightly off (after correcting for my five orders of magnitude error) from my figure before which was based solely on the conversion rates being used by the electroliser, but I'll trust this one more.

So using this value for "hydrogen" kg/unit, 4450 units would be about .39kg, whereas we know that "liquid hydrogen" is exactly .4kg/unit. My current math (which should not be confused with correct math, lol) says that 0.4kg/unit of "liquid hydrogen" / 0.0000876670056457 kg/unit of "hydrogen" is a conversion ratio of 1:4562.72.

So now I have to go back over this thread and figure out where the 4450 number is coming from and why my 4562.72 number is probably wrong.

Regardless, we should make every effort to make the conversion factor between hydrogen and liquid hydrogen exactly correct from a conservation of mass perspective; any gameplay enhancing hand waving should take place on the power consumption and/or rate-of-water-conversion side of things.