Eeloonium - A glowing extension of RoverDude's Kolonization mod

[Cairan]

Discussions have taken place in the MKS/OKS thread regarding the modification of RTG behaviour so that it is more in line with real world characteristics.

Specifically:

- RTGs require a special resource, thermal radioisotopes, produced either from nuclear reactor fission products or by irradiation of specific isotopes.

- RTGs do not have an on/off switch. The clock starts ticking the moment they are built.

- RTGs cannot be reloaded or serviced. The radioisotopes are often fused in ceramics and structurally integrated in the generators.

- RTGs provide power which decays exponentially with the half-life of their radioisotopes and the degradation of their electrical components.

Various posts covering the modification of RTGs within the MKS/OKS mod have led to the need to separate those discussions from the main MKS/OKS dev and release threads.

A quick recaps follows:

Yeah, on the consumable RTG conversion, I wonder if there should be a separate thread for that alone, but RD did say he would handle it.

I don't know how long it takes, as I've never been there, but I do know that Eeloo was mentioned. I think that that would be a good benchmark. A probe equipped with one RTG should be able to just make it too Eeloo and send back some data with energy from it's RTG. I also don't know how these things work in real life, but if it could be set to not "start reacting" or what ever, so that they can be "fired in sequence", that would greatly increase their effectiveness again. I just feel like it would make them a little broken again. If it were possible in real life, I wouldn't have anything to say about it, but if not, I'd rather not see this feature. I feel like if you want an RTG to reliably last longer then a voyage to Eeloo; then you need to build it closer to Eeloo (we should all be using EL here...)

That's not how RTG's work. Unlike a nuclear reactor where the release of energy is triggered by neutrons hitting nuclei, and can be throttled by inserting/removing control rods that absorb neutrons, RTG's rely on nuclear decay which is a purely random process internal to the nucleus. Half of the nuclei in a sample of Pu-238 will decay in every 87.7 year period. Half of the nuclei in a sample of strontium-90 will decay in 28.8 years. There's no starter. There's no throttle. They're less controllable than SRB's -- SRB's don't have an off switch. RTG's don't even have an on switch. There's no point in carrying a can of Pu-238 or Sr-90 with you to refuel, as it will decay at the same rate -- just bring 2 RTG's.

IMHO a physically plausible RTG factory module would involve a nuclear reactor (which could also provide power/heat/etc.) and an immediately adjacent chemical reprocessing module or three that would generate RTG's, xenon, and other products.

Just for the record, Voyager 1 and Voyager 2 are still up and running with their RTGs, after nearly 40 years out there...

It's not just a question of "depleting" the RTG, it's that the output from it goes down exponentially, faster at first and then slower. There is also metallurgical issues with the thermocouples, but a simple exponential curve would do... 1/2 the power output every 400 days, for example. That means your deep space probe can still do stuff, just not all at once or you need batteries to cope with higher loads (this is what New Horizon has done, the RTG provides baseload power, high power capacitors provide power for peak transient loads).

A quick module keeping track of RTG initialization game time and calculate the power at current time could easily set the proper energy output from the RTG. Americum 241 is the best candidate for realworld replacement of Plutonium 238 (which is getting scarce) but it has a quite a long halflife. For gaming purposes, the best realworld candidate is Polonium 210 (of Litvinenko fame...), with a halflife of 138 days.

Some reading material.

Also, I wouldn't use EnrichedUranium in an RTG, the most appropriate in-game resource would be DepletedUranium, if you want to stick to MKS/OKS resources, or Actinides, if you come from the KSPIverse... So if you build a probe using an EL launchpad, you could always use up the DepletedUranium in your PDUs to fuel the probe's RTGs. However, the proper way would be, to stay coherant with MKS/OKS resource flow, would be to separate ThermalIsotopes from DepletedUranium in the Lab module. Then you could put fresh ThermalIsotopes in your RTG, and the moment you hit the "Activate" switch, the clock of power decay would start. To provide easy feedback on how much power the RTG can put out, deplete the ThermalIsotopes exponentially and calculate the power output linearly based on the level of resource left.

A half-life of 900-1200 days would be about right for KSP with a standard sized solar system. For the various 6.4x and 3.2x sizes, you might need something more in the 2500-3500 day range.

Optimal transfer from Kerbin to Eeloo is 4y 184d of transit time. Jool is about 3y 263d of transit time.

4y 184d is almost 1900 days.

I propose the fuel for RTGs should be "Eeloonium" Removes any "real-world" discrepancies from actual physical properties of existing isotopes for obsessive-compulsive people (like me ), yet serves a clear purpose without falling in the way too complex situation of having the full periodic table of the elements in our resource manager display. We have Karborundum and Karbonite after all...

In other news, I'm thinking of writing a Kerbalized periodic table of the elements. Would be hillarious to see what Kerbals have to say about Francium.

I've already been fiddling with adding conversions to the OKS/MKS Lab modules to process the DepletedUranium to extract RareMetals from the EnrichedUranium/Recyclables path the lab's reprocessing function provides. Xenon gas is already removed at the reactor source by internal filters. So if we can have a consensus on fueling RTGs with Eeloonium (or Whateverium), I can integrate it in the revised lab reprocessing I've been working on...

As it stands right now:

- The Lab's Centrifuge function takes Uraninite and turns it into EnrichedUranium plus some Recyclables.

- EnrichedUranium goes into the MKS/OKS PDUs and undergoes fission. Fission yields XenonGas and DepletedUranium.

- DepletedUranium goes back in the lab and the Breeder conversion turns that back in half EnrichedUranium / Recyclables.

All we need to do to produce RTG-specific isotopes is tweak the RecipeInputs and RecipeOutputs so that now the Breeder does this:

* DepletedUranium + (copious amounts of ) Water + Chemicals -> Breeder -> EnrichedUranium + RareMetals + Eeloonium + WasteWater.

Explanation: The DepletedUranium is our proxy for the depleted nuclear fuel, with fission products. The water needed is to keep things cool, along with the chemicals needed to perform the extraction of the various isotopes from the waste. You end up with EnrichedUranium and Eeloonium (my proposed one-size-fits-all fictitious thermal radioisotope) from the PUREX method of Uranium / Plutonium extraction from spent fuel, along with some precious and rare metals. You also have tons of waste water to treat, but fear not, the existing Aeroponics module can grow plants to absorb and sequester the dangerous radioactive and heavy metals with the existing Purify function.

[Errol]

Due to reliance on USI parts for this process, perhaps this should be made as an "extension" to the USI mods? (I know I was the one pushing for a new thread in the first place....)

I'm remembering wombat conversions and the mks multi-purpose kolony modules. Something like that for these RTGs would be cool.

Edited February 22, 2015 by Errol

[Cairan]

Some ideas:

Alternatively, instead of creating a new resource like Eeloonium, maybe Karborundum could be a placeholder as a trace fission products generated in extremely tiny amounts in the PDUs, it already exists in RoverDude's Karbonite+ extension, and we could tweak the RTGs to require a very small amount of Karborundum to provide power. It would still be necessary to have a specific module to deal with the decaying power output, but no need for an RTG-specific resource.

- - - Updated - - -

De to reliance on USI parts for this process, perhaps this should be made as an "extension" to the USI mods? (I know I was the one pushing for a new thread in the first place....)

I'm remembering wombat conversions and the mks multi-purpose kolony modules. Something like that for these RTGs would be cool.

Indeed. That's what I'm thinking. It would relieve RoverDude of dealing with the specifics. With the KolonyConverter module, we have a lot of room for various conversions. I, or someone else, could whip up a decay module.

[Errol]

I think the price per mass for Karborundum very high....the tiny little sample you get in VAB costs 500 000 doesn't it? I wouldn't mind a new "one time" use resource.

[TCFirebird]

Is your primary goal to be able to build RTGs with EL/MKS? Depleting Eeloonium would be fairly easy to implement for RTGs built in the VAB (and wouldn't have MKS as a dependency). I think the challenge with EL/MKS is you'd have to allow filling it once, but then never again.

[Errol]

I think something similar used to exist in the MKS code actually. There used to be this special "MEP" type resource. There were three actually. They represented the specialized machinery that would be required to actually make each module serve a different function, the idea being that each module starts out as an identical shell with different decals. You needed one or two different types I think, in certain modules, some didn't need it, and the amounts were really low, but it had to be there to make the module function.

Somewhat unrelated, but similar. I never got around to messing with EL and MKS together back then, so I don't remember how that worked, but I assume you would have to go fully closed loop, produce the MEP required, and fill on site after building with EL. The same would be the procedure here, except it would be "refining" or what ever the correct verb for making radioisotopes is, rather then MEP production line.

EDIT: If an EL can fill any tank, couldn't we just give the RTGs a none pump-able tank/make Eeloonium none pump-able? That way if a user is playing with EL and builds a vessel with EL, we could generate a warning somewhere to remind him to fill his RTGs.

EDIT2: I just realized, doesn't this understanding of how an RTG functions preclude the the way the PDU actually functions in-game? It is described as a large RTG and a large Battery. Perhaps the description should be updated to list an Inverter as included with the PDU, so that it makes sense that it can be "throttled" through efficiency.

Edited February 22, 2015 by Errol

[TCFirebird]

Using Regolith and Module Manager, I know you would be able to give RTG a limited resource and have it start out active. You'd have to ask RoverDude if it's possible to disable the start/stop buttons using just a part config. That gets you most of the way there.

EL can only fill pumpable tanks. Non-pumpable tanks always start out full (solid fuel is the only stock resource like this). For EL/MKS integration, you'd have to release a modified version of EL (or get the author to modify it) to check for Eeloonium before building an RTG (as well as adding the production in MKS parts).

Decaying power generation would be the most difficult, and you would likely have to write your own resource conversion logic for that because Regolith can't handle conversion rates that change based on time.

[Cairan]

Using Regolith and Module Manager, I know you would be able to give RTG a limited resource and have it start out active. You'd have to ask RoverDude if it's possible to disable the start/stop buttons using just a part config. That gets you most of the way there.

EL can only fill pumpable tanks. Non-pumpable tanks always start out full (solid fuel is the only stock resource like this). For EL/MKS integration, you'd have to release a modified version of EL (or get the author to modify it) to check for Eeloonium before building an RTG (as well as adding the production in MKS parts).

Decaying power generation would be the most difficult, and you would likely have to write your own resource conversion logic for that because Regolith can't handle conversion rates that change based on time.

Indeed, I think Regolith + MM is going to be the way to go. RoverDude is probably going to get involved, I already talked with him about adding some refinements to the breeder output and he was interested. The keyword here is: do not spawn too much new resources.

If only we could specify a conversion rate in terms of percentage of resources instead of absolute values, this thing would be done in an afternoon.