[1.0.5] Atomic Age - Nuclear Propulsion - Red Hot Radiators

[Nothalogh]

@Porkjet

I have a request, there is a massive gap in performance between the KANDL and the LANTR, is there a plausible nuke design that fits in between the two?

[Fraz86]

It looks like CKAN has CTT designated as a required dependency. Since CTT isn't actually required (and designating it as such forces the user to install it), please change it to suggested or recommended.

[Rhedd]

Here's what I've been running. Targets any engine with a mode that runs on LiquidFuel without Oxidizer or IntakeAir, so it picks up stock LV-N, Atomic Age, and the giant nuke in the Taurus HCV pack.

// Convert any engine that runs on straight LF to LH.

@PART
[*]:HAS[@MODULE[ModuleEngine*]:HAS[@PROPELLANT[LiquidFuel],!PROPELLANT[Oxidizer],!PROPELLANT[IntakeAir]]]
{
    @MODULE[ModuleEngine*]:HAS[@PROPELLANT[LiquidFuel],!PROPELLANT[Oxidizer],!PROPELLANT[IntakeAir]] {
        @PROPELLANT[LiquidFuel] {
            @name = LqdHydrogen
        }
    }

    // If an engine that has passed the selectors for LF-only also has an
    //  LFO mode, assume that it's a LANTR and convert that mode to hydrolox.
    //  We'll use the ratio of 10u LqdHydrogen to 1u Oxidizer from Nertea's
    //  Cryo Engines.
    @MODULE[ModuleEngine*]:HAS[@PROPELLANT[LiquidFuel],@PROPELLANT[Oxidizer]] {
        @PROPELLANT[LiquidFuel] {
            @name = LqdHydrogen
            @ratio = 1.0
        }
        @PROPELLANT[Oxidizer] {
            @ratio = 0.1   
        }
    }
}

That script is nice, and I tried it because these really should be running on LH2 and it bugs me that they don't, but the engines don't perform as they should if you just convert them straight like that.

I want to discuss the Nuclear Lightbulb engine, because it's exactly what I imagined using to send my manned mission to Duna. After converting it to use LH2, however, I found that EVERY engine out-performed it. A simple Skipper burning stock RP-1/LOX was far superior in every way, and if I compared it to other engine mods, like the Cryogenic Engines, it was insane to even consider it. It's the type of engine I wanted to use, however, so I did some testing (using Interstellar Fuel Switcher) to find out why it sucked so bad...

I built a craft with one pod, one tank, and a Nuclear Lightbulb running the original RP-1 (liquid fuel). It weighed 29.12t, and engine performance was: TWR=1.58, Dv(vac)=2469.

The exact same ship with the tank switched to LH2 and the Lightbulb modded to use it was just pathetic, but that's to be expected since the best real-world information I can find says you can fit over 11 times the mass of RP-1 in the same space as LH2, so I added more LH2 tanks until the ships were the same total weight. This hydrogen-using ship weighed 29.22t, using six of the tanks used on the RP-1 ship to get there, and engine performance was: TWR=1.57, Dv(vac)=1099.

So, simply converting the engine to use hydrogen instead of liquid fuel lost me 0.01TWR, removed a whopping 55.5% of my Dv, and as an added bonus made my rocket SIX TIMES larger! That's the exact opposite of good.

Compared to a stock conventional engine, it gets even worse. The exact same ship using a Skipper engine instead of the Lightbulb weighs only 12.2t, with engine performance of: TWR=5.47, Dv(vac)=1456.

According to a NASA proposal document on NTP (Nuclear Thermal Propulsion), an engine like this could be expected to have 100%+ Isp over a conventional engine, and even with the Lightbulb's awesome starting Isp of 1500, that means we'd have to multiply it by 2.65 (if using LH2) to get that sort of performance advantage over a Skipper.

Now I don't even want total realism, but, in NASA's words, nuclear propulsion opens up the possibility of practical manned missions to the outer solar system, so I at least want that to be true in my game. Clearly, this just won't do.

I know this mod natively uses liquid fuel, so any discussion of hydrogen is off in left field, but even using the stock fuel it's performance isn't nearly better enough to warrant using it over another engine that weighs much less and provides far higher TWR. And if we do discuss modding it to use H2, why does the same weight of hydrogen give less than half the Dv of liquid fuel, anyway? That's one I can't figure out no matter how I look at it.

Anyway, fuel types is not something I know much about, really, so... thoughts anyone?

[Starwaster]

The exact same ship with the tank switched to LH2 and the Lightbulb modded to use it was just pathetic, but that's to be expected since the best real-world information I can find says you can fit over 11 times the mass of RP-1 in the same space as LH2, so I added more LH2 tanks until the ships were the same total weight. This hydrogen-using ship weighed 29.22t, using six of the tanks used on the RP-1 ship to get there, and engine performance was: TWR=1.57, Dv(vac)=1099.

Welcome to reality. Welcome to the rocket equation.

A lot of people just don't get that you have to look at propellant mass, not volume. This comes up all the time. Remember that performance increases are per kilogram of propellant.

Your d-v losses re probably due to increased tank dry mass and you should look into drop tanks. Also make sure you're using cryogenic tanks and think about using procedural parts if you need larger tanks. Better to have a few large tanks rather than lots of little ones.

[billkerbinsky]

Welcome to reality. Welcome to the rocket equation.

A lot of people just don't get that you have to look at propellant mass, not volume. This comes up all the time. Remember that performance increases are per kilogram of propellant.

Propellant velocity is another very significant factor - and that's going to be higher when the propellant is a low-molecular-weight gas. (Of course, the way this should be reflected in-game is with a decrease in ISP when running atomic engines on LF instead of LH2).

Another area where KSP diverges from reality is in the dry mass of tanks - real-world tanks are lighter.

[Starwaster]

Propellant velocity is another very significant factor - and that's going to be higher when the propellant is a low-molecular-weight gas. (Of course, the way this should be reflected in-game is with a decrease in ISP when running atomic engines on LF instead of LH2).

Another area where KSP diverges from reality is in the dry mass of tanks - real-world tanks are lighter.

Agreed, the problem there is that engines allow you to configure them in unrealistic configurations.

On the other hand, the alternative requires a sophistication that arguably is better suited to simulation than a game. It's easier the way it is now even if it allows you to set up engines that don't make sense.

Tanks on the other hand are readily fixed. If you haven't tried it, Real Fuels does so pretty well.

[Randazzo]

Anyway, fuel types is not something I know much about, really, so... thoughts anyone?

Well, I don't know about all that, but my current Jool-bound vessel weighs in at around 180 tons and gets just shy of 10 km/s dV using the lightbulb and fuelswitched tanks to carry only LF.

[Rhedd]

Welcome to reality. Welcome to the rocket equation.

A lot of people just don't get that you have to look at propellant mass, not volume. This comes up all the time. Remember that performance increases are per kilogram of propellant.

Your d-v losses re probably due to increased tank dry mass and you should look into drop tanks. Also make sure you're using cryogenic tanks and think about using procedural parts if you need larger tanks. Better to have a few large tanks rather than lots of little ones.

Fuel mass, yep. Figured that part out quickly.

Interesting suggestion that I lost Dv because of the dry tank weight. I hadn't considered that, as I was only considering the total wet mass of the ship. Clearly as the fuel gets used the effects of dry tank weight would become more pronounced. Still, each tank weighs 0.5t, so the difference in ship dry weight between the RP-1 and LH2 ships is only 2.5t. I'm not doing math, here, but I can't imagine that a dry weight difference of 2.5t out of a wet total of 29.2 would account for an over FIFTY PERCENT drop in Dv. Am I wrong? Does that sound likely?

When you say "make sure you're using cryogenic tanks", is that a Real Fuels thing? I've looked in to that mod and it sounds cool, but I use an embarrassing number of part mods, and unless it adapts all tanks automatically it means a lot of work on my part, setting up .cfgs for all the bits. I have to do it to put nifty Hot Rocket effects on all my engines, I'm not looking to increase that workload.

Note that all of my values were taken from a simple test ship to observe the difference in Dv when switching fuels. I would never try to build an interplanetary ship with six tiny tanks!

Edited May 27, 2015 by Rhedd

[iVG]

Hey @Porkjet ,

Any plans on an open cycle design to accompany the lightbulb ? Would be interesting to see some kind of a trade off in terms of temperature/thrust but higher ISP. Tell us what you think!

[xtoro]

Just wanted to say that I absolutely love this mod and use the engines often! Good work!

[sashan]

weight difference of 2.5t out of a wet total of 29.2 would account for an over FIFTY PERCENT drop in Dv. Am I wrong? Does that sound likely?!

Welcome to the world of maths man! Now you know the effect of logarithm function in rocket equation. It's realy weird until you grasp it, I'm working with logarithm scales all the time in the university now.

I've faced similar problem with my configs. Ended up dumping hydrogen and just using bimodal NTR with afterburner option. Worked really nice with spaceplanes as well.

[woot]

I use the Lightbulb engine on most of my missions lately, and have noticed a small issue with it: it gimbals in response to control input before it has been activated, often clipping through its fairing even in the early stages of a launch. Is there a way to arrange so it doesn't start gimballing until the rest of the stack below it has been staged?

[RyanRising]

I use the Lightbulb engine on most of my missions lately, and have noticed a small issue with it: it gimbals in response to control input before it has been activated, often clipping through its fairing even in the early stages of a launch. Is there a way to arrange so it doesn't start gimballing until the rest of the stack below it has been staged?

Do you have the Stock Bug Fix modules installed? That sounds like the ModuleGimbalFix, which allows engines to gimbal at all points during a flight so that they'll work when activated by right-click. If you don't have those installed, nevermind.

[woot]

Do you have the Stock Bug Fix modules installed? That sounds like the ModuleGimbalFix, which allows engines to gimbal at all points during a flight so that they'll work when activated by right-click. If you don't have those installed, nevermind.

Ah yes indeed! That likely explains it. Thanks.

[Probus]

Porkjet,

I've added Atomic Rockets to the Engineering Tech Tree: http://forum.kerbalspaceprogram.com/threads/125055, if you want to update CKAN dependencies.

[Terwin]

I suspect you are aware of this already, but in case you are not, the Heat Management AHMS and large radiators attached to my Atomic Age lightbulb engine did not heat up or use any coolant before the engine burned up at 3000 kelvin when I updated to 1.0.4.

I suspect that it is related to the engine not transferring heat properly with the new model and thought it might be useful for you to be aware of this.

(note: in a newer design the parts connected to the stock active radiators are heating up more than the heat sink attached directly to the engine, so that also points to a heat-transfer failure in the engine)

Love the engines, hope they get updated to the new heat transfer model soon!

[Starwaster]

I suspect you are aware of this already, but in case you are not, the Heat Management AHMS and large radiators attached to my Atomic Age lightbulb engine did not heat up or use any coolant before the engine burned up at 3000 kelvin when I updated to 1.0.4.

I suspect that it is related to the engine not transferring heat properly with the new model and thought it might be useful for you to be aware of this.

(note: in a newer design the parts connected to the stock active radiators are heating up more than the heat sink attached directly to the engine, so that also points to a heat-transfer failure in the engine)

Love the engines, hope they get updated to the new heat transfer model soon!

Use the new stock radiators

[futrtrubl]

(note: in a newer design the parts connected to the stock active radiators are heating up more than the heat sink attached directly to the engine, so that also points to a heat-transfer failure in the engine)

That is the way the stock radiators work, they take heat from any part that is over a threshold. No heat transfer failure.

[Randazzo]

I suspect you are aware of this already, but in case you are not, the Heat Management AHMS and large radiators attached to my Atomic Age lightbulb engine did not heat up or use any coolant before the engine burned up at 3000 kelvin when I updated to 1.0.4.

I suspect that it is related to the engine not transferring heat properly with the new model and thought it might be useful for you to be aware of this.

(note: in a newer design the parts connected to the stock active radiators are heating up more than the heat sink attached directly to the engine, so that also points to a heat-transfer failure in the engine)

Love the engines, hope they get updated to the new heat transfer model soon!

Neither mod has been updated for the new heat mechanics. You could use the stock radiators, which at present totally break the heatsink functions in favor of a far less effective function, but I highly doubt it will be able to keep up with the Lightbulbs truly massive heat production.

[Starwaster]

Neither mod has been updated for the new heat mechanics. You could use the stock radiators, which at present totally break the heatsink functions in favor of a far less effective function, but I highly doubt it will be able to keep up with the Lightbulbs truly massive heat production.

Depends on how many you use and whether you actually want to keep pace with heat production or just slow it down enough to keep the engine from destroying itself before you complete your burn.

Also note that ONLY the extendible radiator panels are active. They pull heat from anywhere on the ship and have low conductivity and pass about 1w/kW of heat.

The static panels are not active and only transfer heat from the part they are attached to due to their high conductivity.

The lightbulb engine has very low conductivity so if the Heat Management mod uses passive parts that require the engine to be passing heat using the stock thermodynamic system then it definitely won't work. (the stock active panels actively look for parts that are overheating)

[Randazzo]

The temptation to derail is strong, but suffice it to say I'm fully aware of how the heat management parts work, in what way they got broken, and why they can't manage the lightbulb's heat (at the moment).

Edited June 24, 2015 by Randazzo

[Terwin]

Depends on how many you use and whether you actually want to keep pace with heat production or just slow it down enough to keep the engine from destroying itself before you complete your burn.

Also note that ONLY the extendible radiator panels are active. They pull heat from anywhere on the ship and have low conductivity and pass about 1w/kW of heat.

The static panels are not active and only transfer heat from the part they are attached to due to their high conductivity.

The lightbulb engine has very low conductivity so if the Heat Management mod uses passive parts that require the engine to be passing heat using the stock thermodynamic system then it definitely won't work. (the stock active panels actively look for parts that are overheating)

The design I was talking about has 2 large and 8 medium radiators. When they got above 1600 degrees they were radiating at almost 100% of their transfer rate which seems to max out at a flux of ~200K total (50K each for large, 12.5K each for medium, I suspect they hit their max temp at ~1700 degrees, after which they are radiating heat out as fast as they can suck it in)

The lightbulb was still heating up while over 1800 degrees, but the radiators were pulling out a good sized chunk of the heat that was being produced, greatly extending the time to get to dangerous temps.

On the plus side, the new radiators do not seem to collide with each other when extended, so I can probably put on more if they turn out to be needed.

[AlexTheNotSoGreat]

A really nice mod for only adding 3 parts in my opinion.

One recommendation from me: Make the names look more stock, i.g. "PB-NUK 'Candle' Radioisotope Rocket," not "KANDL Radioisotope Rocket."

[ChrisSpace]

Can you please please put this mod on CKAN?

Also, I might be doing a mod review video on this soon-ish.

[RoverDude]

Neither mod has been updated for the new heat mechanics. You could use the stock radiators, which at present totally break the heatsink functions in favor of a far less effective function, but I highly doubt it will be able to keep up with the Lightbulbs truly massive heat production.

The temptation to derail is strong, but suffice it to say I'm fully aware of how the heat management parts work, in what way they got broken, and why they can't manage the lightbulb's heat (at the moment).

Sorry, but with all due respect you are completely incorrect.

I've been doing some streaming and use the Nuclear Lightbulb as my main interplanetary engine with some very long burns, and have no issue using it with (in most cases) a pair of the medium deployable radiators. I expect if I wanted some insane burn times I could swap to four mediums or even the large ones.

The stock radiators are more than capable of handling the heat generation of this engine.