Jump to content

PSA: Nuclear engine overheating


THX1138
 Share

Recommended Posts

Addendum:

With the central tank empty, and just the X200-8 tanks (partially) full (fuel only), the X200-8 tanks explode after 2.75 minutes of full power. The tanks were down to about 1/3 fuel remaining. So returning this Tug from Jool would require care (requiring short burns).

Note that the delta-V of the full Tug is about 6.7 km/s.

- - - Updated - - -

Ok, so then the issue wasn't so much about the nukes overheating, but of players trying to attach them in the old fashioned way?

No, they are still overheating and causing explosions in less time than I need for my typical burns. But the run time can be extended by putting the engines into good thermal contact with a large heat reservoir. And as you use fuel, the heat reservoir decreases in capacity.

Also, I avoided putting any heat sensitive components (such as batteries) on this test ship.

Edited by Brotoro
Link to comment
Share on other sites

Guess we just need to slightly decrease thrust when the fuel gets low, since lighter tanks get less thermal mass.

Ok, so then the issue wasn't so much about the nukes overheating, but of players trying to attach them in the old fashioned way?

Yep, seems like it's actually not that hard to deal with as long as your ship is designed correctly and the t/w ratio is limited, but that's not what nukes are for anyway. While I like the complexity of the new system, I also partly blame squad, since they hid the absolutely necessary tools in the debug menu.

edit: Brotoro isn't wrong, it's still more tricky to use nuclear engines than before. They probably stay the go-to thing for long range mission while requiring a lot more care. Wouldn't mind a small heat buff (decrease), however it's possible to deal with heat.

Edited by Temeter
Link to comment
Share on other sites

Ok, so then the issue wasn't so much about the nukes overheating, but of players trying to attach them in the old fashioned way?

IMO, it's more about having all this heat generation without any dedicated radiator parts to deal with it.

Something like the heat radiators from KSPI, both the radial foldout variety and the inline types.

I've flown plenty of missions that required extended Nerv burns (on the order of 30 to 60 minutes at full thrust) that will not be possible without radiator parts to get rid of that heat.

Link to comment
Share on other sites

A updated version of my 2.5-meter Standard Nuclear Tug can run 6.8 minutes at full throttle before exploding (the weak point again is the tank the nuke is attached to, of course). That's with the tanks tweaked for maximum fuel load and no oxidizer (it would be nice if Squad gave us the option to tweak a tank to LF-only now that we have fuel-only-nukes). Subsequent burns would be limited to shorter times, of course, as the thermal mass of the Tug decreases.

DEw888u.png

xMeNndi.jpg

This Tug (with no payload) has delta-V capacity of 4.7 km/s.

Link to comment
Share on other sites

Built a test ship that will run for 35 min at full throttle without overheating to critical failure (Tested using infinite fuel). The radiators get to the point where they are radiating away almost as much heat as they absorb from the tank. They run very hot but it heats up very slowly once the radiators near equilibrium. The didn't reach it but appeared to be getting quite close. 15 min run depletes all fuel without problems. No radiator version explodes before it empties half a tank.

Also building something docked in the VAB doesn't stop heat flow but, docking once launched sure does.

6xGqvDJl.jpg

Imgur album that I can't get to embed properly

Edited by Bill Zarr
Link to comment
Share on other sites

What are the settings on the engine? No, I'm not going to download it.

I wouldn't imagine you'd do so. You'd be more likely to write something of professional quality for yourself. ;) (that's really not sarcasm... sort of a backhanded compliment without the backhanded part)

Heat generation is reduced by a factor of 8 and the emissive constant is doubled.

Conductivity is also (theoretically) reduced to make it easier to use them with small tanks, but I need to play with that a bit more.

Link to comment
Share on other sites

Heat generation is reduced by a factor of 8
30? Easy mode. :P
emissive constant is doubled.
emissiveConstant should never be above 1.0. E: Matter of fact, it's probably clamped at 1.0. 1.0 is a perfect black-body, 0.97 ~ 0.98 is a much more realistic value.
Conductivity is also (theoretically) reduced to make it easier to use them with small tanks, but I need to play with that a bit more.
Good idea.
Link to comment
Share on other sites

emissiveConstant should never be above 1.0. E: Matter of fact, it's probably clamped at 1.0. 1.0 is a perfect black-body, 0.97 ~ 0.98 is a much more realistic value.

Thanks for the info, I blindly fumbled about and arbitrarily picked a number that seemed to work. Now I know why it worked. heh

Link to comment
Share on other sites

I probably won't use it, but thanks for helping out people who have trouble with heating.

Also, if someone didn't yet discover the temperature info setting or want a bit more comfort, regex made a small mod creating a switch for ksp's stock toolbar.

http://forum.kerbalspaceprogram.com/threads/117893-1-0-regex-s-Useful-Mod-Emporium

Edited by Temeter
Link to comment
Share on other sites

Also, if someone didn't yet discover the temperature info setting or want a bit more comfort, regex made a small mod creating a switch for ksp's stock toolbar.

http://forum.kerbalspaceprogram.com/threads/117893-1-0-regex-s-Useful-Mod-Emporium

Oh, hey! I like that!

Thanks, regex!

(And thank you Temeter for the heads up.)

Link to comment
Share on other sites

Huh, made an interesting discovery, no clue if its a new one. Tested some spreading with lots of small 1.25m jet engine fuel tanks in different configurations, and it showed some interesting things happening:

The radial heat spreading was stronger than the spread inside of a stack. I actually managed kill my engines and the connected tank at roughly the same time. Maybe pancakes are the future of nuclear travel? Spread is so good the efficiency between radial stacks (didnt tested deep stacks) and big tanks seems to be close to identical.

Might try to copy interstellar's endurance design. A stacked ring might actually be an efficient heat sink, provided the engines are in that ring.

Edited by Temeter
Link to comment
Share on other sites

For my part I found out an inactive KR-2L makes for quite the radiator. I tested

screenshot121.png

a slightly modified transfer stage for a previous Eve lander with a central half kerbodyne tank with a KR-2L flanked by six FL-T800s with nukes on 'em, set up to pull fuel into the nukes from the center so as to get the most delta-v out of it with an option to boost thrust if needed, with around 5:40 burn time in it, and managed 4 minutes 10 seconds at full throttle before stopping to prevent FL-T explosion.

a>

a>

a>

a>

over -22 thousand rad flux...'pretty sure this thing is keeping the rest of the ship alive at this point. The nukes themselves were only radiating around 8k.

a>

Note that the tank is only radiating 6.2k or so.

a>

nearly 31k rad flux.

In a previous un-screenshotted test, I ran it to destruction, pulling off a burn that wound up past Eeloo. That big engine bell turns out to make for quite the heat sink.

Edited by Archgeek
Something was and is quite wrong with my BB code markup.
Link to comment
Share on other sites

IMO, it's more about having all this heat generation without any dedicated radiator parts to deal with it. ...

First off; there's a related thread here.

- - - Updated - - -

Secondly; About the debug mode thermal color code; there's two ways do to it, true color and false color. In true color, as aforementioned, the visual emission spectrum for most metals is black (cool - relatively), red, orange, yellow, yellow-white, then white (some show blue before or after white), by which point most metals are soft enough to melt or break apart like warm cookies. The color order some have mentioned (the 'shower-head' model) is an artificial false-color model based on the most common instinctive associations between colors and temperatures. If there were simply an option to pick between the two schemes ('false-color' and 'true-color'), I think that that would solve the issue.

- - - Updated - - -

Is it me, or does this seem off.

pulled from config file.

"

// heatConductivity = 0.06 // half default

emissiveConstant = 0.8 // engine nozzles are good at radiating.

"

is the heatConductivity suppose to be commented out. Has anyone removed the first set of // and tested the engine?

Only the whole thing is, NERVA's never used radiator when the engine was running!

The fuel, liquid H2, was the engines coolant and the reactor ran at a steady GW output, where the fuel initially cools the reactor before it's becoming hot enough to become the propellant.

So a higher GW reactor needed a greater fuel flow.

Needing radiators in order to run a NERVA, like we now presumably need to, is rather ridiculous. In my opinion.

But if we are going to talk about Bimodal systems, where the reactor runs in power generation mode (10% of GW outpout), then I'm good with needing radiators. As a Brayton* generator can not convert all heat to ec anyway.

As far as NTRs are concerned, I think the heating is a fair balancing point (although, again, a dedicated radiator part would be welcome). The reasons being that LiquidFuel is essentially cryogenic kerosene (by density and other engine stats, IIRC) with the thermal properties of Aerozine 50 (the thermal info according to NathanKell). It's a super-fuel, nothing near cryogenic hydrogen in terms of storage and handling. Realistically, an NTR running kerosene would probably have a terrible isp. Also, while NTRs may actually be quite cool engines (they appear to be about mid-range in terms of chamber temperature) KSP glosses over the storage of cryo fuels, which require some sort of cooling mechanism in order to avoid boil-off.

I can't believe I'm saying this, but I feel the NTR heating is a balanced and interesting gameplay mechanic because KSP is not realistic. If this were Realism Overhaul the way the engine works now would be inexcusable.

A Nerva's shadow shield is built into the engine, right on top of the reactor core where the smallest size shield can be the most effective. I assume the KSP nuclear engine has a particularly massive shadow shield because the engine has a rather high mass.

If Squad wants to screw over the nuclear engine, they have plenty of realistic ways to do it. ...

They certainly didn't have to implement a faulty heat model that indicates and propagates misunderstanding of this particular rocket technology.

Also, while I totally agree with Brotoro and Regex about the whole balance/realism issue, that sounds like a separate thread topic. No one (Squad!) has to explain fission thermodynamics to model it accurately (or even approximately); yet knowing what design type the Nerv is would help clear up some things in the area of expected behaviors.

- - - Updated - - -

Aaaanndd... one more thing. What if we had thermocouples that could be attatched as struts or maybe radiator-like components. They would convert waste heat into electrical charge. Probably not enough by themselves, but with radiators?...

Link to comment
Share on other sites

I like the way they overheat now- I mean, not that they overheat, but how they do, and how the heat transfers to other parts. Much better than the old "just throttle down" way, which was simplistic.

FYI, I've done 10 minute burns with the new model and not hurt anything.

If anything, I just wish they would make bigger LF tanks, because you need to haul a ton of fuel before these make sense vs. A regular engine. To get 10k d/v with one engine and a command pod I needed 15 LF tanks.

Link to comment
Share on other sites

If anything, I just wish they would make bigger LF tanks, because you need to haul a ton of fuel before these make sense vs. A regular engine. To get 10k d/v with one engine and a command pod I needed 15 LF tanks.

The Mk3 LF-only tanks have pretty large capacity if you're looking for that. One large tank is better than many small ones heat-wise, a single big tank has much greater thermal mass than a small one as well as more room to attach radiators wings, and it's the tank attached to the engine that must absorb most of the waste heat.

Link to comment
Share on other sites

  • 1 month later...
This thread is quite old. Please consider starting a new thread rather than reviving this one.

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.

 Share

×
×
  • Create New...