[1.12.x] Kerbal Atomics: fancy nuclear engines! (August 18, 2024)

[Nertea]

I should have mentioned this in the changelog. I did not make NFE integration for this engine, as I need to do some things in order to make it work "properly", namely I need to extend support in NFE to engines with core temperatures of >6000K (the Emancipator should have 20,000K or something). All the mods were already staged for deployment so I only killed the patch.

If you want to refuel it, you can patch ModuleRadioactiveStorageContainer into it. 

 

Also, If you are experiencing problems with contracts:

The problem is with additional files in CryoEngines and/or KerbalAtomics. Download the absolute latest versions (0.4.1 and 0.3.1 respectively) and make sure to delete the old version of the mod!

If you do not want to update or something you can go into GameData/CryoTanks/Plugins and delete everything except SimpleBoiloff.dll

[Nertea]

@AmpCat, @Mikki, are you using the NFE integration patch from NFE Extras? It looks like it should work... 

[Mikki]

1 hour ago, Nertea said:

@AmpCat, @Mikki, are you using the NFE integration patch from NFE Extras? It looks like it should work... 

Yes, now i`ve found the folder in KerbalAtomics Extras, tested in orbit and anything okay.
 Great! Thank you!  I installed only gamedata first... 

[septemberWaves]

I've encountered an issue with CryoTanks. It's not actually an issue with the mod itself; the problem is that cryogenic tanks continue to drain power at high time warp speeds, whilst solar panels seem to no longer work past a certain time warp speed (I think it's 10x warp, though I may be mistaken). This causes the spacecraft to run out of energy quickly during warp and then gradually lose hydrogen.

[Nertea]

2 hours ago, eloquentJane said:

I've encountered an issue with CryoTanks. It's not actually an issue with the mod itself; the problem is that cryogenic tanks continue to drain power at high time warp speeds, whilst solar panels seem to no longer work past a certain time warp speed (I think it's 10x warp, though I may be mistaken). This causes the spacecraft to run out of energy quickly during warp and then gradually lose hydrogen.

Please accompany any bug reports related to boiloff with the following:

• A copy of output_log.txt from the KSP_Data or KSP_x64_Data folder (depending on whether you're using 32 or 64 bit)
• A screenshot of the Cryo Debug window (Press CTRL+SHIFT+P) while the issue is occurring
• An indication of what timewarp level causes the bug
• KSP version and mod version.
• A clear screenshot of the craft so I can see what parts are being involved
• The .craft file if you're not using too many mods (strip what you can, it'll help)

This info has been added to the OP. 

Edited October 23, 2016 by Nertea

[AmpCat]

@Nertea I forgot to put that back in after the recent upload. Works now!

Any thoughts on the manufacturer's tabs not filling in NFT categories? I know it's a small thing, just not sure if it's my build, a bug, or not yet implemented.

[septemberWaves]

Sorry about the inadequate bug report, I was hoping it would be a common issue with a known solution. Since it's not, I'll do some testing later to see if I can either replicate the problem or find a solution.

[Nertea]

10 hours ago, eloquentJane said:

Sorry about the inadequate bug report, I was hoping it would be a common issue with a known solution. Since it's not, I'll do some testing later to see if I can either replicate the problem or find a solution.

No worries, boiloff is a tricky beast with many edge cases because of KSP's resources system, and we're far into difficult territory. Good information is the only way I can deal with fixing the problems.

16 hours ago, AmpCat said:

@Nertea I forgot to put that back in after the recent upload. Works now!

Any thoughts on the manufacturer's tabs not filling in NFT categories? I know it's a small thing, just not sure if it's my build, a bug, or not yet implemented.

For manufacturers' tabs to appears, they must actually be defined as an AGENCY, last time I checked. There are no agencies defined for any of my things at the moment, and I'm not really interested in doing that work without making new logos, and I would rather spend that time elsewhere at the moment. 

[SebastianPahlsson]

If I don't want the boil off feature, how do i delete it / turn it off?

 

[AmpCat]

On 10/24/2016 at 1:11 PM, Nertea said:

No worries, boiloff is a tricky beast with many edge cases because of KSP's resources system, and we're far into difficult territory. Good information is the only way I can deal with fixing the problems.

For manufacturers' tabs to appears, they must actually be defined as an AGENCY, last time I checked. There are no agencies defined for any of my things at the moment, and I'm not really interested in doing that work without making new logos, and I would rather spend that time elsewhere at the moment. 

Gotcha. No worries. Since it was in a previous version, thought I just messed something up.

Edit: Maybe another mod added agencies for NFT?

Edited October 26, 2016 by AmpCat

[AmpCat]

@Nertea I ran into an interesting situation today when playing with these engines and asymmetric rockets. I was going to use the thrust limiter (and eventually, have a mod like TCA do it automatically), but I noticed if I reduced the thrust limiter, the engine quickly overheated and shut down. After looking at the stats, I realized I was never going to be able to add enough heat sinks to overcome the lack of thrust venting for keeping the reactors cool, but now I'm stuck. Is there some way to reduce the reactor heat output as you reduce the throttle limit? I tried lowering the power setting and it didn't seem to make any difference. I drop the throttle limit by just a couple percent and the engine does an over-heat shutdown.

Is this designed to be an all or nothing design, or is there a way to get thrust limiting working?

Edit: In thinking about this and the design of gas core reactors, it stands to reason the reactor runs at a certain temperature to get the efficiency needed. Thus, it requires a certain amount of exhaust to maintain that temperature without getting too hot. This implies if you want to lower the thrust, you also probably need to lower the core temperature in general, which means the engine runs less efficient. Not sure if that level of complication is desired, but it's an interesting thought.

Edited October 27, 2016 by AmpCat

[Nertea]

3 minutes ago, AmpCat said:

@Nertea I ran into an interesting situation today when playing with these engines and asymmetric rockets. I was going to use the thrust limiter (and eventually, have a mod like TCA do it automatically), but I noticed if I reduced the thrust limiter, the engine quickly overheated and shut down. After looking at the stats, I realized I was never going to be able to add enough heat sinks to overcome the lack of thrust venting for keeping the reactors cool, but now I'm stuck. Is there some way to reduce the reactor heat output as you reduce the throttle limit? I tried lowering the power setting and it didn't seem to make any difference. I drop the throttle limit by just a couple percent and the engine does an over-heat shutdown.

Is this designed to be an all or nothing design, or is there a way to get thrust limiting working?

This might be an edge case that needs to be addressed. It looks like reactor power setting can be overridden in engine reactors by the vessel global throttle, so if you are running the vessel throttle at 100% but manually setting power to 50%, it'll still run the reactor at 100%. It doesn't take engine thrust limiting into account, which it should - running the vessel throttle at 100% with a thrust limiter of 70% going on should automatically give you a reactor power of 70%. 

[Temeter]

Hey, Nertea: Your 1.25m nuclear engines have bugged plumes (nerva patch, stubber, smaller trimodal engine), the internal plume trail is at 45 degree.

Noticed it ingame and then just checked everything in my clean install:

Also btw, those cryogenic plumes look awesome in movement!

[Nertea]

14 minutes ago, Temeter said:

Hey, Nertea: Your 1.25m nuclear engines have bugged plumes (nerva patch, stubber, smaller trimodal engine), the internal plume trail is at 45 degree.

Noticed it ingame and then just checked everything in my clean install:

 

Also btw, those cryogenic plumes look awesome in movement!

Weird, I'll ensure that's fixed in tomorrow's update.

[Nertea]

Update 0.3.2 with many fixes

• Updated CRP to 6.0.1
• Fixed exceptions in VAB when using a ZBO tank for holding only non-cryo fuels
• Fixed incorrect dry masses for ZBO tanks
• Improved NFE compatibility for Emancipator engine
• Fixed plume offset for 1.25m engines

[AmpCat]

@Nertea I don't suppose this will address the thrust limit issue? Or is that something I can address myself?

[Nertea]

7 minutes ago, AmpCat said:

@Nertea I don't suppose this will address the thrust limit issue? Or is that something I can address myself?

That needs some head-scratching work in NFElectrical. You reported that issue about an hour after I deployed the latest version of that mod unfortunately so it didn't make it in - if it had been a bit earlier I would have held off. Now it's a full release to fix a small edge case which is a bit vexing. Maybe early next week. 

[AmpCat]

Oh, come on, an hour is plenty! 

No problem. I think for my asymmetric design I probably will want much smaller engines off center for balancing thrust anyway. Thinking of space planes, where in an atmosphere they are fine, but in space the engines/fuel aren't aligned. I'd like more design freedom, so I'm pondering putting balancing thrust engines off center as the CG changes. Some of your smaller cryo engines might be enough.

[Nertea]

13 minutes ago, AmpCat said:

Oh, come on, an hour is plenty! 

No problem. I think for my asymmetric design I probably will want much smaller engines off center for balancing thrust anyway. Thinking of space planes, where in an atmosphere they are fine, but in space the engines/fuel aren't aligned. I'd like more design freedom, so I'm pondering putting balancing thrust engines off center as the CG changes. Some of your smaller cryo engines might be enough.

Haha, if you clone and build the dev branch it *might* work, it's pretty poorly tested though. 

[AmpCat]

Heh. Thanks, but I'll wait. I do enough of my own software to know better. I still need to do some more testing on single Emancipator engines.

[AmpCat]

Sigh. I'm running into a lot of problems with the Emancipator.

It goes back to the thermal management issue. Namely, if the throttle isn't on full, it's not getting enough venting to keep the reactor cool, and the engine shuts down. This means in normal rocket flight, once I've got enough velocity to reach the altitude I want, I shut the engines off, then circularize the orbit. Only now, when I kill the throttle, the engine dies. This means to get thrust again I have to go full throttle, then go back and turn the reactor back on.

It's complicated further using reactors for power. The high heat causing the engine to shut down also shuts down reactors. You can't keep enough thermal management (without running exhaust) to turn the reactor back on, and your ship runs out of electricity, which means now you can't control the throttle if you're on drone control. This gets you into a catch-22 where you're dead in space. Some quick clicking of buttons and hitting full throttle before the heat shuts you down again can sometimes get you back running, if you're fast.

In summary, this means you have to control your throttle by toggling the reactor on and off. I've tried this with several of the KA engines, with similar results. Not sure if you ran into this yet or not, but it's more than just a throttle limit issue.

Edited October 29, 2016 by AmpCat

[Nertea]

There's a few things you can do to make it easier. 

Using the ship throttle overrides the reactor throttle. So you can set the reactor's throttle to 5% or something, then ramp up the ship throttle to 100% - that'll automatically set the reactor throttle to 100%. When you're done, you can kill throttle and the reactor throttle will automatically scale back to 5%. 

You can also increase the emergency shutdown temperature to give you more buffer. A reactor will only start taking damage once it's above the Critical Temperature, which is usually 5-15% higher than the nominal temperature (shown in VAB, for the emancipator it's 2000 K higher). You can safely increase the cutoff until that to give more flexibility (this will also net you a higher Isp)

Additionally, you're choosing probably the hardest reactor to work with :P. It has a heat generation of 1.12 GW! That means it heats up very fast (will probably address that in the next update, it should have more thermal inertia) and can't really be passively cooled by radiators. Even at 1% throttle you'd need 3-4 of the largest HC radiators. The other engines are far more forgiving in their heating rate and capacity to be cooled by a radiator part. 

If you're using reactors for power, there is a specific power setting that gets you max power with the engine. For the Nepture (1.25m) it's... 65 kW. So set the reactor throttle to something that gives you around 65 kW of heat (1-2% normally). That amount of heat is auto-handled by the built-in radiator. For the Poseidon (2.5m) it's 800 kW, also handled by the radiator. 

It's good to have feedback though, this patch has always been pretty experimental. 

[AmpCat]

Thanks for the explanation. It seemed a bit odd that the engine was almost unusable. I figured I was doing something wrong. That being said, it's probably helpful if it worked 'out of the box' so to speak, without fiddling.

I ran a lot more testing today, starting with the lower temperature engines. It seems that if I just start out at launch setting the reactor to 0%, then firing up the engines, it performs perfectly fine. I can throttle the engine as I want, it quickly reaches operating temperature, I can shut it off and turn it back on with no issues with the reactor overheating.

The problem starts to come back, however, when you add another reactor in the mix. Adding something like the 'KerbPower' fission generator in, and if it's running, it will create enough heat to bump the Poseidon or whichever engine over the threshold, causing a shutdown. As you pointed out, bumping the shutdown threshold up a little seemed to help, and for some reason I can't explain, doing so didn't seem to cause the engine to actually run that hot on its own, at least with the reactor set to 0%. I think I noticed something saying 'Actual: 10%' in the KA control panel. So, not sure if there's some other temperature regulation going on there.

I also may not have been clear in my last post when referring to using reactors for power. I was adding a stand-alone reactor, like the KerbPower.

All this being said, a few points:

1. What's the point of running at anything but 0% reactor power? Only seems to affect temperature if the thrust is set to 0, but as soon as thrust is applied, the engine warms up fast anyway.
2. While I love the neat modeling of engine dynamics, I'm also a fan of 'It just works'. I think some tweaking of standard settings is in order so that the engines work fine as is when installed and the rocket launches. I presume you're working on this for future patches, but I thought it worth noting.
3. I'm still a little confused on how the temperature is working for these. Maybe I need to do some more testing, but it seemed to be that even though I set the reactor safeties at 6200K on the last run with the Emancipator, to try and get around the second reactor problem, the engine would still keep itself at 6000K, until I shut the throttle off. At that point it would slowly creep up a little as the other reactor was adding heat to the mix. It would only go down if I turned the other reactor off. This was while having 200kW of active radiators mounted on the ship as well. Should have been more than enough for the little KerbPower.
4. I think if you ran multiple engines (my SSTOs are often quad Liberators.. soon to be Emancipators!) this thermal issue would get exacerbated. Not to mention to trickiness of managing the reactor online state of 4 reactors at once. Have to put it into an action group.
5. Edit: Another quirk is that if you're at 100% thrust, reactor at 0%, there's a note that says 'Effective: 100%'. I presume that means reactor output is tied to the throttle. So if I run out of fuel and flame out, reactor is still at 100%, so it shuts all the reactors on the vessel down (including my power reactors) as it quickly overheats. More of an edge case, but thought I'd note it.
6. Also, indeed, setting the temperature safeties higher than optimal temperature, the engine still runs up to optimal, and not up to the safety. Maybe having the auto-shutdown above optimal by some margin, a standard setting as well.

I haven't actually tried the 1.25m engines yet. I might try the Nerv next, then maybe a cluster of them.

Edited October 30, 2016 by AmpCat

[Nertea]

I think you need to provide me with some screenshots of your setups. For many of your points, it's hard to say whether it's an edge case, a quirk of stock thermal mechanics or actually a bug in my stuff. 

1 hour ago, AmpCat said:

• What's the point of running at anything but 0% reactor power? Only seems to affect temperature if the thrust is set to 0, but as soon as thrust is applied, the engine warms up fast anyway.

It won't affect temperature but it will affect heat and power produced. A reactor running at 50% power will take more time to hit "optimal" temperature. When it gets there it will only produce 50% of power/thrust but will only need 50% of radiator/engine capacity. 

1 hour ago, AmpCat said:

While I love the neat modeling of engine dynamics, I'm also a fan of 'It just works'. I think some tweaking of standard settings is in order so that the engines work fine as is when installed and the rocket launches. I presume you're working on this for future patches, but I thought it worth noting.

What defines "just works" for you though? How much interaction is possible? What interesting things do you do with the system, and how do you interact with it? If you want it to just generate thrust at high efficiency when you run it, there's no point to having the engine simulation, and you should probably just not install the integration patch. If you want something in-between, we've got to think hard about that works. 

1 hour ago, AmpCat said:

• I'm still a little confused on how the temperature is working for these. Maybe I need to do some more testing, but it seemed to be that even though I set the reactor safeties at 6200K on the last run with the Emancipator, to try and get around the second reactor problem, the engine would still keep itself at 6000K, until I shut the throttle off. At that point it would slowly creep up a little as the other reactor was adding heat to the mix. It would only go down if I turned the other reactor off. This was while having 200kW of active radiators mounted on the ship as well. Should have been more than enough for the little KerbPower.

Changing the engine safety won't change the goal temperature, it'll change the turn-off temperature. It'll still try its best to stay at the nominal temperature. This kind of situations would really need screenshots for a good analysis of what's going on. 

It *sounds* like there was not enough heat consumption from the thrust to cool both the engine reactor and the power reactor, and therefore the other reactor's cooling was being cannibalized by the engine. Reactors can't really heat each other up, they can only consume radiator capacity. It's hard to tell though without screenshots. 

1 hour ago, AmpCat said:

• Edit: Another quirk is that if you're at 100% thrust, reactor at 0%, there's a note that says 'Effective: 100%'. I presume that means reactor output is tied to the throttle. So if I run out of fuel and flame out, reactor is still at 100%, so it shuts all the reactors on the vessel down (including my power reactors) as it quickly overheats. More of an edge case, but thought I'd note it.

It's probably a good idea to have an auto-shutdown if the engine runs out of fuel. I can implement that. 

1 hour ago, AmpCat said:

1. Also, indeed, setting the temperature safeties higher than optimal temperature, the engine still runs up to optimal, and not up to the safety. Maybe having the auto-shutdown above optimal by some margin, a standard setting as well.

As mentioned, that's the emergency shutdown. There's no effective way to set the "target" temperature and keep the system stable in the stock thermal paradigm. 

[AmpCat]

I can certainly provide screenshots. It's easy enough to replicate on a launch pad with a very simple rocket.

So, from the top, a KerbPower reactor, a 12K battery, 4x 50kW radiators, a probe core, and then a cryotank, and the Poseidon engine.

I'm learning how these engines work, so pardon if I contradict myself from time to time. Sometimes I make assumptions, then prove myself wrong.

Here's an example of the engine at 0% reactor power and 100% effective power as it warms up. This is where I was suggesting it doesn't make much sense to set the reactor power to anything except 0%, since it scales with the throttle.

Here I throttled down to 13%, and it wasn't enough to keep the reactors cool, so the power reactor (KerbPower) shut down:

The radiators should be more than enough to keep the KerbPower running. I think they're swamped from the Poseidon.

I'm not sure what other screenshots would be helpful. Let me know if there's something in particular you'd like to see. I first noticed this in space, but it's easy enough to replicate on the ground.