Nertea

[1.3.0] Kerbal Atomics: fancy nuclear engines! (June 19)

633 posts in this topic

 

19 hours ago, Stikkychaos said:

I seem to have encountered an issue. I have both Atomics and NFE Patch for NFE. And the engine reactors read nothing, making engines useless :/
http://imgur.com/fsMSoc0

What to do with that?

Scratch that, it works in vacuum.

EDIT: Nevermind that, I still have the same problem, it simply occurs only on SSTOs for some reason. As seen on a screenshot below, I am in vacuum, with power, radiator, fuel (one tank on the engine for test) and yet still the reactor and the engine are not working. 

http://imgur.com/ecan47t

If you're using that patch, you need to stage the part in order to activate the reactor.

Share this post


Link to post
Share on other sites

Hi?

 

I discovering this impressive mod. I'm no quite sure how to use those engines in game. Here is what I gathered :

- Atomic engines uses LH2, which has to be cooled down with energy.

- The reactor has to be started to get any thrust. Thust is depends on the core temp.

- having LH2 running provide core cooling down.

- If core temp goes too high it stops and you don't have any thrust

My question is how do you control core temp when thrust is off ? When using this engine for landing, you don't always thrust and bear the risk of having an engine shutdown. ?

I seems that radiators don't do anything

Thanks for the advice.

Share this post


Link to post
Share on other sites

Using an engine for landing is going to be tough. You will need to have some amount of radiators to keep the reactor hot. If you think the radiators aren't working, ensure you have enough radiators (lots are needed for full heat operation) or make sure they are on... or within 1 part connection if you are using the stock static radiators)

1 person likes this

Share this post


Link to post
Share on other sites

Thx, I'll do some test on Kerbin launchpad.

Share this post


Link to post
Share on other sites

Now that I've made a little script to find configs with unequal numbers of braces and such, here I am continuing my tour. Just FYI - there is a missing curly brace in this file:

\KerbalAtomics\Patches\NTR\hydrogenNTRArk.cfg

2 people like this

Share this post


Link to post
Share on other sites

Been away from KSP for a while, but decided to get back into things a bit. I elected to not install the the NFE patch for Kerbal Atomics, given the struggles I had with the Emancipator last time. While not quite as nifty, it's way simpler and more fun (for me, at east). However, I noticed the Emancipator still won't let you transfer fuel into it without the NFE patch. I made my own little additional .cfg with a module RadioactiveStorageContainer, and it works fine now. @Nertea, were you planning on adding this in to the standard Emancipator and it slipped your mind, or is it intended to not be refuel-able by default?

Also, any thoughts on an IntakeAtm + LH2 (maybe even +oxidizer, just very high ISP?) atomic engine? I was looking at the LH2 mod for the Cutlass, and it seems like a similar small mod might be doable to convert one of them into a nuclear jet you could use on oxygen-less atmospheres.

Share this post


Link to post
Share on other sites

Posted (edited)

@Nertea, found a bug with the Cutlass LH2 mod from Kerbal Atomics, for MarkIV. The IntakeAir requirements for the engine are through the roof! They are at 845.566/sec. I imagine this is because the Lqd Hydrogen is at 140.928/sec. and the file has a ratio of 6:1 in favor of air. Uninstalling the mod it looks like it's supposed to be around 15.135 air, and 2.522 Liquid Fuel.

Should the ratio on the CutlassLH2 be changed from 1:6 (LH2:Air) to 10:1 or so? Not sure what the intent is for air consumption (and not exactly sure how ISP and fuel consumption are calculated, since the times I played with it gave me confusing results).

Edit: Added the following to the config mod to make the ratio 10:1 (Or more accurately, 1:0.1)

		@PROPELLANT[IntakeAir]
		{
			@ratio = 0.1
		}

Getting a quite predictable 14.093/sec for IntakeAir now. ISP remains at 4800. Seems roughly in line with the previous (though I have no idea what the proper real mixture is for liquid hydrogen and air for optimal combustion), so I'm going to leave this here.

I understand it's possible that one Cutlass should require 4 or more 2.5m intakes, but I'm going to pretend it's not.

Edited by AmpCat

Share this post


Link to post
Share on other sites
3 hours ago, AmpCat said:

@Nertea, found a bug with the Cutlass LH2 mod from Kerbal Atomics, for MarkIV. The IntakeAir requirements for the engine are through the roof! They are at 845.566/sec. I imagine this is because the Lqd Hydrogen is at 140.928/sec. and the file has a ratio of 6:1 in favor of air. Uninstalling the mod it looks like it's supposed to be around 15.135 air, and 2.522 Liquid Fuel.

Should the ratio on the CutlassLH2 be changed from 1:6 (LH2:Air) to 10:1 or so? Not sure what the intent is for air consumption (and not exactly sure how ISP and fuel consumption are calculated, since the times I played with it gave me confusing results).

Edit: Added the following to the config mod to make the ratio 10:1 (Or more accurately, 1:0.1)


		@PROPELLANT[IntakeAir]
		{
			@ratio = 0.1
		}

Getting a quite predictable 14.093/sec for IntakeAir now. ISP remains at 4800. Seems roughly in line with the previous (though I have no idea what the proper real mixture is for liquid hydrogen and air for optimal combustion), so I'm going to leave this here.

I understand it's possible that one Cutlass should require 4 or more 2.5m intakes, but I'm going to pretend it's not.

I... think that the Cutlass LH2 bit comes with CryoEngines? But yes good find, I will make that change.

On 3/19/2017 at 2:46 PM, AmpCat said:

Been away from KSP for a while, but decided to get back into things a bit. I elected to not install the the NFE patch for Kerbal Atomics, given the struggles I had with the Emancipator last time. While not quite as nifty, it's way simpler and more fun (for me, at east). However, I noticed the Emancipator still won't let you transfer fuel into it without the NFE patch. I made my own little additional .cfg with a module RadioactiveStorageContainer, and it works fine now. @Nertea, were you planning on adding this in to the standard Emancipator and it slipped your mind, or is it intended to not be refuel-able by default?

Also, any thoughts on an IntakeAtm + LH2 (maybe even +oxidizer, just very high ISP?) atomic engine? I was looking at the LH2 mod for the Cutlass, and it seems like a similar small mod might be doable to convert one of them into a nuclear jet you could use on oxygen-less atmospheres.

It's intended to be unrefuelable by default unless you're willing to put up with the enhanced difficulty of handling it. If you have the skills to add it to the engine though, fair's fair. 

For the second I usually say that anything atmospheric goes in Mark IV. probably not scoped for in here.

Share this post


Link to post
Share on other sites
On 3/18/2017 at 3:29 PM, Nertea said:

Using an engine for landing is going to be tough. You will need to have some amount of radiators to keep the reactor hot. If you think the radiators aren't working, ensure you have enough radiators (lots are needed for full heat operation) or make sure they are on... or within 1 part connection if you are using the stock static radiators)

I can vouch for this. I made a simple rocket to test the Emancipator for atmospheric landing and it takes a lot of heat sinks to run. I couldn't heat sink it enough to get it to run at any throttle other than zero or max for more than a second or two. That makes landing even more challenging. That and refueling the uranium core is frustrating. No plumbing for that kind of fuel! (I kind of liked having to send an engineer out to refuel nukes, but it gets old quick with an Emancipator)

Note: I've not tried that experiment with the NFE patch and recent updates to see if that behavior has changed. 

Share this post


Link to post
Share on other sites

Posted (edited)

14 hours ago, Nertea said:

I... think that the Cutlass LH2 bit comes with CryoEngines? But yes good find, I will make that change.

It's intended to be unrefuelable by default unless you're willing to put up with the enhanced difficulty of handling it. If you have the skills to add it to the engine though, fair's fair. 

For the second I usually say that anything atmospheric goes in Mark IV. probably not scoped for in here.

Sorry, I might have quoted the wrong source. I tend to download all your mods at once, and I might forgot where I got the mod for the mod. I love the LH2 Cutlass!

Could you give a quick summary of how you intended the Emancipator to run (NFE mod included), from an operational perspective? I found it almost unusable, but I admit I may thinking about the engine in a very different manner than it was intended to be used.

For example, how should it be configured at launch? What range of settings is it supposed to operate at? What sort of challenges and struggles is it supposed to be offering to the crew/player? I realize that this is supposed to be a challenging engine to use, but I want to put the LFE patch back in and at this point I don't know what is a purposeful challenge or just a bug that you'd like to know about. 

Edited by AmpCat

Share this post


Link to post
Share on other sites

So any NFE-patched reactor is supposed to be more difficult to use. That's why it's an optional patch and not on by default. Interestingly, the larger the reactor, the harder the reactor will be to operate, because it heats up so fast and requires so much cooling.

There's two real ways to operate the reactors, the simplistic, single-burn way, and the more complex way.

The simpler way is to use the engine like it would be in a RL mission, in a planned, specific fashion. Heat up the reactor a little bit before you start your burn, and cut the power when the burn is done. That way you get maximum performance for the period of operation. Once done, put the reactor in zero or low power mode until you need to execute another planned burn. Use RCS or secondary engines to make fine course adjustments.

The complex way (but really not that complex) to operate them is to have the reactor warmed up to near its nominal temperature before you make any maneuvers, then put it into a low idle at say 1% power. You will need enough radiators to support 1% of reactor power. This should keep the reactor core hot but generating a minimum of extra heat. When you want to make maneuvers, bring the throttle up - in engine mode the reactor will select whichever is higher, the main engine throttle setting or the reactor power setting, and adjust itself to match. So if you set the power level to 1%, then bring the throttle to 90%, the reactor will spool to 90%. When you cut power back to 0%, it'll return to that 1% mode. That will allow you to make smaller maneuvers and even use it for landing. 

You can see how in these situations the higher the reactor power, the 'harder' it is to use. The Emancipator has by far the highest heat generation at 1.125 GW, so you reasonably need ~11 MW of cooling to run it at 1%. That's 11 Large Thermal Control Systems! 

Honestly I will make a user guide eventually.

3 people like this

Share this post


Link to post
Share on other sites

For nuclear engines with the NFE patch, does the engine exhaust provide any cooling? Normally for a nuclear engine most of the heat should end up inside the propellant, or is the heat generation meant to be the waste heat beyond this?

Also, is there any chance of getting a couple reactors designed to operate ion engines? The sizes usually do not match up too well. Maybe one of the normal nuclear engines modded to use Xenon with much lower thrust, higher ISP? It would simulate an embedded nuclear reactor/electric engine combo as opposed to having it as separate, mismatched parts. 

Share this post


Link to post
Share on other sites
2 hours ago, MaxL_1023 said:

For nuclear engines with the NFE patch, does the engine exhaust provide any cooling? Normally for a nuclear engine most of the heat should end up inside the propellant, or is the heat generation meant to be the waste heat beyond this?

The exhaust provides 100% cooling. The difficulty people have is the part before a burn (when the reactor is hot but the fuel is not flowing) or directly after a burn.

2 hours ago, MaxL_1023 said:

Also, is there any chance of getting a couple reactors designed to operate ion engines? The sizes usually do not match up too well. Maybe one of the normal nuclear engines modded to use Xenon with much lower thrust, higher ISP? It would simulate an embedded nuclear reactor/electric engine combo as opposed to having it as separate, mismatched parts. 

So a... reactor designed to operate ion engines is 100% the point of NFE in general. Perhaps I'm not understanding you right?

Share this post


Link to post
Share on other sites

Posted (edited)

9 minutes ago, Nertea said:

So a... reactor designed to operate ion engines is 100% the point of NFE in general. Perhaps I'm not understanding you right?

I think that he just wants to have reactors that power massive amounts of ion engines to not provide very small amounts of thrust haha. 

Edited by HoveringKiller

Share this post


Link to post
Share on other sites
4 hours ago, Nertea said:

So any NFE-patched reactor is supposed to be more difficult to use. That's why it's an optional patch and not on by default. Interestingly, the larger the reactor, the harder the reactor will be to operate, because it heats up so fast and requires so much cooling.

There's two real ways to operate the reactors, the simplistic, single-burn way, and the more complex way.

The simpler way is to use the engine like it would be in a RL mission, in a planned, specific fashion. Heat up the reactor a little bit before you start your burn, and cut the power when the burn is done. That way you get maximum performance for the period of operation. Once done, put the reactor in zero or low power mode until you need to execute another planned burn. Use RCS or secondary engines to make fine course adjustments.

The complex way (but really not that complex) to operate them is to have the reactor warmed up to near its nominal temperature before you make any maneuvers, then put it into a low idle at say 1% power. You will need enough radiators to support 1% of reactor power. This should keep the reactor core hot but generating a minimum of extra heat. When you want to make maneuvers, bring the throttle up - in engine mode the reactor will select whichever is higher, the main engine throttle setting or the reactor power setting, and adjust itself to match. So if you set the power level to 1%, then bring the throttle to 90%, the reactor will spool to 90%. When you cut power back to 0%, it'll return to that 1% mode. That will allow you to make smaller maneuvers and even use it for landing. 

You can see how in these situations the higher the reactor power, the 'harder' it is to use. The Emancipator has by far the highest heat generation at 1.125 GW, so you reasonably need ~11 MW of cooling to run it at 1%. That's 11 Large Thermal Control Systems! 

Honestly I will make a user guide eventually.

Okay, that explains the challenges I was having. I was observing all those features, but thought they were bugs. I was assuming it would automatically handle the reactor power appropriately while I mess with throttle (kind of how it's magically handled without the NFE patch). This makes more sense, that you need to micro-manage reactor power to some degree. Also why my intended use for atmospheric flight (take off and landing, at least), would be problematic. Messing with variable throttles and trying to manage reactor power at the same time gets very, very difficult. It's almost like there should be some.. software or something managing that for us. :wink: 

Also, I've found with such heavy, high ISP engines, to really get the most use out of them you need huge amounts of fuel. Also not terribly conducive to trans-atmospheric use.

Share this post


Link to post
Share on other sites

Posted (edited)

4 hours ago, Nertea said:

So any NFE-patched reactor is supposed to be more difficult to use. That's why it's an optional patch and not on by default. Interestingly, the larger the reactor, the harder the reactor will be to operate, because it heats up so fast and requires so much cooling.

There's two real ways to operate the reactors, the simplistic, single-burn way, and the more complex way.

The simpler way is to use the engine like it would be in a RL mission, in a planned, specific fashion. Heat up the reactor a little bit before you start your burn, and cut the power when the burn is done. That way you get maximum performance for the period of operation. Once done, put the reactor in zero or low power mode until you need to execute another planned burn. Use RCS or secondary engines to make fine course adjustments.

The complex way (but really not that complex) to operate them is to have the reactor warmed up to near its nominal temperature before you make any maneuvers, then put it into a low idle at say 1% power. You will need enough radiators to support 1% of reactor power. This should keep the reactor core hot but generating a minimum of extra heat. When you want to make maneuvers, bring the throttle up - in engine mode the reactor will select whichever is higher, the main engine throttle setting or the reactor power setting, and adjust itself to match. So if you set the power level to 1%, then bring the throttle to 90%, the reactor will spool to 90%. When you cut power back to 0%, it'll return to that 1% mode. That will allow you to make smaller maneuvers and even use it for landing. 

You can see how in these situations the higher the reactor power, the 'harder' it is to use. The Emancipator has by far the highest heat generation at 1.125 GW, so you reasonably need ~11 MW of cooling to run it at 1%. That's 11 Large Thermal Control Systems! 

Honestly I will make a user guide eventually.

I love that complexity of these engines. While I'm pefectly fine with OP engines, it's nice to get some interesting, additional elements out of it. Bit disappointed Squad nerfed the stock nervas heat production after people complained about having to adapt.

(that said, I'm not sure that heat will meak the emancipator balanced either way, but I'm fine with that :D )

Edited by Temeter

Share this post


Link to post
Share on other sites

Posted (edited)

Played with this a bit more tonight. I wanted to get a bit more clarification on the user of reactor power for these. Is it purely to generate heat for the reactor? And I presume if the reactor is below nominal heat, thrust (and ISP?) is reduced?

Also playing with the 1% power setting. So far, seems that still causes issues when having one engine (like the Liberator) and one reactor (like the Garnet) on the same ship. Despite having 4MW of cooling, (which should be plenty for 1% on a Liberator, I think?), the Garnet still overheats and shuts down. I can drop the reactor power to 0% and sit there for a while, since the reactor won't cool.. but it seems with 4 large thermal control systems, I should be able to run at 1% and not overheat a Garnet.

Edited by AmpCat

Share this post


Link to post
Share on other sites

Posted (edited)

*snip*

Edited by Three_Pounds

Share this post


Link to post
Share on other sites
11 hours ago, AmpCat said:

Messing with variable throttles and trying to manage reactor power at the same time gets very, very difficult. It's almost like there should be some.. software or something managing that for us. :wink: 

Functionally, what is a fully automated reactor system? A basic engine module :wink:. At this point you just don't install the patch.

10 hours ago, AmpCat said:

Also playing with the 1% power setting. So far, seems that still causes issues when having one engine (like the Liberator) and one reactor (like the Garnet) on the same ship. Despite having 4MW of cooling, (which should be plenty for 1% on a Liberator, I think?), the Garnet still overheats and shuts down. I can drop the reactor power to 0% and sit there for a while, since the reactor won't cool.. but it seems with 4 large thermal control systems, I should be able to run at 1% and not overheat a Garnet.

My test didn't reveal anything here when I built a ship as you described. See here the Liberator running at 1%, 4 large thermal control systems (4 MW yes) and a Garnet running at full throttle. Complete steady state operation. Is your setup different?

94NPQQw.png

Share this post


Link to post
Share on other sites

Agreed on the 'don't supply patch', I was being silly. :)

I was noticing the problem after a launch burn in orbit, so the engine had been running. I was also using your NFE fixed shorter radiators that are rated for 1000KW of dissipation. And I had a fuel tank. I'll do some more testing this evening.

I also noticed that it seems the only purpose for reactor power at anything over 0% is to pre-heat the engine for a burn. Is this accurate? Since it'll automatically set reactor power to match the highest of throttle or reactor power, throttling up brings the heat up in the reactor pretty well on its own in use, just might be real inefficient for a few seconds as it warms up.

Share this post


Link to post
Share on other sites

Interacting with KSP's heat systems is very complex. Logs won't help in this case, but definitely exact reliable reproduction steps are needed to even make a start

Yes to some extent you can just leave the reactor throttle at 0% if you are willing to accept the inefficiency. The patch also lets the generators in the trimodal engines act like NFE reactors instead of being free energy. 

Share this post


Link to post
Share on other sites

What I meant was possibly a combined part ion engine + nuclear reactor with the reactor precisely sized to power the ion engine at the equivalent throttle levels. It would still generate waste heat, but would remove the requirement for a bunch of different parts and oversized or undersized reactors. It is not a big issue, just something I found annoying when trying to make outer-system craft. 

Share this post


Link to post
Share on other sites

Posted (edited)

Okay, test time! I broke this down into the simplest craft I could (almost). I even took the pretty lights and RCS off. Here are all the parts, from the bottom up:

  • Liberator
  • Octo-girder
  • landing legs from some Space-X sorta pack
  • micro Octo-girder
  • Garnet
  • Octo-girder command unit
  • an additional reaction wheel
  • a Mondo LH2 tank
  • 4x Thermal Control System (large)
  • 3.5m-2.5 adapters on either end of the tank
  • and 8 guidance fins for descent

Pic for reference:

[Old rocket pic deleted]

I'm a little worried about the thermal control, because I can't use them during flight, I don't think. Probably rip off. This may mean I can't actually test what I want to test.

Edit: Okay, changed the ship a bit cause that was not working. Lots of tumbling. Amazing what adding a nosecone (and a bit of weight) to the nose can do for aerodynamic stability. Added the Mk1-2 command pod on top. Also swapped from the Thermal Control System (large) to the VS-1000 from NFT (Results were the same either way, but these I don't have to deploy or worry about in atmosphere):

obLBzSy.png

I tried several runs, and got very repeatable results:

  1. Turn on reactors, both Liberator and Garnet
  2. Stage the engine at full throttle
  3. Turn on Ascent Guidance autopilot from Mechjeb because I'm lazy
  4. Watch pretty rocket lift off.
  5. After primary altitude burn, the throttle is cut to 0, but since the reactor is still at 100%, it quickly overheats and safeties shut it down.
  6. Set reactor power on the Liberator to 0%, then complete orbital insertion burn.
  7. Everything's fine. Now in orbit, reactors behaving.
  8. Set Liberator reactor power to 1%.
  9. Observe temperatures of the Liberator and Garnet reactors steadily climbing.
  10. Observe radiator dissipation skyrocketing.
  11. Alt-tab to document findings. Hear octo-truss holding Liberator explode in the background from the heat and detach the engine and landing gear from the rest of the rocket.

I think this matches my observations earlier, and should be fairly easy to reproduce. I'm not yet worried about landing, so you could probably skip the landing legs and just mount it on cinder blocks or something.

Interesting things noted:

  1. Trying to turn the reactor immediately back on after step 6 doesn't work. It quickly overheats again and shuts down. You have to wait until pretty high in altitude to turn it back on and not have the heat continue to go up. Basically, wait until just before orbital insertion burn to turn the reactor back on. I'm not sure if this is due to atmospheric friction on the way up (going over 1000km/s on liftoff by the time the engine is done with the burn), or something else.
  2. Once in space, even with the Liberator reactor set to 0%, the engine's internal temperature according to the Thermal Debug Window, continues to rise, heating up all the parts around it. This is despite what the 'core temperature' reads, which is dropping. Eventually the ocro-truss holding the engine gets above 2000K and explodes. All 4 radiators are still there, dissipating their 1MW of heat each, supposedly.
Edited by AmpCat

Share this post


Link to post
Share on other sites

Posted (edited)

Okay, here's another problem, which is even simpler to reproduce, and I think leads to the problem above:

  1. Put same rocket as above on the launch pad.
  2. Set Liberator reactor power to 100%
  3. Set throttle to 0%
  4. Turn on Liberator and Garnet reactors.
  5. Stage engine.
  6. Watch core temp on Liberator quickly skyrocket and auto-shutdown.
  7. Liberator starts to cool, Garnet starts to overheat.
  8. Garnet shuts down.
  9. Both core temperatures seem to be dropping.
  10. Hit Alt-F12 and bring up debug window. Go to Physics > Thermal, check Display Thermal Data in Action Menus
  11. Right click octo-truss around engine, watch its temperature slowly continue to rise.
  12. At some point, the internal temperature skyrockets up and you get the overheating warning bars. No idea why, but it's a dramatic change. It's like something else reached a thermal max and is suddenly dumping massive heat into other parts.
  13. Watch the temperature on the truss continue to quickly rise.
  14. Truss explodes, rocket falls over.

Things I'll try:

  1. Remove the Garnet and power with solar cells, see if it behaves differently
  2. Add a LOT more thermal dissipation, see if it behaves differently
Edited by AmpCat

Share this post


Link to post
Share on other sites

Three things:

  1. That's the Emacipator on your rocket picture, not the Liberator, 4MW is not enough to keep it cooled even at 1% for sure. Which reactor are you actually using?
  2. Is this with the latest version of NFE? IE from a few hours ago?
    1. Can you repeat your testing with the latest version if that wasn't the case?
  3. If 2.1 doesn't resolve it, can you trim off the other-mod parts and upload the craft file?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now