[WIP] Nert's Dev Thread - Current: various updates

[Nertea]

Thanks, it's appreciated

NFE Test 5, with fixes to temperature scalars and the MX-4. Also fixes the high timewarp Ec generation bug. 

[Psycho_zs]

This time switching crafts leads to a more explody consequences. A lot more heat is being transferred to ship parts on non-focused vessel.

[Nertea]

:(. Can I get a shot of your test ship please? 

[Streetwind]

Today: NF Propulsion.

Checked:
- All lithium tanks have proper stats
- All argon tanks have proper stats
- One xenon tank has wrong mass, rest ok - that tank actually had a wrong mass previously as well, which I never noticed. Oops!
- One engine had the wrong thrust (you slipped into the next column when reading the table, which happened to have a similar but not identical number )
- VASIMRs properly obey atmospheric pressure

I'm prepping a big pull request with fixed for the tank mass and the engine (thrust, power and heat). Also because I am that kind of person I tweaked all the engines' power ratio number ever so slightly to make the numbers displayed ingame end up exactly as the spreadsheet says.

After you pull that in, Propulsion looks ready for primetime, unless someone else manages to find something I overlooked.
 

I can try help test reactors in timewarp tomorrow, but for today I'd like to get the pull request done before I need to head off to bed.

[Nertea]

Thanks Streetwind - I'll accept the pull when it comes in. This makes balancing much faster

Don't bother about the reactors. Talked with Rover and it's kinda expected behaviour for the way we're handling it, which sucks (reactor plays catchup upon reload, and if HeatProduction > low at equilibrium, explosion could happen). So I'll need to do a number of experiments to try to find a way around it. 

[Streetwind]

There you go. Still learning how github works, but I eventually managed to get all 19 changes into one single pull request

[Psycho_zs]

Just in case, my test ship is here:

http://pastebin.com/yXBbD5xV

It has cooling capacity for MX-1 at 60% throttle, uses a couple of parts from these mods:

[Moderator removed defunct website links]

Edited August 28, 2022 by James Kerman
link removed

[Nertea]

2 hours ago, Streetwind said:

There you go. Still learning how github works, but I eventually managed to get all 19 changes into one single pull request

Merged your commit. Will do a few tests but I'll probably add that to the pile of NFSolar and NFConst that are "done".

2 hours ago, Psycho_zs said:

Just in case, my test ship is here:

http://pastebin.com/yXBbD5xV

It has cooling capacity for MX-1 at 60% throttle, uses a couple of parts from these mods:

https://kerbalstuff.com/mod/148/Modular%20Rocket%20Systems%20-%20Parts%20Pack

https://kerbalstuff.com/mod/1052/Lithobrake%20Exploration%20Technologies

Thanks, might be useful.

What do we think about nuclear reactors producing small amounts of Xe? Since I have no plans for adding ISRU Xe production, it could be a decent way to help improve the economics of using Xe engines in the late game.

Edited December 17, 2015 by Nertea

[Psycho_zs]

Re: Xenon

Why not...

[Captain Sierra]

3 hours ago, Nertea said:

What do we think about nuclear reactors producing small amounts of Xe? Since I have no plans for adding ISRU Xe production, it could be a decent way to help improve the economics of using Xe engines in the late game.

I feel like the rate at which xenon fueled ships would burn through the stuff makes it basically a drop in the bucket. By the point at which you're using reactors to power them, you have so many engines running that I dont think it would make much of a dent, especially since many of us shut down reactors when not in use to save core life. Maybe with a DS4G powered rig it could be genuinely helpful given the insane Isp there, but I think at the end of the day it wouldnt prove very practical.

For engines as efficient as xenon thrusters, why would you even need ISRU refuelling options? Yeah its expensive but if you can afford the engines, reactor(s) and the first fill-up of the ship, you're likely rolling in so much dough it doesnt even matter anymore. Unlike NTRs its not worth it to carry the ISRU rigging out somewhere so you only need the one-way dV, so the necessity for xenon ISRU is sorely lacking.

 

That said, I certainly think its cool conceptually and would love to see it tested out.

[Streetwind]

7 hours ago, Nertea said:

What do we think about nuclear reactors producing small amounts of Xe? Since I have no plans for adding ISRU Xe production, it could be a decent way to help improve the economics of using Xe engines in the late game.

Hmm... define "a small amount". Ostensibly, you could change the conversion of EnrichedUranium into DepletedFuel so that you get less than 100% DepletedFuel, and make up the difference by mass with xenon instead. The question is, is that even useful? If memopry serves, a MX-L has about 5 tons of fuel on board, and takes 6 to 8 Kerbal years to burn through it at 100% reactor throttle. If 5% of the mass you get back is xenon, that means you'll be getting about 250 kg xenon in total... or somewhere between 32 and 42 kg per Kerbal year. That doesn't really sound like it would help.

You do of course have the option of just fudging the numbers and have the reactors produce some xenon on the side, conservation of mass be damned. Not sure that that is the most elegant solution. Though ultimately I'd like to see some form of dedicated xenon refueling method, and here's why.

Quoting myself from an earlier post:

On 9.12.2015, 19:01:08, Streetwind said:

Xenon: 40 funds/kg fuel; xenon tanks: 2.0-2.5 funds/kg fuel.
Argon: 5.886 funds/kg fuel; argon tanks: 1.4-2.8 funds/kg fuel.
Lithium: 0.506 funds/kg fuel, lithium tanks: 4.21 funds/kg fuel.

From a pure gameplay design standpoint (and ignoring the price of other components for the moment), this suggests:
- Argon has reasonably priced fuel in reasonably priced tanks. It's a middle of the road option in every way; it's predestined to be something you use for expendable missions since neither the propellant nor the tankage will break the bank.
- Lithium has very expensive tankage, but the fuel is basically free. It's predestined to be employed on reusable vehicles, which pay just the price of the fuel and recover all the expensive components after the mission is complete. The fact that lithium tends to produce low Isp only underlines this.
- Xenon tanks don't cost very much, but the fuel is absurdly expensive. It's predestined to be used on vehicles that carry only small amounts, and refuel themselves along the way. The fact that xenon tends to produce high Isp only underlines this.

Now of course in practical application, it isn't this clear-cut and easy, especially for a mod that's essentially in-space-only, and at the same time basically begs you to add extremely expensive components like reactors to your ship. But it still gives at least a sense of how the differences in fuel pricing can be used to motivate and reward certain player behaviors, and make the different options feel more distinct. As such, it feels kind of backwards to me that lithium tanks are hard to recover while xenon is the one fuel with no dedicated ISRU refueling option right now.

 

At the same time though, such a thing might be best left for 1.1 - it's about time we finished the 1.0.5 updates, don't you think?

[Nertea]

Well, I'd love to finish them, but that bug of Psycho's is mod breaking in the extreme, and I can't figure out any solution so far, despite 3 hours of testing last night. 

Reactor loads, tries to play catchup, adds time*5000kW of heat in one frame, blows up the everything. Have tried many things to stop it. Few options left. Frustration sets in again. 

Edited December 17, 2015 by Nertea

[Streetwind]

Bleh. Large amounts of bleh. Test version 5 runs just fantastically for me, with the exception of the timewarp thing >__<

How come stock ISRUs do not have this catchup problem? They also continue working while unfocused IIRC... (Honest question, I don't know enough to understand the difference.)

Also, what are the chances that 1.1 will handle it better? Maybe you can at least let people have it at their own risk until then. As long as you zero out your reactor throttle every time you leave a vessel by itself, it should work fine, right? (Yes, I know that's not a good solution.)

Edited December 17, 2015 by Streetwind

[Kobymaru]

2 hours ago, Nertea said:

Well, I'd love to finish them, but that bug of Psycho's is mod breaking in the extreme, and I can't figure out any solution so far, despite 3 hours of testing last night. 

Reactor loads, tries to play catchup, adds time*5000kW of heat in one frame, blows up the everything. Have tried many things to stop it. Few options left. Frustration sets in again. 

Could RoverDude's USI Reactor config work you, potentially?  He is the guy who wrote it, after all...

[Psycho_zs]

Could you reproduce this bug?

By the way, sometimes I see something similar with stock small ISRU. When craft is focused, it heavily starves on ore (drilling too slow), so works only at fraction of power. But when returning from other craft, it instantly shuts down from core overheat.

[Nertea]

I can repro it 100% with great ease now. That's unfortunate but I guess not the worst thing, it would be bad if it was very hard to reproduce.

Rover told me that stock ISRU does not suffer from this problem because the heat production is low at a stable goal temperature. Looking at the stock ISRU, this is somewhat true, but *some* of the NFE reactors have really low heat productions and still cause just as many problems.

 

[Streetwind]

2 hours ago, Kobymaru said:

Could RoverDude's USI Reactor config work you, potentially?  He is the guy who wrote it, after all...

It can work, sure. It's not as nice as Nertea's work though - it shows less info in the rightclick menu, it gives the player less control, and it doesn't simulate a reactor in as much detail. It's certainly an option to use this if the timewarp problem can't be fixed in 1.0.5, but given both options I'd prefer with Nertea's plugin over the USI implementation.

Also, I don't know enough about the modules used to write a config for NF reactors in that vein, unfortunately.

 

[Nertea]

It would have some problems but it would be simple enough. I have a few more ideas to try tonight/tomorrow morning.

However, it does look like all is good in terms of NFP, and I assume besides the low MaxTemps, HC is ok too. In fact I'm going to say that NFE, despite the reactor timewarp problems, is also ready to go. Yay!

[Starbuckminsterfullerton]

On 12/11/2015, 6:52:31, Nertea said:

They may need some changes to their stats. It's just a fuel swap right now. LH2 tanks are almost certainly going to come back as either part of a dedicated LH2 NTRs mini-pack, or as an optional addon to CryoEngines. 

I've been away awhile so there's a lot to catch up on here in the dev thread, but lithium is a great addition to come back to. I can't wait to put it to work.

If it were me, I'd re-task CryoEngines as a general Hydrogen mod, and put the NTR's, the relevant H2 conversion MM patches for other NTR's and ISRU, and the LH2 tanks under one roof. I do hope those tanks return; I modeled my LFO tanks to match their form factor! 

Edit: Reactor Xe production sounds like a good solution, but since most people only warm up reactors when the engines are about to burn I have a hard time imagining that it would make much of a dent in fuel time.  Might work well on a fuel station though.  

What if the reprocessor produced it? 

Edited December 18, 2015 by Starbuckminsterfullerton

[Captain Sierra]

On 12/11/2015, 9:52:31, Nertea said:

part of a dedicated LH2 NTRs mini-pack

I remember a good while ago you had some fantastic paper sketches of bigass rocket engines, some NTR-esque, others not. I may dig through the dev thread for them. At the time you were sufficiently frustrated with NFE problems (the neverending story 'round here ...) and wanted to model something out as unrelated to electrical crap as you could get. Stand by, will report back after some extensive digging.

[Streetwind]

9 hours ago, Nertea said:

(...) and I assume besides the low MaxTemps, HC is ok too.

I've not looked at that at all, but I will give it a quick look-over... probably tomorrow. Company christmas party eating up my entire evening today.

One thing I did notice is that the ModuleManager DLL shipped with it is outdated

[Nertea]

18 hours ago, Starbuckminsterfullerton said:

I've been away awhile so there's a lot to catch up on here in the dev thread, but lithium is a great addition to come back to. I can't wait to put it to work.

If it were me, I'd re-task CryoEngines as a general Hydrogen mod, and put the NTR's, the relevant H2 conversion MM patches for other NTR's and ISRU, and the LH2 tanks under one roof. I do hope those tanks return; I modeled my LFO tanks to match their form factor! 

Edit: Reactor Xe production sounds like a good solution, but since most people only warm up reactors when the engines are about to burn I have a hard time imagining that it would make much of a dent in fuel time.  Might work well on a fuel station though.  

What if the reprocessor produced it? 

That last one is actually a pretty good idea. 

9 hours ago, Streetwind said:

I've not looked at that at all, but I will give it a quick look-over... probably tomorrow. Company christmas party eating up my entire evening today.

One thing I did notice is that the ModuleManager DLL shipped with it is outdated

Well if only the bugs were all as easy to fix as that one :P. 

13 hours ago, Captain Sierra said:

I remember a good while ago you had some fantastic paper sketches of bigass rocket engines, some NTR-esque, others not. I may dig through the dev thread for them. At the time you were sufficiently frustrated with NFE problems (the neverending story 'round here ...) and wanted to model something out as unrelated to electrical crap as you could get. Stand by, will report back after some extensive digging.

You probably mean these two, that are the models that came from those sketches. Also infinitely on hold until I get any motivation. 

[RedParadize]

thats sad, they are quite nice.

[Psycho_zs]

Awesome models!

[Nertea]

https://www.dropbox.com/s/8ap4qmcmm175z2l/NearFutureElectrical0_6_T6.zip?dl=0

Please work :(. I'm kinda at my wits end. This is a bad hack. For the first few seconds after loading a vessel the reactor is forced to generate no heat (while the stupid catchup occurs). This means there will be a big drop in temp suddenly and it will spool back up later. 

It seems to work in my limited tests and is I guess okay.... I haven't seen any overheats at least.

Edited December 20, 2015 by Nertea