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

[Captain Sierra]

Note that HyperEdit has some minor bugs, such as Kerbals clipping into the ground of planets they land on after their ship gets HyperEdited into orbit.

But, as it seems you're content with experimental builds, so it probably won't irk you that much.

Meanwhile, I'm grabbing X.4.12 and checking if any issues I saw before have survived the bugspray...

I am very much tolerant of bugs in alpha builds. Squad gave up that tag with 1.0, so I am very much going to hold that to a much much higher standard. This, certainly not so much.

[Nertea]

NTR patches are getting a significant change. Appreciate comments on this paradigm:

• All patched NTRs get dual fuel modes in LF and LH2
  
• LF NTRs lose ~ 30% Isp (example LV-N gets 575s)
  
• LH2 NTRs gain ~5-10% Isp (example LV-N moves to 875s)
  
• Some exceptions: Kommitz's pack already uses 500-600s as baselines for NTRs, for example, so it stays around there for LF but gets an additional fuel mode at ~900s
  
• Haven't fully decided how to handle dual mode NTRs (eg Porkjet's LANTERN), but they might just get LH2/OX and pure LH2 (small question here - what is the ratio for O2:LH2 in a bimodal NTR? I assume it's not the same as combustion ratios)
  

Well, if you get bored with hyperedit and want to try ground-based launches again, two tips:

- To the traditional litany of "more struts, more boosters", add "more fins.".

- If you're using NFP reactors in your deep-space stage, put them close to the nose at launch time, and put less-dense fuel tanks behind them. Managed to get a ship built around the amusingly excessive 10m-sphere LH2 tank into orbit this way.

Amusingly excessive was exactly my design goal with that . Might rework it to be inflatable later, but for now... heh.

[Captain Sierra]

NTR patches are getting a significant change. Appreciate comments on this paradigm:

• All patched NTRs get dual fuel modes in LF and LH2
  
• LF NTRs lose ~ 30% Isp (example LV-N gets 575s)
  
• LH2 NTRs gain ~5-10% Isp (example LV-N moves to 875s)
  
• Some exceptions: Kommitz's pack already uses 500-600s as baselines for NTRs, for example, so it stays around there for LF but gets an additional fuel mode at ~900s
  
• Haven't fully decided how to handle dual mode NTRs (eg Porkjet's LANTERN), but they might just get LH2/OX and pure LH2 (small question here - what is the ratio for O2:LH2 in a bimodal NTR? I assume it's not the same as combustion ratios)
  

On my hack patch, I'm using the same fuel ratio as your cryo tanks (since that is my method of fuel switch). Realistically the H2 is just really hot, not in any way burnt, so the chemical ratio should be the same as a normal combustion ratio (unless they're going for only partial combustion).

My 0.02 would be leave the LF-LH2 dual mode to the stock NTR (and maybe KSPX as well). Most of the other NTR mods I know that are popular have specific intended roles (Porkjet's are a bimodal and ultra-high efficiency) The stock is most common for SSTOs, which is where I see that LF/LH2 toggle being most useful.

[hoojiwana]

(small question here - what is the ratio for O2:LH2 in a bimodal NTR? I assume it's not the same as combustion ratios)

A quick poke about finds these numbers:

• At LOX/LO2 Propellant Ratio 0.0: Isp = 941s / 925s / 891s. Tankage mass ratio: 14.0%. Engine T/W Ratio 3.0.
  
• At LOX/LO2 Propellant Ratio 1.0: Isp = 772s / 762s / 741s. Tankage mass ratio: 7.4%. Engine T/W Ratio: 4.8
  
• At LOX/LO2 Propellant Ratio 3.0: Isp = 647s / 642s / 631s. Tankage mass ratio: 4.1%. Engine T/W Ratio: 8.2
  
• At LOX/LO2 Propellant Ratio 5.0: Isp = 576s / 573s / 566s. Tankage mass ratio: 3.0%. Engine T/W Ratio: 11.0
  
• At LOX/LO2 Propellant Ratio 7.0: Isp = 514s / 512s / 508s. Tankage mass ratio: 2.5%: Engine T/W Ratio: 13.1
  

The three ISP numbers are for different temperatures, 2900, 2800 and 2600 K.

From http://www.astronautix.com/craft/lannbase.htm and http://www.nss.org/settlement/moon/LANTR.html. They both have the same numbers so presumably use the same source.

Edited May 15, 2015 by hoojiwana

[Streetwind]

Okay, there's some issues here. The radiators doln't have a start/stop button in their rightclick menus, and they don't obey being switch off in the VAB (they always come out active to the pad). However, the action group on/off toggle works just fine.

...And then there's the Tau radiator, which does have the menu button, but since the radiator is nonfunctional I can't test it.

Heatpipe limiter works as advertised. Radiator capability display is correct. Renamed standard radiators present.

M-EXP colliders for radiator attachment off

Toggling radiators doesn't update heatConductivity to the correct value

Tau radiator is nonfunctional (no flux of any kind)

Tau radiator has toggle panels buttons in the rightclick menu even though it is not toggleable

3.75m insulator untextured

[Nertea]

A quick poke about finds these numbers:

• At LOX/LO2 Propellant Ratio 0.0: Isp = 941s / 925s / 891s. Tankage mass ratio: 14.0%. Engine T/W Ratio 3.0.
  
• At LOX/LO2 Propellant Ratio 1.0: Isp = 772s / 762s / 741s. Tankage mass ratio: 7.4%. Engine T/W Ratio: 4.8
  
• At LOX/LO2 Propellant Ratio 3.0: Isp = 647s / 642s / 631s. Tankage mass ratio: 4.1%. Engine T/W Ratio: 8.2
  
• At LOX/LO2 Propellant Ratio 5.0: Isp = 576s / 573s / 566s. Tankage mass ratio: 3.0%. Engine T/W Ratio: 11.0
  
• At LOX/LO2 Propellant Ratio 7.0: Isp = 514s / 512s / 508s. Tankage mass ratio: 2.5%: Engine T/W Ratio: 13.1
  

The three ISP numbers are for different temperatures, 2900, 2800 and 2600 K.

From http://www.astronautix.com/craft/lannbase.htm and http://www.nss.org/settlement/moon/LANTR.html. They both have the same numbers so presumably use the same source.

Thanks!

Okay, there's some issues here. The radiators doln't have a start/stop button in their rightclick menus, and they don't obey being switch off in the VAB (they always come out active to the pad). However, the action group on/off toggle works just fine.

...And then there's the Tau radiator, which does have the menu button, but since the radiator is nonfunctional I can't test it.

Heatpipe limiter works as advertised. Radiator capability display is correct. Renamed standard radiators present.

M-EXP colliders for radiator attachment off

Toggling radiators doesn't update heatConductivity to the correct value

Tau radiator is nonfunctional (no flux of any kind)

Tau radiator has toggle panels buttons in the rightclick menu even though it is not toggleable

3.75m insulator untextured

Ok, how do I reproduce this bug? My test was launch ship (Mk1-2+MK-1+4xGR-1), turn on reactor, observe conduction, open radiators, observe conduction. I didn't see any issues at this point.

I don't really care about the colliders for the M-EXP at the moment

[Starbuckminsterfullerton]

NTR patches are getting a significant change. Appreciate comments on this paradigm:

...

Amusingly excessive was exactly my design goal with that . Might rework it to be inflatable later, but for now... heh.

Switchable is a nice addition, will keep people from complaining about ruining their ssto's and whatnot.

As far as LANTR's go, you're pretty much just hucking LOx into the exhaust stream, so any ratio below whatever you are using for a stoichio mix should be fine.

And yeah, with that fuel tank I think you now hold the honor of largest balls in the modding scene.

[Streetwind]

Ok, how do I reproduce this bug? My test was launch ship (Mk1-2+MK-1+4xGR-1), turn on reactor, observe conduction, open radiators, observe conduction. I didn't see any issues at this point.

1.) Launch new vessel

2.) Turn on reactor with radiators still retracted

3.) Observe conductive backflow from radiators into reactor

4.) Extend radiators

5.) Observe conductive backflow from radiators into reactor

6.) Quicksave and Quickload

7.) Observe conductive backflow from radiators into reactor: it is now different than in 5.!

8.) Retract radiators

9.) Observe conductive backflow from radiators into reactor: it is now different than in 3.!

10.) Quicksave and Quickload

11.) Observe conductive backflow from radiators into reactor: it is now the same as in 3. again, and no longer the same as in 9.!

12.) Extend radiators

13.) Observe conductive backflow from radiators into reactor: it is now the same as in 5. again, and no longer the same as in 7.!

The radiator has the config-specified heatConductivity of 0.001 only when extended. When retracted, it is something else (possibly the default 0.12, but I can't say for sure).

When spawning onto the pad, the radiator is retracted, and thus has the wrong, higher heatConductivity value. When extending the radiator, it should be using the correct, lower heatConductivity... except it doesn't update on toggling, so it continues using the higher one from when the radiator was closed.

As soon as you quicksave and quickload, the appropriate value is loaded and conductive backflow to the reactor drops dramatically (by as much as 90%).

If you then retract the radiator again, then once more it does not update, and keeps the low heatConductivity from when it was extended.

Quicksave/quickload again, and the retracted radiator gains its old, higher heatConductivity value back and conductive backflow to the reactor increases dramatically.

Here's a screenshot: https://dl.dropboxusercontent.com/u/44754370/screenshot52.png

As you can see, two GR-4 radiators on the exact same reactor have vastly different conductive flux values. That's because I forced one of them into the high conductivity state by closing it, saving/reloading, and then extending it again. If I want to I can fix it again by once more saving/reloading now that it is extended, in order to load the correct value.

[Fraz86]

The new heat pipes are an improvement, but they're still too effective when used to make do-it-yourself radiators. On the spacecraft below, the reactor stabilizes around 788K, with the hottest part of the "radiator" at 1477K. You may wonder how the radiator got so hot, given that the temperature ratio is obviously greater than the 1.25 ratio allowed by heat pipes. Well, by placing multiple heat pipes in series, you can still obtain any ratio you want. My example craft uses a series of 3 heat pipes per "radiator," which allows the distal part of the radiator to be as much as 1.95x hotter than the reactor.

Here are the numbers comparing the GR-1 vs the DIY radiator pictured above:

[table=width: 700]

[tr]

[td][/td]

[td]Effective Heat Rejection[/td]

[td]Total Mass[/td]

[td]Total Cost[/td]

[/tr]

[tr]

[td]GR-1 Conformal Heat Radiator[/td]

[td]1250[/td]

[td]0.5[/td]

[td]10000[/td]

[/tr]

[tr]

[td]DIY Radiator (Heat Pipe x3 + Cubic Octagonal Strut x10 + Small Nose Cone x2 + Structural Wing Type D)[/td]

[td]1250[/td]

[td]0.205[/td]

[td]1120[/td]

[/tr]

[/table]

As you can see, the heat pipes still have a distinct advantage over actual radiators, though certainly less than before. I suggest either reducing the allowed temperature ratio (going to a flat 1.0 would decisively prevent abuse of serial heat pipes), or substantially increasing the mass and cost of heat pipes such that there is no advantage to using them in this manner. Personally, I prefer the former, because "heat pipe destination temperature cannot be greater than than source temperature" is an easy rule to explain and remember, and because it definitively solves the problem.

Edited May 15, 2015 by Fraz86

[Nertea]

1.) Launch new vessel

2.) Turn on reactor with radiators still retracted

3.) Observe conductive backflow from radiators into reactor

4.) Extend radiators

5.) Observe conductive backflow from radiators into reactor

6.) Quicksave and Quickload

7.) Observe conductive backflow from radiators into reactor: it is now different than in 5.!

8.) Retract radiators

9.) Observe conductive backflow from radiators into reactor: it is now different than in 3.!

10.) Quicksave and Quickload

11.) Observe conductive backflow from radiators into reactor: it is now the same as in 3. again, and no longer the same as in 9.!

12.) Extend radiators

13.) Observe conductive backflow from radiators into reactor: it is now the same as in 5. again, and no longer the same as in 7.!

The radiator has the config-specified heatConductivity of 0.001 only when extended. When retracted, it is something else (possibly the default 0.12, but I can't say for sure).

When spawning onto the pad, the radiator is retracted, and thus has the wrong, higher heatConductivity value. When extending the radiator, it should be using the correct, lower heatConductivity... except it doesn't update on toggling, so it continues using the higher one from when the radiator was closed.

As soon as you quicksave and quickload, the appropriate value is loaded and conductive backflow to the reactor drops dramatically (by as much as 90%).

If you then retract the radiator again, then once more it does not update, and keeps the low heatConductivity from when it was extended.

Quicksave/quickload again, and the retracted radiator gains its old, higher heatConductivity value back and conductive backflow to the reactor increases dramatically.

Here's a screenshot: https://dl.dropboxusercontent.com/u/44754370/screenshot52.png

As you can see, two GR-4 radiators on the exact same reactor have vastly different conductive flux values. That's because I forced one of them into the high conductivity state by closing it, saving/reloading, and then extending it again. If I want to I can fix it again by once more saving/reloading now that it is extended, in order to load the correct value.

Well that's just so confusing . I'll look into it again.

The new heat pipes are an improvement, but they're still too effective when used to make do-it-yourself radiators. On the spacecraft below, the reactor stabilizes around 788K, with the hottest part of the "radiator" at 1477K. You may wonder how the radiator got so hot, given that the temperature ratio is obviously greater than the 1.25 ratio allowed by heat pipes. Well, by placing multiple heat pipes in series, you can still obtain any ratio you want. My example craft uses a series of 3 heat pipes per "radiator," which allows the distal part of the radiator to be as much as 1.95x hotter than the reactor.

http://i.imgur.com/wRuBvZy.jpg

Here are the numbers comparing the GR-1 vs the DIY radiator pictured above:

As you can see, the heat pipes still have a distinct advantage over actual radiators, though certainly less than before. I suggest either reducing the allowed temperature ratio (going to a flat 1.0 would decisively prevent abuse of serial heat pipes), or substantially increasing the mass and cost of heat pipes such that there is no advantage to using them in this manner, or some combination of the two. Personally, I prefer the former, because "heat pipe destination temperature cannot be greater than than source temperature" is an easy rule to explain and remember, and because it definitively solves the problem.

:|.

Guess 1.0 it is!

[billkerbinsky]

Working as intended: KSP does not simulate heat at 1000x and above. At that point, the heat status of the vessel is just given the tabula rasa treatment... all equalized and frozen in time until you slow back down to x100 or below. Because of this, it doesn't matter that you insulated your tank. At that level of timewarp, KSP just doesn't care anymore.

As this is stock behavior, there's unfortunately nothing we can do about it.

Any clues about how this equalization happens? seems like the tank (with a thermal mass somewhere north of 20k, dwarfing the rest of the ship) was gaining way more heat energy than the rest of the ship was losing.

I'll set aside the inconvenient fact that the tank temperature should have started somewhere closer to 20K instead of 273K ..

[ExavierMacbeth]

Lmao Fraz now its getting excessive.

At that point all your doing is trading part count to move the heat around more. You might as well just start sticking extra radiators everywhere and not bother with a stacked series of heat pipes lol

[Fraz86]

Lmao Fraz now its getting excessive.

At that point all your doing is trading part count to move the heat around more. You might as well just start sticking extra radiators everywhere and not bother with a stacked series of heat pipes lol

The do-it-yourself radiator provides basically the same functionality at 2/5 the mass and 1/9 the cost. That's a huge advantage. Part count should not really be thought of as a balancing factor.

Edited May 16, 2015 by Fraz86

[Captain Sierra]

Quick bug report not related to NFE at all:

The Mk3-9 pod loses all control if it runs out of power, even though there are pilots in it. It appears to be behaving as if you welded a crew can and probe core together and when the probe core loses power, the crew can't do anything (which is counter-intuitive since wtih stock pods, I can still control via RCS w/out power)

The problem spawned because some wierdness happened to my reactor following time warp. Still working out exactly what happened.

[ExavierMacbeth]

The do-it-yourself radiator provides basically the same functionality at 2/5 the mass and 1/9 the cost. That's a huge advantage. Part count should not really be thought of as a balancing factor.

I would be curious. How does the winglet radiator (16+ or so wings attached directly to the reactor) compare with the rig you set up? After all that is the current stock heat cheat people are using. If your pipe it everywhere method is comparable to that anyway then its not a big worry.

Personally I will be using the pipes the other way. To centralize where i need my radiators rather than to pipe the heat out to stock parts. But if there is a significant advantage over whats being used to cool nervas in stock then I could see the need to change it further.

[Fraz86]

I would be curious. How does the winglet radiator (16+ or so wings attached directly to the reactor) compare with the rig you set up?

Good question! Here's what I came up with:

I needed 32x Structural Wing Type C to keep the reactor below its 850K nominal temperature. Thus, we can say that each GR-1 radiator is roughly "equivalent" to 8 Structural Wing Type C. Let's see how the wings compare to our other options below:

[table=width: 600]

[tr]

[td][/td]

[td]Effective Heat Rejection[/td]

[td]Total Mass[/td]

[td]Total Cost[/td]

[/tr]

[tr]

[td]GR-1 Conformal Heat Radiator[/td]

[td]1250[/td]

[td]0.5[/td]

[td]10000[/td]

[/tr]

[tr]

[td]Heat Pipe Radiator (Heat Pipe x3 + Cubic Octagonal Strut x10 + Small Nose Cone x2 + Structural Wing Type D)[/td]

[td]1250[/td]

[td]0.205[/td]

[td]1120[/td]

[/tr]

[tr]

[td]Structural Wing Type C x8[/td]

[td]1250[/td]

[td]0.4[/td]

[td]2400[/td]

[/tr]

[/table]

Basically, the wings were twice as heavy and twice as expensive as my Heat Pipe Radiator. Nonetheless, the wings were still slightly lower mass and 1/4 the cost of the GR-1.

These findings continue to demonstrate that heat pipes are far too effective, and I stand by my recommendation to decrease their allowed temperature ratio to 1.0. However, I would argue that these findings also reveal that radiators are not well balanced relative to stock parts, either. I recommend any of the following:

• Decrease reactor nominal temperatures - this is probably the easiest solution, as the recommended complements of radiators already cool reactors well below their nominal temperatures
  

• Increase reactor heat production, with a corresponding increase in radiator heat transfer rate - this is my favorite option, as it would help to balance against the very tolerant stock heat mechanics (i.e., high maxTemps & emissivity), and Streetwind already established that radiators are capable of dissipating significantly more heat than they're currently transferring
  

• Decrease radiator mass and cost
  

Edited May 16, 2015 by Fraz86

[Streetwind]

The reactors and their associated parts were always incredibly expensive, yes. The larger ones cost as much as building upgrades. Perhaps a price reduction is in order.

However in the interest of releasing this weekend, let's not rebalance the heat/power ratio right now Maybe a reduction of nominal temperature to 750 K can be done.

Also, let me point out that your test is favoring the structural wing setup slightly, because you have a "stealth radiator" in the capsule. You should test with double insulators between the reactor and anything else. That alone might already make you mount 1-3 extra wing parts.

[Captain Sierra]

The reactors and their associated parts were always incredibly expensive, yes. The larger ones cost as much as building upgrades. Perhaps a price reduction is in order.

However in the interest of releasing this weekend, let's not rebalance the heat/power ratio right now Maybe a reduction of nominal temperature to 750 K can be done.

Also, let me point out that your test is favoring the structural wing setup slightly, because you have a "stealth radiator" in the capsule. You should test with double insulators between the reactor and anything else. That alone might already make you mount 1-3 extra wing parts.

I'm fine waiting an extra week or whatever. I want more time to screw with this and work out all of this weirdness (I'm hating the stock heating MORE by working the the EXP release, not less).

[Fraz86]

However in the interest of releasing this weekend, let's not rebalance the heat/power ratio right now

That's a fair point, I suppose. The mod is certainly usable in its current form. I'm really not trying to be a pain in the ass, just trying to provide useful information!

Also, let me point out that your test is favoring the structural wing setup slightly, because you have a "stealth radiator" in the capsule. You should test with double insulators between the reactor and anything else.

On this point, I strongly disagree. On most spacecraft, I see very little reason to surround the reactor with insulators (reactor heat really isn't a threat to the vessel), and the spacecraft would be much larger. Thus, my experiment - with only a reactor, pod, and wings - is unfairly biased against the wings, due to the unrealistically small "stealth radiator." Balance experiments should attempt to duplicate the large "stealth radiator" capacity that will be available (at no extra mass or cost to the player) for most spacecraft designs.

Of course, there are a few situations where an insulator might be useful. For instance, if I were using a lot of high-heat ion engines, I would probably use the following basic layout (from back to front):

Engines - Aft section (high maxTemp parts only) - Insulator - Reactor - Fore section

The insulator is there primarily to protect the reactor from engine heat. As such, I would probably add a heat pipe from the reactor to the aft section (across the insulator, basically making it "one-way" insulator), so the reactor would still be able to use the entire ship as a radiator. I really can't think of a situation in which I would want to prevent the reactor from conducting its heat throughout the vessel.

Edited May 16, 2015 by Fraz86

[Streetwind]

I'd have placed the insulators just to get a scientifically "pure" result as to how much the wing sections actually radiate. Can't really say "x wings equal to 1 radiator" when other things besides the wings perform radiation. Although admittedly, the result will always remain polluted due to the reactor itself radiating a little. *shrug*

[Fraz86]

I'd have placed the insulators just to get a scientifically "pure" result as to how much the wing sections actually radiate. Can't really say "x wings equal to 1 radiator" when other things besides the wings perform radiation. Although admittedly, the result will always remain polluted due to the reactor itself radiating a little. *shrug*

In medicine, we refer to this distinction as efficacy vs effectiveness. An efficacy study (such as those required by the FDA) strives to be scientifically "pure," by excluding as many potential confounds as possible. However, such studies often end up with a highly contrived "ideal" study population and medical care, which bears little resemblance to the real-world practice of medicine. In contrast, effectiveness studies attempt to examine the effect size of an intervention in settings more closely resembling the real world.

Each kind of study serves a purpose. Efficacy is better to prove that something does what we claim. Effectiveness is better to establish how well something works in practice. In this case, I think an effectiveness-type experiment is most useful.

Edited May 16, 2015 by Fraz86

[Nertea]

I don't think I'm going to have time to finish the 2 insulators this weekend to be honest. Might as well test a doubled heat build to see how it goes. I don't particularly like the sheer number of radiators needed to fix stock heating.

[Fraz86]

Might as well test a doubled heat build to see how it goes.

I have a good feeling about this plan!

I don't particularly like the sheer number of radiators needed to fix stock heating.

I'm not sure I follow you. You mean the number of radiators required to dissipate the heat from an LV-N, for instance? If so, I agree, and doubling the heat transfer should go a long way toward correcting that issue as well.

[ouion]

The whole even stock parts disappear.

[RedParadize]

Hello guys.

First, you guys are awesome. This mod is great.

Second, if I may interject about cost balancing, its ok if reactor cost alot, but the large one are a bit too expessive. On the other side radiator should be relatively cheap. Radiator are not state of art technology after all. Their cost should be cheaper than any home made radiator could be, people should feel encourage to use them.

Edit: Oh, I almost forgot. Utility tabs is way too crowded, did you concider carrying your reactor and solar panel to a different tabs? I know you can create new tabs, OSE and KIS did it, I think your stuff deserve its own tab.

Edit 2: The YZ-1 create alot of drag, regadless of its orientation. And when connection break, there is a bug, center of mass and camera start drifting behind the ship...

Edited May 16, 2015 by RedParadize