FreeThinker

[1.8.1, 1.7.3/1.6.1/1.5.1/1.4.5] KSP Interstellar Extended 1.25 Continued Development Thread

Recommended Posts

When I was making a craft for my own use, radiators were on my to do as well as trying to add radiator modules to wings and what not. After like 8-12 custom parts I ended up not finishing due to time restraints and mods dependencies not updating with that stretch of rapid squad updates. I was tempted to dive back down the rabbit hole and start modeling again to fix this but then I thought that I'd never play if I started again. I did see an image someone posted of a cool looking radiator part that is similar to ones on the iss. No one responded/knew which mod it came from when I asked. Are there any pictures/designs of real radiators that could be incorporated into this game/mod that have yet to be made? I have to admit that I am ignorant about reactor/cooling designs IRL.

The post that had the image I mentioned:http://forum.kerbalspaceprogram.com/threads/91706-0-90-Freight-Transport-Technologies-v0-3-1-2014-12-24?p=1519856&viewfull=1#post1519856

I would really love to integrate those models into KSPI Extended if I'm allowed to. I see a lot of models beeing made which are wasted because they are not used. This is a real waste of talent and time while at the same time KSPI is deperate in need for better models and graphics. I think we could make it so much more better looking, but I simply don't have the skills to model.

I'm looking for modelers. If anyone would like to create some nice models which are actualy going to be used, please help me!!

Edited by FreeThinker

Share this post


Link to post
Share on other sites

I'm looking for modelers. If anyone would like to create some nice models which are actualy going to be used, please help me!!

What are you looking for model-wise?

Share this post


Link to post
Share on other sites
It makes sense to scale weight with surface area on radiators, mainly because there's no point in making them thicker, only broader and longer.

I'm not so sure about that. The heat radiators still have to have structural integrity. The further it extends out from the main fuselage the more support you're going to need closer to it in order to keep the radiator structurally stable during low-g engine burns (ones that aren't powerful enough to break the radiators) etc.

Interestingly, this is one of the few trends that works against the "bigger is better" pattern that normally otherwise holds true in real-life rocketry (bigger fuel tanks require proportionally less insulation and cooling to hold cryogenic fuels, for instance, due to the Square Cube Law- a fact that is currently reflected in the relative power requirements vs. volume of the 1.25 meter and 2.5 meter Nitrogen Cyrostats...)

Regards,

Northstar

Share this post


Link to post
Share on other sites
Alternatively one option I'm thinking about is adding some MM script which adds radiators to stock wing parts. Also any SSTO part could have build radiators in it's skin.

Unless you allow the stock parts to have ridiculously high max temps, I doubt this would help.

FractalUK modeled radiation using Stefan's Law, so dissipation is proportional to temperature to the fourth power. For example, the upgraded radiators (max 3500 K) require 169 times less area compared to the normal ones (max 970 K) to dissipate the same amount of heat. Wikipedia says 930 K is already "bright orange." At safe temperatures for the stock parts, that could be thousands of times less effective than the upgraded graphene radiators.

Some of the stock parts do have high max temps, but I believe those are the ablative type heat shields and not for use as radiators (the heat shields peel off with use).

Edited by Absurdist
typo

Share this post


Link to post
Share on other sites
Unless you allow the stock parts to have ridiculously high max temps, I doubt this would help.

FractalUK modeled radiation using Stefan's Law, so dissipation is proportional to temperature to the fourth power. For example, the upgraded radiators (max 3500 K) require 169 times less area compared to the normal ones (max 970 K) to dissipate the same amount of heat. Wikipedia says 930 K is already "bright orange." At safe temperatures for the stock parts, that could be thousands of times less effective than the upgraded graphene radiators.

Well every bit of cooling can help, especialy if the surfaces are large.

Some of the stock parts do have high max temps, but I believe those are the ablative type heat shields and not for use as radiators (the heat shields peel off with use).

Well I would simply not allow them to upgrade and only have a maximum temperature of 970 K.

Edited by FreeThinker

Share this post


Link to post
Share on other sites
I'm not so sure about that. The heat radiators still have to have structural integrity. The further it extends out from the main fuselage the more support you're going to need closer to it in order to keep the radiator structurally stable during low-g engine burns (ones that aren't powerful enough to break the radiators) etc.

Interestingly, this is one of the few trends that works against the "bigger is better" pattern that normally otherwise holds true in real-life rocketry (bigger fuel tanks require proportionally less insulation and cooling to hold cryogenic fuels, for instance, due to the Square Cube Law- a fact that is currently reflected in the relative power requirements vs. volume of the 1.25 meter and 2.5 meter Nitrogen Cyrostats...)

Regards,

Northstar

For foldable Radiators, the 2.5 Mass scale seem to make sense, as you need additional strength to carry all the mass

Flat surface attached radiator on the other hand should scale with Power 2 (surface) as they gain their structual strength from the surface they are attached to. Therefore using all avialble surface space for heat ratiotors pays of since they require less mass / radiator surface.

Edited by FreeThinker

Share this post


Link to post
Share on other sites
What are you looking for model-wise?

Well there are lots of wishes.

More distinct models for the Reactors. The Dusty Plasma and Particle Reactor alsmost look the same.

The Molten Salt Reactor is only mend for small size, and it's dimentions are too small. I need generic modelsfor storing resource.

I need more radiator type models. Preferably radiators which can be stored airodynamicly.

- - - Updated - - -

Would you want models that are more true to real life designs or perhaps more stylized and look cool?

Preferably real life, sleak and streamlined models would be cool as well

Edited by FreeThinker

Share this post


Link to post
Share on other sites

The folding radiators likely wouldn't need any significant extra mass to deal with bending loads. For one thing, bear in mind that these radiators are designed to deal with removing MW/m^2, whereas the sun would only be sending measly kW/m^2 - in any higher-g burns, they should stop sun tracking and align with g-load, and then their own breadth provides strength against the bending moment. Beyond that, it's a bit of a mis-assumption to say that any notable mass addition would be required to deal with any increase in the moment that does exist - for one, as most of the high-power vehicles will be making very gentle burns (<1g, or even <<1g) it is very likely that the existent structure required to hold the radiator together would be sufficient anyway.

In fact, look at aircraft wings - they have to bear a very high load through a cantilever 1/20th of the moment arm of the force, yet the weight of the entire wing is only 1/25th of maximum sustained load - and that is consistent across aircraft of all sizes, as the additional cantilever mass is only in the root, and the wing itself far out-masses that single structural point. Given that the radiators are not load bearing for the craft's structure, and only have to support their own mass under low-g loads (something that is actually very easy to do), even if you want a higher-than-2 power, 2.5 is very disproportionately high.

Share this post


Link to post
Share on other sites

Note also that you are looking to maximize surface area - you presumably want something hollow, with some kind of (low-mass, high flow) coolant circulating? You only want enough interior mass to conduct the heat outward (along the length away from the craft) without melting.

Share this post


Link to post
Share on other sites
The folding radiators likely wouldn't need any significant extra mass to deal with bending loads. For one thing, bear in mind that these radiators are designed to deal with removing MW/m^2, whereas the sun would only be sending measly kW/m^2 - in any higher-g burns, they should stop sun tracking and align with g-load, and then their own breadth provides strength against the bending moment. Beyond that, it's a bit of a mis-assumption to say that any notable mass addition would be required to deal with any increase in the moment that does exist - for one, as most of the high-power vehicles will be making very gentle burns (<1g, or even <<1g) it is very likely that the existent structure required to hold the radiator together would be sufficient anyway.

The sun tracing is indeed silly. Unless you are in a low sun ordit, most of the time , it would be counter productive. It also gets them ripped of faster in the higher atmosphere I think.

Edited by FreeThinker

Share this post


Link to post
Share on other sites
What are you looking for model-wise?

Also, I almost forgot, the exhaust effects of the electric plasma trusters, they are just awefull. I would realy appreaciate if anyone can fix them

Share this post


Link to post
Share on other sites
it also gets them ripped of faster in the higher atmosphere I think.

Radiators breaking doesn't seem to be related to sun tracking.

---

Relevant line is this:

if (radiatorIsEnabled && dynamic_pressure > [COLOR=#daa520]1.4854428818159388107574636072046e-3[/COLOR] && isDeployable) {

The constant seems to be a magic number Fractal cooked up. Unit is atm, it's 150.5125 in pascals. On the huge heat radiator with 800m^2 on one side, that's applying a maximum of 120410 N of shear force. I'll check later how much von Mises stress that applies on a hollow tube to see if Fractal used a reasonable number.

EDIT #2 (I used steel instead of Mo in my previous edit):

Ok, here it is. I used COMSOL 5.0.

Outside diameter: 1.1 m

Internal channel is a 0.5 m x 0.5 m square.

Length: 80 m

Material: Molybdenum

Reference: http://trace.tennessee.edu/utk_gradthes/1633/

7nT6lzw.png

Highest stress is 42.64 MPa near the base. Molybdenum maximum shear stress is ~500 MPa.

sHjCAAx.png

This one is at 970 K, highest is 1.99 GPa.

If anyone can provide more accurate geometry, I'll do another finite element analysis.

Realism aside, Fractal set it too low IMO, unfurled radiators break at what.. 5 m/s on the runway?

---

While I'm at it, here is another number that appeared on the code without explanation:

[COLOR=#0000ff]float[/COLOR] dynamic_pressure = ([COLOR=#0000ff]float[/COLOR]) ([COLOR=#daa520]0.5[/COLOR]*pressure*[COLOR=#daa520]1.2041[/COLOR]*vessel.srf_velocity.sqrMagnitude/[COLOR=#daa520]101325.0[/COLOR]);

1.2041 is the density of air in kg/m^3 at 20 °C and 101.325 kPa. It's supposed to vary with location (air in Eve is not the same as in Kerbin) and elevation.

---

Now, I'm not sure what this one does:

[B]protected static[/B] [COLOR=#0000ff]double[/COLOR] alpha = [COLOR=#daa520]0.001998001998001998001998001998[/COLOR];
...
current_rad_temp = instantaneous_rad_temp * alpha + ([COLOR=#daa520]1.0[/COLOR] - alpha) * instantaneous_rad_temp;

If you simplify the expression algebraically, the whole thing is just instantaneous_rad_temp. Though even with floating points, I don't see how this is different from just assigning it directly.

Edited by Absurdist

Share this post


Link to post
Share on other sites

OK, so a couple bugs I've been sitting on for a while, and wanted to get out there...

First of all, the Plasma Thruster still is showing "0 seconds" ISP for all its fuel-mods in the VAB/SPH. This is an issue for mission-planning and rocket-design, even if it doesn't affect the actual performance...

hJDdFUF.jpg

Second, and more serious, the tweakable-size Thermal Rocket Nozzle that replaced the pre-sized ones is stuck with a minimum diameter of 2.5 meters... This, of course, makes it impossible to build efficient 1.25 meter Thermal Rockets...

ZKSGbD8.jpg

These screenshots were from KSP-I Extended 0.7.14, but there isn't any mention of fixing these bugs in the 0.7.15 changelog (I'll be DL'ing 0.7.15 shortly for the TTJ changes though- and to see if the ISRU refinery fixes for RealFuels I posted made it into 0.7.15)

Regards,

Northstar

Edited by Northstar1989

Share this post


Link to post
Share on other sites

By the way, FreeThinker, let me know when you're ready to start tackling more of the ISRU reactions and fixes (making the Sabatier Reaction produce Methane and Water instead of Methane and Oxygen, and allowing it to operate outside the atmosphere- for instance) I'm REALLY excited to get to some of the In Situ Resource Utilization stuff. Maybe it's just because my dad trained as a Chemical Engineer, and I currently have a buddy who works in chemical reactors, but personally I consider the ISRU reactions in KSP-Interstellar to be some of the MOST IMPORTANT things for it to develop as it goes forward!

And, just because it's inspirational, like all the things we can do with ISRU are, here's the good ole' "Build, Fly, Dream" trailer, which I deeply love... :)

Regards,

Northstar

Edited by DuoDex

Share this post


Link to post
Share on other sites
Second, and more serious, the tweakable-size Thermal Rocket Nozzle that replaced the pre-sized ones is stuck with a minimum diameter of 2.5 meters... This, of course, makes it impossible to build efficient 1.25 meter Thermal Rockets...
This is a Tweakscale problem, Please download Tweakscale 0.51.1

Share this post


Link to post
Share on other sites

[1]I did some further poking around with the antimatter usage stuff. As it turns out, Fractal assumed the antimatter reactors would only have a 30% efficiency or so, which I suppose I can't really argue with. Assuming the real-world value for the speed of light, though, the reactors are still using too much antimatter, because it's only counting the energy of the antimatter's mass alone in the reaction, not both it and the mass of the normal matter it's reacting with (which would ordinarily be half of the energy released). This means the efficiency is more like 15%. It would be trivial to either double the efficiency value for the reactors or simply halve the UsagePerMegawatt number in the reactor mode config, but I haven't bothered with that just yet.

If you like, you can change the UsagePerMW= line in ReactorFuels.cfg for the antimatter modules (this one as well as the existing one) to 5.5555555555555e-15 instead of 1.111111111111e-14 to fix it yourself.

That's quit an increase, hoe exactly did you came up with that number?

Share this post


Link to post
Share on other sites

Quick question since I'm feeling in a plane-y mood: What's the current deal with air precoolers? Last time I checked (which was probably around 0.23) precoolers checked only their attach nodes for intakes which severely limited plane construction and made radial intakes utterly useless. This behaviour was hardcoded in the plugin. Is this still the case? Also which parts are actually recognized as precoolers?

Northstar,

a few pages back you mentioned some new ISRU processed you'd like to add but can't because the KSPI ISRU stuff is all plugin code. Can those processes be added via REGO converter modules? Regolith might not (yet) be capable enough to handle the finer details of resource extraction you need, but conversion is usually based on fixed ratios which it handles just fine (it even handles catalysts in the sense that they need to be present with a certain amount but are not consumed). Using Regolith for conversion processes where possible (maybe even switching over some already existing ones) increases compatibility with other mods, gets rid of custom code and is a lot easier to manage than the current ISRU modules (all you need is simple MM configs).


This is a Tweakscale problem, Please download Tweakscale 0.51.1

Does this include Boris' KSPI-Fix?

Edited by jinks

Share this post


Link to post
Share on other sites

I'm getting NaN with IntakeAtm when I load my aircraft. Here's the log. The fun begins at line 113672. When I delete the KSPI DLL, it removes the NaN error but still problems remain:

mhMmEDxh.png

Here you can see my 5 stages have all be loaded as one on the left, I can't revert from the main menu or from the center UI widget at the top of the screen (no green/blue icon, doesn't drop down). Loading a spacecraft to the launchpad works fine.

If there's any chance of other mods messing around with KSPI, here is a list of updates I've made since I last flew this aircraft without any problems on 2/7:

[3/6/15] Deadly Reentry Continued (Update - 6.5.3 beta)
[3/3/15] KSP Interstellar (Update - 0.7.15 Extended)
[3/3/15] Advanced Fly-By-Wire (Update - 1.5)
[2/26/15] Crowdsourced Science Logs (Update - 2.0)
[2/26/15] DMagic Orbital Science (Update - 0.9.2)
[2/26/15] VOID (Update - 0.17)
[2/24/15] Ship Manifest (Update - 0.90.0_4.1.2)
[2/24/15] kOS (Update - 0.16.2)
[2/19/15] RCS Build Aid (Update - 0.6.1)
[2/18/14] Modular Fuel Tanks (New - 0.22)
[2/18/14] QuizTech (New - 1.2)
[2/18/14] Fuel Tanks Plus (New - 0.4)
[2/18/14] C.O.R.E. (New - Initial Release)
[2/18/14] EVA handrails (New - Initial Release)
[2/18/14] ProbiTronics (New - 0.22)
[2/18/14] Adjustable Landing Gear (New - 1.0.4)
[2/18/14] Stock Part Revamp (New - 1.7.2)
[2/18/14] BoxSat (New - A.02d)
[2/18/14] KerbalMass (New - 0.9)
[2/18/14] Lack Luster Labs (Update - 13.1)
[2/18/15] Community Tech Tree (New - 1.1)
[2/16/15] Distant Object Enhancement (Update - 1.5.2)
[2/15/15] Stock eXTension (Update - 20)
[2/13/15] Kerbal Stock Parts eXpansion (Update - 0.2.8)
[2/13/15] Tarsier Space Technology (Update - 2.5)
[2/13/15] FASA (Update - 5.21)
[2/12/15] Flag Rotate (New)
[2/12/15] Kerbal Engineer Redux (Update - 1.0.15.2)
[2/12/15] Action Groups Extended (Update - 1.30d)
[2/11/15] Soundtrack Editor (Update - 2.1)
[2/10/15] TweakableEverything (Update - 1.8)
[2/8/15] Destruction Effects (New)
[2/8/15] Fusebox (Update - 1.3)

Share this post


Link to post
Share on other sites
Quick question since I'm feeling in a plane-y mood: What's the current deal with air precoolers? Last time I checked (which was probably around 0.23) precoolers checked only their attach nodes for intakes which severely limited plane construction and made radial intakes utterly useless. This behaviour was hardcoded in the plugin. Is this still the case? Also which parts are actually recognized as precoolers?

Well basicly what is does is that is compares the number of air-coolers with the number of open air-intake. It they are equal, your engines won't overheat. If you have too few precoolers, your engines will overheat if you move fast

- - - Updated - - -

Here you can see my 5 stages have all be loaded as one on the left, I can't revert from the main menu or from the center UI widget at the top of the screen (no green/blue icon, doesn't drop down). Loading a spacecraft to the launchpad works fine.

That a nasty problem, I had it in the past too after I hand deleted some part models/texture/interior. Perhapas you did the same, if not, I suggest remove mods one by one until you have found the problem.

- - - Updated - - -

a few pages back you mentioned some new ISRU processed you'd like to add but can't because the KSPI ISRU stuff is all plugin code. Can those processes be added via REGO converter modules?

As long as all input and output resources are storable, it could be implemented using a basic REGO converter. It can even use megawatts.

Share this post


Link to post
Share on other sites

If you like, you can change the UsagePerMW= line in ReactorFuels.cfg for the antimatter modules (this one as well as the existing one) to 5.5555555555555e-15 instead of 1.111111111111e-14 to fix it yourself.

That's quit an increase, hoe exactly did you came up with that number?

That seems to be a decrease to a half of the old number, the exponent changed from e-14 to e-15 change (10^-14 to 10^-15, right?:confused:).

Share this post


Link to post
Share on other sites
That seems to be a decrease to a half of the old number, the exponent changed from e-14 to e-15 change (10^-14 to 10^-15, right?:confused:).

Yes, I ment improvement (in efficiency)

Share this post


Link to post
Share on other sites
Well basicly what is does is that is compares the number of air-coolers with the number of open air-intake. It they are equal, your engines won't overheat. If you have too few precoolers, your engines will overheat if you move fast

That's what it does in ModuleSabreHeating, but have a look at that little prc.isFunctional(). That part is defined in FNModulePreecooler.cs in OnStart() and yes, it still only works for stack-attached intakes. ;.;

Share this post


Link to post
Share on other sites
That's what it does in ModuleSabreHeating, but have a look at that little prc.isFunctional(). That part is defined in FNModulePreecooler.cs in OnStart() and yes, it still only works for stack-attached intakes. ;.;

Is this what you want?

Wp1kS2T.jpg

Notice I'm flying at more than 3 times the speed of sound using 2 radial attached regular air scoops without blowing up the engines!

Edited by FreeThinker

Share this post


Link to post
Share on other sites

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.