[0.90] KSP Interstellar port maintance thread

[Northstar1989]

A little adjustment of the configs, and...

I'm getting there. I got these values by changing the ISP multiplier to 21, and the Thermal Rocket Thrust Multiplier to 2...

The thrust is still a *tad* high (at this ISP, Thrust should be about 211.5 kN if we're basing performance on the Timberwind, as I said in my PM), but it's very close to accurate... Unfortunately, because KSP-I ties Thrust and ISP together, it's hard to tweak the values to *exactly* what I want them to be... Even when I do out the math, it's easy to make some minor rounding-errors and such...

Anyways, if we fix this, make the Atmospheric ISP vary realistically with Mass Flow Rate (by subtracting Exit Area * Background Pressure in the Thrust calculation, which will tie Atmospheric ISP to Vacuum Thrust in such a way as to create the correct relationship with Mass Flow Rate...), fix the displayed ISP-multiplier in the VAB/SPH from 17 to whatever we adjust it to as the default value, and get some CO2-storage tanks rolling we'll be golden.

OK, I know that's a LONG list- but it's good to keep the end in mind. Maybe I can write up a MM patch to allow LqdCO2 to be stored in RealFuels tanks to make all this easier on you...

Thanks as always for the hard work FreeThinker!

Regards,

Northstar

Edited February 11, 2015 by Northstar1989

[FreeThinker]

Anyways, how are we doing for Vacuum Thrust?

Err, not great... We're expecting 91.9375 kN of thrust just to match the Timberwind performance, so we're at about 44.7% of the expected value just using the ISP of 931 seconds (and even worse with 1150 seconds). However, we still have additional Thrust from the nozzle to account for- which the Timberwind reactor also benefited from... With our relatively larger nozzle, we're hoping for about 105.728 kN of Vacuum Thrust assuming a 15% increase in Vacuum ISP from the larger nozzle...

I'm getting there. I got these values by changing the ISP multiplier to 21, and the Thermal Rocket Thrust Multiplier to 2...

The thrust is still a *tad* high (at this ISP, Thrust should be about 211.5 kN if we're basing performance on the Timberwind, as I said in my PM), but it's very close to accurate... Unfortunately, because KSP-I ties Thrust and ISP together, it's hard to tweak the values to *exactly* what I want them to be... Even when I do out the math, it's easy to make some minor rounding-errors and such...

I'm a little confused, I though the goal was 105.728 kN, now it has to be twise of that?

What about the upgraded versions? perhaps it would be an idea to give the upgraded version (1/4 Timberwind =) 211.5 kN and the non upgraded (1/8 Timberwind =)105.728 kN

Edited February 11, 2015 by FreeThinker

[Northstar1989]

I'm a little confused, I though the goal was 105.728, now it has to be twise of that?

I was looking back over the math I used to calculate that, and I accidentally divided the expected Thrust in half (that is, I was balancing against 1/8th the Thrust of Timberwind 75 instead of 1/4th...) I must have been running the numbers a little too late at night, sorry...

Timberwind 75 had a Thrust of 735.5 kN

http://www.astronautix.com/engines/timind75.htm

With a diameter of 2.03 meters, it was [(2.03)^3] / [(1.25)^2] = 4.28 times as large as the "Sethlans" reactor

Which, I've been approximating as 4 times as large to make the math simpler...

If I do the math a little more precisely though...

(735.5) / (4.28) = 171.72 kN

and then correcting for a Vacuum ISP of 1150 seconds from the nozzle (so a 15% increase to Thrust)

171.72 * 1.15 = 197.5 kN

However, I thought we agreed to just go with a 1/4th approximation before (this is what the ThermalPower figures are based on, and the reactor *IS* a lot heavier for its size and likely built with more modern materials than we had in 1992), in which case it's:

(735.5)/4 = 183.875 kN

(183.875) * 1.15 = 211.5 kN

That last bit is where the 211.5 kN number comes from.

We could go with either number, but the ThermalPower production ratings are already based on a simple 1/4th approximation. I'm inclined to stick with Thrust ratings built on the 1:4 approximation so that we get a realistic Thrust/MW relationship. The level of ThermalPower production could reasonably be expected to be a bit higher since our reactor is almost 4x heavier for its volume (in fact 88% as heavy as the full-scale version despite its smaller size- and would be built with 20 years more advanced materials and engineering if it were a present-day design), but Thrust/MW is based on the physics of exhaust temperature and nozzle size...

Sorry about futzing that calculation before. I thought I already listed the *correct* target Thrust (211.5 kN) in several other posts though...

Regards,

Northstar

Edited February 11, 2015 by Northstar1989

[FreeThinker]

I'm not sure, from a game balancing point of view, I kind of like the 1/8 version, the resulting 70kN and 100kN (with double the ISP), it would be a balanced step from the stock 60kN Nuclear Engine in the previous tech node. I think we can make it up with the upgraded version, which we could give the full 1/4 thrust or more. Call me conservative but I think we cannot only focus on the reaslism of the first generation reactors without losing sight of the bigger picture. The bigger picture is that we have upgrades, fusion, antimatter and stock nuclear engines (based on NERVA). In its totality it should feel balanced, otherwise we create a frankenstein monter which might be scienticly/realism accurate, but scare the townspeople. Perhaps it's idea to create a specific Realsim config file later but right now I just want to please the townpeople and prevent them from lynching me.

Edited February 11, 2015 by FreeThinker

[Northstar1989]

Also, what do you want to do about "Exit Area" and calculation of Atmospheric ISP?

As I pointed out by PM, the 0.7657 m^2 approximation was *VERY* inaccurate (based on the performance of an SRB with a similar size/shape nozzle that, as it turned out, had performance with an inaccurate relationship to its actual nozzle model). The actual Timberwind had a nozzle-area of about 0.8 meters, so our scaled-down version should be about 0.2 m^2 for identical performance. However, if we relied on our physical model's nozzle area, and assumed a Nozzle Efficiency of 95% (typical values are 90-99%), we would get an Exit Area of 4.66 m^2 (which would lead to a *NEGATIVE* predicted Sea-Level Thrust...)

So, we could basically go with any value we wanted between those two for our Exit Area. The larger our Exit Area, the higher we should set our Vacuum ISP and Vacuum Thrust, however...

If we go with Fractal_UK's relationship between Vacuum ISP and Sea-Level ISP (which ignores Mass Flow Rate and just assumes 40% of Vacuum Thrust as sea-level for ALL possible Thermal Rocket combinations) we would need an Exit Area of about 1.25 m^2. I would advise that we still need to replace his "ispratio" term with a term to subtract (Exit Area * Background Pressure), though, or we will still get the ridiculous result that a multi-Gigawatt Microwave Thermal Rocket loses 60% of its Thrust at sea-level...

Regards,

Northstar

Edited February 11, 2015 by Northstar1989

[Northstar1989]

I'm not sure, from a game balancing point of view, I kind of like the 1/8 version, the resulting 70kN and 100kN, it would be a balanced step from the stock 60kN Nuclear Engine in the previous tech node. I think we can make it up with the upgraded version, which we could give the full 1/4 thrust or more. Call me conservative but I think we cannot only forus on the reaslim of the first generation reactors without losing sight of the whole picture.

*Chokes quietly*

I play with Real Solar System 64K and RealFuels+Stockalike (which means, the solar system is 58% of real-world scale- even though most parts in KSP are only between 40 and 50% the diameter of real life, and my rocket engines only produce approximately 1/2the the thrust per square-meter of cross-sectional area as in real life). I'm not looking for performance that is nerfed against reality...

Ignoring that issue for a second, and looking at the bigger picture, the reactors are already 4x heavier than they should be (the "Sethlans" is less than 1/4th the size of the 2.5 ton "Timberwind 75", yet weight 88% as much). Isn't *THAT* already enough of a nerf against real-life performance...

The Timberwind engines (includes a reactor AND nozzle) had the following Thrust and mass values:

Timberwind 45

Vac Thrust: 441.3 kN

Mass: 1.5 metric tons

Timberwind 75

Vac Thrust: 735.5 kN

Mass: 2.5 metric tons

Timberwind 250

Vac Thrust: 2451.6 kN

Mass: 8.3 metric tons

You're already talking only 28.4% the Thrust-Weight Ratio (8.52) if you scale down the Thrust 1:4 but keep the excessively-high mass (2.2 tons for a Sethlans + Thermal Rocket Nozzle, and that's not even counting the weight of the reactor fuel!) Now you want to cut the Thrust down to 1:8 instead? These reactors were built for LAUNCH STAGES- but pretty soon you're going to be getting into NERVA-like territory: engines that can barely lift their own weight...

The KSP-Interstellar NTR's are based on technology from 20 years ago (Timberwind/ Particle Bed Reactors) to the near-future (Molten Salt Reactors- which are STILL 10-20 years off), NERVA are 1970's technology. There *SHOULD* be a large gap in performance between the two.

And, as I stated previously, Thrust/MW ratio is based on real-world physics. If you break that, you've got nothing to go by or balance against...

I *STRONGLY* recommend we stick with the correct Thrust:Megawatt ratios for the reactor temperatures and Vacuum ISP (Thrust:MW ratio targets are currently 15% higher than Timberwind because our nozzle is much larger and gives a 15% higher Vacuum Thrust/ISP...)

If you *REALLY want a 1:8 Thrust scaling, then we should cut down on the ThermalPower production instead. Don't forget that Thermal Rocket Nozzles can also be powered by microwaves from Solar Power Satellites in close orbits around the Sun. If you nerf the MW:Thrust relationship to only produce half the Thrust-per-MW that it should realistically provide, then you make those a *LOT* less relatively useful...

In fact, the Thermal Rocket Nozzle is used for every single reactor in the game- so if we don't fix the Thrust/MW ratio (which Fractal_UK had set *FAR* too low compared to real life), then you nerf *every single one* of those as well, and make the entire mod much less useful compared to what it should be...

Regards,

Northstar

- - - Updated - - -

To clarify: currently, the Thermal Rocket Nozzle produces 0.5648 kN/MW with a 3000 K exhaust-stream. In real life, a Thermal Rocket can *EASILY* obtain more than 1 kN/MW with a vacuum-optimized nozzle at 3000 K: the Timberwind reactors all produces around 0.98 kN/MW, and they all had small nozzles optimized for sea-level performance... (larger nozzle --> more kN/MW in vacuum)

It all boils down to this, though: half the Thrust-Weiight-Ratio is *NOT* half as good. An engine with a TWR of 8 can *barely* function as a decent upper-stage (and FORGET about using it as a launch engine) after you add in fuel mass, payload, and thrust-loss to atmosphere. An engine with TWR of 4 can barely serve as an orbital engine.

And just to add a little perspective... A rocket with a TWR of 1.8 on the launchpad can make it to orbit. A rocket with a TWR of 0.9? Not going anywhere soon...

We've already cut the TWR to roughly 1/4th that of Timberwind (such that the reactors can no longer serve as launch-stage engines, after losing 60% of their thrust at sea-level, as the sea-level TWR of the engine itself is now only 3.4- not enough even to lift a launch stage on their own: despite their being modeled around nuclear launch-stage engines with a sea-level TWR of 26.7!) Cutting it to 1/8th (sea-level TWR of 1.7 for *just the engine*) is *NOT* going to sit well with me...

Regards,

Northstar

Edited February 11, 2015 by Northstar1989

[Northstar1989]

One last thing...

All is not right down in Whoville:

I know I wanted to ensure we got enough Thrust/MW for the electric engines, but... *INFINITE* Thrust/MW???

Regards,

Northstar

[FreeThinker]

The KSP-Interstellar NTR's are based on technology from 20 years ago (Timberwind/ Particle Bed Reactors) to the near-future (Molten Salt Reactors- which are STILL 10-20 years off), NERVA are 1970's technology. There *SHOULD* be a large gap in performance between the two.

True, but I don't object to the reality of player eventualy getting access to Timberwind equivalent equipment. I fear we upset game balance when we if I give access to such powerfull equipment too easy. By that I mean the step in CTT techtree from Nuclear engines (Stock Nuclear Engine) to Advanced Nuclear Propulsion/Engines is very small, the accesable technology should refrect this!

If you *REALLY want a 1:8 Thrust scaling, then we should cut down on the ThermalPower production instead. Don't forget that Thermal Rocket Nozzles can also be powered by microwaves from Solar Power Satellites in close orbits around the Sun. If you nerf the MW:Thrust relationship to only produce half the Thrust-per-MW that it should realistically provide, then you make those a *LOT* less relatively useful...

True, but I would like to see a logical technological progression, we I need a good answer to the question what the upgrade path should be. In the current Interstellar, I always found it kind of weird that a reactor could suddenly become upgraded to something completely different with the click of a button. It should at least be something very similiar to the starting reactor. Also upgrading should require some time durring which the reactor is unavailable, preferable in a space dock, but all of that are nice to haves.

But to solve the logical technological progression problem, I think it would be a good idea to Split the first generation reactors into 4 seperate reactor families .

- The Sensan/Akula reactor would be split into a (Timberwind) Particle Bed Reactor and a Dusty Plasma reactor. The Timberwind Particle Bed Reactor would have 2 states, an experimental version (with 105 kN @ 1.25m ) and a upgraded version (with 210 kN @ 1.25m) with lower weight. The dusty plasma reactor would be availabe later and offer and experimental and upgraded state as well . Preferable the upgrade of Timberwind Particle Bed would become availabe before Fusion power becomes availabe.

- Kiwi/Aegletes reactor would be split into Molten Salt Reactors and Gas Core Reactor family. Molten Salt Reactors would focus on long life time while Gas Core Reactor on higher ISP

To implement , I would lower the thermal heat (and temperature) of the experimental versions, lower the ISPconstant to 21 and set the ThermalNozzleTrustMultiplier to 712.5% (as you suggested). I would be better If I got new models but for the moment I can the existing 4 models to create 4 different reactor families . Note the upgraded version would not only increase in heat ouput en core temperature, but also reduce in weight!

Edited February 11, 2015 by FreeThinker

[FreeThinker]

significantly weaker. using 1.25 meter parts: sabre intakes and cooler, omega fusion reactor, 2 small radial radiators, electric generator, and thermal turbojet in creative mode.

before I installed the mod tweaks.

http://i.imgur.com/unoWU9C.jpg

after the mod tweaks.

http://i.imgur.com/4M8TZtK.jpg

can anyone else confirm?

Alright, I have discovered the problem was introduced with 0.5.6 going to find what went wrong

I fixed in in revision 0.6.1 which can be downloaded from KerbalStuff

Edited February 11, 2015 by FreeThinker

[Yemo]

Hey,

imho a balance for "stock" ksp dimensions should be the "standard" value. Otherwise the compatibility with all the other mods, including CTT, would be at risk.

"Realism" values could be part of a separated mm compatibility config, dependent on the usage of Scaling/Realism mods.

@FreeThinker:

About the CTT compatibility: Am I correct to assume that KSPI 0.90 Extended Configuration will be the basis of further development of your KSPI/NearFuture Integration?

So eg could I work on the integration of the SETI-BalanceMod and your KSPI 0.90 Extended Configuration (based on the latters CTT config by Olympic1), without creating issues down the line, when KSPI/NearFuture Integration shifts towards CTT?

That would be much easier than trying to integrate everything at once and would also make the integration less problematic for existing CTT based games, since things like the heat mechanic would be already there.

Thank you for your great work!

[FreeThinker]

About the CTT compatibility: Am I correct to assume that KSPI 0.90 Extended Configuration will be the basis of further development of your KSPI/NearFuture Integration?

They are technically the same, only NearFuture Integration is geared towards better integration with Near Future.

[Northstar1989]

imho a balance for "stock" ksp dimensions should be the "standard" value. Otherwise the compatibility with all the other mods, including CTT, would be at risk.

"Realism" values could be part of a separated mm compatibility config, dependent on the usage of Scaling/Realism mods.

Yemo, I value your opinion- only I think you miss the point for which the Particle Bed Reactor was made. Particle Bed Reactors like Timberwind were designed (in real life) SPECIFICALLY for high TWR and the ability to be used as launch-engines. They ALREADY have a 1:4 Thrust scaling and a 1:4 TWR nerf (so, they have 1/4th the Thrust and 4 times the mass for the Thrust as in real life) with the stats I've been aiming for, and are thus nigh-worthless as launch-engines (and only useful as upper stage and orbital engines).

Nerfing the TWR again will further restrict their function to only acting as orbital engines. You have to look at the function an engine's stats make it useful for when balancing parts, rather than simply looking at numbers. In the case of the Particle Bed Reactors, they were already quite marginal as launch-stage engines in real life (real chemical rockets produce more than 4x the thrust for a comparable diameter, and have 4x the TWR) to begin with- and the nerfs I'm already OK with (keeping the mass high and only imitating real-world Thrust) really limit them to only being useful as upper stages- where a stage TWR of 0.5 might be acceptable, for instance. Nerfing them further will make them only useful once you're already in orbit...

Regards,

Northstar

- - - Updated - - -

True, but I don't object to the reality of player eventualy getting access to Timberwind equivalent equipment. I fear we upset game balance when we if I give access to such powerfull equipment too easy. By that I mean the step in CTT techtree from Nuclear engines (Stock Nuclear Engine) to Advanced Nuclear Propulsion/Engines is very small, the accesable technology should refrect this!

FreeThinker, I think you're used to playing spoiled for Science Points. For myself, and most new or more passive/careful players (I mostly hang around in LEO and make an occasional Mun landing for a great deal of the game), just unlocking Nuclear Engines at all is a MAJOR achievement. The fact that in CTT it requires an *ADDITIONAL* tech-node after that is a MAJOR obstacle. It shouldn't be necessary to nerf the performance because of when the part becomes available- by the time players get to a Particle Bed Reactor (and in the case of a new player, figure out how to manage their WasteHeat) they will already be hungry for higher-performing engines than the stock NERVA- especially if they are playing with any realism mods.

And when the stock aero gets revised, radially-mounted NERVA engines will become *MUCH* more difficult to get to orbit than they are now- due to the destabilizing effects of having a lot of drag on the payload stage of your rocket... Players will be eager to replace a pair of radial NERVA engines with a single stack-mounted Particle Bed Reactor, which with my more *REALISTIC* stats (still 1/4th the TWR of real life, but not 1/8th like you seem to be considering) will only produce about 50% more thrust than a radially-mounted pair of NERVA's...

True, but I would like to see a logical technological progression, we I need a good answer to the question what the upgrade path should be. In the current Interstellar, I always found it kind of weird that a reactor could suddenly become upgraded to something completely different with the click of a button. It should at least be something very similiar to the starting reactor. Also upgrading should require some time durring which the reactor is unavailable, preferable in a space dock, but all of that are nice to haves.

If I'm not mistaken (it's been a while since I upgraded any reactors), researching a tech node still doesn't upgrade reactors out in the field- only the ones you place on your NEW rockets in the SPH/VAB. So, none of this should be an issue. And the tech-progression of having Timberwind-like performance come 1 tech node after NERVA's *IS* logical. I don't think you get how long that tech node will really take most average players to unlock... It's also accurate to real life- the Timberwind Reactors only came 20 years (basically, one generation as far as rocketry goes) after the NERVA- 1992 vs 1972. That's *NOT* a huge gap in time. If anything, the NERVA engines were early/experimental Nuclear Thermal Rockets, whereas Timberwind and SNTP-like engines represent more mature nuclear propulsion...

But to solve the logical technological progression problem, I think it would be a good idea to Split the first generation reactors into 4 seperate reactor families .

- The Sensan/Akula reactor would be split into a (Timberwind) Particle Bed Reactor and a Dusty Plasma reactor. The Timberwind Particle Bed Reactor would have 2 states, an experimental version (with 105 kN @ 1.25m ) and a upgraded version (with 210 kN @ 1.25m) with lower weight. The dusty plasma reactor would be availabe later and offer and experimental and upgraded state as well . Preferable the upgrade of Timberwind Particle Bed would become availabe before Fusion power becomes availabe.

- Kiwi/Aegletes reactor would be split into Molten Salt Reactors and Gas Core Reactor family. Molten Salt Reactors would focus on long life time while Gas Core Reactor on higher ISP

To implement , I would lower the thermal heat (and temperature) of the experimental versions, lower the ISPconstant to 21 and set the ThermalNozzleTrustMultiplier to 712.5% (as you suggested). I would be better If I got new models but for the moment I can the existing 4 models to create 4 different reactor families . Note the upgraded version would not only increase in heat ouput en core temperature, but also reduce in weight!

All of that sounds unnecessary to me. Currently, the Particle Bed Reactors don't unlock until fairly late in the game to begin with, and at a point where being starved for Science Points and adventure will more or less be FORCING most players to start heading to other planets... NERVA should be available for players who want to send an early manned mission out to somewhere like Duna, but Timberwind/SNTP-like performance should be available to players who wait until their space program is more technologically-mature and they are better-prepared for the challenges that actually imposes...

Regards,

Northstar

P.S. I think a LOT of this boils down to a matter of perspective on space programs. Most people, including from my perspective yourself, seem to believe that because we HAVEN'T sent manned missions to places like Mars yet in real life, we haven't been technologically-ready for it yet... In reality, while I would say that something like a manned Mars mission in 1980 would have been a little premature, there's no reason we couldn't have carried one out in the mid-90's after we developed technologies like Timberwind/SNTP nuclear space-capable reactors, and computer-technology started to really mature...

It's more a matter of budgets and political willingness in real life than technology- even I, who am an avid fan of things like Mass Drivers (have you tried my Mass Driver mod yet, by the way? It's linked in my signature...) and Microwave Beamed Power (leave those heavy nuclear reactors on the ground! This is also part of why I think it's CRITICAL we get the right relationship between Megawatts of available power, and available Thrust- because it's rather hard to build even large ground-deployed Microwave Transmitter stations in KSP Career Mode...) strongly believe that the only reason we NEED these technologies is because Congress/politicians aren't willing to fund NASA (or Europeans the ESA) to the point where a manned Mars mission would be thinkable without the cost-savings these technologies enable. Please don't nerf my Particle Bed Reactors- they're the only thing keeping my hope alive that humans in real life will someday ALSO make it beyond Low Earth Orbit...

[FreeThinker]

I tried to achieve the desired result ISP 1150.0s 211.7 kN for the Upgraded Sethan reactor

To achieve this, instead of changing IspCoreTempMult (which I rather not do because of backward compatibility) I changed the reactor Core temperature to 2642.4 and set the Nozzle powerTrustMultiplier at 6.359

Edited February 11, 2015 by FreeThinker

[EvilGeorge]

Hello Freethinker,

I've recently upgraded to your 0.5.6 version of KSPI (and then tried my luck with the 0.5.7 and 0.6). In all of these versions I'm having problems with the tech tree. I have read the forum here and am not much wiser as to how to solve my problem. Because by now I have tried all I could think of, I'm posting here.

I am using an ongoing game started in the 0.90 Boris' KSPI version. I have also used your KSPI version (I think it was 0.5.2) without any trouble.

So. When I load my save after an install of your mod, a pop-up window tells me TechManager is disabled and prompts me to either leave it disabled or to select a working tech tree. Is that normal behaviour?

http://i.imgur.com/bHh3GkX.jpg

Now no matter what I choose, the game doesn't behave correctly.

1. If I choose to leave the TechManager disabled, only the nodes I have already unlocked are visible (it seems like I have nothing to unlock), even though there should still be a few nodes left.
   
2. If I choose the CTT option, the tech tree is just a mess. Some nodes are unlocked, some are locked (obviously the result of shifting tech / creating new nodes or whatever). This means that my current ships are using technology that is not yet "discovered". Some nodes are empty (see picture - 1000 Science for the Experimental Science node, which is empty). I don't want this tree. http://i.imgur.com/CR1Cvts.jpg
   
3. If I choose whichever of the Unofficial KSPI trees (except the 0.25 one, I didn't try that for obvious reasons), the tech tree shows nodes in the progression I am used to, but some of them have the a-new-item-in-the-node number. These nodes require me to pay funds to unlock some of the parts (which I didn't have to do prior to updating KSPI). After paying for them, they function ok. The only problem is that the game doesn't seem to understand that I've already unlocked the upgraded graphene radiators and better generators, so I can only build ships with the pre-upgraded stuff and can't change to the new KTEC generators. As the nodes in question are already unlocked, I can't see how I'm supposed to "unlock" these again :/ This is currently the choice that is working the best out of all of them, but as it severely limits my usage of nuclear power, it's still far from correct. http://i.imgur.com/AZwVLoL.jpg
   

What am I doing wrong? Should I download Olypic's fix to the CTT tree? I don't really want to have to get used to a new tech tree, I would prefer to play with the 'original' one (no disrespect to Olypic or other CTT authors intended).

I could really use some advice on this matter. If you need me to upload my persistence file or whatever, I'll do so.

Thanks for reading so far. Keep up the good work.

[Northstar1989]

I tried to achieve the desired result ISP 1150.0s 211.7 kN for the Upgraded Sethan reactor

http://i.imgur.com/5UQoGHd.jpg

To achieve this, instead of changing IspCoreTempMult (which I rather not do because of backward compatibility) I changed the reactor Core temperature to 2642.4 and set the Nozzle powerTrustMultiplier at 6.359

Good enough for government work.

Seriously, though, in some ways using a lower Reactor Temperature actually even makes some realism sense, as if the nozzle were large enough it would still be possible to achieve the desired Vacuum Thrust and ISP with that temperature (although at the expense of VERY LOW sea-level ISP for a nozzle that large: which I guess we already have as it currently stands...) A lower-temperature reactor could also reasonably be expected to have better endurance- whereas the Timberwind reactors were only built for a relatively short period of high-performance during launch...

Now, onwards to the Thrust: Atmospheric ISP relationship. We already have (thanks to Fractal_UK) a formula where Thrust, rather than Fuel Flow, varies with Specific Impulse- but we DON'T have a formula where Atmospheric Specific Impulse is calculated correctly in the first place based on changes in Vacuum Thrust (i.e. Throttle, total MW of Thermal Power available, and fuel Molecular Mass...)

Could we remove the silly "ispratio" term from the current thrust calculation, and subtract (Exit Area * Background Pressure) instead.

It should look something like this:

Thrust = Math.Max (equation) - (Exit Area) * (Background Pressure)

Notice that (Exit Area * Background Pressure) is located OUTSIDE of the rest of the equation? That's an important point about the format you need to keep aware of... You are basically adding a parallel term to the (Math.Max) function that only kicks in when inside an atmosphere (where Background Pressure =/= 0)

The key thing that creates the proper relationship with Mass Flow Rate, etc. is that (Exit Area * Background Pressure) is a CONSTANT term for a given engine at a given altitude, whereas Vacuum Thrust (the rest of the equation) will vary based on available ThermalPower (particularly variable with a Microwave Thermal Rocket), Vacuum ISP (varies by core temperature, which can vary with a Pebble Bed Reactor like the Sethlans), vessel throttle-setting (throttling down should *REDUCE* Atmospheric ISP as well as Fuel Flow), etc.

Regards,

Northstar

Edited February 12, 2015 by Northstar1989

[Yemo]

Hello Freethinker,

What am I doing wrong? Should I download Olypic's fix to the CTT tree? I don't really want to have to get used to a new tech tree, I would prefer to play with the 'original' one (no disrespect to Olypic or other CTT authors intended).

I could really use some advice on this matter. If you need me to upload my persistence file or whatever, I'll do so.

Thanks for reading so far. Keep up the good work.

Installing new tech trees in an ongoing game is a recipe for disaster. Parts are almost always assigned to tech nodes based on your installation, not on what TechTree you pick in game.

If you eg install CTT, it should be the only tech tree installed (besides stock obviously), the only one selected and you have to start a new game.

No idea why Nertea does not put a fat red disclaimer on it, I already asked him, but got no response.

That is the case with all tech trees, installing multiple ones simply does not work and certainly not for existing saves.

[Northstar1989]

OK, so great work so far FreeThinker! What do you need help with before you'll be ready to release the next version of the Extension Config? (by the way, I'm all up-to-date since you encouraged me to do so before, and will make a greater effort to stay updated while we work on developing this...)

Also, let me know when you're ready for some code for CO2 tanks and Thermal Rockets... (I see Nitrogen, but not CO2, show up as a possible Thermal Rocket fuel- so I assume you haven't added it yet...) I can probably write it up myself, using your previous code as a blueprint...

EDIT: I see that CO2 is showing up in the VAB now. It didn't show up until the very latest version (when you released that bugfix for the .DLL). Weird. Are you making changes not in the Changelog between updates?

Regards,

Northstar

Edited February 12, 2015 by Northstar1989

[FreeThinker]

I could really use some advice on this matter. If you need me to upload my persistence file or whatever, I'll do so. Thanks for reading so far. Keep up the good work.

You created a mess in your techtree, I had the same problem. I advice you te remove the TechManager folder, startup you career, load a safe, close, reinstall Techmanager and select only the CTT techtree.

- - - Updated - - -

Also, let me know when you're ready for some code for CO2 tanks and Thermal Rockets... (I see Nitrogen, but not CO2, show up as a possible Thermal Rocket fuel- so I assume you haven't added it yet...) I can probably write it up myself, using your previous code as a blueprint...

Great that would realy help me, writing and testing MM config files cost a lot of time.

- - - Updated - - -

EDIT: I see that CO2 is showing up in the VAB now. It didn't show up until the very latest version (when you released that bugfix for the .DLL). Weird. Are you making changes not in the Changelog between updates?

Weird is has been in the patch since 0.4.3, my best guess is that you didn't update correctly

[FreeThinker]

All of that sounds unnecessary to me. Currently, the Particle Bed Reactors don't unlock until fairly late in the game to begin with,

Well in the original KSPI techtree this was true , but in the new Community Tech Tree (CTT) techtree there are 3 new technodes for Nuclear Power reactors (Nuclear Power, Large Scale Nuclear Power and High Energy Nuclear Power) and it would be a waste not to use them to the fullest.

(basic) Nuclear Power (300 Science, same as Nuclear Propulsion) could introduce Experimental Molten Salt Reactor and Experimental Particle Bed at 0.625m and 1.25m sizes

Large Scale Nuclear Power (550 Science, essentialy offers bigger reactors) could introduce larger and more efficient Experimental Molten Salt Reactor and Experimental Particle Bed at 2.5m and 3.75 sizes

High Energy Nuclear Power (1000 Science. represent the cutting edge in nuclear reactors)

- allow all Experimental Molten Salt Reactor and Particle Bed to upgrade to their Mature (full power) version

- introduce Experimental Dusty Plasma Reactor (allows magnetic noozle and maxumum fuel efficency) and Experimental GasCore (allow significantly higher ISP themal noozle), which can be upgraded to their Mature version with Fusion technology

All of this could effectly give the player a much smoother technological progression from Nuclear Stock to [Molten Salt/Particle] to [Dusty Plasma/Gas Core] to Fusion.

Thanks to the CTT techprogression we will be able to diffentiate 4 Families of Nuclear Reactors, each excelling in a particular way (and therefore most fit for certain functions)

- Particle Bed reactors excel in high TWR (achieved by high trust and lower weight like Timberwind) , can function as part of laucher

- Molten Salt reactors excel in long lifetime electric power generation using Urananium. Function as reliable, low mainenance power source

- Dusty Plasma reactors offers early adoption of Magnetic Noozles or Direct Conversion Power Generator at maximum fuel Efficency , multifuncyional

- Gas Core reactors excel in generating maximum electric power with short lifespan (using Thorium) and High ISP from Thermal Noozle , function as strongest powersource for electric engines or high ISP thermal rocket

Edited February 12, 2015 by FreeThinker

[FreeThinker]

Could we remove the silly "ispratio" term from the current thrust calculation, and subtract (Exit Area * Background Pressure) instead.

Well I cannot completely remove the ispratio, because it's required for the thermal turbojet. Instead I will make it default to 1 for thermal noozles (effectively making it irrelevant).

It should look something like this:

Thrust = Math.Max (equation) - (Exit Area) * (Background Pressure)

Notice that (Exit Area * Background Pressure) is located OUTSIDE of the rest of the equation? That's an important point about the format you need to keep aware of... You are basically adding a parallel term to the (Math.Max) function that only kicks in when inside an atmosphere (where Background Pressure =/= 0)

Yes I know, but the problem is that the current KSPI thermal noozles make no sence if you want to use them in an athmosphere (for timberwind).

There are several ways to solve it.

A Introduce 4 new noozles with physical smaller exit area and lower vacuum thrust mutliplier.

B add a modifier to the Timberwind reactors to pretend the ExitArea is smaller

C allow the Noozle to be configured in the VAB, a kind of procedural Noozle creator.

D ignore the problem

option C is of cource technicaly the best solution but also the hardest, I would only know how to implment it using the exisitng model, A is easy if someone could make new models (which I can't) but cost memory, B is relatively easy and straigtforward but not entirely correct from a visual perspective. D is the current state.

Edit: I think we should do C , because I also want to reduce the weight of the Nozzle depending on it's it ability to use LFO (Liquid Hydrogen + Oxygen as an afterburner). Perhaps it's also an idea to reduce the weight depending on maximum thrust. Any idea for a realistic model of weight reduction?

Edited February 12, 2015 by FreeThinker

[FreeThinker]

That is the case with all tech trees, installing multiple ones simply does not work and certainly not for existing saves.

I asked anonish (The creator of Tech Manager) to explain how Tech Manager would behave with multiple (conflicting) technodes, he refused to answer my question and told me (and everybody else) to find out themself. Perhaps this is his black humor, but the result is people getting into a technode mess (surprize, surprize) using several techtrees at the same time. I added some technode support to KSPI Extended, to make it less painfull, but it cannot fix everything, there needs to be some changes in the techmanager to fix this!

Edited February 12, 2015 by FreeThinker

[EvilGeorge]

You created a mess in your techtree, I had the same problem. I advice you te remove the TechManager folder, startup you career, load a safe, close, reinstall Techmanager and select only the CTT techtree.

I did exactly that (and removed the techmanager.cfg from my save folder - otherwise it would not prompt me to choose a tree) and it kinda works - the parts are available in VAB, it's possible to use the upgraded generators and radiators. But the tree still wants me to pay for some parts (even though they seem to be showing in the VAB and can be used, so it's safe to ignore, I guess..?) and some of the nodes are still empty, but visible. Did I do something wrong?

I certainly CAN play the game like this now, but is there a way to hide the empty nodes? I feel like I shouldn't even bother you (FreeThinker) with this, when it's not your mod that's causing it ~_~

Edit: After going to the VAB and then back to the science labs, the price for parts in all the nodes disappeared, so that's out of the way. There are still a few empty nodes.

Installing new tech trees in an ongoing game is a recipe for disaster. Parts are almost always assigned to tech nodes based on your installation, not on what TechTree you pick in game.

If you eg install CTT, it should be the only tech tree installed (besides stock obviously), the only one selected and you have to start a new game.

No idea why Nertea does not put a fat red disclaimer on it, I already asked him, but got no response.

That is the case with all tech trees, installing multiple ones simply does not work and certainly not for existing saves.

This is a great advice and probably an obvious one to someone using mods with KSP for a while. However, apart from being new at this, I was just updating KSPI. I didn't really know that I was "installing another tech tree" in the process until everything went bananas

Edited February 12, 2015 by EvilGeorge

[FreeThinker]

This is a great advice and probably an obvious one to someone using mods with KSP for a while. However, apart from being new at this, I was just updating KSPI. I didn't really know that I was "installing another tech tree" in the process until everything went bananas

It's activating additional techtrees in techmanger that cause problems, not installing them. Techmanager should have included some warning not to combine Techtrees mid game or risk serious problems. Now I fear we get more of these problems, people will start experimenting with multiple techtrees (like I did too). They will all eventualy hit a wall and some will come complaining to me.

Edited February 12, 2015 by FreeThinker

[Kielm]

Hello!

First: Great work so far, I really enjoyed playing through a career game with this version of KSPI.

I do have one possible bug report / question however. I've noticed that reactors do not consume resources when not in focus. Even leaving them unfocused at 100% usage, my fission reactors in LKO have used less than 1% of their fuel in 9 years, and fusion reactors can breed full tanks of tritium without using any Deuterium fuel during interplanetary transfers (providing you bring enough Lithium). This does kind of break the immersion a bit, but otherwise everything is working really well!

Thanks for your efforts

[FreeThinker]

I do have one possible bug report / question however. I've noticed that reactors do not consume resources when not in focus. Even leaving them unfocused at 100% usage, my fission reactors in LKO have used less than 1% of their fuel in 9 years, and fusion reactors can breed full tanks of tritium without using any Deuterium fuel during interplanetary transfers (providing you bring enough Lithium). This does kind of break the immersion a bit, but otherwise everything is working really well!

Most players don't even seem to notice or care but yes, this is one of those things on my toDo list. I can do a quick a dirty method like done Science Lab , but if possible it should be implemented with the Backgroup processing, that way, if you ran out of nuclear fuel, you genererator stop producing power and your kerbals should properly freeze (due to lifesoupport heating fallout out). Yes, space is deadly

Edited February 12, 2015 by FreeThinker