[0.25]KSP Interstellar (Magnetic Nozzles, ISRU Revamp) Version 0.13

[Eadrom]

Either you misunderstood him or I am misunderstanding him. It sounded like he just wants to change it so that you must have an antenna and 'transmit' the science rather than having magic science fairies bring it all home.

After getting a nice night of sleep, I see that I totally misunderstood. I thought he was asking for the ability to bring science experiments to the science lab and then be able to transmit them home without transmission losses. I see now he just wanted a button to transmit the science instead of automagically getting the science added as soon as you load the lab. I prefer the automatic method myself. Having to transmit would be a pain if you have a large number of labs. Had 12 labs in one save and that was a pain just swapping around to collect science. If I had to also transmit, that'd take me twice as long, if not longer, to reap my science. Different strokes, I guess.

I seriously can't figure out how to get anything with this pack's engines to even lift off, other than the DT Vista Inertial Fusion Engine. Can someone give me ideas for low tech designs in this mod?

The Methane engine is superb for launch stages. I'm looking forward to Fractal expanding this line of engines. I'd like to see a 1.25m flat lander-style methane engine so that I can make use of it for Duna missions.

Plasma thrusters are in the ion engine range of performance with LF using a 1.25m fusion generator, but really take off as a great engine once you get to space and power it with antimatter. I have also had some success using plasma thrusters as upper stage engines on ships and space station modules equipped with 2.5m or 3.75m fission power systems using Argon as fuel.

The Vista is indeed a nice launch engine, just keep in mind the radiation hazards if you have any crewed ground support vessels.

Thermal nozzles can be tricky. They work well in space, but with the severe performance reductions in atmo, I've only had limited success with nuclear reactors. They can work decently as upper stage engines. Fusion reactors work fantastically for space plane. However, once you unlock antimatter power, and if you can fuel new ships on the launchpad with antimatter, thermal nozzles become VERY powerful engines for a wide variety of applications, not just launching.

Another option you have with thermal nozzles is to build a microwave transmission network in orbit and at KSC. Lob a bunch of 3.75m fission reactor/generator combos into orbit such that you get some power everywhere in orbit. A bunch of rovers with 3.75m fission reactor/generator pairs parked at the space center. Then instead of attaching a reactor to your thermal nozzle, you attach a microwave thermal receiver. Get enough power and you can indeed launch. Power transmitters on the ground will get you up and going and the transmitters in orbit will provide the power to circularize. Keep in mind that Fractal has some tweaking and fixing on the microwave system that hasn't come out in 0.8.2 yet, so you may want to wait until the next bugfix patch comes out before setting up a vast microwave power system. Shouldn't be too much longer before 0.8.2 comes out

TL;DR: Methane, Vista, and plasma thrusters using Xenon or Argon (with a large power source) work well before you unlock antimatter reactors.

Edited December 2, 2013 by Eadrom

[hammyhamm]

For the first part: Have you tried running fuel lines FROM the aluminum-oxide engines TO the electrolosizer/something in its stack? If electrolysis works like normal generators, it's consuming a negative amount of the resource; in other words it's withdrawing negative oxidizer from the tanks. Aluminum flows to the entire ship, much like monopropellant, which is why you're seeing it increase. If you run kethane, set it up like you would set up your converter lines

I don't have any fuel lines running - it has full crossfeed access to the engine via girders so I didn't think there would be a connection issue whatsoever. Will attempt it now as the dude I have to save is stuck in a kethane-free area but I'll likely just redesigned the rocket with kethane in the future (which is a pretty overpowered mod, let's be honest).

[Donziboy2]

TL;DR: Methane, Vista, and plasma thrusters using Xenon or Argon (with a large power source) work well before you unlock antimatter reactors.

Near Future Propulsion has a few Xenon tanks. I think it uses a different Argon thou, but the xenon come in handy.

[Dodecadevin]

I'm having a problem with the Refinery electrolysis on Laythe. The craft is touched down in ocean with a 2.5m UF4 reactor + generator and a single refinery. Electrolysis function activates okay and shows typical power draw, but the resource collection rates aren't what I'm used to: It's producing 0.49 units Oxidizer/sec but only 0.06 units LiquidFuel/sec. Furthermore, resource collection isn't being affected by time warp...I get the same rates at every level of acceleration, no matter what I'd have to wait hours of realtime for this craft to fill up. This is wrong, right?

[forsaken1111]

I'm having a problem with the Refinery electrolysis on Laythe. The craft is touched down in ocean with a 2.5m UF4 reactor + generator and a single refinery. Electrolysis function activates okay and shows typical power draw, but the resource collection rates aren't what I'm used to: It's producing 0.49 units Oxidizer/sec but only 0.06 units LiquidFuel/sec. Furthermore, resource collection isn't being affected by time warp...I get the same rates at every level of acceleration, no matter what I'd have to wait hours of realtime for this craft to fill up. This is wrong, right?

Well the resource proportion you're getting is correct because (as I understand it) water is 1/8 hydrogen by mass and LF is hydrogen in this mod, so you get 1/8 hydrogen and 7/8 oxygen. The timewarp thing does sound like a bug though.

[Eadrom]

Near Future Propulsion has a few Xenon tanks. I think it uses a different Argon thou, but the xenon come in handy.

Yeah, I've heard that. The Kerbal Stock Expansion mod has a larger ion engine and xenon tank as well. I usually use Argon because when I've tried Xenon in the past, the plasma thruster just RIPS through the fuel soooo fast. Argon or even lithium can still provide some decent thrust while not burning through fuel tanks as quickly as Xenon. Though if you only need the thrust for a brief period, then Xenon would work great. There's a couple times where I've use radial clusters of Xenon tanks to great effect.

[QuixoticKoala]

Can someone lay out the step-by-step instructions for reprocessing nuclear fuel? I've been having a lot of trouble getting it to work.

[BigD145]

Methane is going to replace some of my lifter orange tank stuffs. The rest of the engines I have no clue what to do with.

edit

Methane is almost identical (deltaV, ISP, burn time) to Mainsail, just with lower costs. This is according to kerbalengineer and part for part replacement of a lifter with mainsail/orange.

Edited December 2, 2013 by BigD145

[hammyhamm]

Methane is going to replace some of my lifter orange tank stuffs. The rest of the engines I have no clue what to do with.

Imagine the Alcubierre drive as the ultimate middle finger in terms of time taken to reach a destination.

The ISRU stuff is great if you aren't running kethane and are alot more realistic. This mod needs its own wiki/tutorial as many things aren't obvious from the OP outside trawling through this thread for someone with a similar issue.

[db48x]

Here are some graphs for a hypothetical craft, created using graphviz.

First with the science lab turned on:

Second with both the science lab and the thermal nozzle active:

Last, all power to the thermal nozzle:

digraph second {
  graph [outputorder=edgesfirst bgcolor=transparent];
  node [style=filled fillcolor=white];
  edge [tailclip=false];
  rankdir=LR
  reactor -> radiator [color=orange penwidth=0 style="invis"];
  reactor -> generator [color=red penwidth=12];
  generator -> radiator [color=orange penwidth=8];
  radiator -> space [color=forestgreen penwidth=16];
  generator -> "science lab" [color=blue penwidth=4];
  "science lab" -> radiator [color=orange penwidth=4];
  reactor -> "thermal nozzle" [color=red penwidth=12];
  "thermal nozzle" -> radiator [color=orange penwidth=4];
  "thermal nozzle" -> space [color=forestgreen penwidth=8];
}

Edited December 2, 2013 by db48x
clean up the edges a bit

[Kashaar]

Hello, everyone!

I just wrote up a 2000-word, 23-picture report of a Jool mission and inadvertant rescue mission using Interstellar that some of you might find entertaining. You can find it here: The flight and plight of the Rainbow Retriever

And if by chance anyone can tell me what could possibly have gone wrong at the end, or if it's "just" a weird and obscure bug that caused the energy drain, that would be awesome too. But I'm content just labeling it an equipment failure

[Myrten]

I've found another bug:

Microwave receiver dish also works behind - if ship is directly on the opposite side of where dish points, dish will also receive full power just like it would be in front of it.

I've also took a look at receiver's source code since relays acted a bit weird, can you confirm these two things:

1)Power can only be routed through single relay so such connection as in this image won't work right now:

2)If receiver is connected to transmitter via single relay it has to be pointed at transmitter not at relay to receive maximum power even though it doesn't see transmitter at all.

[forsaken1111]

I've found another bug:

Microwave receiver dish also works behind - if ship is directly on the opposite side of where dish points, dish will also receive full power just like it would be in front of it.

I've also took a look at receiver's source code since relays acted a bit weird, can you confirm these two things:

1)Power can only be routed through single relay so such connection as in this image won't work right now:

2)If receiver is connected to transmitter via single relay it has to be pointed at transmitter not at relay to receive maximum power even though it doesn't see transmitter at all.

Not sure about your second point but your first is in line with what he said about the relay system. It will only relay through a single craft.

[Myrten]

Not sure about your second point but your first is in line with what he said about the relay system. It will only relay through a single craft.

OK thanks, I must have missed that, although I hope it will change

Edited December 2, 2013 by Myrten

[Fractal_UK]

OK thanks, I must have missed that, although I hope it will change

The bug you mentioned is fixed in my development version, as is point 2, also I've fixed the calculation of distance losses via relays.

Changing the system to add more relays is potentially problematic because the maximum number of calculations can become enormous with many relays. At present the maximum number of line of sight calculations is N_T*N_R where N_T is the number of transmitters and N_R is the number of relays. So, if you had 50 transmitters and 20 relays, you're talking 1000 calculations.

If you could have unlimited relays, you're talking an upper limit of N_T*N_R! or, using the same numbers 1.2x10²⁰ calculations in the worst case. In practice it won't even be close to this value but you get the idea about how quickly things scale up.

[NotCoach]

The bug you mentioned is fixed in my development version, as is point 2, also I've fixed the calculation of distance losses via relays.

Changing the system to add more relays is potentially problematic because the maximum number of calculations can become enormous with many relays. At present the maximum number of line of sight calculations is N_T*N_R where N_T is the number of transmitters and N_R is the number of relays. So, if you had 50 transmitters and 20 relays, you're talking 1000 calculations.

If you could have unlimited relays, you're talking an upper limit of N_T*N_R! or, using the same numbers 1.2x10²⁰ calculations in the worst case. In practice it won't even be close to this value but you get the idea about how quickly things scale up.

Practice could get there. I am currently building a RemoteTech system that has 3 satellites around every body, including Kerbol. With a transmitter and relay on each that might mean 51 transmitters, and 51 relays. I guess I need to rethink that. What about only transmitters at Kerbol, and receivers and relays everywhere else?

[johnsonwax]

Well, RT has a few tricks. The big one is that it's limited to SOI. Omni antenna only work within SOI (and subbodies) and have limited range, so you can quickly cull the list of craft to check. The dishes target individual craft (trivial) or a body, so you only need to consider craft within that body's SOI, and only those with a dish with sufficient range to reach back - again a way to quickly cull the list. And many of those properties only change when the ship attains focus, so the potential graph can be safely cached rather than constantly checked.

If I were implementing this, I'd simplify and constrain things:

1) I'd make the transmitter a transmitter/receiver and eliminate the receiver as a part. Perhaps make a scaled down version with higher losses, lower max throughput.

2) I'd require all connections be point-point as RT2 is. Give the transmitters a reasonable cone angle. 'Active Vessel' should be a target option.

3) Eliminate the one relay limitation.

Let's say you have a massive ground-based power station. You beam power up to a relay and you want to relay that out to some number of other destinations. You spread that out among your sat net around Kerbin (for constant LOS) and each of these has a transmitter for each destination - Mun, Minmus, Eve, Duna, etc.

Now, you can build a truly epic power network, but it's going to be expensive. It's a lot of transmitters to put up. Or you can limit it by only covering the SOI - a 3 sat network would need 1 for the ground link, 2 for each of the other sats (or 1 if you're willing to incur additional transmission loss), and 1 for active vessel. You can now power your outbound ship, carrying a new ground station for Mun, Duna, etc. and build a similar network there. It can also target active vessel and your active vessel, if armed with 2 transmitters can draw power from each of the networks. You can steadily build up and out this way, and it'll scale very nicely.

RT2 has the identical mechanic, but with some variation due to antenna range. It also has the ability to create new control points so you can have parts of your system disconnected from each other, etc. Very flexible, and not terribly expensive computationally because you only need to work out LOS between a handful of objects at any given time, regardless of the size of your network.

[Myrten]

The bug you mentioned is fixed in my development version, as is point 2, also I've fixed the calculation of distance losses via relays.

Changing the system to add more relays is potentially problematic because the maximum number of calculations can become enormous with many relays. At present the maximum number of line of sight calculations is N_T*N_R where N_T is the number of transmitters and N_R is the number of relays. So, if you had 50 transmitters and 20 relays, you're talking 1000 calculations.

If you could have unlimited relays, you're talking an upper limit of N_T*N_R! or, using the same numbers 1.2x10²⁰ calculations in the worst case. In practice it won't even be close to this value but you get the idea about how quickly things scale up.

Yeah I've realized that this could be the reason, but that's only the case if you check it by 'brute force' method. I think there might be a smarter way to do this with lower algorithmic complexity, I'll check this.

Viable solution might be also limiting maximum number of 'hops' between relays to prevent the worst case scenario from happening.

[VorpalHerring]

You could make it more like RemoteTech, short range omnidirectional transmission with higher attenuation, long range directional with low attenuation.

It could be implemented as a toggleable mode.

[Fractal_UK]

You could make it more like RemoteTech, short range omnidirectional transmission with higher attenuation, long range directional with low attenuation.

It could be implemented as a toggleable mode.

Omni-directional is pointless for power transmission, you'd loose about 99.6% of the power transmitted with a 2.5m receiver only 10m from the transmitter. Over 10km, you'd get a million times less power than that.

[VorpalHerring]

I meant omnidirectional more in terms of forming connections, as a way of limiting the number of connections.

The omnidirectional ones would be shorter ranged to incentivise using directional point-to-point transmitters.

[johnsonwax]

Omni-directional is pointless for power transmission, you'd loose about 99.6% of the power transmitted with a 2.5m receiver only 10m from the transmitter. Over 10km, you'd get a million times less power than that.

Well, the current setup already suffers from that as a transmitter can connect to any and all LOS receivers, regardless of distance. What Vorpal is describing might be best described as a short-range detection/negotiation system, where a transmitter can detect a nearby receiver and automatically orient to it, but it can't detect a long-range one. Long range needs to be manually targeted.

You have the added variable of power loss over distance, which can create a different set of decision points for players.

Realism is already mostly shot to hell here. Microwave power rectenna would be on the scale of kilometers for any reasonable amount of power, and beaming a megawatt to any manned spacecraft we can design in-game is going to cook the occupants of the ship. This is more about how to get something realistic-sounding that is deep enough to do interesting and fun things with, but that won't cripple our computers. A single relay is very limiting, but multiple relays with conversion inefficiencies and power loss, and the cost of hauling up big transmitters for any non-trivial point-point network is hard, but lets you do fun/useful/stupid things. 10 relays to Eeloo? Sure. Just build and cool a 100 petawatt power plant so you can get a couple megawatt at the far end, or come up with any of a million different more efficient designs.

[InfinityArch]

Any chance that we might eventually see beamed power light sails?

[Doomydoom]

It would be nice if the transmitters/relays were assumed to (or actually did) auto-orient to a target. It's a bit silly to have this thing that can't move at all.

[Fractal_UK]

Right, I've finally got this microwave power system functioning how I want it to - namely respecting that all important first law of thermodynamics. Every receiver is now aware of how much power is being received from each satellite by every other receiver meaning that horrible energy amplification effects should no longer be possible. It's a bit difficult to demonstrate this because the power satellite is in a low orbit and moving over the sky quickly but hopefully you can see that as I deploy more receivers we're getting diminishing returns.

The amount of power theoretically available is 69.79% of 61.9 KW or 43.2KW. Whether we get these optimal conditions, however, depends on the angle of reception.

Total Power = 26.24 KW

Total Power = 33.14 KW

Total Power = 34.26 KW

Total Power = 34.84 KW

It would be nice if the transmitters/relays were assumed to (or actually did) auto-orient to a target. It's a bit silly to have this thing that can't move at all.

Transmitters and relays are already assumed to always point directly at the receiver, it's only the receiver that needs to be oriented properly.