Ion Thrusters: Suggest Allowing Leave to Space Center While Throttled Up

[Diche Bach]

Fantastic game 91 hours logged on Steam and I've got a nice constellation of satellites & probe ships in stable orbits around Kerbin, Mun, and Duna and a few others careening off in trajectories I didn't quite intend.

By far and away the finest space game ever made. All other designers should take note of how space games really should be based fundamentally on real physics.

After many attempts to design an efficient long-range ion thruster driven probe, I have finally achieved a nice balance. Five Rockomax Mainsails, plus four solid boosters take it into orbit then on to a Mun encounter. A smaller engine and a 720 Rockomax tank is good for at least a couple of fly-by maneuvers. The final main mission vessel has 1 ion engine, enough tanks and batteries for 7800 xenon gas and 2010 electric, plus the scientific instruments of course. With 6 gigantors and a few other smaller panels this configuration can run 1 ion engine at full thrust indefinitely in good sunlight.

However I'm disappointed at how the game mechanics (still using vanilla no mods installed) make Ion Thruster propelled space craft fairly unusable (or at least less fun) to play with. Specifically, because of their low acceleration performing all but the most minuscule of maneuvers requires that they 'burn' for several minutes; sometimes as much as 45 minutes or an hour.

As such, it seems the only way I'll be able to use my long-range ion driven probes is if I leave the game running at 1x speed while I either stare at the screen for 30 or 45 minutes or put the game in the background. Obviously both viable. However, my suggestion (if it is possible) would be to adjust the game mechanics so that it is possible to leave these types of engines running even while the space ship is not the focus.

I did a search and did not notice any threads that clearly were suggesting this change to game mechanics.

[Ernest3.14]

You can use physics warp to get up to 4x the speed (if your ship is fairly small). The problem with "adjusting the game mechanics" is that is easier said than done. Currently, your ship is put on rails if it isn't the focus, and that requires calculating the trajectory. Unfortunately, you can't just "calculate" the trajectory for a ship that is under acceleration.

[Digger412]

Diche Bach,

At warp levels greater than one (maybe four, don't remember off the top of my head), the KSP physics engine simulates objects as being "on rails". If you were to implement a way to allow acceleration at warp greater than four, the integrity of the physics accuracy would most likely degrade as the physics delta_t would have to be accordingly increased as well (unless you have a badass processor) to maintain simulation speed.

Now if, instead of allowing acceleration under warp, you wanted to instead switch to a different craft or building, I could see this as being more viable. You would (hopefully) still be unable to increase the warp factor, but you could at minimum do other things with your time while waiting. I think the concern with this though is that if something goes wrong with your ion-powered craft (the course changes by accident or the burn needs to be adjusted, staging needs to be performed, Jeb decides that he's curious what an ion engine under power looks like and goes EVA only to get left behind), there is no way to place the fault onto the person at the controls since there was no one there, and you might not be able to revert to a quicksave before the incident (though running the game in the background already does this, it is assumed that you are still at the controls).

Maybe there is a better way to accomplish this, such as pausing the 3D rendering, audio, etc., until a condition is met (like a SoI change or a certain time elapsed) and only performing the physics calculations necessary, then resuming normal operations. You still would not be able to adjust course with this suggestion, but it might be a way to getting around waiting two hours for a burn to complete.

edit: Beat to it by Earnest. Too bad I take a long time to formulate posts.

Digger412

[Diche Bach]

You can use physics warp to get up to 4x the speed (if your ship is fairly small).

Really? All I've noticed is that any time you are applying any throttle with any space craft, it will not allow activation of the physics warp at all.

The problem with "adjusting the game mechanics" is that is easier said than done. Currently, your ship is put on rails if it isn't the focus, and that requires calculating the trajectory. Unfortunately, you can't just "calculate" the trajectory for a ship that is under acceleration.

Yeah I was afraid that was a real constraint. Disappointing to hear you confirm that it is.

Am I the only one who finds this limitation on using the Ion thrusters to be 'annoying?'

If so, then I guess there is really no point in Squad trying to figure out some sort of work around.

But if other players agree, then perhaps it is worth Squad thinking about some sort of workaround to make ion thruster ships more useable? IMO, ion thrusters are more or less useless at present, though I'd be very curious to hear if anyone has a way to use them that mitigates this slow acceleration issue.

At present, with only a fairly small solar system to 'explore,' and no real science to do, this issue with ion engines being less than optimally useful may seem fairly trivial. But as the game evolves and the desire for scenarios or campaigns that model economically-efficient long-range interplanetary or edge-interstellar missions grows, I think it will be realized that this shortcoming in how Ion Thrusters operate is a serious limitation. It is my understanding that, for long-range space craft of the next 50 years, ion thruster technology is likely to be one of the most prominent. Unless I misunderstand how these systems are used and will continue to be used in real life, applying thrust for long periods is typical. And that is not very practical to do in KSP at present.

[Kaleb]

Am I the only one who finds this limitation on using the Ion thrusters to be 'annoying?'

Nope.

One of the first things I did in the game was to create a bunch of ion probes to start unmanned exploration. I fired one up and went to go do something else at the space center (I didn't even know about time warping at the time), and I had *no idea* why I was unable to leave the flight without ending it. The space center option was grayed out without any explanation of why. Eventually I discovered the restriction on leaving (or time-warping) with throttled-up craft, and I've been disappointed with it ever since. I love designing with Ion engines, but I hate trying to fly them for this exact reason.

The other problem with them is that they don't really work with the Maneuver Node system, since it (reasonably) assumes instantaneous impulse burns, rather than long continuous burns. Imagine what you could do if you could set up an ion engine to fire continuously (even at <10% power) and set it on a course towards Duna or Jool. For short distances like the Mun, it may not make sense, but over the long haul, long periods of low thrust may get you there faster and more efficiently.

Unfortunately, I'm not sure what can be done about this, since continuous low-thrust trajectories are still an active area of research, and are much more difficult to compute than standard Hohmann transfers. But I would love for Squad to do something about this, if possible.

BTW, with only one Ion engine, you should only need 1 gigantor array to run it at full thrust (around Kerbin, assuming its unobstructed, and oriented towards Kerbol). Definitely not six. The rough rule of thumb is 1 gigantor per ion engine. However, Gigantors are also heavier than the equivalent amount of smaller panels (1x6 or 2x3 panels) needed to produce the same amount of electricity. IIRC, 8 smaller panels produce the same amount as 1 gigantor, but weigh less.

You can also cut the burn time down somewhat by adding on additional engines, but this requires additional solar panels, which drive up the mass, and reduce the dV. But this approach seems to miss the whole point of ion engines, which is long periods of low-thrust.

[Themohawkninja]

I wish a dev would reply to this, because I recall a while ago that Harv was going to implement a "jump-to" system, whereby you would initiate a burn, hit the button, and the game would jump you to when the burn was done. I don't know if that was ever put into any sort of development, or if the devs still even want to do it.

[Diche Bach]

I wish a dev would reply to this, because I recall a while ago that Harv was going to implement a "jump-to" system, whereby you would initiate a burn, hit the button, and the game would jump you to when the burn was done. I don't know if that was ever put into any sort of development, or if the devs still even want to do it.

Something like that would certainly add much more utility to ion thrusters. Based on how the engine seems to 'pause' for calculations, and then 'resume' play whenever you pass an encounter node, it would seem (to my naïve eye) to be the sort of functionality that is already in the game in some sense.

Ideally, a system whereby one can plot a maneuver with an ion-thrusted vessel, tell it to "stay on trajectory" and then go away and play other flights or whatever while it was chugging along would be great. But that seems like it would be a couple steps more complicated than a "jump-to" sort of functionality.

BTW, with only one Ion engine, you should only need 1 gigantor array to run it at full thrust (around Kerbin, assuming its unobstructed, and oriented towards Kerbol). Definitely not six.

Ah thanks for that info Kaleb. My initial attempts were really just exploratory to get a feel for how the components work together and they were woefully underpowered even with only 1 engine. I tried adding a few of the 1x6 panels and that didn't seem to do the trick and iterations 5 through 8 wound up in the dust bin for various reasons . Finally I just decided to go overkill and put a total of 6 Gigantors _plus_ additional. I think my "Viking 10" model must have about 16 of the 1x6 extending panels, 16 of the small cheapo tack-on panels plus the 6 Gigantors = running one ion thruster at full blast became no problemo!

[Kipeo]

time warp != physics warp.

to use physics warp, hold alt-> and alt-< instead of > and <

Physics does 2x, 3x or 4x, but does full, well, physics. Thrusting and collisions and all, and can be used any time.

Time warp puts you on rails and starts at 5x but has restrictions on when you can do it.

[m4v]

The final main mission vessel has 1 ion engine, enough tanks and batteries for 7800 xenon gas and 2010 electric, plus the scientific instruments of course. With 6 gigantors ...

That's like 3.8 tons of mass without counting the brain, IMO that's a bit big for a probe with one ion engine. If you're going to use ions then you gotta think small, KPS's ion engine is already like 100 times more powerful than IRL counterparts. I also think that the low power of ions have gameplay value as it forces you to plan your missions accordingly, if you want to take advantage of the high Isp ions give, then you have to design a small probe.

Edited July 31, 2013 by m4v

[Scoundrel]

Really? All I've noticed is that any time you are applying any throttle with any space craft, it will not allow activation of the physics warp at all.

Alt+<> (or command+<> depending) will physics warp your ship when out of atmosphere. Apparently KSP can't go faster than that without the physics going nuts, which is why x10+ warp with ion engines will likely never happen.

[Diche Bach]

time warp != physics warp.

to use physics warp, hold alt-> and alt-< instead of > and <

Physics does 2x, 3x or 4x, but does full, well, physics. Thrusting and collisions and all, and can be used any time.

Time warp puts you on rails and starts at 5x but has restrictions on when you can do it.

Hell yeah! 4X while flying ion thrusters is better than nothing at all! Thanks for clearing that up for me guys

That's like 3.8 tons of mass without counting the brain, IMO that's a bit big for a probe with one ion engine. If you're going to use ions then you gotta think small, KPS's ion engine is already like 100 times more powerful than IRL counterparts. I also think that the low power of ions have gameplay value as it forces you to plan your missions accordingly, if you want to take advantage of the high Isp ions give, then you have to design a small probe.

I have a bit of a tendency to over-engineer I guess Thanks for clarifying that I'm well above the optimum maximum probe size. I think now that I know about this physics warping and also that 1 gigantor can power a ion engine I'll go back to the drawing board and see about something nice and light.

I tested out a series of 6 or 8 probes and managed to make some nice progress with them. One dropped off its penultimate stage as an orbiter at Mun right after I lucked into a gravity assist that helped me get the main probe all the way to Duna. However, once I got there, I had to expend all my RCS and eventually that probe ran out of fuel and I left it for dead. None of the others ever had enough power flow to run the engine at full blast, and so when I designed my final one I went a bit overboard on everything with the intent that the vehicle can be multi-functional or something.

Edited August 1, 2013 by Diche Bach

[NERVAfan]

This would be great.

Alt+<> (or command+<> depending) will physics warp your ship when out of atmosphere. Apparently KSP can't go faster than that without the physics going nuts, which is why x10+ warp with ion engines will likely never happen.

Even with the small acceleration applied by an ion engine?

Or could there be an 'activate higher warp' option you had to go into the options menu and check off, so that people who didn't want to deal with it wouldn't have to or run into it by accident, but we could take the risk of weirder things happening if we wanted to try it?

[Scoundrel]

Even with the small acceleration applied by an ion engine?

What would happen is all the forces that affect the vessel would by multiplied by 10, so slight oscillations would be massive wobbles, and the slightest asymmetry or offset would have a huge impact. As more parts are added to the ship, the more physics interactions occur, including some impressive self destruction as ships shake themselves to pieces. A 10 part probe would have few problems, but a 50 part probe would disintegrate. There's some old posts by the devs that explain it far better than I can if you want to dig through the archives here and on Reddit.

[Dunbaratu]

It seems to me that if it's possible to calculate the elliptical path as a time-parameterized equation for on-rails coasting, it should also be possible to calculate what the curved path should look like under a thrust as well. It seems like a calculus problem to just figure out the position vs time equation given two different accelerations - one due to the gravitational primary body, and one due to the craft thrusting in a constant fixed direction (although the thrust is in a constant direction it's not in a constant acceleration because you also have to calculate how the fuel gets burned and reduces the mass as you go). I'm not saying I particularly want to be the one to do the calculus to figure it out, but it does seem like it should be a solvable problem. You know the formula for how the acceleration due to the engine at constant throttle changes over time as the fuel burns ( a = F/m, where m is changing over time by a linear slope downward ), and you know the formula for how gravity pulls depending on current position relative to the primary body, so in principle if you add those two together you should get an equation for the total acceleration from both effects and be able to integrate it twice to get the position equations. Of course this shouldn't be done on the fly, but be done ahead of time by the mathematicians in the dev team, and then the final result could become the new "on rails" equations.

[Scoundrel]

It seems to me that if it's possible to calculate the elliptical path as a time-parameterized equation for on-rails coasting, it should also be possible to calculate what the curved path should look like under a thrust as well... <snip> ...Of course this shouldn't be done on the fly, but be done ahead of time by the mathematicians in the dev team, and then the final result could become the new "on rails" equations.

Hm. Well put. Your logic is sound and it seems like something that, under extremely low thrust, should be doable. I think though that the problem would lie in the interaction with other SOI as the orbit increases - after all, presumably the goal is to actually go somewhere via ion power - which would lead to inaccuracies between rails and physics once out of time warp, would it not? I've already seen such inaccuracies in map mode, though that doesn't mean I think the problem is completely insurmountable... after all, there's loads of good suggestions by far brighter people than me in chaos_forge's VOI thread, so it may just be my ignorance speaking (my maths are limited to ballistics and anything to do with cars/race trucks) when I talk about inaccuracies and the like. Anyhow, I still think you make a good point, so I'll leave this post on that note.

[Psycix]

Perhaps a physics benchmark can determine the acceleration your ship makes and add it into the rail.

[Abram730]

Programmable probes would be better IMO. That is the ability to set a maneuver and have the probe "try" to do it.

[tssn1611]

Hm. Well put. Your logic is sound and it seems like something that, under extremely low thrust, should be doable. I think though that the problem would lie in the interaction with other SOI as the orbit increases - after all, presumably the goal is to actually go somewhere via ion power - which would lead to inaccuracies between rails and physics once out of time warp, would it not? I've already seen such inaccuracies in map mode, though that doesn't mean I think the problem is completely insurmountable... after all, there's loads of good suggestions by far brighter people than me in chaos_forge's VOI thread, so it may just be my ignorance speaking (my maths are limited to ballistics and anything to do with cars/race trucks) when I talk about inaccuracies and the like. Anyhow, I still think you make a good point, so I'll leave this post on that note.

Not that I really have time for this, but just for fun, I tried to preliminarily crank through the calculus using the starting point Wikipedia gave me. Turns out, at least in polar coordinates, if you assume tangential acceleration is zero (i.e. your engine is off), that lets a whole bunch of stuff be constant, meaning a lot of terms end up going away. Once you add tangential acceleration in there (as a function of time, no less), those terms stick around, and the equation is much more complicated. I worked on it until I got to a system of equations I don't think one can solve analytically.

Edited August 2, 2013 by tssn1611

[benjiful9000]

[closed]

Edited July 4, 2021 by benjiful9000