[WIP] Mass Driver

[skykooler]

Well just tried it for the first time and I can say that it is awesome except for you never know which direction it will throw you at!

This will hopefully be fixed when I release the next update, which should be sometime in January.

[7499275]

This will hopefully be fixed when I release the next update, which should be sometime in January.

I didn't realize you were still working on it! thats great news!

[The Ideal Gas Lawyer]

Still hoping for an update but it's not looking so good...

[wasmic]

Still hoping for an update but it's not looking so good...

He said sometime in January, so I'd assume that he's working on it - I don't know how intensely, though. Keep in mind that he also has a real life to attend.

[lajoswinkler]

I'm not in the mood for reading the entire thread, so here's a quick question - can you use this as a rollercoaster, to slide down from the top?

[Hooligan Labs]

Rover does science in space!

I do love alternate ways of traveling to space and have experience with Unity, the Kraken, KerbTown KSP mods. I have some ideas on how to make your mod work, any way I can help?

... Okay, I'm just going to write my ideas down before you answer that. Maybe the Kraken code is almost working! If so, please forgive me.

Create a box-like object and port it in with Unity. This will be the superconducting element that lifts and pulls the spacecraft along the track. The element can be attached to the bottom of the spacecraft. When it is activated between the two rails of the launcher use Lerp (straight track, should be fairly easy to program) or Slerp (arc shaped track, much more difficult) to have the superconductor travel the path from start to finish. To properly reflect acceleration, have the T increments increase at a steady rate. When the end is reached, use the decouple script command to have the spacecraft break away. If there are issues with KSP not having it launch properly, I think you can painlessly set the velocity vector using SetWorldVelocity.

I would strongly recommend having zero drag on the vessel to represent it traveling through a vacuum tube. You probably don't need to model the tube, just assume that it is closed by an invisible plasma barrier with walls made out of transparent ultrastrong graphene. To zero drag, upon entering the tracks make a list of every part, it's drag (a simple Unity value that does just what it says) and node sizes (I think this is what Ferram uses to calculate drag). Set these values to 0. Upon leaving the launcher or breaking away from the vessel, set these back to their original values.

As a final note, it would be most realistic to have the launchers climb the slopes of Kerbin's mountains. Launch pads accessible by KerbTown would make access easy.

That's all I can say for now! Lots of the ideas came from the wiki article on the StarTram.

Edited January 13, 2014 by Hooligan Labs

[roxedboxer]

Cool enough concept..always explodes jet engines, makes my craft flip out, and sends it into about a 120 degree inclination...

[Mein_Gott]

do I see inspiration comming from 'Glimmer Rats'?

[lyndonguitar]

Yey! now we have mass drivers!!! I wonder how many Gs it will give.

The only thing missing now are space elevators and interstellar flight

[MaverickSawyer]

OK, I am going to have to add this once it's updated... Still got another week before the OP's deadline comes up.

[Rokker]

Cool enough concept..always explodes jet engines, makes my craft flip out, and sends it into about a 120 degree inclination...

idk, that sounds more like just KSP in general to me.

[roxedboxer]

True enough...but I would like it to work a Bit better...

[skykooler]

The reason there have been no updates yet is because KerbTown still hasn't been updated for 0.23, and I no longer have an earlier version of KSP to work on (and wouldn't really want to release an update that required an earlier version either). I assumed it would be updated by mid-January, but it still hasn't. So now this is kind of stuck until that happens.

[The Ideal Gas Lawyer]

The reason there have been no updates yet is because KerbTown still hasn't been updated for 0.23, and I no longer have an earlier version of KSP to work on (and wouldn't really want to release an update that required an earlier version either). I assumed it would be updated by mid-January, but it still hasn't. So now this is kind of stuck until that happens.

Sadface

The message you have entered is too short. Please lengthen your message to at least 10 characters.

[skykooler]

I have discovered that "not updated to 0.23" means "some of it broke, but the bits I need for the mass driver still work". So now I'm back on track.

Also, I don't properly understand Krakensbane.

[HerrGeneral]

That's awesome, I'm glad to hear you can start working on it again. That Ace Combat mission was one of my favorites because of the mass driver, so it's great to know that eventually I can launch stuff of my own.

[nahkranoth]

Why would you need Krakensbane? I am also busy developing something similar like you are making (started it before I found your thread) and because I found your thread I decided to take a different path. I can now launch a craft from a cannon and have a super quick acceleration (from 0 to 10.000 m/s in less then a second) and everything stays together.

The problem I had that I thought needed Krakensbane was actually that the friction of the parts differ from eachother so the craft gets ripped apart. I solved it by setting the friction of every part to 0.1 when it accelerates and setting it back when done accelerating. Maybe thats also a solution for your problem?

[skykooler]

The reason I need it is because the floating origin kicks in once you translate more than 2.5 km or so from the origin. The thing is, the mass driver is around 40 km long. As a result, when I try to smoothly move the craft down it it only gets to just past the end of the runway and then takes off into interplanetary space when Krakensbane kicks in, so I need to figure out how to manually reset the origin before that happens.

[Northstar1989]

@Skycooler

Is there any chance I could persuade you to consider changing your design strategy a bit? You're building the mass drive in the wrong place.

What you REALLY should be doing is building a high-altitude launch ramp.

If you build it in the mountains west of the KSC (for simplicity and proximity), you can achieve base altitudes of over 4600 meters.

Throw down a flat concrete pad on top of one of the peaks with Kerbtown, to act as a level foundation, and you could easily build your mass driver there. For *bonus points*, build the platform out of steel instead (with a concrete base) and a couple hundred meters tall for a bit greater benefits in terms of high-altitude launch (listed below).

Vessels could be spawned there with Extraplanetary Launchpads until the Kerbtown spawnpoints are fixed. Here's an example of a launchpad I recently set up at 4630 meters on the flattest spot I could find of the southernmost peak in the mountain range west of KSC, as the start of a larger manually-constructed launch facility:

If you build a high-altitude launch ramp, you will gain several advantages:

(1) You can build a more level mass drive since the ideal trajectory is less vertical at that altitude- ideally just a constant slope to avoid having to mess with curvature at all in its construction/design. Reducing curvature also reduces the normal-force on the craft as it curves up.

(2) Your craft will experience less drag. At 4600 meters, the atmosphere is less than 40% as dense as at sea-level. This will help with your vessels not tearing themselves apart from atmospheric drag during the acceleration, allow them to accelerate more quickly on the ramp, and not lose as much of their hard-earned velocity immediately upon leaving the mass drive if they greatly exceed terminal-velocity.

(3) If you build a shorter, constant-slope launch ramp, you may be able to fire rocket engines while still on the ramp; to achieve greater velocity by the end of the ramp. At higher altitude, your rocket engines will get better ISP doing this (and eventually when Squad changes atmosphere curves to be variable-thrust as promised, and like in real life- better thrust instead of ISP).

(4) The obvious benefit of having less altitude to climb before you're out of the atmosphere. It takes a bit less Delta-V to get to orbit just on that account.

I'd also love to see a mass-drive on the Mun, if you ever get around to it- but due to the much lower gravity there and lack of an atmosphere, that's obviously a much lower priority...

Regards,

Northstar

P.S. I'd help you design/build a high-altitude version- but I have no idea how to design or build stuff in Kerbtown. More importantly, I have no idea how you actually are accelerating the vessel, from a coding standpoint. If you think it's all simple enough to learn though, and want to help me figure it out, just send me a PM- I'd be more than happy to work with you.

Edited February 10, 2014 by Northstar1989

[skykooler]

To address those points:

1) Good point. However, it still needs to curve up somewhat, as the mouth is at 14 km altitude (I was unable to reach orbit from any lower without launching at rediculous velocities), so this is only a change in degree.

2) There is no drag on the ramp itself. The technical reason is that the vessel is on-rails until it is ejected; the explanation is that the interior of the ramp is a near-vacuum.

3) There is no good reason to fire engines while on a mass driver. Any acceleration you get from the driver is essentially "free" in terms of onboard fuel usage, so any further acceleration you want should be accomplished by dialing up the driver itself rather than wasting fuel. Also, see 2) (vessel is on rails).

4) See 1) - mouth of the mass driver is at 14 km, which is unaffected by the altitude of the start.

As for the Mun, a mass driver is on my todo list; they are much simpler there, as they can be straight lines. But I want to get the one on Kerbin working first.

[Northstar1989]

To address those points:

1) Good point. However, it still needs to curve up somewhat, as the mouth is at 14 km altitude (I was unable to reach orbit from any lower without launching at rediculous velocities), so this is only a change in degree.

If you're going to give it all the velocity to exit the atmosphere just from the launch rail, then yes, the top will need to be at quite a high altitude. However if you simply use it as an assisted-launch technology, and the rocket still has to burn its engines at some point, then there's no need to make the mouth nearly so high up...

2) There is no drag on the ramp itself. The technical reason is that the vessel is on-rails until it is ejected; the explanation is that the interior of the ramp is a near-vacuum.

In an enclosed, tunnel-style launch rail with evacuated air, then yes- you would expect no drag. However this thing is clearly open to the air- so I would think drags should be an issue. You're saying that in-game it has no drag as it's on rails though? Well in that case, at least you lose a lot less velocity to drag after release (assuming you release the rocket at a higher altitude due to a higher foundation height).

3) There is no good reason to fire engines while on a mass driver. Any acceleration you get from the driver is essentially "free" in terms of onboard fuel usage, so any further acceleration you want should be accomplished by dialing up the driver itself rather than wasting fuel. Also, see 2) (vessel is on rails).

If you can dial the mass driver up enough without the game bugging out, then sure, there's no need to fire engines. (n real-life, there's an issue of being able to create powerful enough electromagnets without frying all the rocket's on-board systems, as no magnetic shielding is perfect...) You could still get better ISP for rocket burns after release if a higher foundation meant a higher release height though...

4) See 1) - mouth of the mass driver is at 14 km, which is unaffected by the altitude of the start.

I guess a big point I'm getting at: a 14 km mouth with a sea-level foundation is unrealistic and overly-complex. In real life, any great curvature upwards after the rocket has reached a little over Mach (340 m/s in the Kerbal-universe) would be sure to kill the crew of any manned craft due to G-forces... What's more, no modern construction technology would allow a free-standing launch rail that extends that far above the ground and can still support the weight of a sizable rocket without frankly ridiculous amounts of structural reinforcement and size (this is one of the reasons the more serious proposals in real-life suggest staring the launch tunnel deep underground, and placing the mouth in a mountain peak).

So, I feel that building a mass drive with a higher foundation and a lower mouth would be both more realistic. A higher foundation would also allow you to have a higher mouth without unrealistic degrees of curvature... Though, if you want a realistic drive that gives you a significantly greater exit velocity, start the ramp near sea-level and curve it up to a *constant slope* that matches a mountain slope *near the beginning* of the track, where the velocity is still low. This is what the larger open-rail drive plans suggest doing in real-life.

As for the Mun, a mass driver is on my todo list; they are much simpler there, as they can be straight lines. But I want to get the one on Kerbin working first.

You should totally get on it- or better yet, Minmus! The gravity is even lower there, so you wouldn't need as large a mass drive; and the planetary curvature is much greater due to the smaller radius, so you wouldn't need to curve the actual drive nearly as much to avoid obstacles like mesas/mountains/plateaus near the horizon!

I know it doesn't take much Delta-V to escape Minmus, but every little bit helps... Plus, on Minmus, you could EASILY set up a Mass Drive that would shoot vessels out at Kerbin escape-velocities (from that high orbit), perhaps even being used to give vessels a significant portion of their Delta-V to get to interplanetary targets like Duna or Eve...

Anyways, at least start the mass drive at the base of a mountain and curve up the slope, for realism's sake, if nothing else... In fact, that's an even better idea than just starting it on a mountain-top, and would make the base of the drive much easier to get to...

Also, like I said, I'd be more than willing to help with any of these plans- I just don't know how.

Regards,

Northstar

Edited February 12, 2014 by Northstar1989

[Northstar1989]

Take a look at these links for a little background-reading. I think you might understand a little better then what the real-life proposals look like, and why I'm suggesting that for an open-rail, above-ground design (the only type really feasible in KSP) you start at the base of a mountain and match the slope (at more or less constant inclination after the initial curve), instead of starting at sea-level and curving upwards at an unrealistic and deadly pace...

http://www.g2mil.com/high.htm

http://www.g2mil.com/skyramp.htm

http://www.g2mil.com/spaceport.htm

http://en.wikipedia.org/wiki/Mass_driver

Regards,

Northstar

[skykooler]

If you're going to give it all the velocity to exit the atmosphere just from the launch rail, then yes, the top will need to be at quite a high altitude. However if you simply use it as an assisted-launch technology, and the rocket still has to burn its engines at some point, then there's no need to make the mouth nearly so high up...

In stock KSP, assisted-launch is virtually useless, because there's so much drag low down in the atmosphere below 10km - you'd only save a few percent on delta-v.

In an enclosed, tunnel-style launch rail with evacuated air, then yes- you would expect no drag. However this thing is clearly open to the air- so I would think drags should be an issue. You're saying that in-game it has no drag as it's on rails though? Well in that case, at least you lose a lot less velocity to drag after release (assuming you release the rocket at a higher altitude due to a higher foundation height).

It is not open to the air - it is meant to be an enclosed tunnel, and indeed in my dev copy the ship move inside it.

If you can dial the mass driver up enough without the game bugging out, then sure, there's no need to fire engines. (n real-life, there's an issue of being able to create powerful enough electromagnets without frying all the rocket's on-board systems, as no magnetic shielding is perfect...) You could still get better ISP for rocket burns after release if a higher foundation meant a higher release height though...

I can dial up the mass driver plenty high; the current setting is meant to make g-forces no higher than 8 g's in the point of most acceleration (the middle of the turn), so that it is survivable for Kerbals.

I guess a big point I'm getting at: a 14 km mouth with a sea-level foundation is unrealistic and overly-complex. In real life, any great curvature upwards after the rocket has reached a little over Mach (340 m/s in the Kerbal-universe) would be sure to kill the crew of any manned craft due to G-forces... What's more, no modern construction technology would allow a free-standing launch rail that extends that far above the ground and can still support the weight of a sizable rocket without frankly ridiculous amounts of structural reinforcement and size (this is one of the reasons the more serious proposals in real-life suggest staring the launch tunnel deep underground, and placing the mouth in a mountain peak).

There are two reasons for being built the way it is. One is that starting underground is impossible in KSP (otherwise I would have already made it that way); and the other is that starting where it is enables it to be in the same place as the rest of the KSC. As for supporting a sizable rocket - the support spacing I used is roughly the same as the spacing on several large bridges, which carry well over a rocket's weight in cars. Is it perfectly realistic? No. But it's a fairly Kerbal solution I think.

Anyways, at least start the mass drive at the base of a mountain and curve up the slope, for realism's sake, if nothing else... In fact, that's an even better idea than just starting it on a mountain-top, and would make the base of the drive much easier to get to...

But there are no mountains to the east of KSC. I don't want to launch to the west, because then you end up in a retrograde orbit, and I also don't want to make this mod depend on Extraplanetary Launchpads if not necessary, so I'm shelving that possibility until KerbTown regains the ability to spawn ships at other locations.

[Northstar1989]

In stock KSP, assisted-launch is virtually useless, because there's so much drag low down in the atmosphere below 10km - you'd only save a few percent on delta-v.

I wouldn't be so sure about assisted-launch being useless. I've observed pretty significant (maybe 300-400) Delta-V savings just from launching at 4200m with a high TWR rocket (ideal TWR is greater than 2 at high altitude launch). And as I pointed out, the air is already less than 40% as thick at 4200m (in KSP, drag is directly proportional to air density as far as I can tell). Plus, due to the Oberth effect, you'll be wasting a lot less fuel if the mass-driver gives you the first part (maybe 1000-1200 m/s) of your velocity: it will cut a lot more off the Delta-V than the amount of velocity it gives you (think about it- it takes 4500 Delta-V to get to orbit, but orbital velocity is only 2300 m/s. A lot of Delta-V is wasted...)

If you run FAR, the drag will be even less if the rocket is designed properly (long and streamlined).

It is not open to the air - it is meant to be an enclosed tunnel, and indeed in my dev copy the ship move inside it.

Ahh- well all I had to go off is the pictures on the OP when I posted. How did you get the game to not simulate drag then?

can dial up the mass driver plenty high; the current setting is meant to make g-forces no higher than 8 g's in the point of most acceleration (the middle of the turn), so that it is survivable for Kerbals.

I see- so the mass driver itself isn't the limiting factor on acceleration. In that case, no need to burn the rocket while still on the mass-driver...

There are two reasons for being built the way it is. One is that starting underground is impossible in KSP (otherwise I would have already made it that way); and the other is that starting where it is enables it to be in the same place as the rest of the KSC. As for supporting a sizable rocket - the support spacing I used is roughly the same as the spacing on several large bridges, which carry well over a rocket's weight in cars. Is it perfectly realistic? No. But it's a fairly Kerbal solution I think.

If you build up a mountain-side, it will be MORE realistic. Remember that bridges don't go nearly as high up as your mass driver, and as such have a lot less mass in steel to support, as well as fewer oscillations due to wind to deal with...

I used to spend a lot of time working with the West Point Bridge Designer software back in high school. The mass of your bridge matters a LOT to its capacity- and a mass driver that goes 14 km up isn't going to be able to support the mass of an 100t rocket- and probably not even its own weight...

But there are no mountains to the east of KSC. I don't want to launch to the west, because then you end up in a retrograde orbit, and I also don't want to make this mod depend on Extraplanetary Launchpads if not necessary, so I'm shelving that possibility until KerbTown regains the ability to spawn ships at other locations.

KerbTown will get fixed eventually. Work with me to get a mass driver up the side of the mountains to the west of KSC, and then when Kerbtown spawnpoint functionality is restored, you'll be able to launch rockets from there without the need for Extraplanetary Launchpads.

Besides, Extraplanetary Launchpads isn't so bad. It's really just the same concept as Kerbtown's alternative spawn points- which you were already prepared to rely on- except that you somehow have to get the RocketParts to build a rocket to the launchpad, as well as the launchpad itself to the site... (or in the case of a launchpad on Kerbin's surface, edit them in using TAC Fuel Balancer after getting the launchpad and a RocketParts container there in the first place...)

And note that I said 'work with me' again- I'm eager to work with you on a mountainside mass drive, if you can show me how to do it with Kerbtown...

Regards,

Northstar

Edited February 13, 2014 by Northstar1989

[skykooler]

I wouldn't be so sure about assisted-launch being useless. I've observed pretty significant (maybe 300-400) Delta-V savings just from launching at 4200m with a high TWR rocket (ideal TWR is greater than 2 at high altitude launch). And as I pointed out, the air is already less than 40% as thick at 4200m (in KSP, drag is directly proportional to air density as far as I can tell). Plus, due to the Oberth effect, you'll be wasting a lot less fuel if the mass-driver gives you the first part (maybe 1000-1200 m/s) of your velocity: it will cut a lot more off the Delta-V than the amount of velocity it gives you (think about it- it takes 4500 Delta-V to get to orbit, but orbital velocity is only 2300 m/s. A lot of Delta-V is wasted...)

If you run FAR, the drag will be even less if the rocket is designed properly (long and streamlined).

Yes, a lot of delta-v is wasted. Much of that is due to drag. Because drag is proportional to the square of the velocity, this means that going 1000 m/s at 4 km isn't ten times the drag of going 100 m/s, but rather 100 times. Due to this you see very little improvement if you go that fast that low. This is also why you can come racing towards Kerbin at 10,000 m/s and still slow down to 200 m/s before your parachutes deploy. And I don't want to require FAR, because I want to depend on as few other mods as possible.

Ahh- well all I had to go off is the pictures on the OP when I posted. How did you get the game to not simulate drag then?

As mentioned, the craft are put on rails while they are accelerated. KSP does not calculate drag for vessels that are on rails.

If you build up a mountain-side, it will be MORE realistic. Remember that bridges don't go nearly as high up as your mass driver, and as such have a lot less mass in steel to support, as well as fewer oscillations due to wind to deal with...

I used to spend a lot of time working with the West Point Bridge Designer software back in high school. The mass of your bridge matters a LOT to its capacity- and a mass driver that goes 14 km up isn't going to be able to support the mass of an 100t rocket- and probably not even its own weight...

It depends what it's made of. If the Kerbals have nuclear rockets, they might also have large-scale structures made from carbon nanotubes. But I see your point.

KerbTown will get fixed eventually. Work with me to get a mass driver up the side of the mountains to the west of KSC, and then when Kerbtown spawnpoint functionality is restored, you'll be able to launch rockets from there without the need for Extraplanetary Launchpads.

Besides, Extraplanetary Launchpads isn't so bad. It's really just the same concept as Kerbtown's alternative spawn points- which you were already prepared to rely on- except that you somehow have to get the RocketParts to build a rocket to the launchpad, as well as the launchpad itself to the site... (or in the case of a launchpad on Kerbin's surface, edit them in using TAC Fuel Balancer after getting the launchpad and a RocketParts container there in the first place...)

And note that I said 'work with me' again- I'm eager to work with you on a mountainside mass drive, if you can show me how to do it with Kerbtown...

I know Extraplanetary Launchpads isn't so bad - I originally built it before taniwha took it over. But I want to depend on as few mods as possible, because if you depend on EPL then you also depend on Kethane and KAS, and that really starts to eat into your RAM.

Can you model things? If so, KerbTown is pretty simple to work with - you just need to import the model into Unity and export it as a .mu.