Kerbal Space Program

[hobbsyoyo]

The rocket would still go forward in your example. This is how pistons work.

But I should really stress again that I have no idea how the game handles these physics. And in real life the asteroid would feel a thrust but it would be very small. The bigger danger in real life is kicking up regolith and getting hit by rocks - which the game doesn't model at all.

 

[warpus]

Crazy. My mind is blown, I am going to have to read about pistons. I don't understand that at all currently.

No wonder my spaceX application was rejected

 

[hobbsyoyo]

Crazy. My mind is blown, I am going to have to read about pistons. I don't understand that at all currently.

No wonder my spaceX application was rejected

No that was because you're a dirty Canadian. But seriously that would be why.

To be fair, that's not exactly how pistons work but the net effect is the same. For the rocket the momentum is coming from shooting the gas out of the back. It doesn't matter if the gas is contained in a huge container or goes off into deep space. The fact that it is pressurized with a uniform velocity vector and leaves the rocket is the key point. In pistons, they are moved by the change in pressure caused by a small explosion behind the piston (rather than a continuous explosion inside of it as with a rocket). So there is actually a big difference but in the end they both move to the right.

But once that gas leaves the rocket, the rocket doesn't care, it already has the momentum. And the gas doesn't care either and it goes off in every direction, not just the direction it was shot. And vacuum engines are meant to have exhaust pressures as close to absolute zero as they can manage (this means essentially they have taken all of the momentum out of the gas stream) so that gas like I said, is really diffuse and the individual atoms are going all over the place, not just one direction.

 

[warpus]

Hmm but in my drawing on the previous page, it doesn't leave the rocket, as the enclosure it fires into is a part of the rocket.

I didn't actually apply to spaceX btw, I probably don't have the credentials. Plus I heard Musk expects everyone to work crazy hours, and I'm really used to working 37.5 hours a week, no overtime allowed

 

[hobbsyoyo]

It does leave the rocket though. If you assume the enclosure can stretch, the rocket will still move forward because the gas came out the back. In this way it is like a piston. Even if the enclosure is rigid, there will be a strong force trying to tear it apart.

When I was there your offer letter said you had to work a minimum of 50 hours a week. I averaged 55, with many weekends. At the end, I worked 28 days straight and I found myself in the office on a Sunday with a cold at the end of this stretch and I just broke down at my desk. I balled like a baby for a good 20 minutes, I mean I really was wailing. At the end of this fit, I cleaned out my personal belongings from my desk, went home and began searching for another job. There were a lot of other serious frustrations there on top of the high workload and extreme stress. But also a lot of incredible things. I wouldn't trade the experience I got for anything and it helped my career tremendously. I still love the company and think about going back one day but I just don't know that I ever will.

At my current job I get to do all the cool things I did at SpaceX and then some. Because it's a much smaller company, I am given a lot of freedom and responsibility to spread my wings so to speak. And the hours are pretty normal and the pay much better than SpaceX. So I'm really happy; it just sucks because in the end I do want to go to Mars and I miss the cache of being able to tell people I work at SpaceX. My new company doesn't have and doesn't want that same sex appeal. But overall I'm really happy.

 

[warpus]

Don't worry, I tell people I know a guy who works at SpaceX, but that's mostly because I forgot that you don't work there anymore. Sounds like you went through a good life learning experience and ended up at a great job that you enjoy. So, the perfect outcome?

 

[hobbsyoyo]

And thanks to re-use, I actually worked on most of the boosters that have launched since I left.

 

[warpus]

Awesome.

Maybe I should hire you to fix my space plane. Or maybe it's too bulky and I shouldn't make it SSTO, I should just build a thing that can fly on Laythe and refuel and that's it.. That would mean less fuel on board though, and I want to be able to fly from island to island. Some of those distances would be quite long I think

 

[hobbsyoyo]

Meh I'm not great a great aerodynamacist and even less so at KSP

 

[warpus]

I wonder if I should just go to my original plan and build a submarine. What happened was originally my roommate gave me a challenge to build a submarine for the Jool system that can only recharge its batteries using a windmill.

So I was like.. No. But I told him I'd build a submarine that runs on solar power. I was going to build that, but then thought "Wait a second, why not just make a submarine that can fly?". So I started that process by building a space plane, which I was then going to try to throw into the water to see if I could somehow also make it go underwater as a submarine (with the potential use of a mod or two).

But then I was like "Wait just a second, it will be a lot easier to just send a plane AND a submarine." So I'm designing the space plane of this mission basically, but that seems impossible again, so back to the submarine maybe. From what I understand I can use the ore containers as ballast.. or something like that?

Good thing I'm not Elon Musk, imagine the press conference where he explains the above and how he spent billions of dollars working it all out and making it happen

 

[cardgame]

I have a problem with my space plane. I used hyperedit to put it into orbit around Laythe to test the design. When it enters the atmosphere it just spins around like a drunk Russian. And I can never regain control. It just falls into the ocean spinning around, even if I turn the engines on, SAS, try to orient myself, etc. It handles well when I'm taking off from Kerbin, and seems to be maneuverable enough, but it's not maneuverable at all when I'm trying to land on Laythe. So something is up with my design, but I'm not sure what.

Sounds like you're coming in too shallow? I had the best luck regaining control of spinning re-entry planes on Kerbin if I just pointed the nose down and pulled out of it after building speed.

So what I'm saying is, blast straight down through the atmo into the thicker atmo that can support your plane better.

 

[hobbsyoyo]

Sounds like you're coming in too shallow? I had the best luck regaining control of spinning re-entry planes on Kerbin if I just pointed the nose down and pulled out of it after building speed.

So what I'm saying is, blast straight down through the atmo into the thicker atmo that can support your plane better.

I second this suggestion.

 

[warpus]

Let me try to understand what that means. So instead of getting into a low orbit, and burning a tiny bit retrograde, and then waiting as I slowly descend through the atmosphere, I should be burning retrograde like a madman until I am headed almost straight down?

I tried to orient myself in any direction, and my plane just wasn't having it. So I don't think pointing straight down would have worked in my last attempt. But that'd be easier if I was going in like a madman?

 

[hobbsyoyo]

Let me try to understand what that means. So instead of getting into a low orbit, and burning a tiny bit retrograde, and then waiting as I slowly descend through the atmosphere, I should be burning retrograde like a madman until I am headed almost straight down?

I tried to orient myself in any direction, and my plane just wasn't having it. So I don't think pointing straight down would have worked in my last attempt. But that'd be easier if I was going in like a madman?

Yes. The theory is that your design just isn't stable in rarefied air and so you should get down as fast as you can to give your control surfaces something to bite into.

And the reason why you're stable on launch up through high altitude may be due to a stable velocity vector and your massive momentum built up in roughly one direction. But all this is just a theory. And KSP does what it does and we have little/no insight on its internal physics

 

[warpus]

I've also been reminded on reddit that as I burn fuel the centre of gravity moves, of course.. and so it might end up in a place that causes me to spin. But I'm pretty sure I was hyperediting into orbit around laythe with a full tank of gas for testing purposes, so I'm gonna try your guys' suggestion first

 

[warpus]

I wasted the whole day building this submarine

It works though! Here are the waters about 5 km away from the space center. This thing is very maneuverable and will move at 18 m/s.

It's designed to land on the bottom of the ocean in order to refuel.

It works, you can easily land. The whole thing is balanced very well. But the problem is that if you use up too much fuel, you end up not being able to land on the ocean floor again. Meaning that you then can't refuel. The solar panels also don't do anything at the bottom of the ocean, so I'm not really sure how feasible such a scheme would be. It's probably doable, but I'm not sure if just because it works here, that it would work on Laythe.. which is what this was going to be for...

So I'm leaving this project alone for now. In order to make this work I will have to move my balast around and add some more.. and add lots of batteries. But even then I have no idea how I'd get it to Laythe.

The deepest I've gone with it was 920m

 

[hobbsyoyo]

Ok, that's the coolest thing I've seen in a hot minute. Well done. Dude I really wish we lived closer so I could pop in and watch you build and test some of these missions.


After starting a new save, I decided to try something new - unmanned sample return. The only country to have done this successfully was the USSR in the 60's. They attempted to kick out two sample return probes at the last minute just before and after Apollo 11 so that they could claim a moral victory by showing they could get samples without putting astronauts in danger. This was long before they publicly admitted their own failed N-1 manned moon program. Anywho, those two attempts both failed but they kept at it and eventually were able to pull off the feat 2 or 3 times. They returned a whopping ~10 grams of material (compared to Apollo's few hundred) and while these missions were much cheaper than Apollo, I bet the cost per gram of material returned was on par with Apollo.

Anyways, this is Test Shot Baker's return module. I do not have a smaller storage compartment for experiments so in order to make the thing balanced with a low CG, I decided to use a quad-stack adapter to mount my mono-propellant tanks to. I use mini 'puff' mono-propellant engines for ascent and the return to Kerbin. I do not really have enough fuel to do a low-altitude circularization burn so I more or less will have to go straight in to Kerbin. So at the end of the journey the ascent stage will drop and a heat shield will have to do the dirty work of slowing the samples down for a landing on chutes.

Spoiler :

Actually, that's a lot more DV than I remember having so maybe the heat shield is redundant after all. Anyways, here is the descent/landing stage and the whole stack encapsulated in a fairing:

Spoiler :

Below is the nuclear powered trans-Minmus injection stage. You can see that buried in the interstage structure I have attached relay antennas. This cruise stage is fully autonomous and will serve as the relay for the lander which only has short range UHF whip antennas for TT&C (telemetry, tracking and control).

Spoiler :

My first attempt at a launcher was wastefully overbuilt and very expensive. I had dual orange tanks + mainsail in that first configuration. I decided on a whim to ax one of the orange tanks and replace the mainsail with a skipper and low and behold, I retained the same amount of DV but halved the cost. I added a pair of powerful SRB's for good measure too (they are from a mod but I could have used stock ones without issue) though I'm pretty sure they were unnecessary. They just give me margin for my particularly sloppy method of intersecting Minmus. Instead of planning an efficient trajectory, I just go out to Minmus orbit, then burn retrograde until my whole orbit flips over and then I'm flying totally retrograde which gives you a 100% chance of interception if your orbit is mostly circular and somewhat in-plane with the moon. Here is the stack in flight:

Spoiler :

On my first launch attempt everything was going fine until the long coast at which point I forgot to open the solar panels and my batteries died. I'll try again tomorrow and report back. And here's the Luna sample return ship, for reference:

 

[hobbsyoyo]

With a TWR of 10:1, I'm going to have a very difficult time landing that thing.

Edit: I just remembered there was a successful comet sample-return mission and a semi-successful solar wind sample-return mission (it had an accelerometer turned upside down and crashed into the desert but some of the sample wafers survived) in addition to the Luna missions. Both of these other missions were done by the US.

Also, the Luna missions returned a little less than 200 grams in 3 successful attempts, not 10. In any case that's not very much compared to Apollo thought it's infinitely better than nothing.

Edit Edit: Also the Japanese returned a few particles from an asteroid.

 

[warpus]

Ok, that's the coolest thing I've seen in a hot minute. Well done. Dude I really wish we lived closer so I could pop in and watch you build and test some of these missions.

If you wife let you, I would hold a KSP camp at my place, and it's just be us playing KSP and eating pizza.

After starting a new save, I decided to try something new - unmanned sample return. The only country to have done this successfully was the USSR in the 60's.

That's cool.. Maybe that's what I need to do, start re-creating actual missions. I remember a couple years ago thinking that recreating a manned Apollo mission to the moon would be impossible because I'd have to dock things a bunch of times. But now it seems super simple. I suppose once the expansion is out it will have a bunch of cool historical missions set up and ready to go. It would be amazing if they added in future missions too, like a BFR landing on duna, but somehow I doubt they will do that.

I am not sure about that submarine. Maybe I should try creating a smaller one. The problem is that each fuel tank you put on the thing has to have a LOT of ore/balast to balance it out, cause the fuel tanks it seems like fill up with air as the fuel is used up. You'd think it would just be vacuum, but it all seems to be very floaty, so that can't be it. Or can it? Whatever the game is doing is making those tanks very buoyant as you use up fuel. So I could make a small submarine, but it could only go a tiny distance, and it would need a lot of balast anyway. Hmm.. I think I'm going to just install the mod that allows you to have different launch sites. From what I saw on youtube, the most popular mod to do that gives you an option to "launch" a "plane" right from by the ocean, on a little driveway right into the water. Perfect submarine spot, would really speed up development. Up until now I had been driving every single design off the runway and into the water. It can take a while

 

[hobbsyoyo]

You have to backfill propellant tanks with an inert gas as they drain to avoid them imploding due to the major pressure drop caused by a vacuum bubble forming in place of the drained propellant. That's how real rockets work anyways, it's actually really neat that they model that in the game.

Can your submarine fly through the air?


I'll edit in some pictures of the sample return mission later. It was a mixed bag at best.