Jump to content

What did you do in KSP1 today?


Xeldrak

Recommended Posts

7 hours ago, Laie said:

Your vessel is beautiful... (repeating the picture just because of that)

The recipe for low-power-much-wing planes is to *not* do a zoom climb, but keep flying: that is, air under your wings keeps you up and you can gently gain altitude as you accelerate. Here's me doing it with a much simpler, one-NTR flying wing. Your's has twice as much thrust and, for all the additions, I doubt it has twice as much drag. So I think it should be capable of getting there on a similar profile.

Thank you! Yeah, I'll try to do a similar profile. I guess I'm more used to doing a zoom climb given also I've worked a decent amount with more rocket based SSTOs. Thanks for the advice, will definitely try.

Link to comment
Share on other sites

I spent two hours wondering why my new space plane kept crashing into the ground directly after take off.......   Then it hit me after my 7th crash....   I changed w and s around and this was my first time with the new config flying a plane....   rofl   I was tilting towards the ground instead of up every time.   

Link to comment
Share on other sites

10 hours ago, eddiew said:

 

  Reveal hidden contents

For the curious, the wormholey stuff is provided by ESLD Jump Beacons Revived. The planetary system around Grannus is provided by the Grannus Expansion Pack.

The Anemone is a massive, welded, customised part with a beacon module plus all the goodies from ESLD that make it reliable. It is capable of synthesising Karborundum, because I have no other use for this material in my game and didn't want to mine it. My thanks go to @JadeOfMaar for helping work out how to write the configs to get the entire kaboodle running as a single entity without overheating or kersploding.

Synthesis does however require a supply of 5000ec/s, which acted as a tech gateway on getting the one around Iota up and running. The Anemone at Grannus (specifically Bellsama, moon of Nodens) can generate its own power, but for RP reasons I am inclined to require it to have an attached 'command' module with some engineers and scientists to operate. I feel this is a nice way to expand the (already large) GPP system, allowing the Grannusian planets to be accessed as if they are local to Ciro, but without crowding the Ciro system.

Both Anemones were placed with Hyper Edit. That the Iotian one has been apparently knocked out of it's lagrange point by docking is... perhaps predictable. It is of course a fake phenomena, in that it was placed in stationary orbit above the far side of Iota, and there are no gravitational forces helping to keep it stable. At this altitude, its orbital velocity is only 47m/s to even a tiny tiny touch will knock it a few seconds/minutes out of sync. I'm fine with this though, since now I don't have to build a communication satellite for it :P 

/spoilers

 

Yep, that's KSP for you :)  The good news is most of this can be resolved by planning; e.g. ensure your deorbit burn doesn't bring you down in the dark and checking who you've loaded before you launch :wink:  Eventually it becomes instinctive and stuff works more often. For EVA issues, remember to be very gentle and tap-tap the thrusters only, and space will orient your kerbal to the camera.

As for keyboards; anything by Microsoft or Logitech would be my call. That said, the behaviour your describing might be your USB ports. Check under device manager, make sure the box for "Allow the PC to turn off this device to save power" is NOT ticked.


i have a logitech, bought it to replace another logitech that had to many years on it, didnt have this problem before. :(

Link to comment
Share on other sites

Made some progress with the whole stock turboshaft/turboprop engine RUD chapter.

Every time one of these engines blew up the mission report window said that "part x collided with MK1 Liquid Fuel Fuselage". I knew this info was somehow wrong. The MK1 fuel tank is the root part of the engine until the shaft decouples. Later I realized KSP isn't able to communicate that a part of the turbine shaft collided with one of the bearing elements: the "Place-Anywhere 7 Linear RCS Port" -  RCS ball for short. These were all connected to the root part, the MK1 fuel tank.

But I've made some major breakthroughs, at the expense of yet more parts. Engineering in KSP is a lot easier than real life but comes with frequent surprises.

Why more parts? Take a look at the picture. You see the stats of a single Wheesley engine, its thrust is around 110kN. There are four of them. Now take a look at the context menu next to it: around 200kN. This means that a properly designed helicopter can be more fuel efficient than a VTOL.

Z8eMXwj.png

But there was a lot going on. Often my engines exploded during this week. Apparently those RCS balls need even more companions.

Okay, I need to take a step back. We've seen many bearings made from stock parts over the last 6 years. The discovery of the RCS ball changed everything. Most if not all other designs are either obsolete or just fit for low power applications, especially with 1.4.1.

What still works are parts with

  • A: a sphere collider -  there are no parts with lower friction values than these,
  • B: a high impact tolerance, while there are some other parts with spherical or tube-like colliders, they don't have a chance if they have an impact rating of just 6m/s,
  • C: solid joints. Most parts are like spaghetti when centripetal force is acting on them. There are exceptions ... the RCS ball is thankfully one of them.

...

So the RCS ball has become my favorite, throwing all my previous designs using landing gear for bearing elements in the obsolete category. But because of the law of the conservation of misery, we're not done yet. So now I'll continue from where I stopped earlier.

Unity physics are hilariously ill-equipped when it comes to Newtonian physics. I really love how Squad has made it all a little bit more sound.

When you use a sledgehammer on any kind of object or surface, you expect a certain result. Maybe the thing collapses into dust. Maybe the impact leaves a dent. This doesn't happen in KSP.
Instead we have ghost-like behaviour like the colliders of parts phasing through each other if a force is strong enough.
And the only way to remedy this is to place multiple colliders at the exact same spot, to increase the collider integrity.

Right now the bearing at the front of this engine has 12 RCS balls on the shaft, occupying the exact same spot. Around it are 12 balls to contain lateral movement and 2x 12 balls to prevent axial movement. This is just the front of the engine. The rear has a somewhat similar setup. Now calculate the amount of parts needed for one engine and consider that one of my airplanes has three of them.

I'm having fun regardless.

 

Edited by Azimech
Link to comment
Share on other sites

Hey @Laie - Your technique worked beautifully! Thank you for the advice! Image below is taken from a legitimate orbit.

J0THDRB.png

This is just a neat shot I thought.

P7bNpkL.png

Also tried... bigger. HO-229-ish shape. Doesn't really fly, more plummets in slow motion....

BlNxLCw.png

More tweaking!

Link to comment
Share on other sites

3 minutes ago, CatastrophicFailure said:

Quick random question: how do you get the option in the Wheesley menu to Show Colliders?

Unfortunately there's no menu option for that, I used a mod:

Quick one, @qzgy!

Edited by Azimech
Link to comment
Share on other sites

3 minutes ago, qzgy said:

https://forum.kerbalspaceprogram.com/index.php?/topic/149706-13-collide-o-scope-v116/

The most favoritest mod. So helpful for that kind of stuff.

 

2 minutes ago, Azimech said:

Quick one, @qzgy!

Ok, follow on... cuz of course you guys know everything. :wink:

is it possible to simply disable the collider of a part through its .cfg?

Link to comment
Share on other sites

1 minute ago, CatastrophicFailure said:

 

Ok, follow on... cuz of course you guys know everything. :wink:

is it possible to simply disable the collider of a part through its .cfg?

Hmmm.... I don't know.... Thats a good question...

Deleting the module "Collider Helper" in the craft file doesn't change anything at least...

Link to comment
Share on other sites

5 minutes ago, CatastrophicFailure said:

 

Ok, follow on... cuz of course you guys know everything. :wink:

is it possible to simply disable the collider of a part through its .cfg?

No, a collider is "baked in" together with other meshes, textures, transforms etc while building a part in Unity.

Now I'm curious, what would you want to do with that? Because it's not a bad idea.

Edited by Azimech
Link to comment
Share on other sites

Today I sent my first satellite to space it was a elliptical  orbit that was about 23 degrees off kerbin equatorial line. It was pretty easy and I had tons of fuel to spare  but my rocke kept on flipping so after days of researching I came up with the idea that I should use steerable fins and it worked. Also today I did a Mun landing bud didn't have enough delta v to return by about 400m/s this has to do with the fact that asparagus staging on console for me at least doesn't seem to be working so I put a question on the console technical support  hoping to redoe my misson with asparagus staging properly

Link to comment
Share on other sites

23 hours ago, Ace in Space said:

The rockets are still less haphazard than some of the things I've launched, though.

Case in point:

D97328B180B53874754A9AA328E71AB614A6F54D

While my Dres scouter probe has not yet revealed any promising asteroids, I'm going forward for the time being on the assumption that an asteroid-anchored station is a viable plan for setting up an infrastructure there. Which means designing a new hub for the station, one which can latch onto and mine directly from the asteroid. What you see here is that hub component, still attached to its launcher rockets. See, there's a radiator tucked away on the bottom, so I couldn't attach anything down there. And the awkward size and shape of the craft meant I wasn't getting away with shoving it into any kind of cargo bay either.

So I just attached the rockets, which make use of a few new MH parts, directly to the hub arms. They will decouple from them after the hub is in place, leaving the docking ports open for standard station modules to attach. This particular station will need a habitation module and a converter module, unlike a normal station, because it won't have an associated ground base, and this hub design doesn't include living space for kerbals, since it has to accommodate a mining rig instead. I believe I already have a habitation module design tucked away unless I deleted it for some reason, and the fuel and science modules that already exist will work just fine here. I love modular design! :) 

B551BF3FF59622B86150AF429D5A50DCD0BA88AD

But this one is not on its way to Dres. This is just a test craft, and it's headed to one of the asteroids recently sighted near Kerbin, selected by mission control after several rounds of throwing darts at a list. The idea is to intercept and grab it while in solar orbit, after which we don't much care what happens to the test craft and its asteroid. Hopefully, this will satisfy the terms of the World First contract demanding I dock something in orbit of the sun.

It took several carefully calculated burns, the last few of which had to be performed with such precision that I limited the engines to 0.5 strength, I managed to whittle down the minimum distance to 0.1km, which I'm quite proud of.

70599F87A5CD6FF9700257B688ECD81F97234516

Then it just has to latch on as a proof of concept for an asteroid station.

In the mean time I also did a science landing on Ike and transferred the data over to Ruby's science box for a contract, and intend to keep doing Duna landings for a while before sending Ruby back home.

Link to comment
Share on other sites

2 hours ago, Ace in Space said:

I love modular design!

Me too.  My architecture is based on 2.5m.  Using radially-attached boosters, as you do with asparagus, means that you wind up with a lot of power being exerted through the central column.  With heavy loads, this causes a catastrophic collapse.  Plus, you need an attachment point for nukes.  This suggested to me that I needed a top deck with 6-way radially-attached 2.5m tanks that effectively become a payload deck to which you can attach anything; and through which the thrust can be distributed.  If the load is too heavy to fully pack the deck at lift-off, you can pack more on it (consolidation) once in LKO.  You can also then gang several.  That became my Aquila transporter.

When 1.3 wrecked Aquila's fuel sequence, I got expert help to sort it out but not before I went nuts and created GizmoGizmo breathes Brute Force into a 2-stage package.  Once Aquila was resurrected for me by expert help, I looked at Gizmo and saw that I had had a psychotic episode.  Because Gizmo makes everything way too easy.  (Gizmo is basically 7 rockets joined together at the hip and is named after the VAB device that made it possible.)

Is that bad?

:)

Edited by Hotel26
Link to comment
Share on other sites

Playing with kOS again.  Plan is to break it in to modules but at the moment I have a programme that works though a series of modes:

Preflight - set up initial parameters

Launch - Full power climb straight up until it hits 50m/s

Gravity Turn - Calculates a turn angle based on TWR and holds that angle until 100m/s, plan to also include a MaxQ term in there to reduce the angle on anything I want to launch a bit slower

Q limited climbout - Hold prograde and climb until Apoapsis hits the preset target, but limit Q to another predefined value.  Currently doing this by calculating the acceleration required to to hit the required velocity from the current velocity calculate from measured Q, but it's not that accurate, may have to go back to using a PID loop.

Final Apoapsis - Once above 70km burn tangentially to surface to reset the Apoapsis that will have drifted down a bit due to drag

Circularize - Calculates the vector for a circular orbit at the current altitude and the acceleration required to get there from its current velocity vector.  From an initial target of 80km this is getting about a 79.8x79.6 orbit with my current test craft.

Deorbit - Warp to a given Longitude short of the KSC and burn 200m/s retrograde, this is the usuall burn for my SSTOs from an 80km orbit to land without overheating anything, it's deliberately aiming to land a bit long.

Descent - Estimates drag and uses the airbrakes to pull the impact point in line with the KSC.  Drops the landing gear and deploys parachute at 2000m altitude or 1km short of target

Brake - Suicide burn landing, works with or without the parachute deployment.

 

I'm pretty sure I can refine the descent drag calculations a fair bit, currently it works out how far it will go on a third of the current horizontal drag and opens and shuts the airbrakes to pull it back towards the KSC.  When I've seen videos of people doing precision Falcon style landings they've been using the Trajectories mod to calculate impact point in atmosphere, but I want to work out how to do something reasonable myself before trying that.  I also plan to look at steering them a bit further south during the descent, which I should be able to do with the airbrakes.

3 test ships all down with a few hundred meters of each other and no more than a couple of km north of the KSC.  

MzTpS4r.png?1

Edited by RizzoTheRat
Link to comment
Share on other sites

Today I finally launched a rescue mission to save my two kerbals that got stranded in space during my first few days with the game a few weeks back. They were in an orbit about 190k km above the planet with zero fuel.  I’d previously attempted to get out and push but realised this would take absolutely forever. I then tried to send up an un manned probe to do the pushing but I suck at rendezvous and so, even if I’d managed to get my pusher probe into the right position, I failed at the first step!

So I’ll admit i installed MechJeb to help me with the rendezvous. I also watched a Scott Manley episode where he used the claw to grab some space debris. I redisantes my rescue probe to have a claw at the front and used MechJeb to get me to within 40 metres of my stranded kerbals. Then using RCS I slowly pushed my probe towards the craft and after many long minutes they were joined!  I was so happy I had to WhatsApp my work colleagues to tell them :D    .  

Unfortunately I had only given my rescue probe a small fuel tank and so after burning till all my fuel ran out then detaching the claw, my insertion of my lost kerbals was at a very shallow angle which meant they exploded on reentry. 

On my second attempt at reentry I managed to keep my craft at an angle that meant they didn’t explode, but it didn’t slow down quick enough to be able to launch the parashoots and so crashed into Kerban. I tried this a few more times before remembering the airbreak trick some space planes use in videos, spinning around to slow down quicker. 

I was close to sending up a second rescue probe with more fuel but thought I’ll give it one last go but this time I’ll keep the rescue probe attached during reentry. Amazingly this did the trick slowing me enough so that my parashoots didn’t explode. Got down to about 500m/s, detached the rescue probe and enabled my parashoots. After 1 year and 161 days my two lost kerbals we’re finally returned!

Link to comment
Share on other sites

Shameless self-promotion incoming: :)

yzmFYQP.png

I tried to design my first mission with the "Making history" tools. It's a simple STS run to the ISS and deployment of a small satellite. If you're interested, the mission is available here:

TwNguJ2.png

I'll be grateful for any feedback, and hope that you'll have fun.

Thanks!

 

Michal.don

Link to comment
Share on other sites

I've spent today mostly trying to get my gamepad to work with 1.4.1. As you probably all have seen in the tech support forum, 1.4.1 has this nasty issue on Linux, that it doesn't allow joystick input at all, and for gamepads the analogue sticks can't be assigned. Anyhow, I was about to start hacking on a mod that just reads the system's joystick input and sets it as control input in KSP, when I remembered that such a mod actually already exists: Advanced Fly-By-Wire.

So, instead of writing my own mod, I've started to mess with Advanced Fly-By-Wire. Getting it to compile for KSP 1.4.1 and Toolbar Continued didn't require any code changes at all. Everything seemed fine at first, until I realized that just recompiling the mod caused issues with KSP's own input system, which meanwhile also uses SDL. After a lot of guessing what could be the cause of those issues I finally figured out what happened: Advanced Fly-By-Wire was consuming input from SDL, stealing it away before KSP had a chance to use it. A google search and a bit of experimenting later I came up with this nifty patch    , and now the whole thing seems to work perfectly fine.

Long story short: I've got my gamepad working with KSP 1.4.1 and Linux, and it only took me 6 hours or so to get it to work...

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...