The Kraken Program

[Frozen_Heart]

Airbrakes don't seem to be working. They cause plenty of part flexing but are just too strong. They rip things to pieces.

[EladDv]

9 hours ago, kmMango said:

Last I checked, as of 1.0.2  they worked quite well.

I repositioned the legs on the 2.5 m drive slightly. No specifics, just trial and error. I got it to be pretty reliable, though it did have a tendency to want to flip around. Just bolt on a few reaction wheels.

yeah seems consistent with my experimentation it produces a force off center so you get torque and you need to counter it with SAS

[Frozen_Heart]

Have you guys got any links to pusher designs? All mine no longer work.

Edited December 9, 2015 by Frozen_Heart

[kmMango]

The JD drive in the old Tyrene thread works. It's the only one I still have the name of.

[Majorjim!]

Just now, kmMango said:

The JD drive in the old Tyrene thread works. It's the only one I still have the name of.

Does it really? Nobody has posted a link to a working version to test. I remain incredulous until I see it for myself.

[inigma]

I was the second person to develop a k-drive after Frozen and I think the first to understand the inner principal and was thus the first to simplify it and miniaturize it to prove what was happening and make something practical:

http://m.imgur.com/a/6qdBy

Count me in!

[Majorjim!]

Just now, inigma said:

I was the second person to develop a k-drive after Frozen and I think the first to understand the inner principal and was thus the first to simplify it and miniaturize it to prove what was happening and make something practical:

http://m.imgur.com/a/6qdBy

Count me in!

I remember that badboy! So anyone have a working KD to show, and I mean show.

[War Eagle 1]

Wait…Kraken Drives are back?

[Tw1]

Can confirm Ladderdrives are still feasible.

It's not as capable as it once was- that first one was a re-useable Eve SSTO  before they fixed physicsless parts and added heat.

It is still possible to get to space on kerbal power alone, despite various physics changes.

Riskier, seat-and-chuteless designs will be slightly more capable, if harder to use in space. 

  A great way to tour moons on the cheap, or exercise and terrify Kerbals at the same time. 

Edited December 9, 2015 by Tw1

[Iago]

I have been experimenting with k-drives lately. I have reached 60g, and then it tore itself apart.

[322997am]

I have a plane that when the airbrakes open it rushes forward at 1.5g and crashes. Unfortunately I scrapped the design cause I was making a passenger jet and not a krakenplane.

[EladDv]

4 hours ago, Iago said:

I have been experimenting with k-drives lately. I have reached 60g, and then it tore itself apart.

if you can provide craft files designs techniques for new K drives it would be very helpful

[Iago]

It's just the classic landing gear pusher design. Besides, it just disintegrates a few milliseconds after activation, spewing debris across the system.

[Majorjim!]

Just now, Iago said:

It's just the classic landing gear pusher design. Besides, it just disintegrates a few milliseconds after activation, spewing debris across the system.

But it provides thrust right? This is why I am suspicious, nobody is providing craft files. And as far as I know Squad closed that physics loophole.

 

 If yours does indeed provide thrust try it in orbit.

[EladDv]

1 hour ago, Iago said:

It's just the classic landing gear pusher design. Besides, it just disintegrates a few milliseconds after activation, spewing debris across the system.

yeah i agree with majorjim here please provide a craft file since i currently cant reproduce major accelerations (0.2-0.1 g's was all i could get) with regular pusher plate designs i would like to improve and learn from working drives since i dont have the time for a lot of experimentation like i had back in the days

[Guest]

On December 8, 2015 at 11:39:31 PM, kmMango said:

Last I checked, as of 1.0.2  they worked quite well.

I repositioned the legs on the 2.5 m drive slightly. No specifics, just trial and error. I got it to be pretty reliable, though it did have a tendency to want to flip around. Just bolt on a few reaction wheels.

Do you have the craft file?

[Guest]

Because of the new optimizations making ships not collide with themselves, would it be possible to have multiple ships?

[Iago]

Here is my useless K-drive. Only works pro/retrograde in space. And then disintegrates.

[quasarrgames]

Count me in!

I made a k drive back in 1.04, and i think it still works now. This image should show you how to make it

Pros: steady 1g of acceleration, and it's really easy to make (the legs only have to be roughly offset into the bottom for it to work)

cons: it can only push a maximum of 10 tons

[Stevie The Crusher]

The first Kraken Drive development was super fun, but I mostly just tweaked what other people had going to get results. This time I went a little farther. I have revisited old styles, these are what I managed to cobble together: (Was building in 1.0.4, realized I needed to update. Re-did testing on these craft, still work the same.)

 

Push Plate

Functional and very stable in atmosphere, but no effect in orbit. If it stops working, reload client.

Push Plate Atmos

Push Plate Craft

 

Scanner Research

The narrow band scanner can indeed collide with things and spin them, though not for long with this quick setup. Needs work, and a way to convert that rotation into acceleration. Some potential for driving props on helicopters or planes.

Scanner1

Scanner2

Scanner Craft

 

Push Spinner

Looking though the old K-Drive thread, I found a drive by sgt_flyer. Rebuilt a version of it, confirmed it still generates a small acceleration. Miniaturization efforts resulted in mixed results, typically with large G force happening when the rotor jams. (this version doesn't spin constantly, you need to add a slight depth offset to each drive gear until they all push on different plates. I added that in a later version)

Spinner

Spinner Craft

 

Push Cube

This is the one I created and have been working on. It takes the concept from sgt_flyer's rotor blades, and makes them work from any orientation.

Kraken Cube

Kraken Cube Sub-Assembly

 

The resulting structure is highly sensitive to anything clipping inside it. Since it works from any direction, I don't need bearings, but can get away with a small cage. This is my current stable version:

 

Push Cube Atmos1

Push Cube Atmos2

Push Cube Orbit1

 

I was trying to hit mun, but I missed. Steering this thing is difficult.

 

Push Cube Orbit2

 

The properties I've found so far are interesting. It is perfectly reliable, in that it will always turn on when you lower the gear. It will never break, as the force the gear push into the cube with has been tuned so the G-force spikes are within the cage's tolerances. However, the force produced changes constantly, and aims in a wide arc. You need an absurd amount of reaction control to keep this thing pointing even vaguely in the right direction, and the more force you try to generate by squeezing the cube farther makes the problem worse. I've found that you can roll the vehicle while your SAS steers to allow something approaching control.

 

Push Cube Atmos Craft

Push Cube Orbit Craft

 

Here is the next revision 'cage' I have been working on. Higher force output, but it very often tears itself free of the craft and goes on adventures without me.

 

Push Cube2 Atmos

Push Cube2 Blooper1

Push Cube2 Blooper2

Push Cube2 Atmos Craft

Push Cube2 Orbit Craft

 

 

I tried with only one gear at one point, but without the other random forces in play to average out the force direction to forward, you get a lot of lateral movement going on. Miniaturization is going to be difficult, this thing isn't exactly light on part count. The smaller you make the cage, the more focused the forces are towards where you're pointing, but also more likely that the cube escapes, and goes flying off somewhere. If you try to exceed the force limits of the cage, first the force spike will destroy every connection on your craft that is weaker than the cage, and then the cage will burst. Basically anything using this above 10G's is probably going to need special construction, rather than relying on part connection nodes. Even with struts, i'm pushing those regular connections to their limits already.

[Windspren]

I updated the Kraken Tamer list.

[EladDv]

20 hours ago, Iago said:

Here is my useless K-drive. Only works pro/retrograde in space. And then disintegrates.

I don't know what you are talking about I got up to 0.5 m/s^2 of acceleration without breaking apart, what version are you on?

[Iago]

I have 1.0.5. If it doesn't "work" just try again. And remember only point prograde.

[Stevie The Crusher]

Did some more testing, and with enough SAS to generally maintain control, my cube cage design seems to average out at about 4G. It can do 10G bursts, but I've yet to figure out how to make that consistent. Increasing the gear force on the cubes increases the needed SAS, which tends to just balance out to the same G-force afterwards. Loaded up an old craft file for the JD drive, still works. Did some tuning to get it working in orbit again, outputs 14.5G with my test rig. Adding any payload creates rather abrupt diminishing returns on the G-force and reliability, so only useful for personnel and data transport I think.

JD Drive vs Mun

JD Drive Craft

I also tried to remember the old requirements for K-drives, and I think I got it back. You need three things: a suspension part (force), a clipping part (glitch activation), and a pushed part (counter-force). The old designs have a leg ghost through a panel and push against a panel/dock/adapter/SAS. If you apply too much force it can break the leg. If you overly reinforce the pushed part, it can break the leg. If you attach things to the pushed part, it can break the leg. The reason gear have trouble working with this layout is because of the lack of leeway in the suspension column to fit the clipping part. My cube apparently got around this by changing the clipping part from clipping into the suspension, to clipping into the pushed part. So technically my cage acts as the pusher plate, and I have confirmed attaching things to the bottom of my drive is an even worse idea than doing it with the leg drives. Small gear are durable, they are the last part on the ship to break when I did it. Boom. I tried to use this concept to reduce the part count of the clipping part to just a single plate, but you need more than that if you want it to clip, and also stay contained. Might be able to use a single plate if you stand it up and push against the edge, more testing required.

[EladDv]

13 hours ago, Stevie The Crusher said:

Loaded up an old craft file for the JD drive, still works. Did some tuning to get it working in orbit again, outputs 14.5G with my test rig. Adding any payload creates rather abrupt diminishing returns on the G-force and reliability, so only useful for personnel and data transport I think.

i can confirm these reports of 14.5 Gs of acceleration and i can say i have gotten about 20 G's myself on a modified design that includes some more flexible joints, so far it seems like having more legs pushing only reduces the total power of the device but it might be connected to distance traveled by the oscillator which i can increase with multiple legs on different levels and batteries (which are surprisingly flexible as you all know) 

this is the smallest K drive i could muster to produce with reasonable thrust, and no, thrust is not scaled with the ship's' mass

Edited December 15, 2015 by EladDv