The KSP Caveman Challenge 1.3.x - 1.10.x [re-booted]

[Muetdhiver]

On 8/3/2019 at 1:11 AM, IncongruousGoat said:

Actually, we're in the clear here. As per the Jool 5 rules: " Use Normal difficulty or harder, except, any ComNet settings are allowed including turning it off completely. ". As per past judgements on the Caveman Challenge, custom difficulty saves get stuck into the highest category whose difficulty settings are the same or easier across the board as the custom save, so we would get categorized as an "Easy" caveman. But I don't really care - I was planning on using a manufactured save with a pile of starting funds and science anyways. The real challenge here is the Jool 5 itself, not setting up the save.

I like the radiator cup you have there, but I have to wonder about the mini plane idea. Laythe is a tricky beast, and building a light SSTO with caveman tech is hard. Every time I've done a Jool 5, I've had a much easier time getting lander mass down for Tylo than for Laythe.

Hmmm. I like it, especially since it means it might actually get done (I certainly don't have enough spare time on my hands to do a caveman Jool 5 on my own now that I'm out of academia and in the Real World). If we're actually going to do that, I volunteer to put together the Tylo lander.

With comnet out of the way, interesting things can begin

As for laythe : a caveman spaceplane seems most diffcult since the only engines available that fits the bill is something like Juno's paired with terrier (It works on kerbin, so it is doable, however it is also quite heavy).

But what about something like an aligator hinge claw holding a kerbal under an aeroshell while the craft is controlled by a probe core ? This could be used as a pseudo-cmd chair with a jr docking port, so you could hot-swap the engine sections dedicated to each landings and save mass down the line. Eg : jr port, AE-FF1 with the nodes on, 4 hinges to make the cage, put a second AE-FF1 shell on a payload node with a pointy aerodynamic shape with the probe core under. Chute somewhere. Engine block docked underneath. Voilà !

Well. I know what I will try next XD

And hell yeah to a community jool5 thing !

 

here is a first try :

Spoiler

The issue is that no matter what I do, kerbals die on reentry. I do no know if someone knows about a trick for this. Floating, on a ladder, behind a heat shield, no difference. dead.

The Fuslage+popup nosecone with the probecore in is the most efficient in terms of part counts and weight.

Also, Valentina always tunels through the fusselage in time warp, which is very unsettling

Edited August 5, 2019 by Muetdhiver

[IncongruousGoat]

@Muetdhiver

I dunno if that would save mass down the line, though. For all the landers (other than Laythe), all you need for a "chair" is a ladder and maybe a couple batteries to hold the Kerbal in place, which is pretty darn light. I'm especially worried about Tylo; the high dV requirement makes lander mass go up like crazy with payload mass, and we're going to have to put it up in one launch if it's going to work at all because of how wonky docking port staging is. It's getting a bit late tonight, but I'll see if I can test out a lander design tomorrow after work.

[Muetdhiver]

You are right, the mass of such system is not negligible and a terrible idea for tylo.

What one need in any case is a probe core, a ladder, a few batteries, and a solar panel.

For Laythe, adding a fuselage, aligator hinge, chute and airshell adds about 0.4 tons to it. Even more if some sort of heat shield is used. So, terrible for tylo indeed. I tried using radiators for the cage, but the drag is horrendous and the part count is too high for my taste. The reaction wheel is a luxury, so it can be removed. A 1.25 payload bay could be used as a heat shield and aerobrake. Anyone managed to fit a kerbal in there with a ladder ? I tried but unlike chairs, the kerbal does not ignore collisions while grabing the ladder.

Still. If such sheme can manage re-entry on kerbin, it might allow to completly avoid the need for a capsule, which means more DV from the dry mass saving. The issue will be with the jetpack RCS fuel for such a long trip, since each timewarp cost you fuel to put the kerbal back in.

Edited August 6, 2019 by Muetdhiver

[Mr. Peabody]

If you guys manage to pull this off I'll...

...eat my virtual hat. (Horns included!)

[IncongruousGoat]

As promised, here's a Tylo lander design:

It weighs 2.128 tons and has 17 parts, so it can be put in space without any pad assembly at all (hooray). Initial testing was highly successful; I was able to get it down to the surface and back to orbit on my first try with enough of a margin (180 m/s) left for a rendezvous. Annoyances with ladder warping aside, it seems highly promising.

[Muetdhiver]

That is a neat design. 2Tons is more than okay and using bagette is a great choice since they have the best wet over dry ratio.

I tried tons of laythe/return vehicle to little avail. It seems that there are two issues :

1) For some reason the node with the hinge attached to is considered "nacked", creating tons of drag and rocket flipping issues.

2) KSP sees airflow in the "closed" structural element, making Jeb lose the ladder and then burn. My guess is that since it closes with the hinge, in ksp logic it's still open on one end.

I have a few tricks up on my sleve to test out. For example, if the kerbal is on a ladder attached to a radiator, does the radiator cool the kerbal ?

Edited August 7, 2019 by Muetdhiver

[IncongruousGoat]

@Muetdhiver I think I've found a solution to the Laythe problem. We obviously can't use a crew capsule based lander since it would be way too heavy, so we need some kind of ladder-based lander. Additionally, we need the ladder to be enclosed in one of the parts that blocks heating so our Kerbal doesn't overheat like we've been seeing from putting the ladder in a structural fuselage. There are 3 such classes of parts in-game: service bays (too small), cargo bays (we don't have any), and fairings. A "normal" fairing isn't going to work because there's no way to get inside, but there is something we can do. If you take a 1.25m decoupler, put a fairing base under it, offset the fairing base down a ways, and then build an interstage fairing ending at the decoupler, you get a hollow tube with an opening that blocks heating for any loose parts (and Kerbals) inside it. You can put ladders inside by attaching them to the decoupler and then offsetting them into place.

The assemblage looks something like this (built upside-down, for reasons that will be explained later):

I've confirmed that the heat occlusion part of this works great, but I haven't gotten a working lander design yet. I don't own Breaking Ground, so I've been working without hinges, but that hasn't been too much of an impediment since the Kerbal can just enter the lander from the bottom instead of the top. It does require some cleverness on descent, since the craft needs to be pointed nose-first (I think? Heat occlusion is weird), but I've been solving this problem by using a shuttlecock/landing pad arrangement built out of radiators. My current lander design looks like this:

It fits within part limits, it's stable on descent, and can make it all the way to landing with an intact Kerbal. However, it's not aerodynamically stable on ascent, and I'm not super sure it's got enough dV due to drag losses from Laythe's souposphere. I hope it can be made to work with just some re-arrangement of the tanks to move the center of drag back, but the design might be fundamentally flawed.

Also, I found a new bug in KSP. If a Kerbal enters the water while holding onto a ladder and then lets go of the ladder, the physics engine forgets that the Kerbal is underwater and the Kerbal falls to their death on the ocean floor. For our purposes, this means that it's probably for the best if we shoot to land on land on Laythe.

[Muetdhiver]

Damn that's clever.

I tried to use the fairings, either as a structural tube replacement with the pop-can design, or as an envellope around it, but I did not think about the interstage nodes + offsets, so it failed (can't "open" a fairing with a hinge or piston). 

The radiators will be a nightmare on ascent though. I will try doing a pop-can design out of it to trim down drag. Also, it will need a Jr port somewhere to glue it with the other parts. It would also be nice if it can deal with the return Kerbin re-entry.

[IncongruousGoat]

14 hours ago, Muetdhiver said:

Damn that's clever.

I tried to use the fairings, either as a structural tube replacement with the pop-can design, or as an envellope around it, but I did not think about the interstage nodes + offsets, so it failed (can't "open" a fairing with a hinge or piston). 

The radiators will be a nightmare on ascent though. I will try doing a pop-can design out of it to trim down drag. Also, it will need a Jr port somewhere to glue it with the other parts. It would also be nice if it can deal with the return Kerbin re-entry.

Thanks!   If you look closely, the radiators are attached to a decoupler. They're designed to be ditched at the start of the ascent. As for the docking port, you can always just tack that (and a .625m separator) onto the bottom. It's only necessary while en route, assuming we don't want to use this specific Kerbal tube for Kerbin re-renty.

I don't know if I'll have any time to work on this tonight, but I'll definitely be trying to get this lander to work over the weekend.

Oh, and @ManEatingApe - if you think there's enough interest between the three of us, we might want to start that separate thread? Mission planning alone is going to eat up several pages of thread space.

[dvader]

Nice work everyone!

I don't know if I can contribute much due to Real Life being in the way. The next 1.5 weeks are very busy and things will surely come up after that as well. But, I really like this challenge so we'll see if can't squeeze it in somehow.

I've been testing a Laythe space plane and I kind of have something that might work. The plan would be to send Bob down in a badminton ball and then pick him up with the plane. I haven't tried the plane on Laythe yet since I actually remembered that rule about not using cheat menu to test crafts on foreign bodies for once (I did not remember it when testing the badminton ball). I can get the rocket up to ~70km and 1150 m/s on Kerbin which is pretty close so next thing would be to test it on Laythe. The plane weighs just 1.4 tons itself so more fuel could also be added.

It is a two stage plane with a Juno Jet engine which helps with precision landing and also to bring the plane out of the lowest atmosphere. There are no wheels since it lands and takes off from water. Plan would be to land Bob on a beach and then land the plane in the water next to him. That bug @IncongruousGoat mentioned sounds very scary though.

Jeb has no problem holding on during the plane flight but he did at first fall off during the rocket powered flight so I added a solar panel behind his back. I can't go full throttle but have to watch the drag to keep him from falling off.

 

Here he is at almost 70km and almost 1200 m/s. Since Laythe has thinner atmosphere I'd guess this is pretty close to actually working. The next step is testing it with one more fuel tank and then send the whole thing to Laythe for some real practice.

Oh, I just realised I've been designing with KER and stuff installed. Sorry...

EDIT: I tested 3 fuel tanks in the rocket stage and got to 73 km and about 1800 m/s which should be enough for Laythe. I dropped Jeb a couple of times but with practice it should be doable. Next is flying to Laythe for proper testing.

Edited August 9, 2019 by dvader
Added test with 3 fuel tanks

[Mr. Peabody]

14 hours ago, IncongruousGoat said:

Oh, and @ManEatingApe - if you think there's enough interest between the three of us, we might want to start that separate thread? Mission planning alone is going to eat up several pages of thread space.

If you guys could go ahead and do that it would be great. Things are starting to get just a tad off topic.

[ManEatingApe]

16 hours ago, IncongruousGoat said:

Oh, and @ManEatingApe - if you think there's enough interest between the three of us, we might want to start that separate thread? Mission planning alone is going to eat up several pages of thread space.

 

2 hours ago, The Dunatian said:

If you guys could go ahead and do that it would be great. Things are starting to get just a tad off topic.

Done! Let's move all discussion to this thread...
@The Dunatian You're more than welcome to join in

 

[Mr. Peabody]

I'll take the liberty of posting something somewhat off topic:

Who says cavemen don't play ball? (Jake Cave, wearing his "caveman" jersey for player's weekend.)

https://twitter.com/Twins/status/11650766

[Pds314]

On 8/6/2019 at 7:40 PM, IncongruousGoat said:

As promised, here's a Tylo lander design:

It weighs 2.128 tons and has 17 parts, so it can be put in space without any pad assembly at all (hooray). Initial testing was highly successful; I was able to get it down to the surface and back to orbit on my first try with enough of a margin (180 m/s) left for a rendezvous. Annoyances with ladder warping aside, it seems highly promising.

It's amazing how similar our Tylo lander performance is when they look nothing alike.

[Rakaydos]

On 9/2/2019 at 5:31 PM, Pds314 said:

It's amazing how similar our Tylo lander performance is when they look nothing alike.

Agreed.

On 6/10/2018 at 3:19 PM, Rakaydos said:

Ok, so if I havnt missed something, this module has both the wet TWR and the DV rating to go from the surface of Tylo to a low tylo orbit.

The part count requires a build-a-rocket to get it into orbit. Kerbin ground tests says the TWR=1 at about 2 tons, which accounting for atmospheric thrust of the primary engine(16 vs 20) and Tylo's lighter gravity, gives it a tylo TWR just over 1 after it ditches the landing shocks. The Ant engines have a bigger atmo/vac performance gap than the primary engine, so the TWR should be semi reasonable. Gravity losses may still kill the idea, but I dont have hyperedit to conduct simulations.

[Jacke]

I've been helping out a bit in the Community Caveman Jool-5 mission.

Alas, what I've discovered is my skills with Caveman level tech, even maxed out, is rather lacking.  With designs others find workable, I can't complete mission stages.  Even with different designs, I'd find docking in orbit rather hard, mostly because I would need RCS in my spacecraft, but I'd still find it hard to get through close rendezvous without the relative velocity to the target.

I'm not that familiar with other methods like multi-craft ground assembly, so I don't know how well I could do.

While I can put up with the tedium of Nanocrystaline Diamond, without those assembly skills, I don't think I could complete it, as I believe NCD needs going interplanetary.

So, I'm looking for more information on what level of Caveman I could complete.

I can build up to orbital craft and have done so on an NCD run under KSP 1.6.1.  I know I can do a Mun intercept orbit, but I don't if I can make a craft that could do a return mission what with my limits on assembly.  Whether I could manage Minmus or interplanetary is unknown.

[IncongruousGoat]

6 minutes ago, Jacke said:

So, I'm looking for more information on what level of Caveman I could complete.

The only one that absolutely requires orbital assembly is NCD. The lower 4 don't even come close to needing that kind of capability, and Corundum and Diamond are both completely possible to complete without leaving the Kerbin system.

 The only reason this thread is giving any other impression is that A: Corundum and Diamond are more entertaining when interplanetary missions get involved, and B: us old hands have been pushing caveman capabilities to new heights. If you look back through this thread, and especially through the older threads, you'll find that the explosion of interplanetary missions and super hardcore caveman attempts is a recent phenomenon.

[Jacke]

2 hours ago, IncongruousGoat said:

The only one that absolutely requires orbital assembly is NCD. The lower 4 don't even come close to needing that kind of capability, and Corundum and Diamond are both completely possible to complete without leaving the Kerbin system.

 The only reason this thread is giving any other impression is that A: Corundum and Diamond are more entertaining when interplanetary missions get involved, and B: us old hands have been pushing caveman capabilities to new heights. If you look back through this thread, and especially through the older threads, you'll find that the explosion of interplanetary missions and super hardcore caveman attempts is a recent phenomenon.

I'm thinking of aiming for one complete playthrough of a Caveman career.  There's other things in KSP I'd like to try out before the doom of KSP2 arrives and we all get distracted.

Well, turns out there is targeting in Caveman under 30km relative distance and I can rendezvous.  With an RCS design, I believe I could dock.  So I could likely do orbital assembly, certainly simple linear designs, especially with just two parts, or two parts with several tankers  (1st part - Nx tankers - 2nd part).

So, what is needed for NCD?  Is the Mun sufficient or would Minus and even interplanetary be required.  I don't know if I could handle long transfers with a Kerbal on a ladder, so Kerbals outside might be limited to Kerbin orbit.

Edited September 15, 2019 by Jacke

[IncongruousGoat]

2 minutes ago, Jacke said:

I'm thinking of aiming for one complete playthrough of a Caveman career.  There's other things in KSP I'd like to try out before the doom of KSP2 arrives and we all get distracted.

Well, turns out there is targeting in Caveman under 30km relative distance and I can rendezvous.  With an RCS design, I believe I could dock.  So I could likely do orbital assembly, certainly simple linear designs, especially with just two parts, or two parts with several tankers  (1st part - Nx tankers - 2nd part).

So, what is needed for NCD?  Is the Mun sufficient or would Minus and even interplanetary be required.  I don't know if I could handle long transfers with a Kerbal on a ladder, so Kerbals outside might be limited to Kerbin orbit.

Interplanetary is necessary for NCD. There isn't enough science in the Kerbin system to fill up the tech tree at 10% returns. For ladder riding, I managed NCD without doing any ladder riding beyond LKO, so clearly it's possible. Whether that's the better way to do it is a harder question to answer.

Also, for caveman purposes Minmus is a lot easier than the Mun. It's more than possible to do a single-launch manned mission to Minmus, but the same can't be said of the Mun. During my NCD run, I reverted to using probes for biome-farming the Mun because I got so sick of repetitive docking.

In general, though, I would recommend doing one of the easier cavemans. Topaz is good fun without being a grueling multi-IRL-month slog like the higher difficulty levels.

[Muetdhiver]

Speaking of NCD shenanigans, I have put up a summary of the first 19 missions of my NCD run (with all DLC's) there :

Since the start of NCD/Diamonds run is quite grindy and not very exciting, I put it there to avoid using pages and pages of this thread XD.

But here is a short summary : The run has been started a while ago*, but I got bored after farming the KSC (which while optional in Diamond, is quite necessary in NCD). I went back to it mainly because I now have tools for interplanetary caveman navigation that will allow more ambitious missions and avoid multiple orbit to encounter when going interplanetary. Also, with all the Caveman Jool5 prep, I did a lot of ladder riding and got the hang of it after a while, without the problem of losing a kerbal for good from a lack of ladder riding flight experience.

The 19 first missions where all about getting to miniaturization ASAP, and making a piggybank to fund interplanetary fun. Also, cutting costs be recovering as much hardware as possible.

Past miniaturization the plan is to get nearly all the science from outside Kerbin SOI, save for a ladder ride to Mun and Minmus surface. I plan to do quite a lot of on-pad assembly and if possible attempt a caveman grav assist (so far it's very much WiP)

 

*Before that I had two ill fated attempt at NCD.

[Rakaydos]

I keep threatening to make a caveman Eve rover, using 25 launches of a 12-antenna unit (the caveman relay antenna) to make a transmitter that can reach another 25-launch receiver sent to eve orbit, when the planets are mostly aligned.

[Muetdhiver]

Either that or unlock the 2G antenna from a greenolith

Edited September 17, 2019 by Muetdhiver

[Jacke]

1 hour ago, Muetdhiver said:

Either that or unlock the 2G antenna from a greenolith

Aren't the Green Monoliths truly random?  And then the node they unlock is random as well.  That's going to be a tough find on Mun or Minmus in a Caveman career.

[ManEatingApe]

5 hours ago, Rakaydos said:

I keep threatening to make a caveman Eve rover, using 25 launches of a 12-antenna unit (the caveman relay antenna) to make a transmitter that can reach another 25-launch receiver sent to eve orbit, when the planets are mostly aligned.

Community Caveman Mission to Eve?

[Muetdhiver]

1 hour ago, Jacke said:

Aren't the Green Monoliths truly random?  And then the node they unlock is random as well.  That's going to be a tough find on Mun or Minmus in a Caveman career.

Yes. There is one on each body, so that means quite a few rolls. One might get lucky