Cavemen gotta eat: caveman nanodiamond + kerbalism [stock science]

[king of nowhere]

My previous caveman run relied heavily on being able to stay in space indefinitely. Doing rendez-vous without maneuver nodes has that effect.

Kerbalism puts a sharp time limit on how long I can stay in orbit. It also requires bringing more parts and more mass for life support, when both are heavily limited by the caveman challenge. A couple weeks ago I started toying with concepts for a kerbalism nanodiamond challenge; now it's time to start a new report.

To think I announced my retirement from this game only a couple months ago...

Part 0: Thou shalt not get basic science!

At first, I try the challenge with full kerbalism rules. But it turns out, kerbalism deeply nerfs early game science; with rewards capped at 10%, it is impossible to leave Kerbin.

And that's before even counting all the issues with life support and electricity

Spoiler

Using kerbalism has noticeable effects only once one leaves Kerbin orbit. The crew pod is authomatically equipped with enough life support for all orbital missions, after all. I was only expecting real issues once Mun missions came into play.

Yet my first attempt at this challenge was twarthed not by life support, part failure, or radiations, but by something more insidious: science redesign. Indeed, besides adding all manner of rules to how science works, the storage of data and samples, the transmission speed of antennas (all things which could be tackled with some patience), the mod curtails most early game science. Kerbalism does

- remove all the ksc mini-biomes that gave a much-needed science boost to nanodiamond caveman challenges.

- made goo a global experiment; you can only run it once per planet, instead of once per biome. Goodbye to another big chunk of early science.

- flying low crew report has become global; you can only collect it once, instead of once per biome over all Kerbin.

- flying low EVA report can't be collected at all.

- space low EVA report is also global, no longer biome dependent.

As partial compensation:

- flying low baromether is biome-dependent

- you get a new instrument, the geiger counter (but it only gives 0.1 science per biome).

Additionally, there's a bunch of unique experiments available on probe cores and long term habitation modules, but of course they don't matter if I can't get there in the first place.

But I didn't want to just give up on the full experience, so I still tried to get away from Kerbin with the full kerbalism rules.

Getting the easy experiments (goo and reports from early flights around the ksc) doesn't even net enough science to unlock the first node. I could get there by biome-mining, but that's time consuming, so I took a test contract to get early access to decoupler, and built a suborbital jumper with nothing but flea boosters

The flea-based suborbital jumper

And an improved version, that can get to high space. Barely

With that, I brought home enough science to unlock engineering 101, which includes the thermometer and geiger counter. The thermometer is a welcome, if limited, addition to the science options. The geiger counter... it gives exactly 0.1 science points for every Kerbin biome. Plus 0.8 total for crossing both radiation belts.

Still, with that and a few biomes I got basic rocketry, which gives access to the first orbiter.

The main reason to unlock the first orbiter is that the contract pays 1 science

Then I went squeezing every possible bit of early science I could.

This contraption managed to bring Val in low orbit, to collect an eva report. Worth 1 full science point!

This improved version could get to space high, for 1.5 science points!

Alas, that's not enough. I collected all science I can get from Kerbin orbit, and while I still have to mine Kerbin biomes, each one only gives 0.8 science (0.2 for thermometer flying low, 0.1 for crew report, 0.1 for radiation scan, 0.1 for thermometer, 0.3 for EVA report. Remember, neither flying low report, flying low EVA nor goo canisters are available in individual biomes). With stock science, it's 2.6 points.

I run all the math, and determined that if I stripped all kerbin biomes, I would get to 13 science. Not enough to unlock survivability.

I could, potentially, try for a Mun flyby with the tech I have.

This thing, on a lucky day, can get to orbit with 1100 m/s left

1100 m/s from LKO would be plenty for a free-return trajectory, but of course I can't use maneuver nodes. Nah, that's not even on top of my problems right now.

First problem is life support. Kerbals breath, and they produce CO₂. That has to be scrubbed, but the scrubber consumes electricity. The battery lasts approximately 1 hour, then the CO₂ is no longer removed. It accumulates for another couple hours, until it reaches 2%. Then the kerbal starts to suffocate. And I have no way to power up the scrubber. I am limited to 140 electricity given by the pod. Even if I could strap another pod, and somehow achieve enough deltaV despite the extra mass, I'd still be far short of the time needed for a Mun flyby.

Actually, the game does give me an option to generate electricity: the monopropellant fuel cell, which is included as optional in the pod (it adds 100 kg, though, not negligible in caveman). The monopropellant on board would suffice to power up the scrubber for 20-25 hours. But the fuel cell also uses oxygen, and the oxygen on board can only power up the scrubber for a few minutes. Additional oxygen tanks are unlocked with survivability, which I just established I can't unlock.

But ok, I can fix this. Instead of teching basic rocketry, I can get most Kerbin biomes with suborbital jumpers. Maybe. I'm not sure I can reach the poles or the badlands with the available deltaV of the Space 2. Regardless, let us assume that I can somehow get those 2 science points in some way. Then I have to run a Mun flyby with 1100 m/s, lasting at most 3 Kerbal days, and without being able to save-scum. Extremely unlikely.

But to settle things once and for all, I tried cheating the Mun flyby, and see how much science I could potentially get. It turns out, goo container gives less science than it does in stock. Baromether doesn't work in space. Geiger counter is pitiful. I got less than a dozen science points. Even if I somehow managed it, it wouldn't help me at all. I need to unlock basic science to proceed, it gives both bigger electricity supplies and the probe core; but I can't get nowhere near close to it. In those conditions, the challenge is impossible.

Ok, technically not impossible. There is an administrative policy that gives science for money. It requires a minimum reputation, that I can increase by flying tourists at a financial loss. Which would require more missions to pay for. But I calculated, by flying 5000 missions or so, I could eventually get to sell money for tech.

But, even if I had the required patience (it's not actually any more time consuming than assembling the Navis Sideralis Neanderthalensis one small module at a time, though assembling that mothership was a lot more interesting) what would be the point of it?

I therefore settle for using kerbalism, but keeping stock science.

Edited August 24 by king of nowhere

[king of nowhere]

Part 1: Escape from Kerbin

The limitations of life support make Mun missions a lot harder, and Minmus missions impossible early on. To leave Kerbin I need to research basic science; can I find the necessary science points without leaving the home planet?

Landing in polar biomes

1.1) Strip-mining Kerbin

Spoiler

The beginning of a caveman shouldn't be too different. But, once again, I find another complication that has nothing to do with life support: money. I don't get refunds for the crew pod.

Recovering the command pod on the launchpad gives back exactly 9 funds

Early contracts pay very little, and are generally covered by recovering the spacecraft. I lose 600 funds every time I launch and recover a craft, so most early test contracts don't pay enough. I have two contracts that still earn me some money: splashed down tests generally pay 1500 total, and early survey contracts also pay around 1500 total.

To reach the survey location, I increase the range by using the pod as an impromptu wing

I even tested with adding winglets, but those things keep the vessel pointing straight prograde, when I'd need an angle of attack to generate lift. They didn't work

Here I landed on the island runway. I was hoping it would give a science point for discovering an anomaly

All that is to pay the cost of a roller to mine the KSC for early science

With this early science from the KSC, I can unlock basic rocketry, engineering 101, and survivability. Those give access to the early science instruments, and to the first rocket engines.

I am NOT going to use the geiger counter; I was merely curious to see if it still worked with kerbalism science disabled. And I had to post this funny report

When attempting to use kerbalism science, I reached space with a jumper based on the flea booster, but liquid fuel engines allow more control. They allow more precision to fall on specific biomes, and they allow bringing a kerbal in space in EVA.

I already brought a kerbal to take EVA reports in space with this contraption in chapter 0. No idea why, last time it just worked, this time I needed a thermal shield

The key node to unlock is basic science. It does give the probe core, which allows Mun landings. It also gives - kerbalism shuffles some parts a bit - access to hydrogen fuel cells, which last very long and are the first long term source of electricity. Basically, I need basic science to leave Kerbin.

At this point, in my previous nanodiamond run, I got general rocketry, I unlocked a terrier engine with a contract, and I did flybys of Mun and Minmus. Not this time, I can't reach Minmus with the life support I currently have. The Mun flyby, alone, gives 17 science total - and it requires general rocketry, which costs 20 science, so I gain nothing.

On the other hand, I also lose nothing, the technology pays for itself - especially when you also factr in 1 science for recovering a vehicle from Mun orbit, and 1 science from the contract. I still need to tech it to land on Mun, and having access to the terrier engine - and to bigger fuel tanks, for reduced part count - will help. So I did send a Mun flyby. But for practicality, I'll cover it in the next subchapter, and finish first with Kerbin exploration. First mentioning a few more failed attempts.

This thing was supposed to be a rudimental plane to get survey contracts at greater distances. It failed, for it has no capacity to maneuver. No, the gimbaling rocket can do nothing against aerodinamic forces

This was an attempt to reach Mun without general rocketry. The "Munkicker", on the right, should lean on the Munshooter and push it

I experimented that concept with flat surfaces, in another challenge, and it worked. But the long rocket and pointy command pod don't lend themselves to be pushed that way, and they always slipped away. If this kind of pushing can be done with the parts available, it's beyond my skill level.

Anyway, unlocking general rocketry I got an easier time exploring the rest of Kerbin.

Here landing near the north pole. There are four polar biomes on Kerbin, and I needed to get them all

Although I forgot to take flying low EVA reports. Which I could take by jumping on the ground. I missed a handful of science points that way, but I still got enough.

The terrier allowed an improved version of the rocket that brought an EVA kerbal to space. This one could achieve orbit, and therefore get all biomes of Kerbin

In kerbalism, an astronaut on his suit has enough oxygen for 2 hours. Enough to take EVA reports all around Kerbin, but I can't reach Mun. There will be no more EVA reports in this career.

Here landing on the mountains

As soon as Jeb got out of the pod, the pod lost SAS and started tumbling down the ridge. Just a nuisance, but I have to take boring surveys to get the money back

I eventually visited all Kerbin biomes.

1.2) We choose to go to the Mun, even though we're not ready for it

Spoiler

I could wait to go to Mun. If I don't research general rocketry, I can get basic science from Kerbin alone. But where's the fun if I don't push limits?

The problem with going to Mun, as I mentioned in chapter 0, is having a power source. Engineering 101 gives access to monopropellant fuel cell in the crew pod, but it's basically useless. Survivability enables the first oxygen tanks; I did some testing, a ship can survive 20 to 25 hours with that. I can carry more oxygen tanks (though part limitation is always a harsh mistress) but I don't have access to more monopropellant tanks, putting a hard limit to survivability. 20 hours are barely enough for a Mun mission.

First issue: I need to maximize mass to orbit. But the engines I have can't lift 18 tons, and I don't have the lateral decouplers for boosters. Here's my fix.

I strapped a fuel tank, with an in-line decoupler, and attached the boosters to that tank

Initial plan was to activate the Reliant after the boosters are jettisoned, but I discovered that the rocket can work perfectly well even through the fuel tank. Why was I igniting the engine with the fuel tank attached? See, I wanted to experiment if it was possible to destroy the tank with the heat of the rocket; this way I could skip the decoupler. I am that starved for parts. I already have all the caveman limitations, but I also need to add oxygen tanks. Incidentally, the reliant alone could lift that stack, but I was out of parts to put all the fuel tanks I wanted; plus, a higher TWR for liftoff is desirable.

Here I got the first malfunction to the engine. The rocket was lost, Jeb could detach his pod and safely activate the parachute

Malfunctions are another hallmark of kerbalism, but so far I didn't mention them. They are more relevant to long term missions. Engines can malfunction at any ignition, but while on suborbital trajectories, your kerbal will always survive, you just have to launch another rocket.

For orbital missions, I always made sure to stay low enough that I could reasonably deorbit with just the push of the decoupler. Even if that failed, sending a rescue mission to push the pod retrograde is feasible; I can't get a boost to Mun, but a few tens of m/s retrograde, those I can provide.

For this Mun mission... well, I paid the premium price for a high quality terrier engine, but if it malfunctions, the pilot will just die. I can't mount a rescue mission that far with my current technology. I'll just take the risk, it's roughly 1% per mission. Another reason to look forward to authomated probes.

The Munshooter reaches orbit with 2 km/s. Enough for a normal Mun mission, but I'm maneuvering blindly

Timing for intercept good?

Looks like

Spot on!

Getting a Mun intercept is easy. When Mun appears over the horizon, burn prograde and put your apoapsis at 11.6 Mm (Mun orbits at 11.4, but spending longer at apoapsis helps). I've been able to reproduce that very accurately. Of course, now I need to burn radially to avoid crashing on Mun.

I exited in this high orbit. But I have plenty of fuel left, to lower periapsis

Do notice the oxygen and monopropellant level. Ten hours into the mission, I have slightly more than half the oxygen left, and 75% of monopropellant. I could get more time by adding oxygen tanks, but I was limited by part count. By the way, the timer clock uses 24 hours because I created this kerbalism folder by copying the rss modded folder and removing the rss mod. I did that because it was the most recent kerbalism install I had; but I may not have done a perfect job of deleting all remains of rss. So I have a 24 hours timer.

Anyway, I had enough oxygen to return safely, carrying all the Mun space low experiments. The second mission, to get Mun space high science, was a lot closer. I had a contract asking to orbit around Mun, so I slowed down orbit, then accelerated again, but I had to take one full round of the Mun, in a relatively high orbit, and that costed time.

As a result, here I am 16 hours into the mission, starting to run short on oxygen

Here I shut down the fuel cell. I must save some oxygen for, you know, breathing

Tense moments here. If I pick a too high periapsis, I will return to space after an atmospheric pass, and Jeb will suffocate. If I pick the periapsis too low, Jeb will burn.

With the scrubber offline, CO₂ is building up. Here I am at 1%, at 2% Jeb will die: I have maybe another hour. I'm safe

Those two missions brought back enough science to cover the cost of general rocketry, and left me with some useful tools to finish Kerbin exploration.

Aftermath

In the end, all this science farming - getting all Kerbin biomes, and all science from Mun space - barely got me the 45 science required for basic science.

The tech tree, after this chapter

It was close, though not exceedingly so. If needed, I still could gather flying low EVA reports around Kerbin, which I forgot; that was worth 3 to 5 more points. Then I could collect goo observations multiple times, getting an additional 0.1 points per biome. On all KSC minibiomes, that amounts to something. Plus, I missed Mun space high thermometer science, because in that second flyby the thermometer exploded on reentry. All in all, I could have gathered 6 to 8 more science points that way.

Then, if I really was desperate, there are some ponds around where one can get splashed down science in grasslands and desert biomes. But I'd have to be really desperate to try to target one of those tiny ponds while falling down from orbit.

So yes, it was remarkably close. I needed 100 science to unlock this part of the tech tree - which is the bare minimum to achieve Mun landings - and even the most efficient farming would have given at most 10% more than that.

Just goes to highlight how impossible it was to try this challenge with kerbalism reworked science.

Edited August 24 by king of nowhere

[king of nowhere]

Part 2: Probing the way forward

I don't have enough deltaV to land a crew on Mun, and I don't have enough life support to send a crew on Minmus. This chapter involves lots of probes.

I got a relay network, and permanent science stations on Mun and Minmus

2.1) If you believe, they put a probe on the Mun

Spoiler

First things first, now that I have unlocked all science experiments available in caveman, I finish mining the KSC for every bit of science. I even take materials and goo twice, for an extra kick.

The roller used to farm the KSC. Those science Jr are exposed and fragile, but I already tried in my previous caveman, there's no way to protect them and keep the roller functional

I immediately get enough science to tech stability, which gives me the radial decoupler. Then I realize I didn't really need radial decouplers, but I'm no longer starved for science, I can land probes on Mun and return, so it's not an issue.

I already have a perfectly good plan for a Mun lander with my current tech, thanks to my previous run. I only need to adapt it. Minmus would be easier to land on, if I can manage an intercept. But Minmus is far, I'll go there only after I get solar power and can stop worrying about mission duration.

So, my previous Mun lander was this:

The Mun lander probe I used in my previous caveman career

The key feature of that probe is the crew pod; it doesn't have a crew, but it's the only part that provides reaction wheels, necessary to pilot the rocket with the static Reliant engine; designs based on the Swivel and boosters failed to make it because of boosters requiring too many parts and having too little efficiency. Once the first stage expires, the crew pod is jettisoned - at this point it's not worth its weight - and the probe keeps going without any attitude control, using the gimbaling of the terrier engine to point itself the right way. Which is actually more efficient than it looks like; you activate the engine at a minimum, point yourself the right direction, then burn, you only lose a few m/s. Though it's very hard to make precise maneuvers like that, because every time you must turn around the rocket you still end up burning a bit in an uncontrolled direction; another reason to not try a Minmus mission, as I'd need to set apoapsis carefully.

That probe carried a battery, and it was enough for the mission. Here, it's not. Whether an artifact of kerbalism or a bug induced by some of the dozen mods I have (I know some of those are required for kerbalism to work, and some others are required for those prerequisites to work, and some others are useless leftovers that may be messing up the game, but I don't dare trying to sort it out), it turns out I can't hybernate probes. The core uses 1.8 electricity per minute, a full mission will require 2000 electricity, roughly.

But as I said in the previous chapter, another great part unlocked with basic science is hydrogen fuel cells.

The new cargo bay, with the fuel cell. The small tank to the left of the goo is the hydrogen tank, and the round thing between the parachute and the Stayputnik is the oxygen tank. Parachute is off-center for balance

Alas, it still requires three parts, and I was using the maximum available. Where can I save three parts? I can't reduce the fuel tanks, or the docking ports. I could leave behind some science instruments, technically I could just bring back one experiment at a time, but it would be supremely annoying. I'm already giving up on the materials bay - and not even because of mass or part issues, but because it would push the rocket above the 20 m height limit - and I don't have any way to keep the same parts while reducing height without completely destroying aerodynamics. I can skip the battery, but it's only one.

Ultimately, the only way I can keep my science instruments is by removing the landing legs. Here's the three parts I needed to remove, and it also improves the aerodynamic profile. I wouldn't have though to try this if I hadn't an accident in the old caveman, when a lander landed on the side, and it still managed to take off despite laying horizontally.

I figured, if I remove a fuel tank from the probe, and put it among the drop tanks instead, the probe will be shorter, easier to turn around, and more capable of taking off. This simple change also improved maneuverability a lot during descent, I'm glad I did it.

The Mun probe (I should give it an actual name, now that I think of it) ascending with its first stage. It's actually very aerodynamic, despite the probe core in the middle

Another concern introduced by kerbalism is engine ignition limitations. But a high quality Terrier engine has 42 available ignitions, and over 20 minutes of burn time. More than enough for this mission.

The probe reaches orbit... and loses connection. I forgot how annoying those things were to fly

There is actually a big gap in the deep space network (maybe I should call it shallow space network, since it's only level 1?), after a probe leaves sight of the KSC it will take a long time to find the next station. No, I didn't set up any relay network; I don't want to send up a bunch of probes that will run out of electricity and will need to be changed. I'll just pull through until I unlock solar panels - which, incidentally, will also unlock a probe core with reaction wheels and SAS. In this case, aiming for a higher apoapsis and high speed on leaving atmosphere (over 2000 m/s, with apoapsis above 100 km) is enough to reestablish connection to the probe in time for circularization, and it's not too inefficient.

At least, not compared to maneuvering blindly with a probe without attitude control. An efficient Mun landing can be done with 2600 m/s, this probe has 3500 for good measure.

Perfect Mun encounter

Those Mun intercepts are quite easy: wait for Mun to appear over the horizon, then burn prograde aiming for 11.6 Mm apoapsis. Extremely reliable; not only I hit Mun every time, most of the times I'm spot-on hitting farside crater.

True, in many cases it would be preferable to aim for an orbital insertion, instead of a straight-on collision that will force me to spend fuel. But I haven't found a recipe for that, I'll stick with what works. In this case, as I'm aiming to land, I don't need to enter orbit. I can just keep falling until it's time to brake.

At 50 km I start turning the probe around to point retrograde. At 30 km I have a preliminary slowdown that uses up the rest of the drop tanks, leaving the probe more agile for the final descent

You know what, I'm sick of calling this thing "probe". I'm talking too much about it, it needs a proper name. I traditionally call small landers as simple objects, this thing is a stick with a ball on it, what does it look like... a microphone. That's it, from now on this lander will be called Microphone.

At 10 km Microphone is further slowed down to 200 m/s, at 2 km it's slowed at 100 m/s. It's less efficient than a straight suicide burn, but safer

I also leave the engine at a minimum all the time. It keeps attitude control, and it also avoids the 0.2% chance of malfunction every time I ignite the engine. It's still mighty annoying to fly without SAS. But it is pretty reliable. I never lost a Microphone on descent. Well, except one that lost signal because I was hoping the easternmost fringe of farside crater would still be visible from Kerbin. I also lost two Microphones to engine malfunctions. But none was lost to the lack of reaction wheels. By the way, you can see those few kilograms of hydrogen for the fuel cell will last 22 days; that fuel cell is great.

In the end, Microphone gently lays on the ground

In this case, it landed on the rim of a crater, it started falling down, and it eventually exploded. But the descent went perfectly

A problem of coming down directly from a collision course with Mun is that I have very little control over where I land. Still, after that unfortunate accident I learned to control my landing site a bit better.

In the end, Microphone gently lays on the ground. Again.

To leave, Microphone ignites the engine and pulls up with all its might...

And explodes! Ooops!      [The light is different because it's a screenshot from another failed liftoff. It's hard to screenshot the exact time of the explosion, when you're busy trying to avoid the explosion]

Of course I didn't try such a reckless liftoff strategy without testing it. I tested it exactly once, and it worked. But clearly it's not a guarantee.

This Microphone made it. I haven't been able to figure out a pattern - not that I tried too hard to keep track

I performed 9 successful landings with those probes, and four times they were successful. Five explosions. Adding the test as another success, it gives a perfect 50% of succeeding.

Still, I was expecting something like that may happen, and prepared somewhat. See the battery? The one that's not really needed for the probe? I could remove it easily, but I needed to free up 3 parts for the fuel cell, the battery was only 1. So I removed the landing legs, all three of them because clearly having a single landing leg is stupid, so I could afford to keep the battery after all. And now the battery is my backup. The battery allows to transmit data to Kerbin. I send the temperature and pressure reading before leaving, so I recover some science even if the liftoff goes wrong. If the probe core survives long enough, I can even transmit goo observation. No, I can't activate partial transmission, I don't know if it's cause by some mods or by lack of some key tech.

After takeoff, getting back to Kerbin is only a matter of deltaV. I usually have 400 m/s left or so.

Microphone during descent

Microphone is very stable during descent. The open cargo bay and probe core create drag at the bottom, which keeps the probe pointed retrograde. This protects the sensitive Stayputnik from the heat. I never lost one during reentry, despite the lack of control.

By this time, getting money has become more comfortable. Survey contracts for Mun pay over 20000, and all they require me to do is to put a Microphone in polar orbit and take the required experiments as it passes over the right locations before landing.

2.2) Jeb review-bombed this flying hotel

Spoiler

While you progress the game, you are proposed some exploration contracts that push you to do things you haven't tried yet. Orbit Mun. Now land on Mun. Now make a Minmus flyby. Make a rendez-vous.

Kerbalism adds at least another instance: leave a kerbal in orbit for 30 days. The contract pays 2 full science points, and while I'm not starved for science anymore, I took it. If nothing else, 30 days can be a duration of a Minmus mission, if I miss the initial intercept, so this will be a useful trial for a future manned mission there.

I really should have waited solar panels, in retrospect.

I forced Jeb to spend 30 days locked up in this thing. If I did that to convicted murderers, it would be considered excessive phsycological violence

It's not very visible, but there are 11 parts, between oxygen and hydrogen containers, only to ensure electricity for 30 days. This leaves few parts for comfort. I had to tone down the rocket due to the part limit. I wanted to send up two Onion modules, because they have more space (2.3 m³ against 0.8 for the Mk1), and they are pressurized, increasing comfort. But mass, parts and drag limitation conspired against that. I crashed a few such modules around, before accepting I have to only use one Onion (don't worry, Jeb is safe as long as I have a decoupler and a parachute). But adding even a single Mk1 makes the whole ship unpressurized - indeed, I had huge stress issues during my whole DREAM BIG mission because I had one single Mk1 pod in a ludicrously large ship, and didn't knew about that. This time I disabled habitat from the Mk1 pod, hoping it would leave Jeb in the pressurized environment. No such luck. Even the antenna to provide communication was missing signal half the time, due to lacking a communication network. So, you can see the verdict. Living space: cramped. Comfort: none.

It was such a crappy environment, just 10 days into the mission Jeb hit 100% stress and broke something

And Jeb is one of the most stress-resistant kerbals out there. Good thing nothing important was broken: just the pod scientific instruments - which are disabled anyway, since I'm using stock science.

In the end, Jeb made it to the end of those 30 days, but I clearly can't send a crew to Minmus like that. So I learned this important lesson: I need the hitchhicker container before attempting manned landings on Minmus. And of course, docking ports, because I can't send one of those on Minmus without a modular design. But I can collect enough science with probes to get there comfortably.

2.3) Networking!

Spoiler

After many Mun landings, I finally have enough science to unlock electrics!

Now I have the OKTO probe, which - while being a nerfed version of the similar HECS - does include basic SAS and reaction wheels. I could live without, but this makes things a lot easier. And I have solar panels, although the most basic kind.

Now I can start building a relay network. I don't even have to think much about it, because once you get solar panels, the game gives contracts for satellites. Which is another, final reason I wanted to wait this point to build a relay network: why spend money to launch satellites when you can get paid for it?

The basic sat core module

The core basic sat has enough deltaV to reach any orbit, it's cheap, and it's got low part count. Meaning I can add whatever science instrument I'm asked by a contract, or I can liberally add antennas. I put up to 8 antennas on some. Those with thermometers can be left in orbit to get the "science from X" contracts, the easiest way to get money. Even better, after putting them in the required orbit, I moved a couple to low polar orbits on Mun and Minmus; this way they can both satisfy the "science from X" contracts, and the survey contracts. Now that I can earn 20000 funds in a few minutes, I will no longer be tempted to shortcut the money farming with the cheat menu.

2.4) Caveman hungry! Caveman goes to mint ice cream moon!

Spoiler

I no longer have to worry about battery. I still have to worry about malfunctions, but those basic satellites will last several years. Now I must go to Minmus. I am ready, I have the contract for it, and it gives more science than Mun. I feel kinda annoyed every time I land a Microphone, because I know I will have to return later with a manned ship and a materials bay on each of those biomes. While on Minmus I hope to biome-hop thanks to the lower gravity, and be done quickly.

So, I will use Minmus to unlock the necessary technologies to land kerbals on Mun. Landing on Minmus will come later, I will want some bigger, more comfortable ship for the longer trip.

The Minmus explorer will fulfill the contract to flyby over Minmus, grabbing orbital science. No landing will be attempted yet

I'll just follow the contracts, because they give extra money and science. Besides, I need to farm orbital science. I still haven't gotten to bring a materials bay to Mun orbit yet.

Minmus being smaller and cheaper, I do include materials bay, for low and high orbit. And since they are too bulky and fragile to survive reentry, I include an experiment storage unit.

I set an apoapsis of 48 Mm, higher than Minmus to maximize the time spent near apoapsis, but close enough that I'll still be captured by its sphere of influence

Inclination can be fixed with a cheap plane change around the time the probe crosses the orbit of Mun. In this case, by pure coincidence I'm crossing Minmus path near the node. For every other time, here's how to handle inclination in caveman.

First, move your view to see Minmus orbit as a line

Then you can better gauge how your trajectory compares to that. Here's how it must become, this ensures an encounter at apoapsis

I put apoapsis 60° ahead of Minmus, where it will arrive some 8 days later. Maybe it was too far?

But high apoapsis means spending a lot of time near apoapsis

So Minmus catches the probe as it's falling back toward Kerbin

Minmus is a lot harder to intercept than Mun, but it appears I found a good strategy too. Set apoapsis roughly 60° ahead of the moon, aim for a 48 Mm apoapsis. So far I sent 5 probes to Minmus, and I got an intercept four times. The last time I was slightly late, but I could circularize in an orbit slightly lower than Minmus and catch it a few days later, at very small additional cost.

Setting orbit to catch up with Minmus, that one time I missed the intercept

Minmus has a 2000 km SoI, by staying around 45 km apoapsis I can ensure I'll be snatched by its gravity when passing close.

To think the first time I tried this in the old caveman, I resorted to pointing the engine towards Minmus - purely by eye - and firing! I've learned a lot since then.

I also have a tool that calculates orbital times, and I managed to intercept very tiny moonlets with 10 km spheres of influence in a whirligig world caveman (maybe one day I'll complete it and write a report). But it does require some significant effort on my part to measure all the times, so I'm happier if I can reliably eyeball things.

Here, though, I'm having another problem. After getting close to Minmus, I spent fuel to reach low space, to grab science. Of course, that meant I ended up in a high Kerbin orbit, and I didn't have much fuel to start with, so I spent all my leftover...

And barely managed to lower periapsis into the atmosphere

Actually I wasn't very accurate, I got -47 km. 40 km would have been better, but near the end the numbers were moving very fast. And I'm now completely out of fuel. If I will cross Mun on the way down, I have no backup plan, except transmitting what science I can before the probe is lost to space.

With too low periapsis, reentry is harsh.

The science jr exploded early. Here it's the fuel tanks turn

But the cargo bay and experiment storage and made of sturdier stuff, and survive to bring back their precious science cargo!

I then sent several relays and science stations to Minmus. Now I have to land and come back, bringing science. Minmus is easier than Mun, thanks to the lower gravity, so I will attempt to also bring a science jr. It does give a lot of science. I set to use a modified version of the reliable Microphone (though now that I have swapped out the Stayputnik for an OKTO, it doesn't look like a microphone anymore), just swapping out a fuel tank for a science jr and experiment storage.

In this first experiment, I tried to stick the experiment storage on the side. Bad aerodynamics, but it saves a cargo bay. It exploded on liftoff from Minmus

Not too bad, the probe core and antenna survived, so I am using its remains as science station on Minmus, for the "science from X" contracts. I would have needed to land a probe to leave there anyway.

A better version of Minmus lander

I loathe using a 100 kg cargo bay to shield a 50 kg part, but I had significant problems with aerodynamics. I have enough deltaV anyway, I can afford the extra mass just like I can afford some extra drag, this version is easier to fly. But the main change is to put the decoupler under the science jr. It would burn in the atmosphere anyway, and since taking off from Minmus is not as trivial as I was hoping it would be, shedding any possible mass and reducing angolar momentum helps tipping the point upwards.

Discarding the science jr with the fuel tanks

I'll never discover if this would have sufficed to lift off from the ground, though, because I experimented and found something even better:

Opening the cargo bay makes the probe jump

Closing and opening the cargo bay causes the probe to make a small jump on the surface, which is enough to ignite the engine safely, without sliding on the ground.

The last bother was having to keep the spent tanks through all the ascent. I would like an additional decoupler between the science jr and the drop tanks, but I really can't spare any part right now. But I just got enough science for advanced rocketry, which gives bigger fuel tanks, reducing part count. This way I could afford the extra decoupler, and even a couple extra fuel tanks.

Now I have all I need to finish Minmus exploration in comfort. I'll get nearly 300 science, which will unlock what I need for manned landings.

Edited July 12 by king of nowhere

[king of nowhere]

Part 3: I run out of "we choose to go to the moon" references

Putting a kerbal on Mun, obviously.

Final descent phase

Spoiler

The Minmus landings unlocked enough parts to try more complex missions. Next target is manned landings on Mun. Then it will be manned landings on Minmus, and then Duna.

In my previous caveman, I opted for a lander and a booster that would dock in orbit. That's time-consuming, and since I mastered the art of attaching extra drop tanks with a runway-launched rover (plus I have unlocked the bobcat engine with a test contract), I wanted to try that avenue, to keep the mission in a single launch. Alas, I was twarthed not by mass restriction, and not even by not having unlocked wheels (that could be fixed with a couple more Minmus landings). No, the big problem was part count.

In kerbalism, engines can break at any time, as I'm sure you are tired of hearing. While I was running orbital missions, I didn't worry; if a rocket was disabled by loss of engine, I could rescue the kerbal easily enough. While I did the first Mun flyby, I didn't worry too much. Sure, a malfunction would have resulted in killing Jeb, but I only needed to fly the mission twice, and only for a flyby, it was a very small risk.

Now I have to land, which entails many more ignitions, and I have to do it potentially up to 7 times. And I have the means, so there's no way I'm risking a kerbal's life every landing. So I want redundant engines.

First Mun lander project. I was trying to save parts by storing solar panels and some instruments in the inventory

30 parts are a fairly tight limit in any case. Now, instead of a single engine, I need 4 engines, and since those engines need some kind of node to stick, I also need 4 fuel tanks to attach the engines to. Eight parts, when in the past I could use a single engine.

Then I have to carry a gas tank and some food too. And of course I need to bring 5 science instruments, else what's the point of landing? And that leaves very little for the actual rocket.

I had to give up on the plan of a single launch, and go for two launches just to have enough parts. In fact, having the extra engines takes up so many parts, I had to put the science instruments in the booster.

The booster. It docks with the lander, providing an additional 1000 m/s of deltaV. And bringing science instruments, in this case

Onward to Mun

Engine broken

Funny enough, I lost two booster engines, and no engines on the lander, despite the lander engines being more and used more times. Ok, since the booster engine was used a few times and is not critical I didn't bother making it high quality, but still.

Anyway, since the booster is broken, the ship needs to return to Kerbin. I can at least test the reentry.

The rest of the ship exploded. I take it to mean that I really need the thermal shield

Besides the premature end, this lander wasn't adequate anyway. Those lateral tanks create huge drag, making me waste fuel; I probably wouldn't have enough for a Mun landing anyway. I need to fix that.

A key breakthrough came when I realized I didn't need the Spark, I could use 4 Ant engines.

Another experiment. Those baguette tanks look better, but this design is still too draggy

Now we're getting there

The final version has 8 Ant engines, though I could reduce them to 6 if really needed. Four of them are not visible because they are inside the decoupler; they cause a lot of drag, and I was trying to shield them. It didn't work, they still cause drag. Anyway, the rocket flies. I put a nose cone over the docking port, it's the wrong size but the game doesn't seem to care. I discovered that the docking port generates an insane amount of drag, as shown in the picture below.

An experiment to test drag reduction strategies

I built the rocket with normal decouplers, and I put a docking port inside the stack, that will get exposed once the upper part is detached. The above picture shows, the docking port still counts as exposed to the air, and it generates 40% of the total drag of the rocket. The engine causes another 40%. Ultimately, while tapering the rocket by stacking a size 1 tank over a size 0 decoupler, it's still the lesser evil.

On the plus side, since the docking port makes so much drag, I can skip the thermal shield decoupler. I can be sure the pod will never flip and reenter tip-first, so I just leave the thermal shield in the middle of the stack. The heat will melt the rocket, until it will find the thermal shield, no need to decouple anything.

Here the Mun lander is detaching the rocket and the nose cone

That lander can look like a wizard hat, so I'm calling it Wizard. The booster is long and sticklick, it can be the wizard's Staff.

Wizard docks with its booster. The annoying part is, without action groups I have to deactivate each Ant engine individually

A crew pod has food and air for 5 days. As I could spare a part, I decided to include an oxygen tank, but leave the food and water on the booster. The reason is, in case of delays, the kerbal can survive a few days without drinking, gaining that much time for a potential rescue. With the extra food in the Staff, life support is enough for 25 days - an exaggerated number, but I learned to be prodigal with life support. All those supplies only weight a few tens of kilograms.

Staff, having exhausted its fuel, is jettisoned. It leaves behind the science section

Wizard lands on Mun easily. I didn't miss landing legs too much

The experiments are recorded and stored in the storage unit, then the science instruments are dropped

A pity not recovering their money, but I have infinite money at this point, and the Science Jr will burn in the atmosphere anyway.

Leaving Mun, with the power of 8 Ants

Dropping gently on a parachute

The ship didn't even melt, I ended up not needing the thermal shield. After three such landings, I determined I could skip the thermal shield entirely after all. I used the freed part to add a second solar panel.

I considered small improvements, but they aren't really needed. Wizard and Staff are fully adequagte for their jobs, they entail no risks besides falling asleep during descent, and have a good amount of extra fuel.

Now I have to grind some biomes to get some final techs before I can send longer term crewed missions.

Edited July 20 by king of nowhere

[king of nowhere]

Part 4: A new mothership takes shape

Looking forward to a Duna mission, I start building the required mothership: the Paleodiastimoploio.

https://imgur.com/undefinedhttps://imgur.com/undefinedhttps://imgur.com/undefined

Here docking the sunshield, to protect against solar storms

4.1) Habitation modules

Spoiler

I mentioned earlier I wanted to first send kerbals to Minmus. Going to Minmus will be a longer term mission, it could take upwards to a month if I miss the intercept. I want to provide my crew with some comfort for a longer mission, and this will provide an ideal testing ground for the final Duna mission.

I run some more Mun landings, until I unlock Space Exploration. It gives me the Hitchhicker Container, to store my crew in relative comfort for a longer time. Perhaps even more important, it unlocks wheels, which are required to install more drop tanks on my rockets and launch bigger payloads. I used that strategy extensively during my previous caveman run, and I developed several well-optimized vehicles for the task. I directly copied those crafts in this career, they worked very well.

Before setting to work, I need a name. A name that fits with caveman. Navis Sideralis Neanderthalensis was excellent, but I already used it. Let's go with something along the lines of paleolithic? Paleolithic means ancient stone [age], let's go with something like ancient spaceship. Ask google translate how to say spaceship in greek... huh. Diastimoploio. The new mothership will then be dubbed Paleodiastimoploio. Hard to write, but I like it.

Then I contacted a friend with linguistic studies to ask how the hell greek is a indoeuropean language if their words don't look anything like ours. Ancient spaceship = paleodiastimoploio. Was an interesting conversation, but I'm going on a tangent here.

The runway-launched tanker rover docks additional fuel tanks to the rocket, allowing to overcome the 18 tons limitation

Sometimes the tanker doesn't work as intended, but you can always send another

The launch system thus assembled sends the first habitation module

The Hitchhiker Container provides 9 cubic meters of living space, it's got the best space/mass ratio among the parts I have available. It also provides the threadmill, increasing crew morale. I'm putting maximum radiation shielding, which adds another ton to the ship. Radiations will be the big issue in this mission, even with top shielding the crew will only last four and a half years. Alas, I don't have access to either the active shield, nor the radiation decontamination unit. But judging by the experience with Minmus, 4 years should be enough for Duna.

I'd like to send two containers at once, but they are a nightmare of drag. Even sending a single one in orbit is hard, with the parts and limitations of caveman.

Joining the first two hab modules

When going to Duna, I will want at least 4 hab modules. But for Minmus, two will suffice.

4.2) Chemical plants and other necessities

Spoiler

Without enhancements, the crew pods only have two functionalities. I reserved those for air scrubber and pressure control, but now I need external units for water recycler. I am also puttin in water electrolysis, because it's cheaper to carry oxygen as water than to carry oxygen tanks. The hydrogen obtained as byproduct can be used to recycle some waste CO₂ by the Sabatier process.

I am sticking all those chemical plants to a structural fuselage, because it had the best [surface to attach things]/mass ratio. I don't have any of the more fancy cargo bays. Then I also need some basic containers. The large stockpiles of food and water will be on a different launch, but - as I learned at high cost when A'Twin life support failed to work because it lacked a waste container - I need basic containers for intermediate products of the various life support cycles. A container for waste water, else waste water will be discarded instead of recycled. One for CO₂ and one for H₂, to store them temporarily before the Sabatier process can occur. One for oxygen, I'll produce it continuously but I need some temporary stockpile.

Since I'm here, I'm also putting antennas. They won't do anything at Duna, but they will still work past Minmus.

The resulting service module

Those service modules are fairly light, but they take up a lot of parts. And they have even worse aerodynamics than the Hitchhicker Containers. I need drop tanks not much because of the mass limit, but for the part limit.

But since the rocket is light, adding drop tanks unbalances it and causes it to fall

I solved this problem by leaving the tanker rover attached to the service module, to provide a larger base. So, the two drop tanks on the wide side of the Bobcat engine, those are fine. But for the other two drop tanks, I would dock one tank, leave the rover there, send another rover to dock the second tank, then undock both rovers.

Also due to part limitations, I could not afford a decoupler

I resorted to stacking the Terrier engine directly on top of the Bobcat. The Terrier is activated, overheats the Bobcat, causes it to explode. There, no need for a decoupler

I'm calling this solution "very hot staging".

Both service modules are docked. I needed multiple redundancies because I won't be able to run any kind of repair

This was a lot of effort to save a few tons of cargo. In retrospect, maybe I'd have been better off just carrying more water and oxygen and skipping any chemical process. But now that I sent them to orbit, I may as well use them.

4.3) Batteries

Spoiler

Another issue of Kerbalism is that the ship will consume electricity full time, even just for heating. Losing electricity will kill the crew. I considered putting in more fuel cells and hydrogen tanks, but decided to just send a bunch of battery packs. Slightly less efficient, but I won't have to worry about malfunctions. Batteries are light, anyway.

I like the Ant engine placed for easier docking

Both battery modules placed

Those were easy; light, without aerodynamic issues. I only had to send them separately because of part count. So far Paleodiastimoploio is 11.5 tons, but 97 parts.

4.4) Sunshield

Spoiler

Until now, I ignored radiations. But for a longer mission, they are a major problem. While past experience shows a simple water container may suffice as solar shield, I want something more reliable. Especially as Paleodiastimoploio will grow longer. So I build a dedicated sunshield. I could call it umbrella, but sunshield sounds cooler. The marketing department thought it would help justify the price tag.

The sunshield in all its glory

I used large flat fuel tanks because they were the best I had. I have unlocked the technology to build rocket engines, I have the technology to create orbital habitats that could support life for years, but I don't have the technology to build a square metal plate. That requires composites, a 300 science tech. I had no idea square metal plates were so hard to manufacture.

The sunshield will be exposed to the sun, so it's the best place to stick solar panels. Due to part count limitations, I had to install them on the launchpad

There, those should be enough to cover for malfunctions

I like how the sunshield looks silhouetted by the sun

I was really worried about drag flipping the rocket, hence I'm flying straight up (a common strategy for draggy modules). But while drag is high, the sunshield wasn't prone to flipping

Sunshield being made of fuel tanks, I don't have to worry about deltaV. The leftover fuel will be emergency fuel for Paleodiastimoploio

Sunshield installed

I cut out the sides instead of making a perfect hexagon to leave space for the drop tanks.

 

[king of nowhere]

Part 5: A new mothership is complete

Paleodiastimoploio is completed, at least the parts necessary for a Minmus mission.

Paleodiastimoploio full status. Oxygen is low because I'll need to make more from water

5.1) Minmus lander

Spoiler

I want to land a crew on all nine biomes on Minmus, that requires some fuel supplies.

The first rocket I sent broke the engine. It's high enough to use the second stage, by sacrificing the fuel I managed to bring the science to Paleodiastimoploio. Only to discover I lacked the proper docking port on it

The lander itself. Four Ant engines, a simple setup. It will grab fuel tanks as needed

Those round pods are a huge aerodynamic hassle, but I need them because they are pressurized. Docking the Mk1 pod will cause the whole ship to become unpressurized, and that will wreck crew morale. The crew, in turn, will wreck the mothership in stress-induced breakdowns. Better to suffer through some difficult ascents. Even better would be to use a fairing, but I have none.

After I had everything complete, I realized I forgot radiation shielding. Not strictly needed for this mission, but I want to test its effectiveness, so I sent up a replacement.

The replacement also includes science instruments, which I had to place with EVA construction due to part limits, and the last two crew members. The parachutes are only in case of flight failure

5.2) Fuel tanks

Spoiler

For the fuel tanks, I am also keeping the design I devised for Navis Sideralis Neanderthalensis. No point reinventing the wheel.

Upon loading them, I saw they require the Cheetah engine, which I don't have. I briefly consider cobbling up a new system, but I quickly discard the idea. Sure, I could, but I already launched 80 of those fuel tanks, devising a most efficient and reliable launch platform. It's probably the single most optimized vehicle I ever made, I never launched anything else so many times. So, there is value in using it. I drop another couple of Mun landings and unlock heavy rocketry, then I get a test contract for the Cheetah.

That took forever, by the way. I cycled through hundreds of contracts before getting the cheetah. I keep getting test contracts for the Bobcat or the Twin Boar, but Cheetah appears to be very rare.

During those Mun landings, I found this weird graphic glitch

I tried, but the sun through the hole in the ground does not recharge the solar panels

I also almost collided with the mountains next to the canyon. I underestimated how high they were, or how low I was. I almost lost Jeb

Entering inside Kerbin in map mode I got this view. The color palette is magnificent

Before the stackable drop tanks, I need the connectors. I left the holes in the sunshield for them.

That docking port is on the wrong side. I'd need the Service Probe to fix this, but I didn't include one. I will need to repeat the launch

I didn't include the Service Probe because I want to send one that works on Duna too. I don't have antennas that work from Duna, so I need the probe to be manned, and for that I want the Mk1 lander can, which I will unlock with the technology I'll get from this trip. I still can squeeze more science from Mun if needed; I just landed on the last biome, but a good half of the bomes were landed in the early game, with the probe lander lacking materials bay. In some of those biomes that early lander failed to launch (see subchapter 2.1) so I only got very few science from them.

But I didn't take pictures with the correct connectors, so I have to post what I have.

I also have to send up the engine block.

The engine block being docked behind the sunshield

The engine block is made of three Cheetah engines stacked on each other, to provide redundancy in case of malfunctions. For some reason, despite being neatly stacked, they caused terrible drag. But they were light enough, I was able to carry on a vertical launch with four drop tanks added to the main rocket.

Now, finally, the main fuel tanks.

Launching the 5 tons fuel tanks requires the use of six drop tanks, two of them with a Flea boosterfor higer TWR at launch

In this setup, the launcher weights 37 tons, over twice what would normally be allowed by the level 1 launchpad. I made the drop tanks slightly bigger, which gives an extra 100 m/s for the docking. Those bigger tanks increase the chance that the rocket will capsize on the launchpad. I need to dock both flea boosters on the sides before I can start the individual tanks.

The flea boosters are dropped early on, but they provided much-needed early acceleration

Second drop tanks dropped

When the final set of drop tanks are dropped, the rocket is already in the high atmosphere

It runs out of fuel for the first stage soon after; the Cheetah takes over. It's got low TWR at the beginning

The rocket drops tanks from the tip, just when it's high enough to no longer need a nose cone

A second drop tank is discarded from the tip just before achieving orbit

The fuel tank approached Paleodiastimoploio. It's got a small Ant engine in the cargo bay to help with docking

Docking. This better shows the shape of the sunshield

Finally, the engine is jettisoned and deorbited, leaving the fuel tank neatly docked to the growing ship

That's an extremely efficient system, the staging is very well placed. I'm especially appreciating that I devised a system that doesn't leave orbital debris. I wasn't as careful in this career, and I littered the low orbit with pieces of spent Mun landers.

Navis Sideralis Neanderthalensis required 80 fuel tanks, but for my purposes four will be enough. I want to reach Minmus and then take a dip outside of Kerbin's SoI; 2 km/s should suffice, I'm carrying over 3, and that's without counting the emergency fuel in the sunshield.

Finally, some food, water and nitrogen. This launcher was aerodynamically challenged, but light enough to allow a fully vertical launch

Last I sent up the crew, but I already included them earlier in the Minmus lander subchapter.

Paleodiastimoploio is ready for its maiden voyage

Life support looks good; that single large container gives food for two and a half years. Water supposedly will last less, but it will be recycled. Oxygen will be produced from water by electrolysis. Two small nitrogen cans will last 25 years. I'm so used to my colossal mothership, I sometimes forget how little it can take...

Main issues will be stress and radiations, though. Besides grabbing some science and leveling the crew to level 3, this mission will provide some hard data on the resilience of the system.

 

[king of nowhere]

Part 6: Paleodiastimoploio's maiden voyage

Paleodiastimoploio goes to bring a crew to Minmus, then take a dip in solar space. Besides colllecting some science, the mission serves as a general test for the future Duna mission. No major issues are found, though a few things can be improved.

6.1) To Minmus

Spoiler

Paleodiastimoploio weights 50 tons, and the Cheetah engine provides 125 KN of thrust, for a TWR of 0.25. Too little to make the Minmus transfer burn in one go, especially without maneuver nodes. I have to perform a couple consecutive apoapsis raising.

A close-up of the ship

Radiation report

Unfortunately, a gradual apoapsis raising entails crossing the radiation belts. Here I'm only getting a few % of radiation, but with the crew subject to continuous low levels through all the trip, I will want to avoid it when going to Duna. So, first issue found: provide more thrust.

Jettisoning the first drop tanks

I made a perfect Minmus transfer, but I forgot the plane change maneuver. Oooops!

I can fix that. Circularize, make sure to cross Minmus orbit

Now Minmus is further ahead than Paleodiastimoploio, so if I enter in a 45x45 Mm orbit I will catch up. Ultimately, I'm glad for the accident; it gives me the chance to test longer term survivability in terms of supplies, stress and radiation exposure.

Solar storm. Note how radiation is 4.7 rad/h, while habitat radiation is 7.3 mrad/h

This shows the sunshield doing its job. The solar storm is completely negated. I was unsure it would work so well, since there are small holes between the fuel tanks. The 7.3 mrad/h can't be avoided, they are the background radiation lowered by the shielding (which at the highest difficulty level negates only 72% of incoming radiations). Without access to decontamination units or active shields, the crew has a hard time limit for how long it can stay in space. Which is why I want to reduce as much as possible time spent in the radiation bands.

Note that by past experience, the game glitches if warp speed is greater than x1000 during a storm, the crew gets slightly irradiated. I don't want to check if the bug is still active; I also would rather not slow down the game dozens of times, I think I'll just lower solar storm frequency. Solar storms are potentially a huge issue, but once you have a safe way of dealing with them, they are no more than a waste of time..

6.2) At Minmus

Spoiler

Without other issues, Paleodiastimoploio arrives at Minmus. There, it drops an unnamed lander.

The lander carries two people at once. I want to bring everyone on the surface, to promote everyone to level 3

The lander arrives on the surface with minimal fuel spending

Those low thrust engines are quite dangerous, my instinct is to come down with higher speed, but here I must brake in advance.

Since fuel is limited, I reach some biomes by rolling on the ground

Slow, but effective for short distances.

Coming back for the second fuel tank

Here I had an issue. Without the lander, Paleodiastimoploio has no control - I plan to add a dedicated pod, later, but I didn't think it would have been relevant. Anyway, while the solar storm raged, I was unable to orient the sunshield properly, and the crew took damage. Also the crew in the lander took damage. I will need to devise something for the Ike landing, Ike is exposed to solar storms.

I still had to grab science from greater flats. But I placed the components in the wrong order to discard the spent science bay. Here improvising

Docking the science container on the ground. Not the most elegant solution, but I could drop that dead weight before going orbital

Fuel was a bit short, I almost lost the lander. I should have carried more fuel; barring that, I should have skipped one biome, it's not like the 2.5 science I get from crew report is all that important.

I also ended up with limited resources in the lander. Just like I can't transfer fuel between tanks, I also can't transfer food and oxygen. In this light, using electrolysis to replenish oxyge stocks everywhere is good. But I will need to provide small supply packages that probes will be able to grab, in case they run out of their own.

6.3) In solar orbit

Spoiler

Now I will leave Kerbin SoI for a short time, to grab some additional science and provide enough xp for the crew to advance to level 3.

Leaving Minmus. Still half a drop tank available, then I can drain the connectors, then I have some emergency fuel in the sunshield

Orbiting Kerbol

Unfortunately, I came out on the wrong side, orbiting lower and ahead of Kerbin. Hard to aim, without maneuvers. So, this orbit will bring me farther and farther from Kerbin. I need to spend some fuel to rectify that issue.

Burning towards Kerbin

So, just like I did in the previous caveman, I pointed the ship towards the planet... no. No, I wasn't so crass. I did learn a bit since then.

I have an orbit slightly inside Kerbin, meaning I'll move a bit faster. I could raise orbit, ensuring I move slower, and this way I will surely meet Kerbin. But it will take at least 6 months. Barring that, I could burn radial-out, until Paleodiastimoploio is moving away from Kerbol. I'll do that. Notice how pointing radial-out in the navball is the same as pointing towards Kerbin. But at least I know when to stop burning: when the altimeter says I am no longer moving downwards.

It required some 300 m/s. I ended up consuming what's left of the drop tanks, and one connector. I set the connectors to be drained separately, even though it imbalance the ship, because that way I don't have to launch both connectors again.

To circularize, I use aerobraking in high atmosphere. Paleodiastimoploio is not made for heat endurance, so I am keeping a 60 km periapsis. Still, the sunshield is great at aerobraking

As I take several passages through the atmosphere, I am also spending more time in the radiation belts. I would be a real joke to return from Duna, only to lose the crew during circularization maneuvers. So I will include a couple of hardened reentry pods that can bring the crew on the ground with minimal radiation exposure.

The first engine lasted more than it should have, but eventually it broke from repeated use. Here's why I carried 3

And with that, Paleodiastimoploio circularized back into a 90x90 orbit.

The maiden voyage was successful. Radiation levels were as predicted. Water consumption was minimal. Crew stress was under control. Structural stability was good. Ship performance and maneuverability was good.

The mission also highlighted a few issues. I have to increase thrust, to avoid passing the crew through the radiation belts multiple times during the transfer burn. I have to include heat-hardened reentry pods, to avoid passing the crew through the radiation belts multiple times when returning. I have to make sure that the Ike lander will be radiation-hardened. I will need to prepare some food supplies that crew pods will be able to access.

Now I have to refurbish the ship and ready it for Duna.

 

Edited July 31 by king of nowhere

[king of nowhere]

Part 7: Preparing for Duna

Going to Duna requires a lot more preparation. Besides,  it also requires a dedicated lander. Paleodiastimoploio is expanded to take care of those needs.

Paleodiastimoploio now has everything it needs, except the fuel tanks

7.1) Harnessing some sled dogs

Spoiler

As I established in the previous chapter, I need more thrust to avoid multiple passages in the radiation belts. For this, I will bind two Poodle engines to the back of the sunshield. There's a small problem with that, there is no docking port where I need; hence I will send Bill in a spacewalk to install a couple of docking ports. To send him up outside of a cockpit, I will once more resort to a tried and true designed I developed for my previous caveman.

Bill goes out of the cockpit. Jumps on the truss

Grabs a ladder, goes into the cargo bay

The truss is jettisoned, and the ship goes to space on remote piloting. Now, for weird reasons, Bill is not shielded from the airflow, even though he's in a cargo bay. He generates a ludicrous drag. The lone astronaut creates more drag than the whole rest of the spaceship. To compensate that, the shuttle has a lot of extra deltaV.

Quite a common situation in a caveman run, really.

The only real difference introduced by kerbalism is that Bill has oxygen for two hours. That's enough to complete the mission with a reasonable margin, especially if I can spend deltaV to force an earlier encounter.

Bill installs the docking ports where needed, then gets inside his ship and returns

Now that the ports are in place, I send the engines. They are drag incarnate.

I launch them with extra deltaV, courtesy of the drop tanks rover

Sigh. The whole "this payload has terrible aerodynamics, so I send it up vertically with extra deltaV" is starting to get stale. Ok, how about this: to avoid the rocket flipping, I put the draggy payload at the bottom, and the main rocket body on top. To have the poodles as close to the rear as possible, I put even the second stage engine in front.

Like this

The rocket reverts in mid-flight after jettisoning the first stage. That's a novel solution

The Poodles are docked to Paleodiastimoploio

I considered unlocking fairings. If I landed on all remaining Mun biomes that I partially explored with unmanned probes, I could just get one more tech. But it would be a lot of work, adding drop tanks is faster.

7.2) Bits and odds

Spoiler

Following a bunch of smaller parts that don't deserve their own chapter.

The command module, on the launchpad

I want a dedicated crew cabin for a pilot to command the ship, and a probe core to at least keep stability in any case. Not strictly needed, I could have used any of the other command modules I'm about to send. But I like it. And it's habitale space. I wanted a two-seater, and the Mk2 Command Pod is better than the Pea pod. I needed to get the advanced flight control tech for the Mk1 Lander Can anyway.

Speaking of habitable space, I added a third Hitchhicker container. No pictures for it. This time, I managed to stick 4 docking ports to it without blocking the exit hatches by the incredibly complex trick of staggering them 45°. I don't know why I didn't think of it earlier.

Escape pods and service probes

Another thing I established I would need is some sort of reentry pod, so that the crew wouldn't have to aerobrake gradually, passing through the radiation belts every time. Well, I could have added 900 m/s of fuel instead, but two pods are lighter. In the picture above, shown as they jettison the first stage.

I am incredibly frustrated with how draggy those round command modules are. It's ridiculous. A sphere is decently aerodynamic, but those things are worse than a flat surface. I swear, a Hitchhicker causes less issues! They also have ridiculous surface, which is irrelevant in the stock game but requires more radiation shielding in kerbalism - the mass of the additional shielding more than makes up for any mass saving.

But hey, those things have good heat tolerance, so I need them as reentry vehicles.

Since they are there, they can also double as Service Probe(s). I won't have communication, so a service probe that must move stuff around needs a pilot. Those Pea pods need a bit of fuel and small engines for final course corrections, I may as well put them to use. The Ant engines are wonderful for the purpose of orbital maneuvering, they have more than 100 allowed ignitions.

And of course, the Service Probes need some small dedicated fuel tanks, for practicity. Any fuel not used will still be good for the main engines

Fuel tanks are placed. Also, there were no other screenshots showing how the command module and Service Probes are placed

Next item on the list is science. Materials bay and goo canisters can't be restored in caveman, so they will be discarded after use. Again, I recycled the design I had for Navis Sideralis Neanderthalensis.

The science stuff, going orbital

That payload has an even worse aerodynamics than most of the other horribly draggy stuff I'm launching, but it's also light. The first rocket I sent flipped three times, but there was enough fuel to still reach orbit.

I sent a total of 12 science containers. I plan to send a rover and scout some additional biomes, just because a normal lander feels uninspired.

The science containers are docked with Paleodiastimoploio

Since I'm there, I also used the leftover fuel in the second stage to deorbit some space debris. I was leaving too much litter in the low orbit

Now it's time for some redundancy. Paleodiastimoploio already has exaggeratedly high standards in that regard, but its landers and probes don't. What if one of the reentry pods breaks the reaction wheels? Or a parachute?

Four service kits are added to the ship

Those kits are lightweight, and contain everything a vehicle needs: reaction wheels, life support, parachutes, food. Just dock and take one. I may have issues with the docking if a vehicle breaks its reaction wheel; but then, I have two Service Probes for one such occasion.

All the stuff I'm docking is starting to confuse the game. Occasionally, Paleodiastimoploio will start shaking. If I'm not fast to time warp (which cancels relative movement within the same ship) it will break apart.

Quite spectacularly, at that

Well, this is a nanodiamond run, but for that specific problem I consider myself fully allowed to reload.

I tried fiddling with autostruts - or their lack thereof - but couldn't find any conclusive solution. At the time of this writing, I seem to have stabilized the structure by splitting that pile of 8 science containers - they seem especially bad. But I have no certainty that Paleodiastimoploio won't disassemble again the next time I load it; for safety, I now save the game before getting the mothership in physical range.

Next item on the list: more food and water.

I could send multiple smaller containers in a more linear fashion, but this reduces part count

Radiation will kill the crew in four years and a half. No point bringing food for a longer duration. Including the smaller containers with the service kits, I loaded food for five and a half years, which will cover possible "food has been dumped overboard" issues.

Finally, I need some more Cheetah main engines.

Delivered as a neat pile that can be docked to the sunshield when the main engine pile there gets exhausted

That should cover all I need for travel and reentry. Now it's time to design the rover.

7.3) A rover for the red planet

Spoiler

As I mentioned, a simple rocket lander seemed lame, so I decided to make a full rover.

A rover will need solar panels, a fuel cell to survive the night, science instruments, and of course full redundancy. I even included the Geiger counter, I will finish the challenge with the science I return from Duna, so it's no longer cheating. I tried to go for a large crew, but in the end just building a rover for a single person proved difficult enough with the limitations of kerbalism and caveman together.

The rover, as it should be. The docking ports will dock fuel tanks

Of course, that thing has 50 parts, and there's not even a launcher yet. So I stripped everything I could from the basic design - basically, I only left what could not be added with EVA construction.

A stripped-to-the-bone Duna rover, ready for the launchpad

And now Bill has to work on adding everything else. Getting the wheels properly lined up was a pain. And then I discovered I could not install the engines inside the trusses in EVA construction, so I had to remake the rover and start again.

To provide better access to the top, I have a new version of the Brontosaur rover

Four kerbals are required to lift an Oscar-B fuel tank. Good thing I have exactly four

After the rover is completed, drop tanks are added. Here I realize the wheels block access to the docking ports, except two of them. So I have to revise the design of the drop tanks a bit, by sticking them to each other.

Like this

All those tanks may be overkill for a 3.5 tons payload, but I can't test this thing, I want to be sure it can reach a stable orbit on the first try.

Wait, I forgot to name the rover. I can't keep writing "the rover" in the report. I stitched it up together, like a Frankenstein monster, maybe I could merge it with the name of an actual rover and go with Frankensteverance? Naah, I don't like that sound. It's X shaped, maybe X Without Wings? But I stopped liking Star Wars decades ago. And I already used the X shape as inspiration for a name in Right Answer, for one of my Jool 5. It's short and squat and sort of round, and it will be even rounder with the addition of drop tanks. It really looks like something that has been squashed flat. So perhaps Hamburger? I like the sound. It keeps with the tradition of naming small landers after mundane objects. Hamburger it is, then.

Hamburger was very difficult to control. But the bloated first stage was enough to push it past the low atmosphere

I realize just now that I had no contingency plan in case the launch failed. Which normally is cool, but this time I had Jeb piloting - I figured I could use all the attitude control I could get. I dodged a bullet here. Sure, I could lose Jeb and hire another pilot and move on, but it's not the same thing.

Hamburger in its final shape, approaching Paleodiastimoploio

And docking on the front

7.4) Moar bits and odds!

Spoiler

Hamburger can land on both Duna and Ike, but it still needs its fuel tanks. Four 1-ton tanks give 2000 m/s. Some will be spent to soften the landing; then there may be some service kit added to the rover, making it heavier. And I totally sacrificed aerodynamics, to the point that even Duna's atmosphere is a serious obstacle. In the face of that, I would not want anything less.

Ike is a lot easier, two drop tanks providing little over 1 km/s will be enough. I can even loan one of the smaller drop tanks for the Service Probes, if needed. So, I will launch six 1-ton drop tanks.

In two batches of three, here the docking of the second batch

I moved them around for better balance, including docking two of them to Hamburger already.

Finally, Hamburger has unique needs: to last the night, it has fuel cells, and electrolysis apparatus to make hydrogen during the day. What if some of that breaks up? The service kits don't have fuel cells. So I'm sending a couple such kits tailored to Hamburger.

Here shown with the second stage

Only one of them has parachutes, for Duna. If that one breaks, I will add one of the regular service kits, of which I have 4.

And that's all, I'm ready for Duna! Now I only have to carry enough fuel.

Paleodiastimoploio, status

In the end, this thing is heavier than I was hoping. Then again, I already docked a 5-ton fuel canister, and the two fuel connectors hold another 5 tons by themselves. Plus the sunshield is holding another couple tons. Anyway, I have to bring all that stuff to Duna and back, in four years, without maneuver nodes. I made some detailed calculations of how much fuel I will need, but I'll spare those for the next chapter, which would otherwise be very short as "I sent up the same fuel tank a few dozen times, now I'm done, yay!"

Edited October 14 by king of nowhere

[king of nowhere]

Part 8: Giving the rocket equation its due

Sending up all the fuel for Paleodiastimoploio.

Also, Paleodiastimoploio is finally finished. At 230 tons and 550 parts, it's way bigger than I was expecting

8.1) Propellant for the propellant god

Spoiler

Now that I sent in orbit everything I need for the ship to work, I must perform some calculation for how much fuel I will need.

Normally, 1 km/s is enough to reach Duna and aerobrake. Not so when you can’t set maneuver nodes. I need a sufficient amount of fuel to account for the flight plan. But since every 5-ton drop tank sent to orbit requires 40-50 minutes of work, I’d also rather not carry too much of an excess. Hence, some calculation is needed.

Going to Jool, I eyeballed the Kerbin ejection burn, and ended up with way too low a solar apoapsis, spending lots of fuel to raise apoapsis in solar orbit. This time it won’t happen. I will make a tool to tell me exactly which speed I should have depending on altitude, so I plan to get a Duna intercept right. Should be 1100 m/s. Let’s make it 1200, for safety.

Of course, while I plan to pass near Duna, it would take a huge amount of luck to get an intercept. Instead, I will have two options. Option 1, do what I’ve been doing with Minmus: circularize orbit until I am slightly lower than Duna, or slightly higher, and overtake the planet. That is only practical if I miss the planet by a small amount. Option 2, I will do what I did in coming back from Jool, taking the exact time of passage and using that to calculate how to modify my orbit to meet Duna in the next passage. The encounter must be at the next Duna orbit, at the latest, because of the radiation deadline.

Both options require circularizing in solar orbit. I checked, it’s 950 m/s. Let’s make it 1 km/s. On the plus side, arriving at Duna I will have a very low intercept speed, so I can mostly neglect that cost. Let’s call it 2500 m/s to get there.

In the return trip, I will have to do the same, on the opposite direction. So, in total, I want 5 km/s.

Drop tanks can be modeled as fuel. After all, they get ejected from the spaceship during engine operation. A 5 ton drop tank can be modeled as having 5 tons of fuel that will be burned with Isp 355 s, and 665 kg of metal that will be “burned” with Isp 0. In total, I can consider the whole mass of drop tanks as fuel with Isp 310 s.

Paleodiastimoploio, after removing the fuel tanks I already sent, weights 55 tons. Applying the rocket equation, 5000=3000*ln(Mw/55), I get Mw= 275 tons. I’d have to launch 40 drop tanks. Ok, less than the 80 I used for Navis Sideralis Neanderthalensis, but it took me weeks for that one; I’d rather not repeat the full experience. Let’s try some less precautionary estimates, taking into account that I will shed a lot of mass on Duna.

The 55 tons of mass of Paleodiastimoploio are divided as such:

5.5 tons Cheetah engines

3.8 tons Poodle engines

7 tons drop tanks for the rover

3.5 tons for the Hamburger rover

4 tons for the disposable science kits

1 ton of spare kits for the rover

1.7 tons of spare kits for the reentry vehicles

1.5 tons of fuel for the service probes

3 tons of fuel, mostly in the sunshield. I will count that as dry mass and save it for emergency.

2 tons of food and water

2 tons for the command module

3.3 tons for each Hab module

The remaining 10 tons are mostly the reentry vehicles, the sunshield, and the service modules. I’m not counting them because I absolutely cannot afford to drop them.

I expect to have expended at least a couple Cheetah once at Duna, so let’s count 2 tons shed. I may be able to leave without the Poodles, but it may be more efficient to perform the Duna eject burn with them; I can certainly drop them for the last burn, which will be in solar orbit. Let’s average that to 2 more tons shed, it’s 4 so far. The drop tanks for the rover will be spent, and if they are not, they are extra fuel I can use. The disposable science kits also will not be there for the return trip. I may want to bring the rover back for a museum if I have extra fuel, but I can safely discard it, and its spare kits. It’s 19.5 tons shed with that. The fuel for the service probes, either I will have used it, or I will be able to recycle it. And I will have eaten 1 ton of food.

All in all, I can assume 30 tons of mass that will need to return from Duna. In an emergency, I can drop the command module and pilot Paleodiastimoploio from the Service Probes (I keep referring also to reentry vehicles, but the two are the same), and drop a habitation module too. I could even drop two habitation modules if I was really desperate. That would remove another 8 tons; in addition with the 2 tons of fuel in the sunshield, it can already be considered a very safe margin.

So, with 30 tons of dry mass at Duna, needing 2.5 km/s to return, 2500=3000*ln(Mw/30), I will need 40 tons of drop tanks. Before that, I will need another 2500 m/s to carry to Duna 55 tons + 40 of drop tanks, 2500=3000*ln(Mw/95), I get Mw= 217 tons, requiring 160 tons of drop tanks.

It means 30 drop tanks. Ok, factoring in the exact mass of the empty tanks and docking ports, I will need 28 tanks, but I prefer a nice round number. It will take a couple weeks (it took three, but only because I was busy), but it’s already a lot better than 40. Which is itself a lot better than 80.

8.2) Tanks for the tank throne

Spoiler

I start work on docking 30 drop tanks in the two lines

32! Kerbalism does cause the occasional malfunction

No, still 30. I freely admit I just reverted flight for those couple rockets that malfunctioned. Docking all the fuel tanks on the launchpad is a chore, I saw no point in repeating it

As I launch yet more fuel tanks, I gradually discover some quality of life ehnancements. For start, I discover that I can use 2x time warp while driving the rover. No, it’s not obvious; sometimes time warp destroys stuff moving on the ground. I also optimize the rover trail: I used to turn right near the spaceplane hangar and taking the crawlerway, but near the end there is a terrain glitch very dangerous to wheels. Now instead I keep going past the SPH, and I go down afterwards. I discovered that those ramps are not dangerous if taken diagonally.

Like this

Paleodiastimoploio keeps exploding, though

That's a rather annoying fixture. Sometimes it explodes, sometimes it doesn't. Sometimes it starts shaking, but if I quickly time warp, it stops. If Paleodiastimoploio doesn't explode within the first minute, it's generally safe, which makes me think it will work during the mission.

However, to avoid having this issue at every docking, I soon decide to create a train of fuel tanks in orbit, and then send the train to dock Paleodiastimoploio at once. This way, I won't have to load Paleodiastimoploio every time. Not only avoids the risk of RUDs, but the ship is complex enough that it takes some time to load and lags the game noticeably - though nowhere near my major ships.

I start docking the first two fuel tanks

Earlier in this project I built bigger drop tanks, which I now call the superheavy configuration (I can’t always get creative). In the previous nanodiamond run, I used the smaller heavy configuration. It turns out, the heavy boosters are adequate to carry drop tanks in orbit with some extra, and they are a lot safer to drive as rover – the superheavy has a relatively high chance of exploding upon activating the decouplers, or capsizing the rover if you fail a turn, or breaking wheels if you hit a ridge too hard, or capsizing the rocket once docked. I reserve the superheavy boosters only for the most challenging payloads – the ones where I need some aerodynamic-unfriendly modifications.

Like, the drop tanks only have a small battery, because they are supposed to be docked within a few orbits; to create a train with them, I had to place solar panels and antennas on a few of those launchers. And since the launcher is already very tight at 30 parts, I had to make room for the additional stuff by removing non-essential parts.

Like the nose cone.

For those, I used the superheavy boosters. The heavy was enough for everything else.

More drop tanks are added

I'm keeping the last part of the launcher for two practical reasons: first, those contains still a bit of fuel, which will be needed for the final docking to Paleodiastimoploio. Second, they also contain the reaction wheels needed to turn around the ensemble. At some point, after realizing all those probe cores are draining an ungodly amount of electricity, I decided to deorbit some of the more spent launchers, but I immediately regretted it due to how sluggish the tank train became.

Eventually, it grew to 22 tanks

I already placed 8 tanks on Paleodiastimoploio before deciding on this course, so this is the end.

The tank train has become quite hard to manage. Despite autostruts, the tanks wobble. The attitude control is very sluggish (damn square cube law increasing the torque more than the reaction wheels), and the wobbling means it's hard to point at anything precisely in any case. It will be a problem for docking, but Paleodiastimoploio can still turn around easily. Maneuvering with the engine works well enough, at least.

I had the tank train parked in an orbit 3 to 5 km lower than Paleodiastimoploio. This way it wouldn't have risked a collision, but it would have been cheap and easy to rendez-vous.

By lucky coincidence, as the final drop tank is docked, Paleodiastimoploio is passing close

The tank train maneuvers to reach the spaceship. It ended up at 152 tons and 260 parts

By the way, did I discuss docking without maneuver nodes and close approach marker?

Get in orbit. Fix inclination, by eyeballing - it won't be perfect, but will be within a few km. Get in an orbit slightly higher or lower than your target, to slowly gain or lose ground. Slowly is the key word here, else you may just miss it entirely. You should be able to get within a few kilometers. When you reach under 7 km of separation, point towards the target on the navball, and get a speed of 35 m/s towards it. That's enough to overcome orbital drift. Less speed is dangerous, you risk drif moving you too much, and you'll end up needing another burn towards your target, spending more fuel than you saved. It's probably possible to get into even closer orbits, but kinda hard, because the game does not show apoapsis and periapsis of your target. Anyway, with around 100 m/s, and a dozen rocket ignitions, you can perform all the docking.

The first time it was loaded in physical range, Paleodiastimoploio greeted me by exploding. Obviously

As anticipated, docking was hard, but made possible by Paleodiastimoploio itself being able to move with precision, thanks to Jeb

I also need to rotate the tanks correctly, else they'll interfere with the engines exhaust

After docking, I leave behind 15 tanks and detach the other 15 to dock on the other port

Lining them up wasn't easy, Paleodiastimoploio is a lot less mobile with the first tank line installed

But it was accomplished. The engine broke at the last ignition

Finally it's time to remove what's left of the launchers; they make up 150 parts and 35 tons, quite a lot!

Using a functioning engine to deorbit the broken one

Various launchers dropping in the atmosphere

This operation was slowed down by a glitch. When you deorbit burn a spent launcher, it will move out of physical range. But when you change vessel with [], and I know of no other ways to change vessel, you will still change to the spent launcher, even if it's outside of physical range. It's like the game included it among the vessels you can cycle with [], and doesn't remove it when it gets out of range.

Problem is, every time this happens, Paleodiastimoploio itself goes out of physical range. It takes a minute or two to load it again - and the time increases with subsequent loading, to the point that after a few times it's better to reload the game.

The drop tank train took a second to load. It seems the processes of the life support and chemical plants are what slows down the pc. Unfortunately, asking the crew to please stop breathing is not a viable solution.

Wait, it's actually a very viable solution, I'd just have to deinstall kerbalism. Except that the whole point of this career was to use kerbalism.

Anyway, Paleodiastimoploio is ready

I considered changing the name to Mesodiastimoploio, since it's a more advanced version. But if I did, I'd have to eventually reach Neodiastimoploio, and I can't have a name proclaiming "new" in a caveman career.

I did a preliminary test, and the engines work At TWR 0.28, the 1100 m/s burn will require six minutes; not good, but manageable. I still need to bring Jeb back on the ground, to avoid him getting irradiated too early, and then I will move on the crew at the final moment.

Oh, and I still have to figure out a few formulas on my datasheet for the ejection burn.

I can't wait to try and see how well Paleodiastimoploio works.

Edited August 25 by king of nowhere

[king of nowhere]

Part 9: To brave the open space

Paleodiastimoploio reaches Duna. Everything goes according to plan, except for one slight, grave miscalculation.

Duna is visible above the tank train

9.1) Tools and calculations

Spoiler

My original plan was to launch at the transfer window, then use the next orbit to refine the encounter. This would have required circularizing orbit in solar space, and spending one Duna year to meet the red planet.

I realized I can do better. If I launch after the transfer window, I will pass well behind Duna on the first orbit, and I will have to catch up in the next. This means that I won’t be required to fully circularize orbit in solar space, but only to raise it just enough to ensure an encounter. I may even be able to get some Oberth effect, or even some aerobraking done. And I will spend less time in space with the crew, with the radiaton clock ticking.

The Kerbin-Duna transfer is a 640 days orbit. Launching in the launch window, I will reach solar apoapsis when Duna will pass there. Then, if I don’t change anything, I will return to the same spot 640 days later. Duna will return to the same spot 802 days later. If I want Duna to be in front of me, so that Duna will reach my spot in 640 days to minimize the need to burn in solar orbit. But I’m on Kerbin, which moves differently than Duna; I can’t just launch 160 days earlier. How to calculate it… Ok, got it.

By the alexmoon tool, I can see that there is a 920 days interval between two Duna transfer windows. This tracks the relative movement of Duna and Kerbin; in this time, Kerbin will make a full orbit relative to Duna, returning in the same relative position. So, to miss Duna by 160 days, I want to miss it by 20% of an orbit, 72°. It means I have to leave 20% of the time before the launch window. 20% of 920 days earlier, so 180 days before the transfer window. And at that time Kerbin will be 72° behind the ideal transfer geometry.

But I won’t be very accurate with the apoapsis, and I want to make sure I will have to raise solar periapsis a bit to reach Duna. This will safe guard against needing to lower periapsis, which will cause to both spend more fuel and increase intercept deltaV. So we’ll start somewhat later than that. Let’s say 120 days before the transfer window, that should be safe.

Of course, I am stuck with a 24 hours clock, so it will be 30 days for me. In my 24 hours clock, the next transfer is at 1:290, and I will go at 1:260.

Finally, if I want to perform the transfer burn with any accuracy, I need to gauge 1100 m/s exactly. No, I can’t just keep accelerating until I reach 3400 m/s, because Paleodiastimoploio will be going away from Kerbin by the time the burn is done, so its speed will get lower. For this, I built the final tool – just a few lines of calculations in my excel datasheet, really. It calculates the speed I should have at every altitude. If I want a 1100 m/s burn, I enter 3400 m/s, which is the speed I should have if the burn was istantaneous, and I enter the actual altitude. The tool then calculates the sum of kinetic and gravitational energy and returns the speed I should have at that altitude. So, when I’ll be in a Kerbin escape trajectory, I will pause the game, input my current altitude, and check with the tool if I should accelerate some more or if I’m good.

For fun, I tried to integrate the gravity formula myself, but I discovered that my calculus is very rusty, and I had to get some hints from the internet.

9.2) Didn’t think this through

Spoiler

Paleodiastimoploio has a TWR of 0.3; it will take around 6 minutes to perform the transfer burn. In the last chapter I said it was fine, but it really is not. With an orbital time of 32 minutes, starting the burn 3 minutes earlier means a 34° angle. Not only cosine losses are quite high (that part, I was ready to accept) but the burn will be less accurate overall, especially because I don’t have any reference for where the theoretical istantaneous burn should actually happen.

I could send another couple of Poodle engines, but I got a daring idea. I said that raising apoapsis gradually is off the table, because it would expose the crew to multiple passages through the radiation belts. But! What if I sent Paleodiastimoploio alone to raise apoapsis, then I’d bring the crew later? For this plan, I need a crew shuttle with lots of deltaV, which I achieve by strapping drop tanks to it. I rendez-vous with Paleodiastimoploio, but don’t dock. I raise apoapsis on Paleodiastimoploio, until I achieve an orbital time of 64 minutes. Or 96. This puts Paleodiastimoploio in a resonant orbit with my crew shuttle, so that it will go back to periapsis right when the shuttle is passing there. So I can dock – using all the extra fuel to speed things up – and finish the transfer burn.

The shuttle designed for the task. It's got a lot of deltaV once the six boosters are strapped on

Only halfway through launching the shuttle I realized there were several problems with this idea. The main one is, I will dock with Paleodiastimoploio at periapsis. A bit later, because docking maneuvers still take some time. To finish the transfer burn properly, I’ll need to start burning before periapsis, not after. If I make the final burn after periapsis, it will alter my direction.

I also realize this plan will require the crew to be in orbit in an unshielded vehicle, exposed to radiatons for a while. Just the background level, but it still hurts over time. Putting shielding on the shuttle would lose too much deltaV.

As I’m wondering whether it’s worth salvaging this plan, I realize I forgot to put solar panels on my shuttle.

And electricity will run out in one hour

So I abort the mission.

And I send up another couple of Poodles, see 7.1 for the launcher. Their mass will be negligible, unlike the cosine losses I’d have suffered without them

Too bad for the elliptic rendez—vous, it was a good idea.

No. No, it really wasn’t.

9.3) All aboard! Next stop, Duna

Spoiler

After sending the crew in a shuttle with less fuel, but with solar panels, I am ready to start. I need some more calculations for the timing, though.

The alexmoon planner says I have to make the burn 152 degrees from prograde. I could eyeball, but this transfer burn is important, everything I can do to improve its accuracy is welcome. I’m helped by the fact that the prograde-retrograde direction respective to the planet falls exactly on the terminator line.

Unfortunately, the line is not defined enough to use as a reference. You try figuring out where it falls exactly in this!

So I tried to use solar panel output. If I have the ship pointing prograde, when it passes over the terminator, the panels on the back of the sunshield will have 100% output.

Turns out that’s not an ideal answer too, because those solar panels are tolerant of slight misalignments, they keep 100% yield for almost a minute. But I can still fix this: I chronometer the time those panels will spend at 100% output, then when half that time has passed, I will know they are perfectly perpendicular to the sun.

Before implementing this plan, though, I discovered a simpler way.

The two drop tank trains are so narrow, they almost make up a pinhole gnomon!

With the sun equidistant from both tanks, and the ship holding prograde, I can confidently eyeball I'm 180° from prograde, with a margin of error of no more than a couple degrees

Ok, here I am 180° to prograde. I must center the burn at 152° to prograde. Wth an orbital time of 1916 s, angular sped 0.188 °/s, it’s 149 seconds from now. With the improved TWR of 0.5, it will take 220 s to perform the burn, I must start 110 second earlier. So, burn in 39 seconds, at 06:08:44.

I wanted to also include a screenshot for using the tool to determine when to shut down the engines, but what am I going to show? Paleodiastimoploio with engines shut down?

Finally I can discard the spent tanks

To install the extra Poodles, I was forced to keep a few tons of empty mass. This increased consumption by 4 or 5%. Taking cosine losses from lower TWR would have resulted in at least 20% increased consumption, though.

And radiation is 0%. The passage through the belts was so fast, the crew got less than 1% damage. Considering that every 1% damage means 20 or so less days that the crew can spend in space, that's even more important than saving fuel.

As for how the burn went...

Wow, it was incredibly accurate. So accurate, in fact, that the 50 m/s I included to account for inaccuracies made me overshoot Duna instead. Well, that, and hitting Duna at its periapsis; on average, my burn would have been flawless. I had to burn 80 m/s retrograde to reestablish a proper Hohmann trajectory. As for inclination, Kerbin and Duna are similar enough, there's no need for a plane change. Besides, I actually tried to fix inclinations, but without a functional tracking station I am limited to eyeballing, and I concluded it's not accurate enough to improve on my current orbit.

I am so proud of my work with a datasheet. When I go read my previous diary, and I see my encounter plans were something like "set an orbit to cross the target, and wait for a chance encounter", I marvel at how much I learn.

Paleodiastimoploio tanking a solar storm with no increased habitat radiation

I turned storm probability way down, because during a storm I must time warp at x1000, else the game glitches and irradiates my crew regardless of the sunshield. And those storms are very frequent, and last for seven hours each. They are just a huge bother. But turning them off completely felt wrong.

9.4) How I shoot myself in the foot

Spoiler

To calculate the encounter, I need to take the exact times at which Duna, then Paleodiastimoploio, will pass at the encounter point.

The passages. As well as a radiation report on Paleodiastimoploio

Duna passed at the intercept point in day 307.5, Paleodiastimoploio passed on the same spot in day 329.25 (more accuracy with the hours is pointless, as I can’t be sure I have the exact same spot anyway, and the method is robust enough to account for it). A difference of 21.75 days, or 87 Kerbin days. Duna has a orbital period of 801.5 days, to syncronize I must raise Paleodiastimoploio to a 714.5 days orbit. Entering that in the tool, I have to raise periapsis to 18870 Mm. Having intercepted Duna in its periapsis is working against me here. But the initial fuel calculation accounted for an even less efficient maneuver, I should be fine.

As for radiation, at the current rate I should get around 40% irradiation when reaching Duna, consistent with previous calculations that I should have time to bring the crew back, if barely. My higher tech kerbalism missions didn't have this problem because the active shield cancels background radiations, and decontamination units revert their damage.  But without those technologies, the small amount of background radiations will kill the crew in roughly 5 years.

Traveling is fine, just a couple of minor accidents.

A solar panel broke. I got 24 for a reason

The reaction wheel of the probe core broke. I didn't need it anyway

Why does it seem like I'm a bit too far behind the planet?

After a long orbit around the sun… I miss Duna by 4 days

I can barely catch a glimpse looking out of the window at maximum magnification

I am flabbergasted. All my tools, all my calculations up to this moment far exceeded any accuracy expectation, and now this. What went wrong? Duna actually got into the same spot after 800 Kerbin days, that part was correct. The timing I took for the passage was good.I used this same calculation to return from Jool, twice, and both times I got within 6000 km of Kerbin. I double-checked all tools and calculations, all is fine. Desperate to figure out what went wrong, I cheat all structures to level 3, and I find a discrepancy between my orbital time and the one calculated by the tool. But I used Duna to tune the tool, and it calculates its orbital time exactly…

I used the values in the wikia. Does this mean that the wikia has wrong data?

....

NOOOOOOOOOOOOOOOO!!!!!!!!!!!

The Duna page in the wikia, which I used to get its orbital time, reports the periapsis and apoapsis as distance from the center of Kerbol. Unthinkingly, I put those values into the equation, which was set to work with distance from the surface, as the game uses. And it returned a too-high value. Trying to figure out why it did, I came across the solar radius. If I remove a zero from the solar radius, the orbital time for Duna becomes correct. So I concluded that the solar radius was missprinted in the wikia, removed that zero, and used the tool.

I even considered that I used the same tool, with the same numbers, to return from Jool; but, lacking other explanations for why the orbital time of Duna returned by the tool was wrong, I concluded that coming from Jool, very high orbit, the solar radius must have been negligible on an orbit so long. In fact, I was so convinced my data was correct, I was about to post in the forum on how the game wiki has the solar radius wrong.

What actually happened is that without a zero, the solar radius becomes negligible enough, so the orbital time calculated for Duna becomes correct within a reasonable error margin; I didn't bother writing more than four significant digits for every number, I assumed any small inaccuracy to be caused by that. But this oversight removed 260000 km of radius for the spaceship, throwing off the calculations by 16 Kerbin days. Thus fixed, my tool reports I’d have needed a periapsis of 18400 Mm. At this point I’m curious, and I already got the level 3 structures, so I check… Yes, I get a very nice intercept with this value. I get a Duna encounter with periapsis between 18300 and 18450 Mm, so my calculation was pretty spot on. The tool worked perfectly. And the methodology is robust enough to account for an error of up to a couple days.

I did a lot of difficult calculations and I did them very well and I was so proud of myself. Except for the part where I calibrated the tool with the wrong data.

Now I know how those engineers that crashed a probe on Mars due to an error with units of measurement felt. Well, except the part where I crashed an actual probe that costed actual money and most likely sunk my actual career. But the whole "I am a dumbass and should have totally seen it", that's there. It's not an error of unit of measurement, rather an error of reference point, but just the same, it's a silly error on how your data is measured.

Sigh. Well, let’s reload and see if I can at least get the challenge done with one instance of reloading.

9.5) How it should have gone

Spoiler

Luckily, even if this was supposed to be a nanodiamond run - therefore, no save-reloading allowed - Paleodiastimoploio still has the nasty habit of occasionally exploding for no reason. This, in turn, gave me the habit of saving regularly, at least when that cursed mothership is in physical radius. So I could reload to the previous apoapsis, and make the correct burn, with a 18400 mM periapsis.

Then I waited another two Kerbin years. Malfunction time is determined randomly when a part is spawned, so the same parts broke at the same time.

I also got a stress-related malfunction that broke a Geiger counter. I brought two for one such event

This time, the encounter looks right

I can see Duna from the window, but this time I'm moving straight towards it

I got a 15 Mm periapsis, reasonably close

Inclination is also not too bad, I will fix it from apoapsis once I'm in orbit. Radiations are excellent, 38%; if the return trip takes as long as this one - and it should actually take a bit less, because I will be in a lower orbit - I should be good.

After some deliberation, I decide to go for aerobraking. I'm slow enough that I know even the fragile parts can take the heat. My main issue is that, the challenge including a "no reload" policy (well, except for catastrophic damage), I can't find the right periapsis by trial and error as I've always done. I know Paleodiastimoploio is extremely draggy due to its sunshield, and don't want to accidentally crash on the planet. I decide to play it safe; I've been aerobraking on Duna often enough, and I know that before 30 km of altitude, there's very little drag. A 35 km apoapsis will slow me very little; I may have to still spend some fuel to finish capture, but I'll still save something, and I won't risk braking too much.

I changed periapsis by pointing inward and burning. By the fuel gauge, I spent 150 m/s on it

One thing I did NOT think about was avoiding Ike. Fortunately, I narrowly miss it, or I would have been forced to make expensive corrective maneuvers

As I take science, I start discarding science modules. Those are the main sources of spontaneous explosions, I can't wait to get rid of them

I exited with a 3000 km apoapsis, which works wonderfully to fix incliation

Having experimentally determined that a passage with a 40 km periapsis will lose some 20 m/s, I slowly aerobrake the rest of the way to a near circular orbit.

The original plan called for arriving at Duna with 95 tons of vehicle left, and I have 115, having saved 20 tons of precious fuel. They will make the return trip easier. Radiation levels are as expected. Food consumption was a bit bungled, I should have added some more. If I had a stress accident where I lost 10% of my food, it could jeopardize the mission. Anyway, I still should have enough to return, even if one such accident occur.

As for stress itself, the crew is holding up surprisingly well, considering that they miss many of the amenities unlocked by higher tech, like artificial gravity, plants, a cupola, the ability to call home; then again, my multi-decade missions gave me a bit of a skewed perspective on how much of a problem stress is. Water levels are very good, I ended up not needing the Sabatier reaction - I could have saved the weight and part count.

Next, exploration begins.

Edited August 28 by king of nowhere

[king of nowhere]

Part 10: The alleged rover

The Hamburger rover is used to explore Duna and Ike. It suffers from a lot of shortcomings, and gets bugged too.

No, I'm not unhappy with the design. I worked under the constraints of caveman and kerbalism, I did what I could with what I had. It's already a small miracle that I managed to bring a rover to Duna in the first place.

IVA view from Duna. Unfortunately, I need to better see the navball, so driving in this fist person perspective wasn't really viable

10.1) On Duna

Spoiler

Before circularizing Paleodiastimoploio, I detach Hamburger. Before it's ready for landing, I need it needs to grab all the drop tanks, the parachutes and the science modules.

It results in quite a bit of maneuvering around

The final result is not very pretty, those science modules are rather cumbersome. But I will drop one in the lower atmosphere and one immediately after landing

The parachutes are being surprisingly effective

Landed in the middle of the central sea

In retrospect, dropping the rover like that was't a good idea. I saved a bit of fuel, but I had no control over where I landed. I ended up perhaps in the worst place possible to explore other biomes. Not only the middle of central sea is rather far from other biome boundaries, but it's also one of the lower elevations of Duna. I will have to go uphill, with a rover that's not really designed for it.

That's far from the only problem I have. Before leaving Kerbin, I gave higher priority to the large food stores on Paleodiastimoploio, because I wanted them to be used first, and the smaller containers to be reserved for the rover. Well, turns out priority only works with fuel, not with regular eating. Those food containers are already half empty. I still have enough for 40 Kerbin days, though, so food is not going to be an issue. It could be on reentry, when I will need to get into the pods and the storages will be mostly empty.

But there is an even worse problem. See the pic with the parachutes; one of the engines is stated as having no fuel. Though the fuel should be transfered through all the ship. But during the trip, Hamburger was struck by a bug. The craft file became scrambled, and now there's no communication between the central pod and the four arms. I had a similar issue with the DREAM BIG, where half the ship got disconnected from the other half and could not transfer fuel, even though the tanks were adjacent. Luckily, for now I have fuel for all engines, it will be a problem for later.

Since it's a bug, I could cheat in a replacement, but I won't for two reasons. First, Hamburger needed to be assembled on the launchpad due to part count limitation. Or upgrading the VAB, even though I'm not going to need it again. I'm not going to do any of that as long as the rover remains somewhat usable. Second, it would reset the counter on engine usage.

And engine usage is a much bigger issue than I anticipated. I figured, each engine has 25 allowed ignitions, and I can alternate the lateral ones in pairs and the central one. I only need full power to ascend from Duna, otherwise I can work the engines in groups and achieve over seventy ignitions, enough to do anything. But I didn't consider how taxing all the docking and undocking to grab the drop tanks and science kits would be. That alone ate dozens of ignitions. I spent some of the allowed ignitions already upon launch from Kerbin, as Hamburger reached Paleodiastimoploio with its own engine. Now I'm running short on ignitions, and the bug is preventing me from switching the engines as I'd need to save uses. I just hope none of them malfunctions prematurely.

Moving uphill

The closest biome boundary has midlands and highlands some 40 km to the northwest, but I need to reach 5 km of altitude. Hamburger really doesn't have much power, those wheels are puny. I used to consider them so strong in my first rover, but that was less than 2 tons. It didn't need to take off from Duna, or to bring heavy science instruments around. Each of those science modules is 290 kg.

In those conditions, Hamburger can only climb up to 5° slopes. The drive required hours, even at 2x warp. It entailed a lot of zig-zagging to keep going 5° upslope.

As a further complication, Hamburger also doesn't have much power generation on its own. I'm limited by the tiny solar panels, and the need to strap a lot of drop tanks and science modules around the rover, blocking the sunlight. Anticipating this, I planned the rover to use fuel cells to drive, and take pauses to recharge the hydrogen storage. Those pauses took upwards to one Duna day. I had two of them. In the end it took 30+ hours for a trip that Leaping Mantis could have accomplished in twenty minutes.

At least there's an advantage. All my cirvumnavigations with rovers, I had to stop and save frequently, because I'd crash the rover regularly. I'm not allowed that in nanodiamond, but it won't be necessary. Those puny Rovemax S2 can't go fast enough to risk any damage. Worst case I'll break wheels, and will be forced to abort the ground exploration.

Yeah, righ

Ok, I have to reload again. I had to reload multiple times while driving. Duna is surprisingly treacherous, with sharp angles jutting out unexpectedly, and masked by the uniform coloring.I already failed at the strict "no reload" policy when I missed the intercept earlier, no big deal.

Finally, I visited some more biomes and shed some science modules

So far, I went northwest some 50 km. Next biome is maybe 100 km to the west. It would require going downhill most of the way, back into the central sea, then climbing up again. Conveniently, it would also be a mountain; Hamburger has terrible drag, I will want to launch it from a mountain.

However, I realized I don't want to drive such a long road, going up and down multiple times, just to get another biome. Besides, I landed in this same exact location (close enough, just a couple dozen km to the west; meaning Hamburger would have to face an even longer trek) during the Marco Polonium mission. And I did search for a close mountain on the equator, because the small lander had the same drag problem. And it was a very long and uncomfortable road, and at the end was mount Inadequate, thus called because it was only 4000 meters, inadequate for my purposes. Here I am at 5000 meters of elevation, only 6° of latitude, probably fixing inclination won't be any more expensive than launching from a lower elevation.

So I left Duna.

Drag is an issue even in this thin atmosphere

Drag is such an issue, at 200 m/s I had to throttle down and go straight up, least drag overcomes gravity drag.The drop tanks contain a generous surplus of fuel to account for this.

In one picture I show Hamburger back in orbit, and the path taken. For some reason, the discarded science modules are counted as rovers

Rejoining Paleodiastimoploio

10.2) On Ike

Spoiler

I put down the spent parachute pack - I don't discard it yet, it contains useful food - and get one of the small tanks for the service probes. This allows using the central engine, to save ingitions on the lateral ones. I grab a new lander kit - I want the extra food in case I fumble and get stuck in a weird orbit - together with the remaining drop tanks and science modules. One of the science modules was on the bottom of a pile, and required moving around other stuff. All in all, I used up lots of ignitions that way.

 

Halfway through the process

Hamburger ready for Ike

Now another connection broke, the upper engine can't access fuel

I checked that fuel transfer is allowed. No, according to kml editor, there is no connectivity between parts. I was already stuck with 2 usable engines out of 5, and that was still enough to handle Ike. But with only the lateral engine working, I can't land anymore. This time I need some creative debugging.

I cheat in place an engineer, as I can't take EVA in caveman, and install the engines directly attached to the fuel tanks

This for now fixes the engines. Though I will need to be very careful, I only have 7 ignitions left

I intercept Ike the same way as I intercept Mun. Flawless, also this time

Wait a moment, why I have less than 1000 m/s available despite using an additional tank? I really only had two drop tanks for Ike? Who is the moron who thought two drop tanks would be enough to reach Ike from low Duna orbit, land, and return?

Of course, it's me. Landing on Ike is easy, so I didn't calculate or test anything. I just figured 1000 m/s would be enough. You can always count on me to mess up the really simple things. Well, as long as I can return to orbit - and I surely can, after I get rid of all the extra mass of the science modules - I can send a Service Probe to carry Jeb back.

On the plus side, by landing directly from the intercept trajectory, I was able to use one single ignition for the whole maneuver.

IVA view from Ike

Notice how it's markedly different from the view on Duna, when Val was driving. It's like this view is taken from a higher perspective, I see less of the sky and more of the ground. Does the game actually takes into account different height from different crew members? Maybe I will test on it.

On Ike I also set to drive until I find two new biomes; I landed near a boundary, so the first biome is easily reached, but the second required a 40 km hike.

This speed is quite dangerous, and I must brake immediately

On Ike, the lower gravity - and lower mass, due to less fuel - changes completely the driving experience. I could go in a straight line, and I didn't even need the fuel cells, solar panels provided enough power. I still occasionally broke the rover, some slopes are fairly steep and Hamburger is pretty crappy. Being unable to use hotkeys forced me to turn the reaction wheels to only SAS mode, meaning it's much harder to recover if the rover starts capsizing. Anyway, mostly an unremarkable trip.

Once I discard the science containers, I gain 100 m/s. I gain some more by discarding the survival kit, food be damned.

I gain some more by accelerating with the wheels as much as I can before activating the rockets

By the way, notice how the engine is no longer perfectly centered on the docking port. Due to the unnatural way the parts were attached, that engine got twisted around at some point when I loaded a game. Now I also have to deal with uneven thrust! At least all those extra reaction wheels I broke to cover for malfunctions can stabilize, mostly.

The available deltaV is barely enough to get away from Ike...

...to a suborbital Duna trajectory

I spend half my remainig deltaV to raise periapsis to 43 km, and slowly aerobrake. I wanted to be extra sure the 18 m/s I had left would be enough to raise Hamburger in a stable orbit.

After aerobraking, I ended in an orbit close enough to Paleodiastimoploio

I also run out of fuel in one of the tanks. When I discovered that one engine had lost connection with its tank? That moment, one tank spent more fuel than the other. So now one of the tanks is completely spent, and I have very asymmetric thrust. But I am in space, I don't need big thrust.

All in all, using Hamburger was a miserable experience. But I'm not faulting the poor rover. It had to fit into the 30 part limit, while including the kerbalism parts, which can't be moved by EVA construction. And then the whole rover, which was an unprecedented aerodynamic nightmare even for the standards of this mission, had to be launched to orbit with the limitations of caveman. And then the kraken raged against the poor rover. The only improvement I could have made would have been a single RCS propeller, to spare ignitions during all the docking/undocking maneuvers.

Anyway, from here I can rescue Jeb with a Service Probe.

The Service Probe grabs Hamburger. Notice the lack of power generation

Those Service Probes have a small tank for 250 m/s, plus an exchangeable tank raising it to 700 m/s. More than adequate for their needs. Their engines are rated for over 100 ignitions.

On the downside, I didn't include a science container, so I have to carry the heavy Hamburger all the way back to Paleodiastimoploio to recover the science. And they don't have solar panels.

Wait, the lack of solar panels is not a bug, but a feature. I already noticed in the past, Service Probes are only active for short time intervals. They don't need continuous power generation. I gave them some generous batteries instead. Solar panels malfunction, batteries don't, so I picked the reliable options. I only have a few instances to use those ships anyway: first, when returning to Kerbin, Paleodiastimoploio will enter orbit on remote control, while the probes will take a slight deviation to land directly with the crew, sparing it from having to cross the radiation belts. The probes will be on their own only for a few hours at most, and the battery covers that. Or they will be needed to move stuff around Paleodiastimoploio, again in this case the battery is enough.

Or they can be tasked with recovering a troubled Hamburger as far as Ike. For this they need solar panels. it's definitely an oversight on my part. But the battery did last long enough to rescue Jeb, even though I had to shut down the CO₂ scrubber to save some electricity for the reaction wheels. Good thing I could at least leave Ike, or I don't know if a rescue mission would have been possible without moving the whole mothership.

I brought everything back to Paleodiastimoploio, I got rid of the bug-riddled rover, and moved around the remaining load in a more symmetric position.

Status after spending 20 Kerbin days around Duna

I have plenty of fuel, thanks to having saved on the outbound trip. I'd like some more food, but it's sufficient. There's only a few part malfunctioning. Even crew morale is holding, Val got a bit stressed by spending a long time in Hamburger's cramped confines on Duna. Time to go back. After spending so many weeks preparing this mission, it's playing out in a few days.

 

Edited August 30 by king of nowhere

[king of nowhere]

Part 11: It's over already

Paleodiastimoploio returns to Kerbin without particular problems.

Recovering the science

11.1) Return to Kerbin

Spoiler

To get the best outcome for the return trip, I must repeat the same process I used when leaving Kerbin.This time I won’t have all the tank train to act as a gnomon, I will have to measure solar panel efficiency. As I adapt the tool for the final speed to Duna, I also discover I made not one, but two mistakes. First, I gave Kerbin a radius of 300 km, not 600. Second, I forgot to include the 1/2 factor in the formula for kinetic energy - rather, I forgot to carry that factor when I inverted the formula. Fortunately, the first error overestimates how much speed should decrease with distance, and the second error underestimates how much speed should decrease with distance, and taken together they sort of canceled each other, at least for small differences in altitude, hence my first transfer was still reasonably correct. Now I fixed the tool, anyway.

The Duna-Kerbin window was one week ago. Going now would require lowering solar periapsis below Kerbin’s orbit. Else, I could wait a bit more around Duna and aim at making two orbits around the sun before catching Kerbin. I probably have enough radiation tolerance left to do this, but I absolutely have enough fuel left to do the former, so I’ll take it as the safer option.

By the alexmoon tool, in a 100x100 orbit I will need a 670 m/s burn at an angle 120° from retrograde. With the single engine I have, it will take 511 s for the burn. But Duna does not have radiation belts, I can split the burn, first 200 s then finish.

So, from the moment I am 180° to retrograde, my orbit being 185x51 with a time of 3316 s, burn must be centered 553 s after 180° to retrograde. I will start 453 s after that event. The solar panels remained at 100% efficiency for two full minutes, so I am 180° to retrograde after one minute of having the solar panels at 100%. I will start the burn 513 s after the solar panels reach 100% efficiency, for 200 s, in the apoapsis raising.

At 10:54:42 then, stopping at 10:58:02

All this does not consider than I’m in an elliptical orbit, so I’ll surely need some correction. I could instead circularize my orbit, I doubt it would be cheaper. In my current orbit I may have to spend more in course correction, but to circularize I would have to spend more for sure.

The apoapsis raising burn was interrupted by the docking ports on the drop tanks refusing to undock, which forced me to use KML editor again. At least this time I was prepared for the whole fuel priority getting scrambled.

End of transfer burn, looks good

The Cheetah engine I'm using is the same that pushed Paleodiastimoploio away from Kerbin. It's overshoot its allowed burn time, but it keeps working. Still, I no longer need all those heavy engines

The transfer burn was spotless!

I am once again surprised by how accurate those burns can be while done blindly, with the only help of an excel datasheet. Here Paleodiastimoploio has a periapsis slightly lower than Kerbin, which would be great normally, as crossing the orbit increases the useful time to get an intercept. But in this case I will lower solar periapsis below Kerbin, so I need to raise that periapsis a bit, to ensure I'll still cross Kerbin's orbit afterwards. I only spent 20 m/s on the correction burn.

Those maneuvers are so accurate, in fact, that I'm starting to consider doing a caveman grand tour. Then I remember how many drop tanks I had to bring for Navis Sideralis Neanderthalensis, and consider that I would have to bring many more. I further consider how big was Bolt, who made a grand tour on drop tanks alone. And how unstable a ship made of dozens of detachable modules is, and how often Paleodiastimoploio is exploding. And how those problems would only be compounded on an even bigger ship.

But who knows, maybe I'll do it one day.

A couple more parts malfunctioned. Another solar panel will follow later. Still, I always overshoot the needs for redundancy

First crewmember reaches 50% radiation damage. I'm well into the return trip, so I will be fine

The two images taken to time the syncronizing orbit

I missed Kerbin by 50 hours, so I will require a syncronizing orbit of 418 Kerbin days. Periapsis at 12960 Mm.

Status after that burn.  Paleodiastimoploio is still in excellent shape. It's running out of both food and radiation tolerance, but it's got plenty to return

Turns out waiting another year to get a Kerbin encounter would have been possible with the current food stores, if barely. Also with the current radiation damage, I'm taking about 17% for every Kerbin year. I definitely could have carried a bit more food.

First sighting of Kerbin

Captured by Kerbin

I forgot to take a picture, but I entered Kerbin's SoI in a 4500x90000 km orbit. I also had roughly 60° of inclination, hard to see in the picture, which is good because it gives me the certainty of missing Mun. 90000 km apoapsis is outside of Kerbin's SoI, this happens when intercept speed is very low. I spent 120 m/s (measured by the fuel gauge) to bring that periapsis to 90 km, and in the process of burning retrograde I also completed capture.

11.2) Recovering crew and science

Spoiler

The crew has enough radiation tolerance left that I could keep them onboard while aerobraking, but I want to use the systems I worked hard to design. So, I move everyone into the Service Probes, which double as escape pods.

Alignment bug strikes again. Commonly associated with kerbalism. Won't interfere too much. I hope.

One problem I didn't consider: I need to deorbit two ships at once. I can't leave one in orbit, because they don't have solar panels. When I deorbited three ships simultaeously at the end of the A'Tuin mission, I distanced them 20 minutes from each other to have the time to follow each one of them. But here I can't plan maneuvers. Only one thing to do, dock the escape pods to each other.

Also, those pods lack a science container. I will have to send up another flight to recover the science later.

The docking, off-center because of the alignment bug. Also, final radiation level. I had one more year and a half I could spend into space

Reentry. I was pretty confident about those pods being able to take the heat (while the fuel tanks and batteries exploded), but I don't think I actually tested them

It pains me to admit, because in all my mission my Service Probes have maintained an excellent record of perfect functionality, but in this case they were pretty crappy. Actually, they worked as service probes, but in their triple role as escape pods and rescue vehicles they were severely lacking.

I only detached the pods when it was time to deploy the parachutes

The first time, upon undocking, both pods exploded immediately. Looks like a consequence of the misalignment bug. I had to reload all the way back to undocking from Paleodiastimoploio, and that also fixed the alignment bug. It's not a given that reloading will fix that, or I'd have done it sooner.

Meanwhile, I take advantage of Paleodiastimoploio being in a high orbit to fix inclination. I was hoping all the parts I undocked would ease the structural tension, but it still explodes occasionally.

Paleodiastimoploio got captured by Mun. Can't be lucky every time

But it wasn't an issue, because I have a crapton of fuel left. 2500 m/s. How did I use up so little fuel? Apparently, I prudentially calculated 2500 m/s both ways, but maneuvers were very precise, I saved some 500 to 700 m/s both times. The rest is emergency fuel in the sunshield.

I set Paleoadiastimoploio in a nice equatorial orbit, I gradually aerobrake most of the way, and finally park it in a 90x90 orbit.

Sending a new ship to recover the science

Which is enough to finish the caveman challenge

But not by a crazy amount. I thought bringing science kits and a rover was a fancy luxury because a simple landing was uninspired, but they turned out to be key to finishing the challenge. As it is, I have roughly 150 science points I could still recover from Mun, so bringing science kits to Duna was definitely necessary.

I am ready, again, to say goodbye to this game.

Until the next time I will think of a nice challenge to try.

Edited September 1 by king of nowhere