totm january 2020 Community Caveman Jool 5 mission

[ManEatingApe]

The goal of this challenge is for multiple players to collaboratively plan, build and fly a Jool-5 mission.
The catch? We'll be attempting it Caveman style - meaning Tier 1 buildings, 30 part limit, 18t weight limit, primitive tech, no maneuver nodes, no patched conics, no resource transfer, no EVA and no mods (except visual & audio).

Caveman challenge and Jool-5 challenge threads linked for guidance and inspiration.

A lot of patience, practice and ingenuity will be required to pull off the tricky ground and orbital assembly necessary to construct a craft capable of the mission, to say nothing of actually flying the contraption to the moons of Jool and back.

Anyone is welcome to make any contribution small or large to the effort.

Fair warning: In order to succeed it will take considerable planning, orchestration and team work. If you're more of a lone wolf or lean towards instant gratification then this is not the challenge for you. On the other hand if you are in (or aspire to) project management or middle management then this may be right up your alley (really selling it here I know...)

 

Download shared save game from here

The shared save game is available as a public GitHub repository. To contribute fork the repo, then open a pull request.
I haven't quite worked out how to serialize access during mother ship assembly, suggestions welcome.

To help keep everything on track the mission will take part in 4 phases, details of each phase in the respective spoiler.

[DONE] Phase 1: Mission Definition

Spoiler

Objectives:

• Agree the rules of the challenge
• Agree contribution model

Rules:

1. Vanilla KSP 1.7.3, no DLC
2. Normal difficulty, CommNet off
3. Jool-5 challenge rules apply
4. Caveman challenge rules apply, except:
   • Testing using HyperEdit/debug menu is OK (but obviously no shenanigans during mission itself)
   • Adding funds to the save using debug menu is OK (effectively sandbox mode except for Tier 1 buildings)

Contribution Model:

1. Public GitHub repo

[DONE] Phase 2: Mission Planning

Spoiler

Objectives:

• Agree mission profile
• Order of moon landings
• Location of parking orbit of mothership in Jool system
• Agree total number and approximate lander designs
• Agree approximate mothership design
• Agree return vessel design

Decisions:

• Single Kerbal on a ladder for the outbound journey and moon landings
• Re-use same lander design for Tylo, Vall, Bop & Pol

Mission plan:

1. Launch Mothership core (MIKE) to LKO
2. Circularize MIKE orbit to 80km x 80km
3. Launch 7 × Fuel tenders (FOXTROT) then dock with MIKE
4. Launch Tylo lander (TANGO) then dock with MIKE
5. Launch Bop & Pol lander (BRAVO) then dock with MIKE
6. Launch Vall lander (VICTOR) then dock with MIKE
7. Launch KILOLIMA extra fuel then dock with MIKE
8. Launch Laythe ascent vehicle (LIMA) then dock with MIKE
9. Launch Kerbin return vehicle (KILO) then dock with MIKE
10. Launch Crew delivery vehicle (CHARLIE) then rendevous with MIKE. Crew jetpacks to MIKE and grabs ladder.
11. MIKE ejection burn from Kerbin to Jool via Hohmann transfer
12. MIKE capture into high elliptical Jool orbit (burning as close to Jool as possible to maximise Oberth effect)
13. MIKE fix Jool inclination to 0° then separate landers.
14. TANGO+BRAVO+Kerbal adjust Jool PE to match Tylo then capture into low Tylo orbit (burning as close to Tylo as possible to maximise Oberth effect)
15. LIMA+KILO adjust Jool PE to match Laythe then capture into low Laythe orbit (aero-braking to conserve dV if possible)
16. VICTOR adjust Jool PE to match Vall then capture into low Vall orbit
17. TANGO land on Tylo then return to orbit
18. TANGO rendevous in low Tylo orbit with BRAVO
19. BRAVO lands on Bop and Pol
20. BRAVO rendevouses with VICTOR in low Vall orbit
21. VICTOR land on Vall then return to orbit
22. (a) VICTOR goes to low Laythe orbit (b) Jeb lands on Laythe
23. LIMA rescue Kerbal then return to orbit
24. KILO eject directly from low Laythe orbit (to maximise Oberth effect and Laythe's orbital velocity boost)
25. (a) KILO intercept with Kerbin (b) KILO Fairing/decouple device used to protect valiant Kerbal from re-entry
26. Jeb uses personal parachute to glide to a safe landing (bonus style points for landing at KSC)

[DONE] Phase 3: Mothership Assembly

Spoiler

Objective:

• Participants take turns adding components to the mothership in LKO
• Share knowledge on lander design and assembly tips & tricks

Notes:

• Ladder orientation (parallel or perpendicular) makes a difference to aerodynamic drag.
• Adjust craft naming priority in the VAB in order to keep the mothership name the same during orbital assembly.

[DONE] Phase 4: Mission Execution

Spoiler

Objective:

• Break the mission into several parts.
• Decide who will fly each part
• Actually fly the mission!

Participant Badge:

Photos:

Orbital Assembly

@ManEatingApe Craft design, orbital assembly and Vall mission https://imgur.com/a/gcxb6rz

@ManEatingApe Orbital assembly, docking TANGO with MIKE https://vimeo.com/360111385

@IncongruousGoat KILO launch: https://imgur.com/a/hU6aFug

@dvader 2xFOXTROT Launches https://imgur.com/a/NfgvDOH

 

Jool 5 Mission

@Muetdhiver Bringing Jeb to orbit and transfer from Kerbin to Jool https://youtu.be/vonjAX5r3_0

@dvader VICTOR KILO LIMA Setup https://imgur.com/a/btKYyBe

@Muetdhiver Bop & Pol landing + Vall transfer https://youtu.be/9UKmKqCkktw

@IncongruousGoat Tylo landing: https://imgur.com/a/N0b0kz9

@dvader LIMA - Laythe Landing and Rescue

Screenshots https://imgur.com/a/uZsrkKi
Video https://youtu.be/DV9TxOlBg2o

@IncongruousGoat Jool->Kerbin transfer: https://imgur.com/a/Gv76004

@ManEatingApe Landing on Kerbin https://vimeo.com/380077525

 

Statistics

[GAME]
Unique craft designs: 11
Launches: 24
Total cost: 264,523
Net profit: 3,636,056
Mission duration: 24 years, 317 days

[REAL LIFE]
Contributors: 7
Mission related posts: 367
Mission related DMs: 68
Duration: 4 months, 8 days

[GITHUB]
Commits: 65
Tags: 30
Pull requests: 27

 

Edited January 7, 2020 by ManEatingApe

[JAFO]

1 hour ago, ManEatingApe said:

The catch? We'll be attempting it Caveman style - meaning Tier 1 buildings, 30 part limit, 18t weight limit, primitive tech, no maneuver nodes, no patched conics, no resource transfer, and no EVA.

You forgot to mention the restrictions on mods that affect gameplay and/or make things easier... for the benefit of readers who haven't tried Caveman before, it's probably worth clarifying those rules as well.

[ManEatingApe]

16 minutes ago, JAFO said:

You forgot to mention the restrictions on mods that affect gameplay and/or make things easier... for the benefit of readers who haven't tried Caveman before, it's probably worth clarifying those rules as well.

Thanks for the reminder - rules updated!

@Muetdhiver @IncongruousGoat @dvader Regarding rules, I'm suggesting:

• Union of Caveman and Jool-5 rules
  • That would mean no testing craft on Laythe or Tylo using debug menu. We'd have to do our best on Kerbin and accept the risk.
  • Raising funds by completing contracts. A little grindy but higher kudos. (We could re-visit if things get too grindy)
• Commnet disabled (to allow probe controlled Kerbal-on-a-ladder designs)
• Allow MH and BG (this would unlock potential cranes and construction equipment on Kerbin and prop designs for Laythe) but no problem if folks have objections to DLC

 

Let me know what you think.

Edited August 9, 2019 by ManEatingApe

[Jacke]

I've come no where near Jool and my Caveman campaign (aiming for Nanocrystaline Diamond with minimum science) didn't get out of LKO.  I'm also very rusty in KSP vessel design and can't really help at these levels.  But I am interested in following along.  And I do have a few observations.

You may want to make a base Caveman savegame to start from.  Perhaps with all the tech tree filled out already to unlock all parts.

Do try to design both without and with the DLC parts.

Once you get some polished craft, could you share lots of screenshots and perhaps share them on Kerbalx.

[ManEatingApe]

48 minutes ago, Jacke said:

You may want to make a base Caveman savegame to start from.  Perhaps with all the tech tree filled out already to unlock all parts.

Absolutely, once the rules are solidified, I'll create a base game and we can take it from there. Updated the OP to make this clear.

48 minutes ago, Jacke said:

I've come no where near Jool and my Caveman campaign (aiming for Nanocrystaline Diamond with minimum science) didn't get out of LKO.  I'm also very rusty in KSP vessel design and can't really help at these levels.  But I am interested in following along.  And I do have a few observations.

There's lots to contribute before even leaving LKO (in fact probably the bulk of the work). So plenty of opportunity to help out around Kerbin if you like!

Edited August 9, 2019 by ManEatingApe

[Muetdhiver]

Indeed. Experience with on pad assembly, for example would be really neat. (I can do some half backed stuff, but it was not very good TBH XD)

Also, we will need a way to navigate the jool system in our terms, i.e. avoiding crapy encounters.

 

[Mr. Peabody]

@ManEatingApe, I must commend you on the the excellent and entirely thorough thread that you have created. Great work! This thread certainly has my endorsement.

Edited August 9, 2019 by The Dunatian

[IncongruousGoat]

6 hours ago, ManEatingApe said:

Union of Caveman and Jool-5 rules

• That would mean no testing craft on Laythe or Tylo using debug menu. We'd have to do our best on Kerbin and accept the risk.
• Raising funds by completing contracts. A little grindy but higher kudos. (We could re-visit if things get too grindy)

This one I'm going to have to disagree with. The second bullet point just adds grind (farming survey contracts is not exactly hard), and I (and probably @Muetdhiver) have broken the first one already. Plus, I like being able to test things. If this were being done as part of a proper caveman run I would agree, but it's not, and the nature of this challenge means that we're going to need to build some down-to-the-wire landers, i.e. the kinds of things that need lots of testing. Also, I've lost a Jool 5 before to a bad Tylo lander design and I don't care to repeat the experience.

6 hours ago, ManEatingApe said:

Commnet disabled (to allow probe controlled Kerbal-on-a-ladder designs

Given that I'm the one who proposed this in the first place, I totally agree.

6 hours ago, ManEatingApe said:

Allow MH and BG (this would unlock potential cranes and construction equipment on Kerbin and prop designs for Laythe) but no problem if folks have objections to DLC

I'm going to wait and see what everyone else thinks for this one. The problem here is that I don't own either DLC, and while I'm not entirely against picking up one (or maybe both, though I can't see how MH would help here) for this, this would probably be the only time I got any use out of it. If there's popular demand for BG, I'll go with it, but for now I'm going to keep planning and testing in stock.

 

As for the other rules (the ones that currently lack suggestions), my recommendations are:

• KSP version 1.7.3 throughout. Less chance of the save breaking, less trouble updating things, easier to co-ordinate, and makes it easier to judge the attempt if 1.7.4/1.8 change something in a weird way.
• Follow Jool-5 rules. The do-the-trip-in-a-pod rule was removed, and with that out of the way it's not hard to follow the rules, with all the restrictions we're working under.
• Difficulty setting normal, other than CommNet. Quicksave/quickload are going to be vital, and re-entry heating above 100% will make Laythe substantially more difficult than it already is.

[dvader]

20 hours ago, IncongruousGoat said:

This one I'm going to have to disagree with. The second bullet point just adds grind (farming survey contracts is not exactly hard), and I (and probably @Muetdhiver) have broken the first one already. Plus, I like being able to test things. If this were being done as part of a proper caveman run I would agree, but it's not, and the nature of this challenge means that we're going to need to build some down-to-the-wire landers, i.e. the kinds of things that need lots of testing. Also, I've lost a Jool 5 before to a bad Tylo lander design and I don't care to repeat the experience.

Since it is a community effort it doesn't really fit within the rules of either Caveman or Jool-5 anyway so I personally don't mind if we relax the "no cheat menu, not even in sandbox" condition. It doesn't really change the difficulty since the tricky part is planning, building and flying the crafts with a Kerbal on a ladder. 

Also, I do remember that it was harder to keep Bob on the ladder around Mun than around Minmus for some mysterious reason (he kept drifting off). I guess there might be similar issues or worse around Tylo, Vall or Laythe so test flying the missions in place is advisable since the physics engine can cause unexpected behaviour not seen around other bodies.

As for Funds and raising cash, I'm fine either way. It is not too much work but it is also not the important part. If someone wants to run through the missions it would be nice for the feeling of completion but it isn't necessary if no one wants to do it (I can't, at least not for the next week).

By the way, what about "no copying of craft files"?

[IncongruousGoat]

1 minute ago, dvader said:

By the way, what about "no copying of craft files"?

Either way is fine. Nothing we're building is going to be over 30 parts anyways, so remaking lander designs and whatnot isn't a huge burden.

[ManEatingApe]

@IncongruousGoat @dvader Thanks for the feedback, I'm happy to accept your suggestions.

The proposed rules would then be:

• Vanilla KSP 1.7.3, no DLC
• Jool-5 challenge rules apply
• Caveman challenge rules apply, except:
  • Testing using HyperEdit/debug menu is OK (but obviously no shenanigans during mission itself)
  • Adding funds to the save using debug menu is OK (effectively sandbox mode except for Tier 1 buildings)
• Normal difficulty, CommNet off

I'll leave this proposal stand until Monday August 12, if I don't hear any objections/suggestions by then we'll proceed with these rules.
 

49 minutes ago, dvader said:

By the way, what about "no copying of craft files"?

This touches on what I wanted to resolve next - contribution model. My initial suggestion is this:

• Create a new career mode save game, using the debug menu to unlock the tech tree up to and including level 5 then adding plenty of funds, say 10M.
• Create an initial savefile numbered "0"
• Upload the savegame to DropBox and include the link in the OP
• Each time a player makes a contribution, they add a new save incrementing the number by one e.g. "1.sfs", "2.sfs" and so on.
  (this preserves the history in the savegame in case we need to rollback)
• Players then send me the latest file and I update the DropBox link in the OP
• Players can include their craft files in the save game - this is a community effort, so I'd encourage sharing each other's designs.
• Craft files would be prefixed with their creator's username to allow easy sorting
  (e.g my craft files would be "mea - foo.craft", "mea - bar.craft")
   

Some other thoughts:

• How do people feel about GitHub instead of DropBox - we could submit contributions as PRs instead
• We need a process to serialize access when building the mothership to avoid conflicts
  e.g imagine the following scenario
  • ManEatingGoat and IncongruousApe are both working on version 5 of the save file.
  • ManEatingGoat adds a part to the mothership and saves it as version 6
  • IncongruousApe adds a part to the mothership and also saves it as version 6
  • Now we have a tricky problem!

 

Edited August 10, 2019 by ManEatingApe

[IncongruousGoat]

1 minute ago, ManEatingApe said:

• How do people feel about GitHub instead of DropBox - we could submit contributions as PRs instead

Good idea! Distributed version control is what we need, and distributed version control is what Git/GitHub are for. We can just make a private GitHub repo, hook everyone's GitHub accounts up to it, and we'll be off to the races. Assuming we can get everyone else on board with this plan, of course.

[Muetdhiver]

I have made bit of work on a return vehicle based on the Laythe lander from IncongruousGoat.

1) I fixed the ascent stability and found a nice way to boost overall DV from the launch. Simply put, my craft was too light and made the rockets to top over. Which is a classical problem. What I did is put an FLT400 on top with the nose cone, which is then ditched in space.

2) The lower stage plus top tank is nearly enough (or likely enough with a really good turn) to get in orbit. In short : the craft gets in LKO with a fully filled FLT400, which is neat.

3) The capsule does just fine on rentry, tested with AP = 1000km, kerbal was fine. I assume we'll do a personal chute landing soviet style to land => we'll have to level up the dude.

Spoiler

 

Also, I had a look for the mother ship.

If we can keep the overall landers + return stage masses at 9tons, a transfer stage like this would work :

Spoiler

The poodle + tank is for sure possible with on pad assembly. maybe even a bigger tank. The four drop tanks can each be done with a regular lauch, which is not too crazy. The two FLT800 are the payload mass (9tons).

It has TWR of ~1 which makes ejection not too painfull.

2400 DV can be pushed a tad by droping the side tanks first. That's enough for ejection and an eliptical jool capture. What do you guys thing about something like this ?

 

If anyone has a idea to minimize kerbal drif with warp, it would be a godsend !

Edited August 11, 2019 by Muetdhiver

[ManEatingApe]

2 hours ago, Muetdhiver said:

3) The capsule does just fine on rentry, tested with AP = 1000km, kerbal was fine. I assume we'll do a personal chute landing soviet style to land => we'll have to level up the dude.

Neat, nice work testing the Kerbal atmospheric re-entry system. Unlike other challenges just getting the parts to orbit can be tricky!

2 hours ago, Muetdhiver said:

Also, I had a look for the mother ship.

If we can keep the overall landers + return stage masses at 9tons, a transfer stage like this would work...

This is a good start, the Poodle is a good fit for the mothership. I think the best plan is to design the landers first, the mission profile second then work backwards from there as these choices will affect the requirements for the mothership.


Any suggestions/observations for the rules/contribution model?

[Muetdhiver]

For the rules I think no DLC is a good idea since not everyone has them.

github is also a nice idea and the rules generally suit me. I would advise againt grindy stuff, this is not the goal here.

One thing that we might want to consider (if enough are interested) is doing some streaming for stuff like transfer/capture and then the different landing. It would be nice, even if it'll just be watchin and chating

As for the mothership, I tend to guestimate the landers mass & stuff, eyeball the hardware and then iterate.

 

[IncongruousGoat]

So, I've been doing some investigations into Kerbal drift physics. First, some results on multi-orbit drift (experiments conducted in LKO):

1. Kerbals start moving radial out w.r.t. the ship, regardless of which way the ship is oriented.

2. Having the hatch/ladder facing (i.e. perpendicular to) normal or anti-normal results in the Kerbal's orbit having very nearly the same period as the ship's orbit, with a different eccentricity. This is bad for reducing drift over part of an orbit, but good for reducing drift over multiple orbits.

3. Having the hatch/ladder facing radial in results in the Kerbal smacking into the ship, reducing relative velocity to near-zero and changing the direction of the relative velocity vector to retrograde. This is good for short-term (1/3 to 1/4 of an orbit) drift, but bad for drifting over multiple orbits since the Kerbal's period is shorter than the ship's, resulting in the Kerbal falling increasingly "behind" the ship in orbit.

4. Prograde, retrograde, and radial out are all weird. I would avoid them.

Next, some results on interplanetary transfer drift (tested on a Kerbin->Jool-crossing transfer):

1. Hatch direction seems to have less of an effect here. Relative velocity at letting-go is so much lower that the whole ramming-into-the-hatch effect I saw in LKO didn't seem to have any effect at all (at least not over the amount of time I spent free-floating but not in time warp).

2. I did notice some variation between the different directions, but it doesn't make any sense. Radial in, radial out, retrograde, and anti-normal were all in the 1.2-1.3 km range, with radial in consistently the best. Normal was 2.7 km, and prograde was an inexplicable 40.3 km of drift. Right now, I'm putting those last two down as anomalies caused by weirdness. I would test some more, but it's getting late and running these tests takes a long time.

Some general conclusions:

Drift seems to be caused simply by the Kerbal's position relative to the ship at letting-go time. This means that drift (relative to size of orbit) is going to be a lot lower for bigger orbits, even if overall drift is higher. This also means that the ramming-into-the-ship effect is only going to be noticeable in low orbits. This also also means that putting the ladder as near to the center of the ship as possible should help reduce drift, but I haven't investigated whether this is the case or not. Future research is needed.

 

As a side note, I was trying to get @dvader's warp-reduction trick to work earlier and wasn't seeing any noticeable effect - and, based on my current understanding of drift physics, it shouldn't work in the first place. More research is needed into this.

[Pds314]

"primitive tech," what does this mean? Are we allowed to use jets? Liquid fuel rockets? Decouplers? Are we allowed to use Nervs? Ions? Where is the line being drawn?

Edited August 13, 2019 by Pds314

[Muetdhiver]

Primitive tech means no facility upgrades, so only T5 and lower.

No patched comics nor manoeuvre nodes either.

@IncongruousGoat nice testing. On my side I tried tricks like using landing legs (does not work), pistons and hinges to pin the kerbal still before warping, but it does not help in any way.

Edited August 13, 2019 by Muetdhiver

[Jacke]

It also means that Kerbals can only exit the vessels on Kerbin's surface.  Which means to get EVA reports, Kerbals have to ride outside of the capsules at launch and at all points during flight, like during interplanetary transfer.  Thus all this talk about Kerbals on ladders and drift during time warping.

[ManEatingApe]

I've updated the rules as discussed and created an initial base game with 10M funds and tech tree unlocked up to and including level 5. Clone it from the GitHub repository linked in the OP.

We should think about the mission profile next. Looks like we want to send a Kerbal on a ladder to Jool. Open questions:

• 1 Kerbal or more?
• Keep Kerbal on a ladder for Vall, Bop and Pol landings also?
• Ladder or pod for the return journey?
• Parking orbit for the mothership?
• Sequence of moon landings?

@IncongruousGoat Great research into the drift, extremely thorough!

 

@Pds314 What @Muetdhiver said. The tech tree is limited to what can be unlocked using the Level 1 R&D facility, meaning up to and including the Level 5 (90 science) nodes.

Edited August 13, 2019 by ManEatingApe

[Jacke]

What's the science difficulty setting, BTW?

 

2 hours ago, ManEatingApe said:

We should think about the mission profile next. Looks like we want to send a Kerbal on a ladder to Jool. Open questions:

• 1 Kerbal or more?
• Keep Kerbal on a ladder for Vall, Bop and Pol landings also?
• Ladder or pod for the return journey?
• Parking orbit for the mothership?
• Sequence of moon landings?

I've had a Kerbal ride the offset separator to LKO and that was demanding enough.  Going interplanetary and worrying about drifting Kerbals makes me think you want to just have 1 external Kerbal.

However, you might want a pilot Kerbal in a pod for Crew Reports and other stuff.  Scientist for the external Kerbal; who also rides all the landers down to all 5 moons, maybe resets experiments like the Goo and the Science Jr.

And after getting through all of the Jool system, I suspect having a 2nd pod seat for the Scientist on return would be a good idea.  Don't want to screw things up at that point what with drift.

Where in the Jool system, I've not done meself.  I'd read a number of them in the Jool-5 topic.  Many work with leaving the mothership at one of the moons, but that would cut into Crew Report science.  So I'd say it's up to whether the mothership itself will have the delta-V to go to all the moons itself.

Edited August 13, 2019 by Jacke

[dvader]

Wow, there's a lot to catch up with.

@ManEatingApe GitHub is a great suggestion. Should be easier than dropbox. There's still an issue if two people try to do the same thing at the same time but we could just use the forum for coordination.

@IncongruousGoat Nice work! It fits pretty well with what I remember. I remember having a preferred direction when letting go of the ladder and I think it was ladder facing radial out / Bob facing radial in. My main concern was ghosting into the ships body which can be really dangerous if you de-warp while inside. Drifting away seemed safest. With quicksave/reload, ghosting and explosions are less of a concern though. One of the reasons I pressed warp right after letting go was to make sure the orbit didn't change before going to the tracking station. If the Kerbal touched any part of the ship, the orbit changed drastically so I'd keep the area around the ladder clean. I never tested (or considered) if keeping the ladder close to the center of mass makes a difference but it is worth investigating. Having the ladder off center makes it a bit easier to catch the Kerbal though since you can move the ladder around the CoM. The Kerbals RCS fuel is a limited resource. I practiced catching the ladder without using any by moving the ship instead of the kerbal ("view" from the Kerbal while controlling the ship was easiest I think).

For interplanetary, I learnt two important things.

First, always check that Ap+Pe of the Kerbal equals Ap+Pe of the ship, preferably from the tracking station so the orbit does not change by accident. If Ap+Pe is the same for both, then the orbital period is the same and they will return to the same spot even years later. For interplanetary, there will be a diff in Ap+Pe but you can easily estimate the diff in km one orbit later.

Second, always set the Kerbal as the "Target" of the ship by double clicking. Even if the Kerbal drifts more than 100km away, the Navball will show the direction. (From what I remember, 100km is the limit when you can't see him anymore).

@ManEatingApe I'd suggest:

• 1 Kerbal only. You don't want to play catch with more than one.
• Jool 5 requires walking or swimming on the surface so no pod for Vall, Bop, Pol. I'd suggest no pod at all since we can only use it for the return flight anyway. @Muetdhiver's Kerbal in a fairing looks cool and very light weight.
• For my Jool 5, I used Tylo for parking. I don't know if it is optimal but it seemed good enough. I don't think we want to park further in than Tylo since there is always a risk of hitting Tylo by accident (not much of a real risk thanks to quicksave/reload of course).
• If parking at Tylo, I'd suggest starting with Tylo, do  Bop+Pol with one craft. Send the Laythe lander to Laythe (unmanned), do Vall, then Laythe and return to Tylo. The order of Bop+Pol and Vall+Laythe can be switched.

[ManEatingApe]

1 hour ago, Jacke said:

What's the science difficulty setting, BTW?

Difficulty setting is normal, however Science isn't strictly necessary as the entire available tech tree has already been unlocked. Although for bragging rights I'd like to at least return an EVA report from the surface of each moon, so we should consider bringing a science container on the mother ship.
 

1 hour ago, Jacke said:

 Going interplanetary and worrying about drifting Kerbals makes me think you want to just have 1 external Kerbal.

1 hour ago, dvader said:

1 Kerbal only. You don't want to play catch with more than one.

Agreed, more than one Kerbal sounds painful. I'm concerned about running low on EVA fuel with all the rendezvouses required, so that's the rationale behind a pod for the return journey. Completely willing to chance it with ladders all the way though!

1 hour ago, dvader said:

For my Jool 5, I used Tylo for parking. I don't know if it is optimal but it seemed good enough. I don't think we want to park further in than Tylo since there is always a risk of hitting Tylo by accident (not much of a real risk thanks to quicksave/reload of course).

A highly elliptical orbit around Tylo has lots of advantages in terms of dV as you can eject to Laythe, Vall, Bop and Pol really cheaply for around 20 m/s.
I have used this approach successfully before for a Jool 5, having stolen it from @5thHorseman

Downsides are that it's very very tricky to rendezvous with elliptical Tylo orbit, so I like your idea of sending the other moon landers on ahead unmanned and rendezvousing with them at the respective moons.
 

Edited August 13, 2019 by ManEatingApe

[Jacke]

1 hour ago, ManEatingApe said:

Difficulty setting is normal, however Science isn't strictly necessary as the entire available tech tree has already been unlocked. Although for bragging rights I'd like to at least return an EVA report from the surface of each moon, so we should consider bringing a science container on the mother ship.

Speaking of trajectories and bragging rights, do you want to do an aerocapture at Jool or Laythe or both?

[IncongruousGoat]

On 8/13/2019 at 5:43 PM, Jacke said:

Speaking of trajectories and bragging rights, do you want to do an aerocapture at Jool or Laythe or both?

Neither. Jool aerocapture is a non-starter even under the best circumstances because of how brutal the heating environment is. I remember trying it once just after 1.0 dropped (it had been standard before 1.0 because no heating) and having a miserable time getting it to work even after I wrapped my whole ship in an aeroshell. Laythe isn't much better, unfortunately - entry velocity off of a transfer is going to be something like 3 km/s, and whatever conglomeration we come up with for a mothership is going to have all the aerodynamic stability and thermal resistance of a pile of overcooked spaghetti.

On 8/13/2019 at 3:58 PM, ManEatingApe said:

Downsides are that it's very very tricky to rendezvous with elliptical Tylo orbit, so I like your idea of sending the other moon landers on ahead unmanned and rendezvousing with them at the respective moons.

I've always favored a low Vall orbit for these purposes, since it's reasonably cheap to get to, trivial to rendezvous with, and in a nice central location. It's not quite as good in delta-V terms as an elliptical Tylo orbit, but the orbital mechanics are a lot simpler, which is good in our case.

I do like the unmanned lander delivery idea. I've used it very successfully for a past Jool 5, and I imagine it would work similarly well here.

 

 

EDIT: Also, my theory about putting the Kerbal at the ship's CoM appears to be completely wrong. Drift seems to work the same regardless of where on the ship the Kerbal starts.

Edited September 14, 2019 by IncongruousGoat
Spelling