totm january 2020 Community Caveman Jool 5 mission

[Jacke]

2 hours ago, IncongruousGoat said:

Neither. Jool aerocapture is a non-starter even under the best circumstances because of how brutal the heating environment is.

And at best there's only the 2.5m Heatshield available, for a craft that will likely need to be minimum mass.

What about a close-to-Jool initial periapsis for the Oberth Effect for the Joolian-Orbit Insertion, with the apoapsis near to whichever moon is chosen for parking ?

[IncongruousGoat]

6 minutes ago, Jacke said:

What about a close-to-Jool initial periapsis for the Oberth Effect for the Joolian-Orbit Insertion, with the apoapsis near to whichever moon is chosen for parking ?

You assume we're going to be able to control the transfer to that degree of accuracy. We might, but I'm not counting on it.

[Muetdhiver]

I would just like to point out a few things :

1) Tylo parking orbit. While it's great for a regular jool5, rendez-vous are tricky like @ManEatingApe said. Even more so in caveman with the lack of approach indicators. I worry a bit about doing multiple warps and correction burns with the guy on the ladder. That might prove costly for the RCS fuel.

2) a single kerbal because two is going to be herding cats.

3) ideally the capture should be done as a reverse assist with Tylo. Though without patched comics it's going to be "interesting". 

I won't be able to do much testing for the coming two weeks but I'll follow closely.

Edited August 14, 2019 by Muetdhiver

[Jacke]

16 hours ago, IncongruousGoat said:

You assume we're going to be able to control the transfer to that degree of accuracy. We might, but I'm not counting on it.

I think it'll need two course correction burns: first after leaving Kerbin SoI, early to mid-transfer to get the pass at Jool close enough, then second right after entering Jool SoI to adjust the Jool pass.  Then burning at periapsis to get an encounter with the parking moon.

Edited August 14, 2019 by Jacke

[ManEatingApe]

On 8/14/2019 at 5:24 AM, IncongruousGoat said:

You assume we're going to be able to control the transfer to that degree of accuracy. We might, but I'm not counting on it.

15 hours ago, Jacke said:

I think it'll need two course correction burns: first after leaving Kerbin SoI, early to mid-transfer to get the pass at Jool close enough, then second right after entering Jool SoI to adjust the Jool pass.  Then burning at periapsis to get an encounter with the parking moon.

Without patched conics, even getting an intercept is an achievement and they tend to be...suboptimal
Here's one of mine from a caveman to Duna mission - you can see that the inclination is way off and the craft just about grazed the edge of Duna's SOI making the capture burn very expensive.

I'd recommend that folks clone the save, use the debug menu to place a test craft into orbit and give it a try, in order to get a feel for the difficulty involved.

My suggestion is that we massively over-provision the mothership with dV, 2200 m/s for the Hohmann to Jool and say another 2000 m/s for the "tweaks" needed.

 

Edited August 15, 2019 by ManEatingApe

[Muetdhiver]

I think we can avoid the extra DV by either :

Do an encounter over multiple with a Pe correction at Ap to catch up Jool after a number of orbits. (This can be very accurate if done down to the second). 

Or go the extra mile. Launch as best as possible, correct inclination at mid transfer. Then solve the craft orbit, Jool's orbit and the Lambert  problem for the encounter and calculate the correction burn. Correct. Encounter.

option B is harder but would also be rather kickass IMO. I have an ESA package reconfigured for KSP (I use it for multiple grav assists planning, but it can surely do simple Lambert just fine). Just need to whip up some code to do it I suppose. If interest is there I could try to do it when times permits and do some tests with probes. But not before two weeks.

Edited August 15, 2019 by Muetdhiver

[dvader]

19 hours ago, Muetdhiver said:

Do an encounter over multiple with a Pe correction at Ap to catch up Jool after a number of orbits. (This can be very accurate if done down to the second). 
 

Yes, this is the way to do it. Visually find the AN/DN node in the tracking station, intentionally miss Jool for the first orbit, get the exact time when Jool passes your Ap, get the exact time you pass the Ap and adjust the orbit to get a perfect hit after one Jool orbit. It works really well even for small bodies like Bop/Pol.

19 hours ago, Muetdhiver said:

Or go the extra mile. Launch as best as possible, correct inclination at mid transfer. Then solve the craft orbit, Jool's orbit and the Lambert  problem for the encounter and calculate the correction burn. Correct. Encounter.

 

option B is harder but would also be rather kickass IMO. I have an ESA package reconfigured for KSP (I use it for multiple grav assists planning, but it can surely do simple Lambert just fine). Just need to whip up some code to do it I suppose. If interest is there I could try to do it when times permits and do some tests with probes. But not before two weeks.

Yes! Much more kickass! But, I assume you need some input parameters for Jools orbit and Jools current location in the orbit. You could measure the on-screen angle between your major axis and Jool using a protractor but I don't know how to compensate for the eccentricity of Jools orbit. Assuming a circular orbit might be good enough though.

[ManEatingApe]

@Muetdhiver and @dvader I volunteer you both as navigators.

Solutions using protractors, duct tape, twine or sextants leveraging the new HD star field are encouraged!
Advanced equation solving is also supported as long as the input values are available in the game UI (e.g. orbit AP and PE)

Would anyone like to volunteer some lander designs? With the Kerbal-on-a-ladder approach, we should be able to make a tidy Tylo lander around the 2 ton mark.

 

Edited August 16, 2019 by ManEatingApe

[Muetdhiver]

I think that @IncongruousGoat has already done a nifty Tylo lander with a mass ~2 tons.

There was also a staged spaceplane proposal for Laythe.

For option B it will require the orbital parameters of Jool. This is easy to get (measuring tools, epoch 0) so long as we work with coplanar trajectories (no need of knowing An/Dn). Alternatively I have already the parameters in a python dict from previous works with pykep. I have to get a look at the Lambert solver. If I am Correct, this is easier than it looks, since we get speed, altitude (kerbol) and time in game without upgrades. As long as the trajectories are coplanar I'm quite sure this is enough.

[dvader]

I tried a Laythe landing and return with the space plane. As they say, tenth times a charm and it worked! Some caveats apply:

1. The first string of failures was due to this weird staging issue where SAS would turn off when staging the wings. This resulted in the rocket tumbling unti I could turn SAS on again. I "think" this happens when the probe core isn't "root" but I am not sure. The problem disappeared after I relaunched with a new plane where the probe core was root. This would be a real issue if it happened "live" but I am not sure what is causing it or how to make sure it doesn't happen. Sometimes I could regain control but sometimes the Kerbal would fall off.

2. I had 42 m/s left when reaching orbit which isn't much of a margin. I think it can be improved with more practice but I don't know by how much. From my Kerbin testing I thought I'd have 100-200m/s left so I may be doing something differently this time.

Here's a video where I cut together the successful attempts:

Spoiler

 

 

@Muetdhiver Right, so we get Jools params from the current time and starting position at epoch 0? Sounds OK to me. It sure sounds like we have enough to make it work then. Could you post your script once you're done with it? I've never tried pykep but it would be fun to try it and see how it works.

[ManEatingApe]

I feel things are starting to come together!

So far we have:

• @IncongruousGoat Compact Tylo lander
• @dvader Laythe ascent/descent (great video by the way, really like the simplification enabled by water take-off!)
• @Muetdhiver Orbital rendezvous scripts

I propose we simply re-use @IncongruousGoat's Tylo lander twice for the remaining moons.
A Tylo lander needs about 4,600 m/s which for example is plenty to leave low Tylo orbit, rendevous with Vall, land, return to orbit, then eject and rendezvous with a fresh craft for Bop & Pol, then onwards to Laythe.

 

Edited August 17, 2019 by ManEatingApe

[joshudson]

On 8/15/2019 at 12:56 AM, ManEatingApe said:

Without patched conics, even getting an intercept is an achievement and they tend to be...suboptimal
Here's one of mine from a caveman to Duna mission - you can see that the inclination is way off and the craft just about grazed the edge of Duna's SOI making the capture burn very expensive.

I'd recommend that folks clone the save, use the debug menu to place a test craft into orbit and give it a try, in order to get a feel for the difficulty involved.

My suggestion is that we massively over-provision the mothership with dV, 2200 m/s for the Hohmann to Jool and say another 2000 m/s for the "tweaks" needed.

 

Seriously? I got my Laythe encounter with a 2500m/s ejection burn and less than 500m/s all the way to Laythe. I don't have the robots expansion, but otherwise I'd volunteer to fly your intercept flights for you.

Edited August 17, 2019 by joshudson

[ManEatingApe]

1 hour ago, joshudson said:

Seriously? I got my Laythe encounter with a 2500m/s ejection burn and less than 500m/s all the way to Laythe.

I did enjoy your mission (linked here for convenience - 10th post on this page of the Caveman challenge)

Just bear in mind that we need more than a arbitrary encounter for the mothership. It must end up in a orbit of our choosing (say for example either Tylo or Jool at 0° inclination) to make the rest of the mission possible.

Intentionally trading excess dV for an easier life when recovering from suboptimal encounters is one approach, repeated F5/F9 is another, like most things the optimum trade-off is probably somewhere in the middle.
 

1 hour ago, joshudson said:

I don't have the robots expansion, but otherwise I'd volunteer to fly your intercept flights for you.

The mission is vanilla KSP, no mods, no DLC expansions.

Edited August 17, 2019 by ManEatingApe

[joshudson]

I've had further practice. I can hit either extreme of Jool's SOI easily and correct inclination on the cheap out there. It should be less than 500 delta V to an equatorial encounter with any moon. But those capture burns are going to be rough.

[IncongruousGoat]

4 hours ago, ManEatingApe said:

I propose we simply re-use @IncongruousGoat's Tylo lander twice for the remaining moons.

A Tylo lander needs about 4,600 m/s which for example is plenty to leave low Tylo orbit, rendevous with Vall, land, return to orbit, then eject and rendezvous with a fresh craft for Bop & Pol, then onwards to Laythe.

Sounds good to me. Plus, it's really convenient to launch, requiring no pad or orbital assembly at all.

Edited August 17, 2019 by IncongruousGoat

[ManEatingApe]

1 hour ago, joshudson said:

I've had further practice. I can hit either extreme of Jool's SOI easily and correct inclination on the cheap out there. It should be less than 500 delta V to an equatorial encounter with any moon.

Great! Any tips/tricks to share? An Imgur or video tutorial would be amazing

1 hour ago, joshudson said:

But those capture burns are going to be rough.

Agreed - getting that intercept close to Jool will make all the difference.
Here are some rough figures capturing from a Kerbin Hohmann transfer to an highly elliptical Jool orbit:

• PE at 250,000m = 220 m/s
• PE at 62,500km (Tylo) = 615 m/s
• PE at 110,000km (Bop) = 1,335 m/s
• PE in the sticks (1,000,000km) = 1,800 m/s

We'll need more dV on top of this to circularize this orbit. Using Tylo's Oberth effect we can do this for about 300 m/s.
Doing the same maneuver at Tylo orbit but outside its SOI costs about 900 m/s (ouch)

46 minutes ago, IncongruousGoat said:

Plus, it's really convenient to launch, requiring no pad or orbital assembly at all.

This is a great feature. Simplifying logistics will help the mission succeed.
I was hoping to get away with almost no pad assembly, perhaps only a frame to hang everything off, or the central transfer section.

Edited August 17, 2019 by ManEatingApe

[Muetdhiver]

2 hours ago, IncongruousGoat said:

Sounds good to me. Plus, it's really convenient to launch, requiring no pad or orbital assembly at all.

Speaking of which : the best part from the MH DLC for caveman is the bobcat engine. It frees quite a few parts from the sparks and nose cones. Like 6 to 8 XD

but yeah, tangent appart, launch convenience is super important.

Getting the poodle up is going to require on pad assembly though.

Edited August 17, 2019 by Muetdhiver

[joshudson]

So I went back and rebuilt the Jool rock thrower again because those numbers weren't making sense to me. I thought it had 7500-8000 delta v. It has 8994. I really did eyeball everything.

It really is just practice though. But due to a horrendous piece of coding, if you launch ahead of the transfer window and intercept Jool by having its SOI collide with you as you're dropping back down to Kerbol, you can cut a chunk out of that capture burn .

But Inclination change burns in the sticks cost a couple hundred m/s because you're moving at 10s of m/s so it's actually more efficient to just make a radical burn towards the plane you want to be in, and another radical burn to stop your motion when you reach it. You could accomplish the same thing with an elliptical orbit all the way out to the sticks though.

[ManEatingApe]

I've been thinking on and off about the mission profile over the last couple days.
My proposal is:

1. Assemble mothership: 3 x Tylo landers, 1 x Laythe ascent, 1 x Kerbin return
2. Hohmann transfer from Kerbin to Jool
3. Capture into high elliptical Jool orbit
4. Fix inclination and adjust PE to match Tylo
5. Capture into high elliptical Tylo orbit (burning as close to Tylo as possible to maximise Oberth effect)
6. Unmanned Laythe ascent and Kerbin return craft go to Laythe orbit
7. Unmanned lander goes to Vall orbit
8. Mothership with remaining 2 landers descends to low Tylo orbit
9. Tylo lander descent/ascent
10. Rendevous in low Tylo orbit with remaining lander
11. Land then return from Bop and Pol then go to Vall
12. Rendevous with Vall lander
13. Land then return from Vall
14. Go to Laythe, using propulsive braking to slow down in atmosphere enough
15. Unmanned Laythe lander rescues Kerbal and returns to orbit
16. Kerbin return vehicle ejects directly from low Laythe orbit (to maximise Oberth effect and Laythe's orbital velocity boost)
17. Fairing/decouple device used to protect valiant Kerbal from re-entry at interplanetary speeds
18. ???
19. Profit

What do people think?

Edited August 19, 2019 by ManEatingApe

[IncongruousGoat]

4 hours ago, ManEatingApe said:

I've been thinking on and off about the mission profile over the last couple days.
My proposal is:

1. Assemble mothership: 3 x Tylo landers, 1 x Laythe ascent, 1 x Kerbin return
2. Hohmann transfer from Kerbin to Jool
3. Capture into high elliptical Jool orbit
4. Fix inclination and adjust PE to match Tylo
5. Capture into high elliptical Tylo orbit (burning as close to Tylo as possible to maximise Oberth effect)
6. Unmanned Laythe ascent and Kerbin return craft go to Laythe orbit
7. Unmanned lander goes to Vall orbit
8. Mothership with remaining 2 landers descends to low Tylo orbit
9. Tylo lander descent/ascent
10. Rendevous in low Tylo orbit with remaining lander
11. Land then return from Bop and Pol then go to Vall
12. Rendevous with Vall lander
13. Land then return from Vall
14. Go to Laythe, using propulsive braking to slow down in atmosphere enough
15. Unmanned Laythe lander rescues Kerbal and returns to orbit
16. Kerbin return vehicle ejects directly from low Laythe orbit (to maximise Oberth effect and Laythe's orbital velocity boost)
17. Fairing/decouple device used to protect valiant Kerbal from re-entry at interplanetary speeds
18. ???
19. Profit

What do people think?

I like it overall, but I'd recommend the following change:

4. Fix inclination & separate the landers & return vehicle
5. Adjust the PE of the mothership to match Tylo & capture into Tylo orbit (burning as close to Tylo as possible to maximise Oberth effect)
6. Adjust the PE of the Laythe ascent and Kerbin return craft & capture at Laythe (run concurrently with 5)
7. Adjust the PE of the Vall lander & capture at Vall (run concurrently with 5 & 6)
8. (empty)
 

Edited August 20, 2019 by IncongruousGoat

[Muetdhiver]

I think the mission plan is sound, and I think the changes are sensible.

My only worry is the accuracy/planing of the ejection from Laythe to return kerbin. This will require maths.

 

[ManEatingApe]

19 hours ago, IncongruousGoat said:

I like it overall, but I'd recommend the following changes...

11 hours ago, Muetdhiver said:

I think the mission plan is sound, and I think the changes are sensible.

My only worry is the accuracy/planing of the ejection from Laythe to return kerbin. This will require maths.

Thanks for the feedback!
I've merged the suggestions into the plan and updated the OP with the revised list (under the 2nd spoiler).

@dvader @Jacke @joshudson @The Dunatian or anyone else - thoughts/suggestions?

Edited August 20, 2019 by ManEatingApe

[Jacke]

This is me at my least Caveman, but delta-V is base stock in 1.7.3, so....

What's the delta-V sums of the planned mission (from LKO since orbital assemble) and how much padding will be needed for Caveman performance?

[Pds314]

Tylo is really twitchy to land on. Especially when you have to eyeball everything...

Only the engine survived my attempt to build a Tylo lander with caveman.

 

[Pds314]

What kind of partcount/weight capabilities does each lander, etc need to have? Like, is a 2 tonne Tylo lander too heavy? Is 10 parts okay? 20? Where can we rely on Kerbal jetpacks if ever?

What about physical form factor? Does it need to be <Mk1? Mk1-ish, in between Mk1 and 2.5m? What about other vehicles?

Also WRT 4600 m/s vacuum Delta-V being enough to land on Tylo with a sane TWR and then return to orbit, then leave orbit and go to Bop and Pol, how? Tylo ground-level orbital speed is like 2145? You're gonna land, take off, leave Tylo orbit, escape Tylo orbit, and land on Pol, leave Pol, land on Bop, leave Bop, and return to the mothership with 4600 Delta-V? When it necessarilly takes 4290 even with impulsive burns to do a Tylo landing and orbit? And 900 more to leave Tylo SOI???

Or maybe I'm confused. Is the idea to have three separate Tylo landers? But only land one of them on Tylo and use the other two as long range lander-transfer stage combined?

Edited August 21, 2019 by Pds314