I compliment Squad on the insane of difficulty of Eve.

[sedativechunk]

Hey KSP community,

I retired from this game, oh, two~ years ago after putting in over 900 hours of my life into it. I've played it for an hour here and then since when updates came out but never really got deep into it. Well, I got sick a couple weeks ago with a bad head cold and decided to play again to pass the time (I'm also bored senseless of some of the other video games out there right now). I mean really play, building new rockets and things, getting deep again. One thing I never successfully accomplished was putting a kerbal on Eve and return him (or her) safely to Kerbin. It's been in the back of my mind for the two years that I've taken off from this game and now I felt like attempting it.

I just wanted to say I hugely compliment Squad on the difficulty of Eve. For someone that's now put over 1,000 hours into this game and I STILL haven't done it. It just says something. It's nice when a video game has an actual huge challenge or feat like this. Even with some of the new engines and mining, it's just such a RIDICULOUS challenge to try and do. You have to be extremely patient, dedicated, and eager to land a Kerbal to and from Eve. And that is with stock parts and gameplay restrictions in place. You have to build, test things, test some more, and really plan it almost like a real-life NASA mission or something. 

I put maybe 40 hours in since I started playing again just for Eve. I blew up about 6 or 7 ascent vehicles I built in a spectacular explosion trying to land them on the surface with reentry heat on. Finally land the ascent vehicle. Landing objects near one another is a feat in itself, so I put a rover on the surface. I blew up two of those trying to land them straight on the surface. Finally had the ascent vehicle and rover. Send two Kerbals down, drive them 180km (which took hours....) with my rover to the ascent vehicle. I miscalculated the TwR on the ascent vehicle, barely lifts off, waste tons of fuel, crashes back down to Eve. I'm back to square zero.

To make this more grueling, as mentioned, I have reentry heat on, and I'm also playing with TAC life support mod and EVE atmospheres. So I had to land numerous probes down to find safe spots to land because I can't see through the atmosphere. Besides the life support (and some big wheels for my rover), all gameplay and rocket design is stock, I want to to the landing and ascent with stock parts. But I just had to say something about it here at the sheer difficulty of Eve. Pic of the descent vehicle and rover if we can still post images below.

[MR L A]

Well I mean I'm on 2k hours and I haven't done this either... I'm a habitual career mode restarter. I should probably get around to it at some point.

You need something like 11k dV to get to orbit from eve too

Edited October 2, 2018 by MR L A

[Laie]

1 hour ago, MR L A said:

You need something like 11k dV to get to orbit from eve too

Nah, not that much. Here's a very simple design that can easily go from the shores to orbit. It has a generous safety margin and, overall, isn't especially efficient. The bare minimum is more around 5-6km/s.

 

5 hours ago, sedativechunk said:

and I'm also playing with TAC life support mod and EVE atmospheres. So I had to land numerous probes down to find safe spots to land because I can't see through the atmosphere.

You may want to add Scansat to the mix, then.

[WWEdeadman]

Well, I can't exactly say how many hours I put into KSP in total, as my steam hours are only the shorter time I put into 100% stock stuff, but I'd estimate around 1000+ hours all in all, and I have gotten a Kerbal back off Eve exactly once before. I am currently going for for a second time. It probably is the single hardest manned mission you could possibly attempt in the stock system.

[Atkara]

The best way I could come up with, to get Kerbals off the surface of Eve, is this:

Spoiler

It's meant to come down almost empty, get fueled up on-site, take 3 Kerbals onboard and launch. So you need something like a miner on wheels on the surface and something to grab the upper stage and bring it back to Kerbin. But it gets the job done gloriously. With a careful ascent profile, it can extract Kerbals from Eve's Lowlands.

And yeah. Eve is... what it is  

[juanml82]

3 hours ago, Laie said:

Nah, not that much. Here's a very simple design that can easily go from the shores to orbit. It has a generous safety margin and, overall, isn't especially efficient. The bare minimum is more around 5-6km/s.

 

You may want to add Scansat to the mix, then.

Can you land and control that thing? I has no control surfaces, the dart in the second stage lacks gimbal, there are no reaction wheels and the cockpit has a 1200 °K heat tolerance IIRC

[Aeroboi]

5 hours ago, MR L A said:

Well I mean I'm on 2k hours and I haven't done this either... I'm a habitual career mode restearter. I should probably get around to it at some point.

You need something like 11k dV to get to orbit from eve too

11K dv is old pre 1.0 dv requirement information. Today the Dv chart says 8K dv. If you search online you find this newer Dv chart.

But truth be told, 8k dv is bogus also. It's just a estimate number thrown in for whatever average variety of rockets a player could make. The short answer is that there is no clear requirement in Dv needed.

A efficient Eve ascent lies in optimized Aerodynamics to increase the ascent vessels terminal velocity. A higher terminal velocity would mean that it can go faster through air with the same thrust. So if you have a very aerodynamic rocket you have a higher terminal velocity so you can increase your TWR.
The more TWR you have the more rapidly you go faster. This means you get out of the thicker atmosphere faster which means less time wasting fuel on aerodynamic and gravity losses. It also means you can do your gravity turn sooner because your allowed to go a lot faster on a lot lower altitude which in turn also means you spend less Dv on gravity losses because of it.

My result is that I can get a vessel into orbit for less then 5800 Dv from 200m ASL https://imgur.com/a/Zc6s7nB

But the truth is, that's likely not the lowest amount of Delta-V either, although I'm sure you can't really go a full leap lower.
So there is no specific requirement in Delta-V. The Delta-V with which you can do it depends on the factors I explained.

All this is not to say that optimizing aerodynamics and TWR is easy in and of itself. Especially if you have never done this before. If you can pull it of at all then you've passed a major challenge.

Edited September 27, 2018 by Aeroboi

[Atkara]

28 minutes ago, juanml82 said:

Can you land and control that thing? I has no control surfaces, the dart in the second stage lacks gimbal, there are no reaction wheels and the cockpit has a 1200 °K heat tolerance IIRC

I don't know how he lands it -the main reason I opted for the plane solution. The cockpit does have an internal reaction wheel, but I'm not sure how well it steers by itself during the aerospike stage -Laie propably knows.

As far as temps are concerned though, if you don't go Kerbal mode during ascent, you can bring this cockpit out of Eve -with scorch marks, yes. But safe. I know this well, because the only times the inline variant I use (with the same 1200°K heat tolerance) exploded, were when I was trying to determine the lowest possible periapsis for the aerobraking phase -never during ascent.

[Aeroboi]

23 minutes ago, Atkara said:

I don't know how he lands it -the main reason I opted for the plane solution. The cockpit does have an internal reaction wheel, but I'm not sure how well it steers by itself during the aerospike stage -Laie propably knows.

I've seen a video of Laie's vehicle and the aerospike stage is already reasonably well above the atmosphere so the Mk-1 cockpit reaction wheel torque is sufficient. Also, if the gravity turn is optimal your at prograde and there's little required to keep it oriented.

[MR L A]

3 hours ago, Aeroboi said:

So there is no specific requirement in Delta-V. The Delta-V with which you can do it depends on the factors I explained.

Isn't this true for all atmospheric bodies though? But we have a generally accepted dV figure for Kerbin and Laythe

[Aeroboi]

7 minutes ago, MR L A said:

Isn't this true for all atmospheric bodies though? But we have a generally accepted dV figure for Kerbin and Laythe

Yes of course.

On eve it matters a lot more so there's a lot more to gain if you optimize the ascent rocket. The margins for Laythe and Kerbin are a lot smaller because there's less atmosphere to travel through. But on i.e. Kerbin you can also make orbit with 3050-3200 m/s instead of the recommended 3400 dv. It just isn't profitable to do this because you need a lot of TWR. That means you need more engines which equals more funds cost and less Dv in orbit to do other maneuvers.

Dv also doesn't necessarily mean much on Eve because you want the highest payload for the lowest cost which doesn't always mean that a low dv rocket is better. Payload fractions on Eve are low because you throw away more rocket then on any other place so cheap parts with more weight and more Dv can be more profitable. I just had to make my pointers because the 8k estimate seems bogus IME. I find high thrust and aerodynamic rockets do very well on Eve. Although it is a pain to get the right engine combination to achieve it.

I'm not sure why 8k Delta-V is a generally accepted number for Eve. The rockets I created to lift off of Eve have always been 5.5-7k and never towards 8k. IIRC the 8k number was first introduced in the early 1.0.x versions and Squad have tweaked aerodynamics over the versions thereafter. Kerbin and Laythe estimates are expectedly closer because drag and gravity losses are much less of a concern at those places.

[Laie]

6 hours ago, juanml82 said:

Can you land and control that thing? I has no control surfaces, the dart in the second stage lacks gimbal, there are no reaction wheels and the cockpit has a 1200 °K heat tolerance IIRC

 

(also paging @Atkara)

This was done for the Douglas Adams Challenge, I was optimizing for part count rather than mass: hence the Mk1 cockpit which doesn't need a nosecone, and a few other quirks.

The LFBs have a wide bottom (try to surface-attach a part in VAB to see just how wide the collider really is). The lifter alone can stand on a ten degree slope without problems: if you want to do a quick test using Hyperedit, you don't need any landing gear.

You have to dial in your gravity turn during the first stage; by the time of separation you must already be locked to prograde, and stick to prograde for most of the second stage. I'm using Mechjeb, but SAS will do just as well. The third stage can then make up for a lot of early mistakes. The lifter is quite forgiving of going too high (you may end up with a 120km AP and still make it); going too low, however, is likely to end in fiery death.

[bigcalm]

2 hours ago, Aeroboi said:

I'm not sure why 8k Delta-V is a generally accepted number for Eve.

I think it's because it's about right!  Yeah, you can do it with less but you'll have to be really low drag, and if it's not really low drag, it can take you more than 8k.

I love the challenge of Eve.  Yeah, it's pretty easy to slap a craft together that has 8k delta v and is reasonably aerodynamic.  However, you have to get that to the surface of Eve.  That means heatshields, it means parachutes, it means having a landing leg arrangement that doesn't explode (hah, good luck with that on just stock parts!), and having a way to eject those extraneous bits before ascent.  An aerodynamic craft is generally long and thin, but that's awful for landing because it might topple over if you land on a slope.  So you make the landing bit as wide as you can but then it won't fit underneath a heatshield.  And then you realise you also need a way to get the pilot, high up in his lofty perch ladders to get to the ground, which results in an Acme Corporation style ladder section that has to be ejected before ascent and not cause you any aerodynamic issues.

And yeah, once you've done all that hard, hard work, figure out how to get it off Kerbin and off to Eve.  Eve ascenders are the hors categorie of Kerbal Space Program, and so many many times I've thrown away everything after spending hours on the design because it just doesn't work, the margins are too tight and I simply can't fit an ISRU and drill anywhere without causing stupid problems with the descent phase or ejection phases.

Few Eve ascenders are perfect.  All are epic and praiseworthy.

 

[purpleivan]

Ah... the magnificent obsession that is Eve ascent vehicles.

I've had a few over the years.

1st design... complete failure and needed a pretty flat landing site too.

2nd attempt... similarly useless, but was a little more stable on landing.

My 1st successful design (Jeb looks pleased with it too).

Eve Party boat... a team of four sent to Eve and back (along with a deck of cards for the trip).

This was followed by my most lightweight design. I screwed up the fuel lines on the first one sent, so had to send a corrected one as a rescue vehicle.

Val took a trip in a much more recent, ISRU based lander.

Jeb decided he wanted in on that action, so took a more aerodynamic version to the purple planet.

Edited September 28, 2018 by purpleivan

[Mikki]

My personal efforts and successes regarding Eve are basically close to zero.

I never had the courage to send Kerbals down the velvet hell in my careegame.

Well, there is still alot to endeavour for me.

"Damn!"

[Foxster]

I'd say the biggest cause of failing to get off Eve is the temptation to add stuff to a craft that doesn't quite work. I mean, it makes sense to add another engine and a bit more fuel when you are just 100dV short of making orbit, right?

In fact, quite the opposite is the way to go. You need to use the slimmest parts, the smallest possible number of stacks, with not one unnecessary part, and optimal engines. If the craft won't quite make orbit then look what you can take away. 

[FinalFan]

1 hour ago, Foxster said:

I'd say the biggest cause of failing to get off Eve is the temptation to add stuff to a craft that doesn't quite work. I mean, it makes sense to add another engine and a bit more fuel when you are just 100dV short of making orbit, right?

In fact, quite the opposite is the way to go. You need to use the slimmest parts, the smallest possible number of stacks, with not one unnecessary part, and optimal engines. If the craft won't quite make orbit then look what you can take away. 

I can't say I speak from experience (soon...), but it seems to me this is mostly but not entirely true.  This is due, if I am right, to the differences between Eve and Kerbin that can trip people up if they are designing according to habits developed on Kerbin.  

Generally, adding a lot of fuel and just enough engine is the best way to add dV on Kerbin, but on Eve the atmospheric thrust penalty is much greater so launch TWR is much more valuable (yet paradoxically going too fast is also even worse).  So shedding weight as you suggest has not only the benefits we are used to but Eve's conditions (in theory as I understand it) significantly exaggerate the benefit of improving TWR to get into a decently reduced atmosphere faster (unless TWR is already quite high).  So going the extra fuel and engines route can also work but you probably should add more engines than you would expect to with habits developed for launching from Kerbin.  

Adding engines is more expensive than adding fuel, but the simple fact of making an Eve return vehicle means you should probably just pay whatever it takes to get the payload you need.  Needless to say, removing things is usually cheaper than adding them, so payload you don't need can be discarded on the planning floor.  

[MaverickSawyer]

I have attempted designing an Eve Ascent Vehicle once, and only once. I never did use it legitimately, instead cheating it to Eve to ensure it could reenter, land, and take off. And since I've not quite figured out the whole ISRU bit, it had to land fully fueled. Thankfully, there was a set of engines found in the OPT parts pack that worked extremely well on the surface of Eve, and those were used as my first stage, but I've never used them again as I felt they were a little too cheaty... 

Doing this legitimately is on my bucket list, but I suspect I'll need a better computer before I can attempt this properly.

[Foxster]

On 10/2/2018 at 12:09 AM, FinalFan said:

So going the extra fuel and engines route can also work but you probably should add more engines than you would expect to with habits developed for launching from Kerbin.  

It might work to add fuel+engines but, as you hinted, because it is Eve it also adds a lot of drag losses and it is this that is the biggest challenge on Eve. That extra fuel+engine+nosecone+coupler+ can end up gaining nothing or worse. 

Having built more Eve craft than is healthy, I can say that the add-stuff-until-it-flies approach that you can get away with on Kerbin does not work on Eve. Or at least not if you want something that is practical to transport to Eve and to land. What you want is a craft made with the skinniest parts, the highest isp for the <50km engines, the absolutely minimum payload, all the landing stuff dumped before takeoff, not a single extra bit that doesn't contribute to making orbit, and lots of experimentation with your gravity turn because it will be different to the one used for Kerbin. 

[Aeroboi]

On 10/2/2018 at 4:20 AM, MaverickSawyer said:

I have attempted designing an Eve Ascent Vehicle once, and only once. I never did use it legitimately, instead cheating it to Eve to ensure it could reenter, land, and take off. And since I've not quite figured out the whole ISRU bit, it had to land fully fueled. Thankfully, there was a set of engines found in the OPT parts pack that worked extremely well on the surface of Eve, and those were used as my first stage, but I've never used them again as I felt they were a little too cheaty... 

Doing this legitimately is on my bucket list, but I suspect I'll need a better computer before I can attempt this properly.

You can make a 1-3 crew eve ascent vehicle plus launcher and airbrakes/heatshielding (the whole shebang) for 100-150 parts I guess. Some people do even less. The I guess part means I don't know how good you are at keeping part count low but you can. To me that's in the realm of at least 10-15fps on pretty much all semi modern computers. I hereby assume no computer is ancient but I'm sorry if yours is.

Remember if part count is really a issue you can just go for the larger parts like mammoth engines or even overpowered vector engines and the larger Kerbodyne tanks. It may not be the most efficient way but through this route you won't end up with a complex asparagus design and consequently a high part count.

The ISRU bit really helps keeping part count low. If you fuel feed all the tanks fuel to the first stage you can rocket SSTO from Kerbin. Then based on the amount of Delta-v in orbit you can go directly to Gilly or Minmus to refuel. Or alternatively you do not have this fuel and you will only need a small LV-N liquid fuel stage to get you there.
Then refuel and go to Eve. This means you wont need a separate launcher off of Kerbin.
If you had enough Dv you could even do a rocket powered descent to Eve. That way you wont even need heatshields.

[MaverickSawyer]

18 minutes ago, Aeroboi said:

You can make a 1-3 crew eve ascent vehicle plus launcher and airbrakes/heatshielding (the whole shebang) for 100-150 parts I guess. Some people do even less. The I guess part means I don't know how good you are at keeping part count low but you can. To me that's in the realm of at least 10-15fps on pretty much all semi modern computers. I hereby assume no computer is ancient but I'm sorry if yours is.

Remember if part count is really a issue you can just go for the larger parts like mammoth engines or even overpowered vector engines and the larger Kerbodyne tanks. It may not be the most efficient way but through this route you won't end up with a complex asparagus design and consequently a high part count.

The ISRU bit really helps keeping part count low. If you fuel feed all the tanks fuel to the first stage you can rocket SSTO from Kerbin. Then based on the amount of Delta-v in orbit you can go directly to Gilly or Minmus to refuel. Or alternatively you do not have this fuel and you will only need a small LV-N liquid fuel stage to get you there.
Then refuel and go to Eve. This means you wont need a separate launcher off of Kerbin.
If you had enough Dv you could even do a rocket powered descent to Eve. That way you wont even need heatshields.

Good points... But, that's all farther down the line. I've still not successfully landed a crew on Duna and returned them yet.

 

Baby steps, man. Baby steps.

[Aeroboi]

25 minutes ago, MaverickSawyer said:

Good points... But, that's all farther down the line. I've still not successfully landed a crew on Duna and returned them yet.

 

Baby steps, man. Baby steps.

XD. Jolly good luck with it

[StrandedonEarth]

My 2-seat Eve Ascender turned into a monster, with enough dV for Surface-to-Kerbin when you build to over 8km/s dV. Includes sacrificial landing legs.

On 9/27/2018 at 2:57 PM, bigcalm said:

it means having a landing leg arrangement that doesn't explode

Mount them on the end of a truss and let them explode

On 9/27/2018 at 6:47 AM, Atkara said:

And yeah. Eve is... what it is

 

[bigcalm]

A brief guide to Eve Lander problems, and solutions.  I'm ignoring the most obvious in that if you craft explodes on entry and you don't have a heatshield, you should know to put a heatshield on.

My Craft flies sideways after initially ascending ok

 - Your craft is most likely not streamlined enough, or has some property that is causing the craft to fly sideways.  This is likely not fixable by adding tons of reaction wheels, but it depends on the severity of the problem.  It's also possible that once you clear the lower-atmospheric treacle it will become controllable again, so throttling down whilst ascending through the lower atmosphere can help.  Here's an example of a failed one from me -- too unaerodynamic:

 

On atmospheric entry, the craft spins around and immediately blows up due to overheating.

 - Essentially, this is because your centre of mass is a long way from the heatshield.  This is difficult to avoid because you want your craft long and nicely aerodynamic for the ascent.  There's two possible solutions - pad the bit nearest the heatshield with some (ejectable) mass, or use additional heatshields at the top which act like very very draggy fins and stop it from spinning round.  An example --

 

Despite having over 8k delta-v, I can't make orbit

  - This can be a number of things, but most likely:

    -- Your craft is not aerodynamic enough.  General hint - avoid Mk2 plane parts.  Also, if you're using the adjuster tool to make it "look" more aerodynamic, that's not how KSP calculates drag!

    -- Your craft has poor thrust to weight ratio in at least one of your stages.  If you have MechJeb installed then in the VAB, the "SLT" number is the one you want to pay attention to - this is the surface level thrust, i.e. the thrust you will achieve at sea level (set the body to Eve, though, eh?).  Kerbal Engineer Redux can also give you the same information.

  -- Aerodynamic instability in one or more stages.

 -- Ascent profile.  Very generally (and again, aerodynamics weighs heavily in this), you should ascend to roughly 25k above Eve before making your gravity turn - basically, get out of the treacle-like lower atmosphere before steering!

Though it is tempting to add fins and stuff, be careful.  This is a successful one-man ascender --

However, the addition of basic fins on the upper stage changed this from being able to make orbit from sea level with 1000ms spare, to one that could not make it to orbit from a starting altitude of 2000metres.   On Kerbin, you can generally brute-force your way through the lower atmopshere with enough engines.  On Eve, you probably can't.

Ejecting re-entry stuff causes explosions

 -- This is either when you're in the descent phase, when you've dropped below the exploding-due-to-entry-heat phase, and want to get rid of extraneous heatshields, or on the surface when you want to get rid of parachutes and landing legs.

 -- Separatrons are good (in one of the earlier versions of KSP Separatrons basically didn't work at low altitudes on Eve, but this appears to be fixed since at least 1.3.1).  Spinning whilst staging can also help.

 -- Ejecting the bottom heatshield is recommended after inflating the parachutes.  This is mainly because before this, your lovely aerodynamic rocket will move much faster than the ejected draggy heatshield, causing you to crash into it.

 -- It can be helpful to lift off and then eject any landing gear structure if that's possible, depends on your first-stage engine though - at the same time normally works.

 -- Landing legs exploding on landing is almost inevitable, even if you load up on stupid amounts of parachutes to have a softer landing.  Have lots of landing legs and some will likely survive, or a substructure that you attach your landing legs to that will take the place of the exploded landing legs.  Alternatively, if you aren't opposed to mods, there's some much tougher landing legs available in other part packs - certainly better than the Mk2 at least.  Here's an alternative that just uses a bunch of girders instead of legs --

 

Testing on Kerbin

 -- mimicking the conditions of Eve is very tricky on Kerbin, but you might want to try the following tests.

 -- With the ascent vehicle you should generally be able to take off, get to Kerbin orbit, land back on Kerbin (or at least, get a few feet from the surface as you probably won't have landing legs on that bit), and then get to orbit a second time.

 -- Are you able to fly stably at a constant low altitude whilst picking up speed?  Do you ever reach a speed where everything explodes?

 -- You can sort of mimic entry conditions by first putting it in orbit of Kerbin, accelerating to roughly 3km/s, and then doing a little burn at apoapsis to set your periapsis deep inside the atmosphere - this will give you an entry speed into the atmosphere similar to that you'll get from low Eve orbit.

 -- Depending on how you feel about it, you could Hyperedit to Eve Orbit or Eve Surface for testing.

Edited October 3, 2018 by bigcalm
and a couple of clairifications.

[Foxster]

9 minutes ago, bigcalm said:

 

 -- Landing legs exploding on landing is almost inevitable, even if you load up on stupid amounts of parachutes to have a softer landing.  Have lots of landing legs and some will likely survive, or a substructure that you attach your landing legs to that will take the place of the exploded landing legs.  Alternatively, if you aren't opposed to mods, there's some much tougher landing legs available in other part packs - certainly better than the Mk2 at least.  Here's an alternative that just uses a bunch of girders instead of legs --

Some good stuff in the post above. On this issue...aircraft landing gear can make an excellent, strong and just better alternative to landing struts.