Please Help! Reentry failure - what am I doing wrong?

[IILEOII]

I'm very new to the game, trying to understand the logic and physics behind many things, but I really can't see what I'm doing wrong here.  3 command pods launched on a suborbital trajectory. Each is separable by a decoupler from the others and all have 2x radial parachutes (the top one has an m16 instead). Once the 3 pods reach apoapsis, I aim them retrograde to reentry and trigger the decouplers and prime the chutes of the bottom 2 (set at 0.75 and 1000 as per tutorial guidance). I then align the final command pod retrograde and turn off the SAS. Despite the headstart of the other 2 pods and the same reentry angle, the final pod catches them up and shoots past them, slamming into the ground before speed drops enough for chutes to successfully deploy. The other 2 pods are slowed enough by reentry for the chutes to deploy.
I MUST be doing something wrong for the final pod not to work, but I can't for the life of me see what it is. Video below is one of the many trials I've had - could anyone please take a look and tell me what's going wrong?
Thank you!

https://youtu.be/sQHOsALWRYk

Edited June 2, 2018 by IILEOII

[Gilead]

I believe the radial mounted chutes add more drag to your capsules, thus slowing them more than the capsule with only an Mk-16.

Also, I would suggest that you use the 2 kinds of chute (normal and drogue chutes) in conjunction. The drogue chutes slow less but can be deployed at higher speeds, slowing you enough for normal chutes to deploy safely.

Also, on a unrelated note, your ship is way too powerful at launch. It would be more efficient to fire first only the side boosters and engage the middle one after depleting and getting rid of the side ones.

Edited June 2, 2018 by Gilead

[HebaruSan]

When your re-entry trajectory is that steep, you don't spend much time in the atmosphere. A more horizontal trajectory would bleed off more speed.

Also consider ditching the external heat shield; it adds more mass without adding drag, which makes it harder to slow down, and you don't need it for the speeds you're reaching.

[ElWanderer]

3 hours ago, Gilead said:

I believe the radial mounted chutes add more drag to your capsules, thus slowing them more than the capsule with only an Mk-16. 

I think that'll be dwarfed by the drag from the extra capsules having an abrupt 0.625m to 1.25m transition (capsule top to decoupler) and an open 1.25m node (decoupler to nothing). Admittedly, I can only view the video at poor resolution on my mobile phone screen at the moment, so I'm somewhat guessing at the parts. Either way, I agree that the extra capsules slow down just in time because they have considerably more drag than the prime capsule.

Solutions include making the prime capsule more draggy and/or lighter (if you have a heat shield, remove all the ablator... or even the entire part), adding chutes that open earlier (drogues) or pitch over more during ascent so that you have a flatter trajectory. The last one gives you more time to decelerate. I think these have all been mentioned already.

I would add that this type of craft is inherently risky - if the individual capsules get too far away from each other, those outside the active craft's physics bubble will be deleted. Admittedly, that's something like 23km now, so you should be okay.

My other advice would be to orient to normal before firing the decouplers. That brings everything down with a more similar periapsis, but on slightly diverging courses. That reduces the chance of running into your own dropped stages. That said, (argh, out of place quote follows, stupid mobile interface)

5 hours ago, IILEOII said:

The other 2 pods are slowed enough by reentry for the chutes to deploy.
I MUST be doing something wrong for the final pod not to work,

this may not be appropriate for this particular craft if the extra pods have tourists and can't orient back to retrograde for reentry.

[Laie]

On the pod you control, the chute symbol is displayed in red, meaning it's "not safe to deploy". That's why it doesn't open. I can't tell you with any kind of certainty why the parachutes on the other pods work.

I can tell you from sad experience that this kind of MIRV approach is inherently unsafe. The game likes to "destroy" parts that are unattended in the atmosphere, and while sometimes it does the right thing with items several kilometers away, I've seen others vanish without a trace at distances as short as 500m. This can really ruin your day.

[GrouchyDevotee]

The launch is too steep. You are hitting the ground at over the speed of sound.

Flatten out the launch for a shallower entry, this will allow your pod to slow down, also adding a pair of radial mount drogue chutes that can deploy before the main will allow you to slow down further.

[GDJ]

Have to agree with the others: Too steep of a re-entry angle. Shallow it up, spend more time in the upper atmosphere.

[pincushionman]

Not sure what you’re trying to accomplish with the MIRV configuration, but if it’s “get multiple Kerbals to space and back simultaneously for level-up/contracts” you’d be better off keeping all three pods connected.

Or better yet, use one command pod and a crew cabin. Only one craft, so no physics-range problems - note that if you do a more-horizontal flight as suggested, your draggier capsules will go out of range and be destroyed.

I myself use such a craft in my save.

[bewing]

Welcome to the forums. As Laie said, going for a MIRV approach is unsafe because of the way the game handles multiple objects. Unless the objects stay within about 25km of your current vessel, the game may delete them and your kerbals.

ElWanderer is exactly correct about why your first two RVs slow down -- it's a concept called an "open node". In the editor, an open node is represented with a green ball. In flight, when you decouple parts of your craft from each other -- the places where they were attached become open nodes. The game assigns high drag values to open nodes. It is also true that the two radial parachutes have significantly more drag than the MK16 -- and that will slow those two RVs down even faster.

If you want to use multiple capsules like this, it may work better for you to wait until they have reentered and are below 10km altitude before you activate the parachutes and decouple all the pods.

 

[MisterKerman]

Rather than create a thread I'll derail this topic if that's okay!

- - - -

My strategy for re-entry/landing on Kerbin with capsules carrying the mission's payload of scientific data from the Mun/Minmus have been pretty brutal. I just generally dip into the atmosphere at about 35K at retrograde and burn off my remaining fuel in an attempt to slow myself down, then staging the spent engine to separate. The capsule always makes it down in one piece.

I feel like accuracy basically isn't reasonably possible with this approach.

The capsule itself makes it back in one piece, but for my own morbid OCD I would love to have some sort of elegant technique to actually touch down at KSC. Is there anything you guys do to mentally ballpark re-entry from an extreme inclinations? I'd love to have the chance to fine tune re-entry and trim altitude using nothing but the lift capabilities and reaction wheels of a command pod. It would really really make my day.

 

<3

Edited June 4, 2018 by MisterKerman
Grammar

[Kesa]

2 hours ago, MisterKerman said:

Rather than create a thread I'll derail this topic if that's okay!

- - - -

...

Welcome aboard!

With a heat shield, you should not need to burn off your velocity.

Dropping from such altitude, your landing site is chaotic (very sensitive to initial conditions). If you want to pinpoint land on an planet with an atmosphere requires you co circularize first (so one or two pass at 40-50km, then kick your periapsis back past 70 km). From there, it's a matter of trial and error to find the good trajectory to land where you want (that trajectory depends on the craft). Some control surface or lifting surface improves your precision a lot, but even the lifting body and the variation of drag depending on the orientation of a rocket can help, especially if acted upon early on (very small aerodynamic forces in high atmosphere, but there effect carry on and multiply over time).

You also might be interested by the mod called "trajectories"

[CrashyMcCrashFace]

You were still going almost 300 m/s at 1000m which doesn't give enough time for the chute to deploy and open.

But yeah, being in control of a ship vs an uncontrolled identical ship, I've noticed they act very different in the atmosphere.

[MisterKerman]

1 hour ago, Kesa said:

Welcome aboard!

...snip...

I figured as much. With suicide trajectories from extreme inclinations, it seems you get what you get and should just thank your bum that the capsule makes it back in one piece.

[Xd the great]

Well, heat shields, higher PE, no need to burn once you have a good trajectory set up.

[Snark]

On 6/2/2018 at 9:28 AM, IILEOII said:

I MUST be doing something wrong for the final pod not to work, but I can't for the life of me see what it is. Video below is one of the many trials I've had - could anyone please take a look and tell me what's going wrong?

Okay, so a few things going on, here.

The other two pods are slower because they've got those big decouplers stuck to them, which is really draggy and slows them down a lot.  That's why they were okay, and this one wasn't.

I note that going MIRV-style like that doesn't actually accomplish anything-- there's no point to it, unless you simply wanted to do it to "look cool".  For example, once you ditched the SRB... why did you separate the capsules?  Could have just activated their parachutes and left the three capsules stuck together, and I bet you would have been just fine. 

(Not that there's anything wrong with doing something just-for-the-heck-of-it.  It's what KSP's mostly about, after all.    Here's an example of a MIRV reentry with happy ending.)

Anyway:  Various ways you could solve this problem.

• Launch on a shallower trajectory, rather than straight up, as various folks havee suggested.  Gives you lots more time to slow down.
• Make that last capsule draggier.  For example, when you launched, you could have put a decoupler on top of it, too, and put a nosecone or something on it.  Then all three capsules would have been the same and would have been fine.
• Make the capsules lighter.  One reason that that last capsule is still going 300+ m/s when it gets down close to the ground is that it's too heavy.  Get rid of the heat shields.  You don't need them.  A Mk1 command pod can re-enter from a suborbital trajectory with no problems-- heck, it can even re-enter from low orbit without trouble.  Getting rid of the heat shield will shave 300 kg off the mass of the craft, which is a lot for a small craft like this-- you'd slim down from 1200kg to only 900kg.  Given that you were going only slightly too fast for the parachute to deploy, I suspect that simply doing this would have solved the problem for you.
• Add a drogue chute.  That's overkill in this case, but would do the job (and works nicely for bigger ships, too).  Drogues can deploy at double the speed of regular chutes, meaning they can save your bacon as long as you're not going over 600 m/s when you faceplant.

Any one of the above strategies would solve the problem, I expect.

 

[Kryxal]

Another possibility that doesn't require any change is to jettison the heat shield at about 15km up.  If you're aligned retrograde (and you should be), the thing stays practically glued to you, but due to how aerodynamics work in-game, drag roughly doubles.  It seems the jettisoned part doesn't occlude at all just because it's in front of you...