Jump to content
  • 1

Service bay reentry


Fraktal
 Share

Question

In the first half of this year, I did a lot of reentries with a Mk1 pod and a matching service bay underneath and it always went flawlessly, with the craft properly orienting retrograde and staying there even without SAS. When I fired up the game earlier this week after several months, however, I'm suddenly finding reentry nigh impossible with this arrangement: if it orients even 10° away from perfectly retrograde due to the pilot not yet having hold-retrograde for SAS, the whole craft violently flips prograde and plows into the ground at 600 m/s. Attaching a heat shield to the bottom of the service bay not only makes the issue even worse, but the heatshield actually generates so much lift that it nearly flips the rocket after first stage separation.

Turning on the center of mass indicator in the VAB shows that the center of mass refuses to budge from the pod, even with a service bay and a heatshield full of ablator underneath. This most definitely did not use to be the case a few versions ago. What happened? Did an update modify the Mk1 pod's aero characteristics? Before anyone asks, I do not have FAR or any mods that affect aero.

 

Granted, the Mk1 pod also didn't use to start violently spinning at the beginning of EVA that makes returning nearly impossible. And since I updated to 1.4, the "edit vessel in VAB/SPH" button on both the Launch Pad and the Runway is permanently dark and unclickable.

Edited by Fraktal
Link to comment
Share on other sites

Recommended Posts

  • 1

A couple ideas...

  • Add a heat shield - even if you zero out the ablator. It's still going to bring your CoM down a bit and I think it will help with drag. The bottom of your Service Bay has an open node which is draggy. Adding the heatshield will "cover" that node.
  • Consider clipping your pod into the top of the service bay so the two have a smooth joint rather than a gap. I'm not sure if it will be different aerodynamically, but it'll lower the CoM. I often think of clipping as cheaty, but in this case you're clipping into empty space, so it seems believable.
  • Add drogue chutes. They'll slow you down and help the pod orient correctly
Edited by Tyko
Link to comment
Share on other sites

  • 1

Tried them upside down, still no issues.

I think you should set your parachute settings back to default, as that's the only difference I can see. 250m/s is "safe" deployment speed and you can get away with a bit higher than that sometimes.

Something odd must be going because no steering or manual input was required on my part, once the drag forces had it; it flew straight as an arrow with a slight spin due to no SAS (I tried it again with SAS stability hold on and it forced the nose off course by 10 or degrees, but other than that everything went fine.). I know it sounds counter-intuitive but turn SAS off during re-entry if all you have is stability hold, it's just fighting what you want to happen.

Edited by Rocket In My Pocket
Link to comment
Share on other sites

  • 0

Your pilots skill level greatly affects the SAS ability of your craft.

Perhaps you could try copying the craft file into a sandbox game and see how it runs with a level 5 pilot.

I usually build a vey simple chute/Mk1 pod/heat shield upper stage for my first orbit.  If you still flip to prograde on reentry you can add drogue chutes and "wobble" your craft left and right after the flip to help bleed off speed before main chutes.  Another possibility would be to aerobrake by dropping your Pe to around 55km, and then burn the last of your fuel retrograde at that point.

The heat shield isn't really required but I'm a stickler for safety.

 

Edited by James Kerman
Added last 2 paragraphs
Link to comment
Share on other sites

  • 0

Command pods are very heavy compared to similarly sized components, because they're supposed to be general purpose. I figured if I lived in KSP's universe the Mk1 pod would contain life support, small reaction wheels, batteries, controls, heavy plating to at least take some heat and shield against radiation, and so on. So they will out-weigh most other components. This was true in KSP 1.3.1, where I flipped many command pods attached to Mk1 crew cabins and killed many tourists.

One 'solution' I came up with was to attach the Mk1 pod to the inside of the 1.25m service bay after attaching other things to the outside. It isn't realistic and it clips into components above it, but it works in KSP. I wouldn't do it in reality unless there was a 1.25m service bay with double the height. But no really, that Mk1 pod is massive and it will force centre of mass closer to it.

Link to comment
Share on other sites

  • 0

What's in the service bay out of curiosity?

If it's science stuff, you can just nab the results with a Kerbal and bring them back to the pod, no need for the bay or the shield. (You will need your Astronaut complex to be at level 2 though.)

Slapping more reaction wheels into the bay for better SAS authority is another possibility. Or focus on getting your Kerbal Pilot to level 2 for Retro hold, or bring along a probe core with Retro hold. As far as why it's different than you remember, idk? When had you last played, what version was it? Not much major has changed unless you were playing before the "atmo-soup" update.

Link to comment
Share on other sites

  • 0
4 minutes ago, Rocket In My Pocket said:

hat's in the service bay out of curiosity?

If it's science stuff, you can just nab the results with a Kerbal and bring them back to the pod, no need for the bay or the shield. (You will need your Astronaut complex to be at level 2 though.)

Alternately, take an experiment storage unit up and have it move the science data over. You get that at the same time as you unlock the Science Jr kit. No need to have a kerbal retrieve the science with that.

Link to comment
Share on other sites

  • 0
7 minutes ago, Gordon Fecyk said:

Alternately, take an experiment storage unit up and have it move the science data over. You get that at the same time as you unlock the Science Jr kit. No need to have a kerbal retrieve the science with that.

It's not exactly a small part though and now you've either got to have it above the pod possibly getting cooked, or below it definitely getting cooked. Not sure if you'll be much better off compared to the service bay situation.

I honestly don't know though, I really haven't found much use for them since they were added so late; so I don't have much experience re-entering with them.

Worth a shot maybe?

Link to comment
Share on other sites

  • 0

I started KSP this January, shortly before the 1.4 update.

 

Had Steam run an integrity check, said 83 files didn't match, which was probably me having renamed the folder with the stock ships. Got rid of a persistent little bug with RCS Build Aid that's been plaguing me since the 1.4 update and the pod's sudden pitch spin when going EVA isn't as strong now, it but did nothing else.

Did a short suborbital test flight a few minutes ago. Hit atmosphere at 2000 m/s from an apoapse of 80-something km, prograde flip occurred at 1135 m/s via a yaw to the left (SAS hit its limit), hit water at 514 m/s. Both during the ascent and the descent, I aligned perfectly retrograde at 60 km and turned SAS off, craft immediately started rotating away from retrograde even in that nearly-nonexistent atmosphere. Service bay contains three goo canisters, one thermometer and one barometer, for a total payload of 0.16 plus the bay's own weight of 0.1 versus the pod's 0.84.

 

I'm explicitly building the craft out of low-tech parts, so no experiment storage unit yet and I'd prefer to bring the instruments back if at all possible; it's not a career game but I want the craft to be career-compliant for the early game (26 parts, 16.872 tons), which means saving scientific instruments if at all possible because that stuff isn't exactly coming out of Jeb's pocket money.

Edited by Fraktal
Link to comment
Share on other sites

  • 0

As far as the heatshield goes -- the heatshield now has two attachment nodes. They are at slightly different heights, and determine whether there is a shroud on the heatshield or not. However, the two setups have very different drag characteristics, and that may be causing some of your problems when you are using heatshields (which you don't need to do).

The yawing/flipping issue sounds to me like a CoM imbalance of the parts inside the service bay. If you strip everything down to just the service bay plus contents, and then turn on the CoM indicator in the editor, I think you will be able to adjust the positions of all the internal parts so that the CoM is in the center -- and then your RV will have a much easier time reentering retrograde.

Additionally, of course, drogue chutes can stop you from hitting the ground at 514m/s. Also, a couple of properly positioned basic fins can help your RV glide with much more control and bleed off speed -- so being a little more creative with aerodynamics can completely fix this problem.

 

Link to comment
Share on other sites

  • 0
29 minutes ago, Fraktal said:

I'm explicitly building the craft out of low-tech parts, so no experiment storage unit yet and I'd prefer to bring the instruments back if at all possible; it's not a career game but I want the craft to be career-compliant for the early game (26 parts, 16.872 tons), which means saving scientific instruments if at all possible because that stuff isn't exactly coming out of Jeb's pocket money.

You play however you want, I'm certainly not telling you how to enjoy the game but you don't need to save them, they aren't that expensive in the grand scheme of things, and you'll be rolling funds by the early-mid career game. If you enjoy the engineering challenge though or you're planning on a "hard-mode" style career with rewards and funds turned down, then by all means; you do you.

I'm pretty sure I've re-entered similar builds though in early career before space EVA is possible. I'll have to test it out if I get a chance, see if it's just you or something has changed.

Update: Built what you described and it re-entered fine on my end with SAS completely disabled. Chute deployment was safe at around 10km, what kind of re-entry profile are you using? Steep or shallow? I was at about 30-35km Pe. Or maybe we built it differently?

Spoiler

EE45021BE3A4ED3A0BA5984B87D3AAB6672984C3

 

Edited by Rocket In My Pocket
Link to comment
Share on other sites

  • 0
21 minutes ago, Rocket In My Pocket said:

[experiment storage units are] not exactly a small part though and now you've either got to have it above the pod possibly getting cooked, or below it definitely getting cooked.

I've had good success with the things. They are great for taking an extra copy of science data for instruments that benefit from more than one experiment, saving a return trip and not needing a second crew part. You could even get silly and send a probe with one or two of them to the Mun and back, making for a very light craft, and use 0.625m heat shields to protect them on re-entry.

I don't have the game available to me right now or I'd build a quick example: A 1.25m service bay with 1.25m nosecone and parachute on top, 2 x storage units and an OKTO probe plus batteries inside, 1.25m heat shield on the bottom of the service bay. Attach science instruments, solar panels and an antenna to the landing stage below that. Lighter than a Mk1 pod and just as capable of taking two copies of Mystery Goo, seismic, gravimetric, even atmospheric fluid readings.

Link to comment
Share on other sites

  • 0

Note that the goo canisters are on the top of the service bay, ie. hanging upside down (I usually put them on the bottom but I recently had a landing where the canisters clipped through the bay and exploded on touchdown, so I decided to try this configuration instead). A few minutes ago, I relocated them to the bottom of the bay, which brought the center of mass significantly closer to the center of lift. However, it still wants to flip and this time, I lost control at 1600-something m/s due to having tried to manually stay on course, followed by falling sideways for several kilometers and hitting the water at 245 m/s, still above parachute release speed (I set the chute to start opening at 50 atmospheric pressure and fully open at 1000, which gives me safe descent speed by 500 m unless I'm coming down over a mountain). According to aero data, the service bay's drag was close to 30 while the pod was at 14.7.

Edited by Fraktal
Link to comment
Share on other sites

  • 0

The craft in question:

Spoiler

LnonRuI.jpg

Reaction wheel torque is at maximum and I'm not carrying any monoprop. The thermometer is on the wall of the service bay opposite of the barometer, the pod doesn't have anything radially attached to it. Parachute is set to arm together with final stage separation. Reentry profile is an 80-something km apoapse on suborbital after 3447 m/s (first stage doesn't have nosecones so it won't reach orbit due to drag loss, which is intentional; this is a suborbital rocket, after all), hitting atmosphere at 1900+ m/s and crashing into the ocean a short distance beyond the Korea-shaped peninsula east of the KSC.

Also, a bit of trivia. I tested a modified version of this rocket where the science bay is replaced with another command pod upside-down as a trainer craft for non-pilot personnel. If I fly sideways to maximize drag, one of the pods explodes from overheating even if reentering from suborbital but if I separate them via opposing-facing decouplers, both pods flip retrograde and reenter properly.

 

Update: Fired up a different save and loaded a different craft with an almost identical part configuration (service bay is 90° rotated and there's an antenna at the tip of the pod, but otherwise it's identical). Result is the same, so it's not a craft bug or save bug. It can reenter, but only with SAS set to retrograde hold because if it deviates even 5° away from retrograde, the drag on the service bay overpowers SAS in less than a second.

Edited by Fraktal
Link to comment
Share on other sites

  • 0
3 hours ago, Tyko said:

Add a heat shield - even if you zero out the ablator. It's still going to bring your CoM down a bit and I think it will help with drag. The bottom of your Service Bay has an open node which is draggy. Adding the heatshield will "cover" that node.

Tried it with the .625 shield and it only made the problem worse, but the 1.25 shield did the trick. Displays the overheat bar during reentry but it doesn't burn away, so all's good. Did see an anomaly where the shield, rated for 9 m/s impacts tops, exploded from landing on it at 6.7 m/s, but it didn't repeat.

I kinda didn't have a choice at using a service bay because the science gear kept burning up about 60% of the time if I mounted them externally, even at the very tip of the pod.

Quote

Add drogue chutes. They'll slow you down and help the pod orient correctly

Tried that approach too, they didn't open at all.

 

 

Funny bit of trivia. WIth all this messing about with service bays, I got curious at how badly would two Mk1 pods attached to the ends of a service bay would work as an improvised rover for harvesting science at the KSC. Answer: surprisingly well. It's impossible to steer while moving, but it rolls way faster than a kerbal on foot, does not overstress and break wheels, does not break stuff from tipping over and holds course much better than a single pod. Good thing kerbals don't get nauseous from spinning around at 120 RPM for a minute straight, or Jeb and Bob would've barfed all over the VAB.

Edited by Fraktal
Link to comment
Share on other sites

  • 0

Would you try one more thing?

Use your RCS buildAid in editor, set it to parachute and temporarily remove the parachutes, see if it shows any rotation. Sometimes this happens and usually is solved with using the translate gizmo set on absolute and click on any one of the 3 arrows without moving it so the part snaps to the dead center.

Link to comment
Share on other sites

  • 0
2 hours ago, Rocket In My Pocket said:

Hmm...wait this is a sub orbital rocket? You aren't just going straight up to 80km then back down are you? That's never a good idea.

Put some arc to your trajectory so you have more time for drag to slow you down.

Of course I'm not flying vertically. I land a couple hundred kilometers away from the KSC.

Link to comment
Share on other sites

  • 0
18 minutes ago, Fraktal said:

Of course I'm not flying vertically. I land a couple hundred kilometers away from the KSC.

That's hard to picture off the top of my head, anyway we could get a shot of the trajectory?

Y'know crazy as it sounds, I used to do sub-orbital probes with jet-engines facing down; to slow them down as they descended so the chutes can pop. Before I learned it was easier to do much shallower trajectories anyways.

Do you have Airbrakes yet? That's always an option too, or two Elevons set to deploy opposite eachother make a great "red-neck" airbrake.

Link to comment
Share on other sites

  • 0
On 9/21/2018 at 4:31 PM, Fraktal said:

The craft in question:

  Hide contents

LnonRuI.jpg

 

 

Assuming a pure stock game: place an Mk1 pod. Attach an Mk16 parachute on top, a 1.25m service bay below, and a 1.25m heatshield under the bay.

Drain the pod monoprop. Then using the offset gizmo in absolute mode, shift the service bay one 'tick' up, leaving no gap between the edge of the pod and the service bay. The bottom of the pod ends up inside the service bay, becoming the new curved ceiling of it. It makes no difference for drag calculations, but the CoM is moved down, which is the most important consideration here.

That combination is very stable in retrograde reentry. Even when plunging in at >3100 m/s from near the Mun to a Pe of 15km, with SAS off and tumbling around, it will still orientate itself retrograde before heat causes damage, and it will slow down enough for chute deployment well before reaching the surface (assuming proper chute settings).

 

Video inside the spoiler of how to build it and how it points itself retrograde even when reentering the atmosphere tumbling:

Spoiler

 

If it doesn't work for you like that, something in your game is no longer stock.

 

Edited by swjr-swis
highlight video offered
Link to comment
Share on other sites

  • 0
On 12/16/2018 at 2:55 AM, JeyJeyKing said:

New player here, having the exact same issue.

Did you look at @Tyko's suggestion (and my video, which amounts to the same thing)? The answer hasn't changed since this question was asked - it works without fail.

Welcome to the forum btw.

Edited by swjr-swis
dutchisms
Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...