[0.20] Deadly Reentry 2.3 - reentry heat, plus thermal and g-force damage to parts

[Ratzap]

So is there any reason my normally designed rocket gets G-Force damage before takeoff? It's really starting to piss me off.

Have you never noticed that KSP sort of dumps your rocket on the pad and it jiggles and sways? In 0.20 I've had things snap off and the crash screen would say 5.6G max experienced or something just from that drop onto the pad. Since the drop and jiggle still happens I'd say that's your problem. It happens to me too. I've been messing around with mission controller and some of my rockets for those simply exploded on the pad before I got to do anything. Revert to VAB, come out and try again :-/

Having removed deadly reentry for the time being the random explosions have stopped.

[corumanime]

Which is why we need options to disable or configure the amount of g damage, and still use the heat part which is what we all want it for.

Has the author abandoned this mod? OP hasn't been updated since June. :/

[NathanKell]

The author is busy with some bad stuff in RL, per the ModuleManager (or was it Modular Fuels?) thread. Will be back as soon as can though.

[birrhan]

I understand that author is busy with bad stuff IRL, but I'm having some serious issue with FAR compatibility using chutes and heat shields on larger-than-normal craft (16 tons or so).

Most times it can get through the firey part of reentry. The issue comes with the chutes. Any combination of chutes induced DBG (death by g force)--drogue or main chutes, in tandem or in sequence. I have deployed at 1000 m/s and at 500 m/s. All induce g force mediated disassembly very very quickly.

So. Tips or tricks would be appreciated. I understand that the mod is designed to be FAR compatible, but there are a whole host of issues that still crop up, this being the most recent. (the 6.25 m heatshield is still really buggy too).

[togfox]

I have deployed at 1000 m/s and at 500 m/s. All induce g force mediated disassembly very very quickly.

If you have fuel left you can thrust so that your speed is reduced to something less than 500m/s. That might reduce g forces. I realise chutes are intended to save on fuel - and drogues are intended to slow one down with out stopping them but if you're deploying at 500m/s then that is 1,800 km/hr or 1,100 miles/hour!!! If there is an atmosphere then let the atmosphere slow you down by having a shallower re-entry. I deploy stock chutes ~250 m/s where possible.

[sirklick]

birrhan - 500 m/s is over 1100 miles per hour (way too fast for a parachute), the stock parachute has too much surface area for use at that velocity. There are a couple of possible solutions to your problem.

1. Slow yourself with a burn (100m/s or so) and then deploy main 'chute.

2. Use the lower resistance parachute first and use a radial main

3. Come in less steep and give yourself more time in atmosphere to slow down.

You also may want to look into mod parachutes since KSP is severely lacking in parachute diversity. Drogue parachute ex. http://forum.kerbalspaceprogram.com/showthread.php/28063-Radial-Drogue-Parachute

[birrhan]

If you have fuel left you can thrust so that your speed is reduced to something less than 500m/s. That might reduce g forces. I realise chutes are intended to save on fuel - and drogues are intended to slow one down with out stopping them but if you're deploying at 500m/s then that is 1,800 km/hr or 1,100 miles/hour!!! If there is an atmosphere then let the atmosphere slow you down by having a shallower re-entry. I deploy stock chutes ~250 m/s where possible.

Understood, it just seems excessive. I mean, ialdabaoth has stated earlier that chutes ought to be deployable below 1000 m/s without damage. And he's also stated that DRE is FAR compliant. Given those two things, I see a bug.

[jrandom]

Understood, it just seems excessive. I mean, ialdabaoth has stated earlier that chutes ought to be deployable below 1000 m/s without damage. And he's also stated that DRE is FAR compliant. Given those two things, I see a bug.

I've easily deployed one drogue and four main radials on the Mk1-2 pod w/ heat shield going 800m/s. Some creaking, but nothing rips off. (This is w/FAR, BTW. It's been so long since I've played stock atmosphere I can't comment on chute behavior there.)

Edited August 22, 2013 by jrandom

[birrhan]

birrhan - 500 m/s is over 1100 miles per hour (way too fast for a parachute), the stock parachute has too much surface area for use at that velocity. There are a couple of possible solutions to your problem.

1. Slow yourself with a burn (100m/s or so) and then deploy main 'chute.

2. Use the lower resistance parachute first and use a radial main

3. Come in less steep and give yourself more time in atmosphere to slow down.

You also may want to look into mod parachutes since KSP is severely lacking in parachute diversity. Drogue parachute ex. http://forum.kerbalspaceprogram.com/showthread.php/28063-Radial-Drogue-Parachute

Already using drogues with radials. Or just radials. or just drogues. They never get a chance to inflate.

For now I want to keep things as stock as possible. There really needs to be a radial drogue. Though I feel like it will run into the same problem.

[birrhan]

I've easily deployed one drogue and four main radials on the Mk1-2 pod w/ heat shield going 800m/s. Some creaking, but nothing rips off. (This is w/FAR, BTW. It's been so long since I've played stock atmosphere I can't comment on chute behavior there.)

This is with FAR as well. As you can see, we're just shy of 3x heavier than a fully kitted Mk1-2 pod (about 5.5 tons)--around 15 tons. So something's got to give....

so I fitted a retrobooster module underneath. it adds about 8 tons, bringing us to 24 tons (eeek!) but GREAT SUCCESS--Kerbal style landing.

[birrhan]

I just checked the screenshots from my last munar mission in 0.20.2, the only one with FAR + DRE + KER. I don't remember the exact details, but I have a shot of my capsule doing 3061 m/s at about 31 km. The orbit is 10000x10.3 km at this point, and the heat shield just lost 22 units (out of 1000). To bad I don't remember to what I set the periapsis, but I guess 15 km, or a wee bit more?

The next shot 36 s later shows me at 15.8 km, with a periapsis of -3.5 km, half the heat shield is gone and I'm still going 2450 m/s. The temp is 880 at the shield and I'm pulling ~4 g. In the next shot, 10 seconds later, I'm at 13.1 km with 2000 m/s, pulling 5 g and the heat shield is pretty much gone, but the heat is down to 470°. Two seconds later, the shield was spent and heated up to 1470°. At this point I'm still pulling 5 g at 12.5 km, 1820 m/s. At 2.5 km, the three radial chutes can be seen partially deployed, I was doing 245 m/s with 2 g.

That was propably the hottest reentry I ever did with DRE+FAR, maxing out both the g-margin and the heat shield. Riding the edge, a real nailbiter! I guess I came in at the typical speed for munar returns of about 3150 m/s. If you bleed off some speed with an aerobrake at 35--45 km, you won't loose much of your heat shield and get more elbow room with the g-loads.

So I'm doing essentially this, and it definitely doesn't work. Pod is a Mk 1-2 pod with a docking clamp, 2 radial shields, a tiny battery, an RTG, and a radial mono tank--total weight when the shield gives out is 5.3 tons.

Apo=5.0 million km (0.5 mun distance), peri=12 km, speed 2988 m/s. Ablation went out at 13 km and immediately got destroyed by G force damage (speed was 1465 m/s).

The pod will actually survive a peri of about 22 km, same apo, same speed (actually a little faster). the killer is that when the ablation shield gives, SAS overreacts, starts pitching into the flames, and the chutes burn up before I can even deploy them. It happens VERY quickly, less than a second, I don't have time to manually control it back, even with Caps lock engaged. I'd say I'm still doing 1700 m/s at around 15 km when the shield goes there.

I'm asking you because you've actually done it. Hopefully it's not a problem with SAS being too ginger nowadays in 0.21.1.... and if so, what's a good way to combat this?

[vardicd]

There is an airbrake part included in the b9 aerospace pack. it mounts on the side of parts and is a godsend for slowing down during reentry. it will even work just two or three attached directly to the command capsules, if that's all that's reentering. just set it to an action group and deploy. keep them open until chutes are fully deployed. its made so many of my craft that couldn't survive, make it down.

[p1t1o]

Every problem with DRE (w/FAR) can be solved in the following ways:

higher in-atmosphere periapsis

come in from a lower apoapsis

move centre of mass toward heatshield (you are essentially an aircraft with all the drag at the front, this is less than stable without taking this into account)

increase aerodynamic stability (eg: add fins or brakes to the rear)

otherwise increase control authority (rcs, gyros etc.)

deploy chutes at a lower velocity or employ brakes or drouges to help slow down.

If you cant achieve this, you need a re-design. In this way, if you are deorbiting something small, you can even avoid the "fiery" phase entirely.

[Lexif]

So I'm doing essentially this, and it definitely doesn't work. Pod is a Mk 1-2 pod with a docking clamp, 2 radial shields, a tiny battery, an RTG, and a radial mono tank--total weight when the shield gives out is 5.3 tons.

Apo=5.0 million km (0.5 mun distance), peri=12 km, speed 2988 m/s. Ablation went out at 13 km and immediately got destroyed by G force damage (speed was 1465 m/s).

Is that

- Pod w. shield & docking port

- RTG, roundified monoprop tank and small battery stuck near the top, balanced(!)

- 2 radial parachutes?

Because that is 5,5 t for me, but I'm not sure if Ioncross Crew Support adds some mass to the pod.

I'm going to try it this evening and tell you the result. But if you flip over, it surely is because of the crap that's stuck to the pod. With only a shield, a docking port and three parachutes, it's aerodynamically stable with the shield forward, FAR takes care of that. I don't even use SAS once I'm feeling the drag.

Edit:

Every problem with DRE (w/FAR) can be solved in the following ways:

higher in-atmosphere periapsis

(...)

If you're burning up, a lower PE will help. If you're experiencing g-force destruction, a higher PE is needed.

Edited August 27, 2013 by Lexif

[p1t1o]

If you're burning up, a lower PE will help.

Thats odd, thats the opposite of my experience, it also makes little physical sense (agree with the g-force statement though).

[asmi]

The solution to the chute problem is to activate it once subsonic. No chute can survive when deployed at 3M.

[Ralathon]

Thats odd, thats the opposite of my experience, it also makes little physical sense (agree with the g-force statement though).

The idea behind burning up with a high Pe is that you spend so long slowing down your heat shield gets used up. After that bad stuff happens.

It seems that DRE models the loss of the ablative shield based on altitude or it is nonlinear with the deceleration, since I experience the same, my heat shield runs out at slower speeds when I keep the Pe high. The effect isn't massive or anything, but enough to be noticeable.

[BubbaWilkins]

So I'm doing essentially this, and it definitely doesn't work. Pod is a Mk 1-2 pod with a docking clamp, 2 radial shields, a tiny battery, an RTG, and a radial mono tank--total weight when the shield gives out is 5.3 tons.

Apo=5.0 million km (0.5 mun distance), peri=12 km, speed 2988 m/s. Ablation went out at 13 km and immediately got destroyed by G force damage (speed was 1465 m/s).

The pod will actually survive a peri of about 22 km, same apo, same speed (actually a little faster). the killer is that when the ablation shield gives, SAS overreacts, starts pitching into the flames, and the chutes burn up before I can even deploy them. It happens VERY quickly, less than a second, I don't have time to manually control it back, even with Caps lock engaged. I'd say I'm still doing 1700 m/s at around 15 km when the shield goes there.

I'm asking you because you've actually done it. Hopefully it's not a problem with SAS being too ginger nowadays in 0.21.1.... and if so, what's a good way to combat this?

FAR has had some drag code updates recently, so you may experiencing more drag induced deceleration g-forces. You need to spend more time in the upper atmosphere bleeding off speed gradually instead of punching through the upper layers and hitting the dense air in the lower atmosphere. Better to establish a lower orbit and utilize a much shallower re-entry curve. You'll have lower temps overall, less g-forces, and your craft will be less likely to get all explodey.

Every problem with DRE (w/FAR) can be solved in the following ways:

higher in-atmosphere periapsis

come in from a lower apoapsis

move centre of mass toward heatshield (you are essentially an aircraft with all the drag at the front, this is less than stable without taking this into account)

increase aerodynamic stability (eg: add fins or brakes to the rear)

otherwise increase control authority (rcs, gyros etc.)

deploy chutes at a lower velocity or employ brakes or drouges to help slow down.

These are all secondary to the primary anti-explodey rule of FAR/DE - which is SLOW DOWN before you start pushing the G-limits.

The idea behind burning up with a high Pe is that you spend so long slowing down your heat shield gets used up. After that bad stuff happens.

It seems that DRE models the loss of the ablative shield based on altitude or it is nonlinear with the deceleration, since I experience the same, my heat shield runs out at slower speeds when I keep the Pe high. The effect isn't massive or anything, but enough to be noticeable.

Drag increases exponentially with Speed.

Drag produces friction with the atmosphere.

Friction produces heat.

Heat eats your shielding.

What you are seeing is that the higher PE allows you to bleed more speed off in the upper atmosphere. This results in you going slower when you hit the more dense atmosphere thus lowering temperatures throughout and increasing the duration your shield remains intact.

Edited August 27, 2013 by BubbaWilkins

[birrhan]

The solution to the chute problem is to activate it once subsonic. No chute can survive when deployed at 3M.

I'm not sure you understand. It never gets a chance to deploy; when the shield overheats, the craft wobbles, tipping the (still packed) chutes into the fire, at which point they disintegrate.

[jrandom]

I'm not sure you understand. It never gets a chance to deploy; when the shield overheats, the craft wobbles, tipping the (still packed) chutes into the fire, at which point they disintegrate.

Make sure you haven't made your pod top-heavy or it will flip around every time. The trick is to get the radial chutes attached above the center of mass while keeping the center of mass as close to the head shield as possible.

[birrhan]

FAR has had some drag code updates recently, so you may experiencing more drag induced deceleration g-forces. You need to spend more time in the upper atmosphere bleeding off speed gradually instead of punching through the upper layers and hitting the dense air in the lower atmosphere. Better to establish a lower orbit and utilize a much shallower re-entry curve. You'll have lower temps overall, less g-forces, and your craft will be less likely to get all explodey.

These are all secondary to the primary anti-explodey rule of FAR/DE - which is SLOW DOWN before you start pushing the G-limits.

Drag increases exponentially with Speed.

Drag produces friction with the atmosphere.

Friction produces heat.

Heat eats your shielding.

What you are seeing is that the higher PE allows you to bleed more speed off in the upper atmosphere. This results in you going slower when you hit the more dense atmosphere thus lowering temperatures throughout and increasing the duration your shield remains intact.

So I didn't explicitly state it, but this is coming back from deep space. Without a retrograde burn we hit atmo at 3600 m/s. I can slow this to 2900 m/s, which puts us in an elliptical ~5,000 x 22 km orbit. This is within the tolerances that Lexif had quoted, and why I was asking him for more specifics about how he worked his magic. It appears you need to get to a point that the heatshield never goes over max temp (1400 IIRC), or if it does, to make sure your craft is balanced to a T. See below. So, while I agree with everything you say in principle, some situations don't allow for ideal reentry profiles. Aerobraking is particularly troublesome for the FAR/DRE combo, and aerobraking is how I imagine it would have been done IRL (and how you are suggesting it be done). That said, even at optimal (slowest) speeds, you need to bleed off around 500 m/s to get into Kerbin orbit (<3100 m/s). Which requires a substantial amount of atmosphere. And heat. Just to get an elliptical supra-Minmus-apoapsis orbit (40,000 km?). Multiple passes are just fine, it's the initial capture that sets constraints on the upper bounds of what's allowed.

Is that

- Pod w. shield & docking port

- RTG, roundified monoprop tank and small battery stuck near the top, balanced(!)

- 2 radial parachutes?

Because that is 5,5 t for me, but I'm not sure if Ioncross Crew Support adds some mass to the pod.

I'm going to try it this evening and tell you the result. But if you flip over, it surely is because of the crap that's stuck to the pod. With only a shield, a docking port and three parachutes, it's aerodynamically stable with the shield forward, FAR takes care of that. I don't even use SAS once I'm feeling the drag.

Yeah, but bleed out the mono, that'll get you to 5.3. Turns out I had 10 units left, I bled out the remainder, and it balanced nicely. That became a non issue, because I found a sweet spot where, even though ablation burned off, I was slow enough to do a suborbital hop up then back down, and temp never went critical (1370 or so--max is 1400?). So the shield stayed put, it never wobbled, and I was able to deploy chutes at ~3000 m altitude, 600 m/s, deployed at 200 m/s, and lost everything to g force damage--except the pod.

Razor's.

Edge.

You were probably right about the slight imbalance though. Good call. Thanks.

[BubbaWilkins]

So I didn't explicitly state it, but this is coming back from deep space. Without a retrograde burn we hit atmo at 3600 m/s. I can slow this to 2900 m/s, which puts us in an elliptical ~5,000 x 22 km orbit. This is within the tolerances that Lexif had quoted, and why I was asking him for more specifics about how he worked his magic. It appears you need to get to a point that the heatshield never goes over max temp (1400 IIRC), or if it does, to make sure your craft is balanced to a T. See below. So, while I agree with everything you say in principle, some situations don't allow for ideal reentry profiles. Aerobraking is particularly troublesome for the FAR/DRE combo, and aerobraking is how I imagine it would have been done IRL (and how you are suggesting it be done). That said, even at optimal (slowest) speeds, you need to bleed off around 500 m/s to get into Kerbin orbit (<3100 m/s). Which requires a substantial amount of atmosphere. And heat. Just to get an elliptical supra-Minmus-apoapsis orbit (40,000 km?). Multiple passes are just fine, it's the initial capture that sets constraints on the upper bounds of what's allowed.

Yeah, but bleed out the mono, that'll get you to 5.3. Turns out I had 10 units left, I bled out the remainder, and it balanced nicely. That became a non issue, because I found a sweet spot where, even though ablation burned off, I was slow enough to do a suborbital hop up then back down, and temp never went critical (1370 or so--max is 1400?). So the shield stayed put, it never wobbled, and I was able to deploy chutes at ~3000 m altitude, 600 m/s, deployed at 200 m/s, and lost everything to g force damage--except the pod.

Razor's.

Edge.

You were probably right about the slight imbalance though. Good call. Thanks.

My main point isn't that your re-entry profile isn't optimal...its suicidal. By your own admission, you were travelling at 600 m/s at 3000m of altitude while Lexi was nearly half that

At 2.5 km, the three radial chutes can be seen partially deployed, I was doing 245 m/s with 2 g.

The bottom line is you were still going to fast when your chutes fully deployed at 500m. The resulting jolt ripped your craft apart. So you either need to figure out how to slow down before, during, or after re-entry. It's that simple.

[Lexif]

So I tried your setup, Birrhan. Nothing was destroyed.

First, about the craft. After flying, I just had a look at the CoL-indicator in the VAB. I balanced the masses, but the CoL is pretty far set off to the side by the RTG. With just shield/pod/radial parachutes/docking port, the CoL is just a tiny bit higher than the CoM, and centered. With this combo, the CoL travels pretty far up (which should be good, right?), but also to the side. Here is a pic:

Then, I didn't drain the tank, but mass wasn't the problem.

I had a look and my old standard ship with an SAS below the docking port and three radial parachutes, and it comes in at 6180 kg, about 700 kg more than this capsule setup.

So, how did it go? I went for a 5200 km x 12 km orbit. The capsule wanted to roll over, but the internal SAS was able to keep it below 15° deflection. I lost all of the ablative at about 13 km, like you, but the temp on the shield never went above 1090° because at that point peak heating was over. Maximum deceleration was about 6.2 g, but that didn't do damage. I pulled the two chutes out at about 5 km (I landed at about 500m above the sea), doing maybe 400--500 m/s. Didn't get the exact numbers, but at 6.2 km I was doing 554 m/s, and shortly after the chutes half-deployed, I was at 4.2 km doing 370 m/s. At 550 m above ground, before full deployment, I got 160 m/s. The only groaning sounds came when the chutes fully came out. An after flight inspection by Hadster Kerman, my dumbest Kerbal (and therefore, most qualified test pilot) revealed "critical damage" to the RTG and "severe damage" for the batteries. The parachutes and the pod showed moderate damage at most.

Here is an imgur-album with some snapshots. (How do I embed the album so nicely, as some people seem to be able to do?)

http://imgur.com/a/bBDl4

Edited August 27, 2013 by Lexif
Added CoL picture

[Lexif]

My main point isn't that your re-entry profile isn't optimal...its suicidal. By your own admission, you were travelling at 600 m/s at 3000m of altitude while Lexi was nearly half that

The bottom line is you were still going to fast when your chutes fully deployed at 500m. The resulting jolt ripped your craft apart. So you either need to figure out how to slow down before, during, or after re-entry. It's that simple.

Yes, I deployed the chutes when I went somewhat below 600 m/s and that slowed me down enough to survive full deployment at 500 m (or 1000 m above sea level in my case). The parachutes won't burn off once the thermal effects are gone, there's no reason to wait to get below Mach 1.

[jrandom]

I've been having some luck just deploying the drogue chute as soon as I drop below 800m/s. It fully deploys at 2km and slows the capsule down enough that when I deploy the radial chutes and they open at 500m, there's no g-force damage done to the craft.