Investigation into new decoupler failure.

[Midnight Star]

Sometimes, the large fuel tanks simply fail and everything goes to hell.

*jumps into the middle of the thread while only reading the first 10 or so posts*

I\'ve gotten a rocket up without a decoupler up and it works fine. As soon as I strap on that decoupler, when the inner tanks are almost empty, the decoupler fails and everything blows up into smitherings. The tanks themselves don\'t blow up, it says that the decoupler fails.

[cardgame]

I haven\'t had any of these problems... attach cross-struts between any hangers-on and attach secondary supports between the tops/bottoms and the center of your rocket. Very little, if any, wobble.

[Ziff]

It has very little, if anything, to do with a gimbaled engine. It is a combination of a known bug (old part strength on new decoupler) and the amazing amount of thrust we can now easily obtain. Mount a few of those new engines, turn off ASAS and fly straight up, watch your G-forces as go above 10km. They will slowly start to climb, and at around 4-5G\'s that decoupler will go byebye. If you don\'t input any commands while in orbit you can also prove it has nothing to do with the gimbaled engine and everything to do with G-forces, and not Max Q as others have mentioned.

Also, even without decouplers you can still experience failures. On several larger rockets I have had a full tank with an engine under it push right up through 2 empty tanks, on more than one occasion even with what I believed was adequate bracing. But again, it was in extremely high G-force conditions.

-Ziff

[samstarman5]

I was just starting to think maybe the large engine that is gimbaled might be a factor, but thanks for the details, Ziff!

[rkman]

It has very little, if anything, to do with a gimbaled engine.

A little, yes but definitely something. The extreme wobble that multiple powerful gimbal engines cause, does not help to keep a craft in one piece. It will also fail without gimbals, but with gimbals it fails sooner.

I\'ve gotten a rocket up without a decoupler up and it works fine.

I also tested rockets (with four large engines) without decouplers, and fuel tanks fail fairly consistently.

[grynspan]

Silly thing told me there was only the first page of discussion... never mind...

[Randox]

Another thing that contributes to problems with the large decoupler in combination with the new large engines is the fact that there is only a gimbal version of that engine.

From 015 and before we know that it is usually not wise to put only gimbal engines on the first stage (if it\'s a heavy rocket) because all that gimbaling puts very strong lateral forces on the craft, which can cause the craft to break up during ascent.

Because there is no non-gimbal version of the new large engine, the most powerful rockets that can be build in 0.16 do experience the strong lateral forces that could be avoided in 0.15 by using some non-gimbal engines instead of only gimbal engines. (though that\'s nothing that a bit of copying and cfg editing can\'t solve)

The lateral forces do seem to be significant. I know my rocket looks like its shaking pretty bad by the time it fails, if I don't scale back the thrust. Also important is that it causes at least my rocket to jump. The g force meter will just bounce between say 1.5 and 2.25 more than once per second, which means very very high jerk values (m/s^3). I feel like it might also be these high jerk values that cause the part to become crushed, and it could probably handle higher g loads if teh rocket was flying smoothly.

And struts aren't always the answer. I rebuilt my large rocket to clean up the struts, making an identical ship but with honestly better strut placement, and it could never make it into orbit. Even using more struts, it could never make it, which means I am stuck making modifications to my first version because by some fluke it can actually get into space.

I also get sessions where it just keeps exploding, mush sooner than it should, for no reason. When I tried to take a video of it so I could watch a particular failure (one of my fuel tanks always seems to fail before the separation ring, and I want to know which one it is), the rocket flew right up into space at full throttle, taking abuse that should have blown it to pieces.

What I want to do is scale up the strength of the ring a bit, but I can't even find that part in the directory >.>

[rkman]

What I want to do is scale up the strength of the ring a bit, but I can't even find that part in the directory >.>

The large decoupler is in \Parts\decoupler1-2\

[bsalis]

And struts aren't always the answer. I rebuilt my large rocket to clean up the struts, making an identical ship but with honestly better strut placement, and it could never make it into orbit. Even using more struts, it could never make it, which means I am stuck making modifications to my first version because by some fluke it can actually get into .

Agree. Big parts need more struts but placement matters. I found that with a persistent fault point, that after several struts, if it's not fixing it, remove them all and start again with another aproach. I believe the reason is because of how struts redirect stress forces around the craft. So struts can redirect stress to a weak point instead of away.

[Ziff]

Agree. Big parts need more struts but placement matters. I found that with a persistent fault point, that after several struts, if it's not fixing it, remove them all and start again with another aproach. I believe the reason is because of how struts redirect stress forces around the craft. So struts can redirect stress to a weak point instead of away.

It has a lot to do with rocket design. If you are using the new rocket parts the same way you did the old ones, you need to start calculating your thrust to weight ratio. Slapping 3 or 4 large tanks with an engine and then putting 4 more copies around it is not a good way to use the new parts. I have put some seriously large landers up into orbit with only 2 of the new tanks and engine in the ascent stage, the rest are 1m parts.

[vexx32]

May we see a few designs, please? I can't seem to make them work for me xD

[bsalis]

May we see a few designs, please? I can't seem to make them work for me xD

An example is here

[CrashnBurn]

The large decouplers were not properly scaled. The strength is identical to the small decouplers. Given the weight and diameter increase a minimum of 10x strength increase is indicated, with 20x + more reasonable..the weight is way out of proportion to the strength. Can you make a rocket with the weak decouplers? Yes, I've done it. But that doesn't mean they are properly scaled.

[Braun]

For me, both couplers break when breaching around 130meters per second - Which appears weird no matter how you look at it.

[lstutzman]

The large decouplers were not properly scaled. The strength is identical to the small decouplers. Given the weight and diameter increase a minimum of 10x strength increase is indicated, with 20x + more reasonable..the weight is way out of proportion to the strength. Can you make a rocket with the weak decouplers? Yes, I've done it. But that doesn't mean they are properly scaled.

I agree 100%. I believe when the new scaled up parts were created, they simply forgot to scale up the breakingForce and breakingTorque from the default values. It only makes sense that you if have new larger parts with thicker material, they should be stronger as well.

Also, most of the successful big lifter designs I've seen, look like an explosion in a girder factory ( I borrowed that one from someone here . NASA, RSA, ESA, JSA - none of them have this much extraneous support structure on real designs.

[Ziff]

May we see a few designs, please? I can't seem to make them work for me xD

Sorry, I somehow missed this post before.

[

This makes a great comparison to the design Bsalis posted, and helps show part of the problem with those new engines. No offense meant Bsalis, I actually like your design. I just gave up on using the new engines in lateral stages when I realized how inefficient they are.

The Discovery-Nova craft has a total mass of 1154tons and gets 115tons into orbit. The lander itself weighs 56tons.

My IPX-X4 craft has a total mass of 478tons and gets 120tons into orbit. The lander weighs 54tons.

Edit: How do you get the attachments to post? I uploaded a copy of my .craft file but it did not appear in the post. Hmm. Linked with dropbox, for now.

IPX-X4: https://dl.dropbox.com/u/93355756/IPX-X4.craft

Edited August 15, 2012 by Ziff

[spikeyhat09]

I think it happens when the 1200 thrust of the new engine "crushes" the wimpy new decoupler, which despite being bigger, has the same specs as the old on. I can usually fix this for my rockets by putting 2 decouplers on top of each other

[bsalis]

This makes a great comparison to the design Bsalis posted, and helps show part of the problem with those new engines. No offense meant Bsalis, I actually like your design. I just gave up on using the new engines in lateral stages when I realized how inefficient they are.

No offence taken. Asparagus style staging is more efficient, and yeah, the 1m engines are better. The aerospike in particular is a bit overpowered IMO. But sometimes it's not about efficiency... sometimes it's simply about MOAR. So now I present to you, my next project...

It's not finished. Like... no BOOSTARS!? Oh, and it's not orbit capable, because it, always, explodes. Often after parts fall off. All minor problems in the Kerbal Space Program. Will get there eventually... Dare Stupid Things. Keeps me busy too.

Oh, yeah, on-topic. The big decouplers are tin cans. We can expect more tweaks I think. IMHO, Aerospike OP, small SRBs OP (after being too weak in 0.15!), big T30s have terrible in-atmo ISP.

[Ziff]

The aerospike in particular is a bit overpowered IMO. But sometimes it's not about efficiency... sometimes it's simply about MOAR. So now I present to you, my next project...

Yeah, I agree with you about the aerospike. I half feel like it's cheating, so I tend not to use it. That is one huge rocket! Wow. So have you modified the part.cfg file for the large decoupler or no? I upped the strength to 120 I think and it pretty much stopped the decoupler from being crushed. I hope you get that thing working, I would love to see it in orbit!