Rockets falling over, engines dropping off, oh my!

[NoXion]

Hi everyone,

I've been playing around with KSP for a couple of months now, but it's interesting how long the learning curve is. Although I've achieved both stable orbits and a powered landing (although I fell over... ), I still have a long way to go before I can achieve consistent results, both in terms of design and piloting.

A symptom of this that I'm finding particularly irritating is the tendency for my larger rocket designs to start listing fairly rapidly, despite having multiple beefed-up stabilisers and a symmetrical design. Sometimes I'm able to wrestle back control, but by the time that happens I've so used up so much fuel and gone through too many stages to do anything other than a sub-orbital arc.

Another problem is the seemingly inconsistent strength of parts - some of my large rockets manage to keep hold of their engines despite wobbling wuite badly on the way up, while other times the rocket takes off using its boosters - and leaves the main engine on the launch pad! o_O

I can't provide any details of the designs I'm using because little I've made so far I've considered worth saving. I'm a bit of a perfectionist. But I would appreciate some general tips from experienced builders of larger rockets. I have a large supply of custom parts, all of them awesome work by the way.

[foamyesque]

Well, leaving the rockets behind on the launch pad is because of the sticky pad bug, not anything you're doing. You can work around it by putting a decoupler under each stack that would normally rest on the ground. Stage them so that the decoupler and the engines igniting occur at the same time, and as long as you have a decent TWR, you'll lift off and be clear before your rocket exhaust makes them explode.

[NoXion]

Well, leaving the rockets behind on the launch pad is because of the sticky pad bug, not anything you're doing. You can work around it by putting a decoupler under each stack that would normally rest on the ground. Stage them so that the decoupler and the engines igniting occur at the same time, and as long as you have a decent TWR, you'll lift off and be clear before your rocket exhaust makes them explode.

The thing is, I've had engines fall off my rockets at times other than take-off. Another problem I encounter is with really powerful stages; if I'm over-generous with the throttle, it can get 'stuck', which greatly increases the chances of my Kerbalnaughts exploding in mid-air or crash-landing, due to being forced to either try and wrestly back control until the fuel runs out or attempt to eject the faulty stage (which can be a dangerous since there's a good chance the ejected but still-burning stage will simply accelerate through you).

Thoughts: The throttle sticking issue appears to be a natural consequence of my tendency to over-engineer things - One of my latest designs, and the one with which I can say I am most pleased with performance-wise, is the Kratos MkI, which has as the lowest stage a cluster of nine E-600 engines with 'big red' SAS modules. I reckon the best solution here is to just take care with the throttle. With great power comes great responsibility. However, in the event that a stage does get stuck in such a manner, the thought occurs of including a more explosive coupler or even a dedicated rocket motor to ensure a safe seperation distance.

Here's an image of the Kratos MkI, perhaps the first truly stable large rocket design of mine:

This thing is powerful enough to get into LKO without expending any stages:

[rdfox]

I try to avoid the E-series engines; they're impossibly powerful for their weight and fuel consumption, but that's your choice.

One thing that can happen is vibration sets in (pogo or otherwise), eventually breaking engines off. One way to deal with that is using the EAS-4 struts to help lock the engines down. Another trick you might try to help deal with the sticky-pad bug is using SRBs or the 'stage-and-a-half' concept, where you have extra power for the early part of flight, then jettison them and continue on only your main engines, reducing weight and fuel consumption after getting past the phases of flight that require maximum power.

As for tipping over, that happens most often with top-heavy rockets, particularly ones with low thrust-to-weight ratios. You may need to find a way to move weight further aft, or add a Yawmaster to manually assist the SAS. Of course, extra-long rockets also tend to have a lot of sway due to stack flex. (My VK Nova Mark Two is incapable of a vertical climbout during first stage burn; it tends to settle in about 2-3 degrees off vertical in a random direction, then straighten up after first stage separation.) If that's the problem, you pretty much have to live with it... or find a way to make the rocket shorter and/or stiffer.

[foamyesque]

The 'stuck throttle' issue sounds like your engines are tearing their connections free, severing your control, but continue to bulldoze your rocket forward; it's exactly the same effect that occurs if you fire a stack decoupler before a liquid stage has finished burning. Struts, careful throttle management, or avoiding turbocharged engines are also possibilities for avoiding it.

[Sordid]

Another problem is the seemingly inconsistent strength of parts - some of my large rockets manage to keep hold of their engines despite wobbling wuite badly on the way up, while other times the rocket takes off using its boosters - and leaves the main engine on the launch pad! o_O

That's because the weight of the rocket is crushing the engine (~breaking it off from the rest of the rocket). What you can do is move the boosters downwards so that the rocket rests on them and not on the main engine. Example:

When building very large rockets, always try to distribute the load between as many parts as possible. If you put all the weight on a single part, you'll break it off.

[foamyesque]

Vertical snapping is a huge help with that; you can turn it on by changing a flag in th settings.cfg file.

[NoXion]

Thanks for the replies and advice everyone.

So what kind of values are 'realistic' for rocket performance? I notice that the Omnipotent Bertha is even more powerful than the E-1000, although admittedly it's physically bigger as an engine.

[Lawn Dart]

I've also had success using the lunar landing legs to hold the rocket up and away from the pad. Looks silly but it works.

[rdfox]

Thanks for the replies and advice everyone.

So what kind of values are 'realistic' for rocket performance? I notice that the Omnipotent Bertha is even more powerful than the E-1000, although admittedly it's physically bigger as an engine.

It's not the thrust level that's implausible for the E-series, it's the rate of fuel burn and the thrust to weight ratios. You'll note that the Omnipotent Bertha burns fuel a lot faster and weighs a lot more than the E-1000.

Right now, truly realistic values wouldn't work very well due to issues with game balance (we're still early in testing, after all), but Sunday Punch and NovaSilisko's packs are pretty good, and the Big Effen Engines 'feel' pretty good by that measure, too. Anyone else got suggestions on that?

[NoXion]

It's not the thrust level that's implausible for the E-series, it's the rate of fuel burn and the thrust to weight ratios. You'll note that the Omnipotent Bertha burns fuel a lot faster and weighs a lot more than the E-1000.

How implausible are we talking here? nuclear salt-water rocket, or closer to anti-gravity? Because if it's too implausible then that kind of spoils the fun for me.

[Millitron]

The thing is, I've had engines fall off my rockets at times other than take-off. Another problem I encounter is with really powerful stages; if I'm over-generous with the throttle, it can get 'stuck', which greatly increases the chances of my Kerbalnaughts exploding in mid-air or crash-landing, due to being forced to either try and wrestly back control until the fuel runs out or attempt to eject the faulty stage (which can be a dangerous since there's a good chance the ejected but still-burning stage will simply accelerate through you).

Thoughts: The throttle sticking issue appears to be a natural consequence of my tendency to over-engineer things - One of my latest designs, and the one with which I can say I am most pleased with performance-wise, is the Kratos MkI, which has as the lowest stage a cluster of nine E-600 engines with 'big red' SAS modules. I reckon the best solution here is to just take care with the throttle. With great power comes great responsibility. However, in the event that a stage does get stuck in such a manner, the thought occurs of including a more explosive coupler or even a dedicated rocket motor to ensure a safe seperation distance.

Here's an image of the Kratos MkI, perhaps the first truly stable large rocket design of mine:

This thing is powerful enough to get into LKO without expending any stages:

Using more struts to connect the lower stages to the upper stages will spread the forces across the whole rocket, so when a new stage lights, the huge change in acceleration doesn't tear the rocket apart.

[Millitron]

I've been having this problem with Sunday Punch's latest pack. My engines keep falling off the two-meter tanks, and I think it has something to do with the 2 meter 1-2 adapter. I can see that adapter shake around pretty violently just before the engines break, or even pop right off.

[sirmonkey]

I've been having this problem with Sunday Punch's latest pack. My engines keep falling off the two-meter tanks, and I think it has something to do with the 2 meter 1-2 adapter. I can see that adapter shake around pretty violently just before the engines break, or even pop right off.

go into the cfg for those parts and play around with the Mass on them. the default setting is to low and is causing them to shake around.

[rdfox]

Or just apply three struts to connect the adapter to the tank, and two to connect the engines to the tank. Makes the parts rock-solid.

And I'm not sure, Noxion. You'd have to ask Doctor Evo, he's the one who's good at doing detailed analysis of engine performance in-game.

[DoctorEvo]

I'd need to see the .cfg.

It also depends what fuel tank you're using with it... I'd need to see that .cfg too.

This is part of the reason why I wish the addon developers would adopt a standard .cfg spec for propulsion components. That way, it'd be easy to tell if an engine or fuel tank is unrealistic right off the bat, and also ensure that mixing parts from different sets or developers wouldn't cause ridiculous performance (in other words, guaranteeing parts behave exactly as they were designed to no matter what other parts you use them with).

Of course, it sounds to me like the E-series is just plain overpowered in general, no matter what fuel tank it drinks from.