New parts are unstable.

[AlternNocturn]

So, when I try to build a rocket with the new larger parts, my rockets tend to be unstable and hard to control. They wobble a lot, so much so in fact that one of my rockets shook itself to pieces just from the force of the vectoring engines during my gravity turn. The impression I\'m getting from the new parts is that they\'re far from finished. There\'s not even an SAS scaled for the new parts. Which is understandable since they had a lot to do in just a short time that they just couldn\'t get them done in time. That\'s perfectly fine, but I think I\'ll hold off on using the new engines, command pod and fuel tanks for now. They look really good, and are very well made, but I\'m pretty sure that they\'re going to be very hard to use until the team works on them a bit more. I really do appreciate the work you guys at Squad do and I\'m still extremely grateful the new parts even made it in at all.

[JebThrillMaster]

My 2 rockets seem to be doing just fine, if you need something to get a good flight ill have a download [ftp=ftp://www.dropbox.com/s/cprivvlgoh4fl8e/Titan%202.craft]here[/ftp] to one of my rockets that can get to orbit of Kerbin and back.

[_Aramchek_]

I had just the opposite experience, the new stuff is very stable so far.

[Kymlaar]

I\'ve run into the instability as well. It seems to be if they have a heavy load above them, or if you combine sizes. The parts looked like they shifting back and forth at their connection points, kind of like this:

I found that if I stuck to all one size it worked out for the best. I do wish I had understood the scaling thing though, as I was hoping to be able to visually adjust the size of some components to match others.

[Jarnis]

Well, I got the 3M parts to behave almost acceptably by using the 1M decouplers (3M decoupler only for 3-person pod attachment) and struts. LOTS of struts. 8 struts between each stages over the engine, additional struts between 1st and 2nd stage and tons of cross-struts between side-by-side 3M tanks of first stage.

Still wobbly, but no longer gets ripped apart even with 400ton+ takeoff weights (my 3M-based test setup for future interplanetary missions)

3M decoupler is problematic tho. 1M decoupler also is very weak for separating 3M parts which are so much heavier, so you have to stage carefully. Also radial decouplers are still fairly weak. Easy to cause unintended collisions...

Edit: With some serious tinkering managed to get 730 ton monster off the pad... had to do lots of little tweaks to prevent it from tearing into bits... but in the end I have 73 tons of rocket on 180km circular orbit, ready to insert to an interplanetary trajectory. This is fairly close to the maximum as long as we have this tower; the rocket is already twice as high as the launch tower and can\'t fit more stuff radially without hitting the tower

[Kymlaar]

I noticed some of these same issues in the past when I was using NovaSilisko\'s mod parts in ways they weren\'t designed for, specifically the larger parts. I\'m wondering if the wobble might be an artifact caused by parts that are too large, and that\'s why I experienced it back then.

Thanks for the tip concerning the coupler, I\'ll give it a try.

[Jarnis]

After further testing I\'m fairly certain that this is caused by two factors;

- 3m decoupler is not as 'firm' as the 1m one, it lets parts shuffle around more (which can then be amplified to a rocket snapping in two)

- Massive 3m liquid engines being gimballed is... problematic. Even a tiny movement of such a massive engine jerks the ship around. I\'m sure if we had also a non-gimballed large 3m engine and would use that with some fins for steering instead, things would work somewhat better.

Still, you can work around these issues with struts. Lots of struts. Also make sure you do not overdo the first stage; wobble gets worse if you have 'too much' thrust and start hitting the atmospheric drag big time at low altitudes.

Additionally, never ever start the automated gravity turn before 20km or so; You want the first stage gone before you start turning...