Jump to content

[1.3] Kerbal Joint Reinforcement v3.3.3 7/24/17


ferram4

Recommended Posts

Ah, I see you changed the values for the angular limit. Set them back to zero and it won't to stuff like that anymore. Actually, just don't change any of the angular limit values. Only mess with the linear and angular "drive" values. I saw that same thing happen to many, many rockets. An alternative solution is to play with the angular drive settings and see if you can bring it back to an upright launcher.

There is a damping value added in there that is removing energy from the system, but it's part of the problem with numerical simulations having to run things in discrete time; all motions will be less stable when run with discrete timesteps than if they were run with continuous time. Unfortunately, for that situation the timestep was large enough to make that thing unstable. So yeah, it is a limitation of the technology, and there's really no way around it, besides shorter timesteps and more lag (or more joints; try rebuilding your vehicle and connect the girders to each other with struts and see what happens).

Link to comment
Share on other sites

Ferram, thanks to KJR I'm able to put together some fun, real-world inspired designs with hardly any struts at all. This one has only 2 sets, to stabilise the SRB's.

7pp7ccq.png

MK8y8li.png

Edited by Jack Wolfe
Link to comment
Share on other sites

@Galane: That rocket has more non-stock parts than just Restock. Some kind of SAS module keeps me from loading it. To check the issue. You are running version 1.3 though, right?

Version 1.3 is what I had in with that rocket. Could be a reaction wheel from KSPX on it somewhere. The only KSPX parts I have installed are these folders large_shortMonoTank large_tripleFuelTank large_sasModule mini_asas mini_sas

Edited by Galane
Link to comment
Share on other sites

Having gotten KSPX, I could not reproduce the issue. Every launch I did of the booster resulted in a successful launch and no sticking to the pad. Download 1.3 again, reinstall it and see if the issue still occurs. If it doesn't, I suspect some other mod is causing an issue, possibly involving an exception being thrown initializing physics on a part. From what I can see, KJR isn't causing this.

Link to comment
Share on other sites

Here's a much simpler craft to poke around with quantum struts. http://pastebin.com/Dmwd3wDR Definitely all stock parts except for the four Quantum Struts, assembled specifically to demonstrate the effect.

Pop it into orbit with hyper edit. Try to rotate it in any direction. It won't move. Hit 1 to turn off the Quantum Struts. The Cubic Octagonal Struts they're mounted on go flying off at high speed and now the vehicle can rotate - with the Quantum Struts hovering next to the vehicle.

Hit 1 to turn the Quantum struts back on and it once again cannot rotate in any direction.

And now for the really weird part. The problem isn't with the Quantum Struts, it's with the little Cubic Octagonal Struts and some interaction between them, KJR 1.3 and the Quantum Struts if they're attached to the little girders and turned on. The first version of this test craft had the Quantum Struts between the two adapters between the fuselage sections, with none of the little girders. No problems at all rotating the vehicle.

How I found this was I replaced the robot arm on my Kethane drill rover with one made mostly of DROMOMAN parts and I wanted to brace it (so the docking port could be used in Kerbin orbit to refuel) without leaving any strut ends on it. Quantum Struts to the rescue! But to get them to reach over the arm to the port they had to be a little higher so I stuck on a couple of the Cubic Octagonal Strut girders, stuck the QStruts on them and gave it an orbit and landing test.

I'll try some other part as a QStrut mount on the test craft and see if something like the Small Hardpoint shows the same effect.

Soooooo, the problems with Surefoot's rocket might not be with the Quantum Struts but with the parts the struts are attached to, but only when Quantum Struts are attached to those parts and turned on.

Alan Aerospace Recycling and Packaging's legal team is advising applying for a process patent for this "Quantum Locking Effect". Used intentionally it might have some useful purpose.

Link to comment
Share on other sites

How heavy are those top stages? Those nuclear engines can't be light. I'm thinking you are a little ambitious with the weight you are placing on that rocket. If it's top heavy, and is that long, it exacerbates the wobbly issues for sure. I'm really curious though, how many tons?

Thing is, it's stable without KJR (notice the amount of struts :) and these are all extra-heavy novapunch variety) ! I have launched much worse, really. The payload is 132t, i use its own engines to assist the launch so actually it's lighter than this, so the weight itself is not so much a problem. And my issue is an ever-expanding wobble that comes out of nothing, as if the launch pad was shaking nonstop, what you see on the screen is one extreme of the wobble, it will move from one side to the other like this, without ever coming to rest. This is clearly non-realistic behaviour, here a lot of movement is created out of nothing.. Or see my previous design which disintegrates itself *in orbit* (with no force applied to it).

@Ferram: i will reset the values - so angularDriveDamper and angularDriveSpring are the values i should play with, right. I changed the values to try and fix the issue with the 6-orange-tank design i posted above. So angDamp and other angular limit values have to stay at 0 ?

Soooooo, the problems with Surefoot's rocket might not be with the Quantum Struts but with the parts the struts are attached to, but only when Quantum Struts are attached to those parts and turned on.

Yeah i replicated the issue with a 100% stock model (i posted it above). Not related to q-struts in this case.

You would probably need clamps placed on those high stages as well. But since clamps are being a bit annoying in 0.22 not sure if it'll make things worse.

Yeah i think phantom forces are running through the whole setup. I'll try just to move some clamps around.

Edited by Surefoot
Link to comment
Share on other sites

They work fine with 0.22 (and did with 0.21). They have the gamedata/ structure, no problem.

(edit) success!

NezBByz.jpg

Removing clamps, moving one set around the top load, and launching vertically until the 2nd stage helped. Has to fly manually during launch though... So too much clamps = phantom forces, then wobble,then kraken.

I'm pleased as by itself this cargo engine module is hard as a rock, has got dV to go anywhere in the system with a 250t payload attached in front of it (maybe more, i'll try 300t), and has got a set of ion engines for small corrections during inter planetary travel. Now working on the 2nd part, the cargo payload itself...

Edited by Surefoot
Link to comment
Share on other sites

Woot look at this baby

IRJloyA.jpg

And it's kinda stiff thanks to Ferram :) (i use FAR too, makes it a challenge to launch a supertanker..)

Issues remaining:

- the toroidal tank hubs are still squishy

- the B9 airbrakes "dance" all around upon re-entry or aerobraking (they still do their work though)

Also upon launch i got another "dance of death" issue, this time with procedural fairings, and a matriarch engine lying on the ground (no launch clamps).

Edited by Surefoot
Link to comment
Share on other sites

@Galane: That's probably due to the fact that physics are technically turned off for the small cubic struts. I don't know how that factors into how things work out, but it does mean that using them to do anything physics-based will most likely cause problems.

@Surefoot: Playing with the non-drive settings generally results in nothing good, as far as I can tell. The drive settings tend to be better at controlling the behavior of the joint.

Yes, the toroidal fuel tanks are still squishy; they weight practically nothing and are rather small, so they shouldn't take that much in the way of force in any realistic situation.

Further testing indicates that setting "AngBounce = 1" seems to fix the surface attach joints, but causes the off-center-under-load stack attach joints to reappear. So I'll change some code to allow separate tuning of the different attach joints and we'll see what happens.

Edited by ferram4
Link to comment
Share on other sites

For the toroidal tanks i mean the attach plate, basically it's just a plain metallic disc, shouldnt be squishy like this. Something large and light could be kinda flexible or squishy indeed, but not small solid parts. I guess it comes from the physics engine itself, we are in a situation where things that should flex (large hollow parts..) are rigid and things that shouldnt (small full parts) become sponge-like..

Link to comment
Share on other sites

To people using Quantum Struts - be extra careful to not have them hit any cubic or hexa struts (parts that are massles physically). It makes things behave crazily, independently of Ferram's mod.

Honestly I was a heavy user of QS before 0.21, but starting from that version it seems to bring more issues than benefits. So I've dropped it off - and guess what - I never missed it as bad as I thought I would.

Link to comment
Share on other sites

Add me to the list of people who are really happy about this mod. I'm especially blown away that it worked for me right out of the box with no immediate configuration changes and has no noticeable performance hit.

Thanks for this!

Link to comment
Share on other sites

Add me to the list of people who are really happy about this mod. I'm especially blown away that it worked for me right out of the box with no immediate configuration changes and has no noticeable performance hit.

Thanks for this!

The same goes for me. I'm really happy that I can build taller rockets without to strut the hell out of it. :D

Link to comment
Share on other sites

Honestly, after trying this mod it's better than I even imagined it to be. My rocket design principles are now so much more like real life, and instead of building asparagus carbuncles, you can now build a proper rocket that is thin and tall and cuts through the atmosphere like it should. It's just a magnificent achievement, thanks again Ferram

Link to comment
Share on other sites

I'm absolutely satisfied with this plugin,especially with the option to customize the cfg for our own needs.

With this my rockets shall never fall apart again,unless i want them to :P Great work ferram4, this is now one of my all time favorite mods/plugins :)

Link to comment
Share on other sites

That never happened for me, I know however 0.20(?) added two ways of doing symmetry, there's kinda relative to the whole ship and relative to part. In your case this would be part, because the clamps attach at equal distances from top/bottom instead of at the same place. Either that's what's wrong, or you rotated the part.

Link to comment
Share on other sites

That never happened for me, I know however 0.20(?) added two ways of doing symmetry, there's kinda relative to the whole ship and relative to part. In your case this would be part, because the clamps attach at equal distances from top/bottom instead of at the same place. Either that's what's wrong, or you rotated the part.

Well this seems to happen no matter what I do. On the screenshot I simply added a tank part (without rotating it or etc) and attached 2 clamps in symmetry.

Link to comment
Share on other sites

Adding my name to those that really enjoy this mod. I play with KSP Interstellar and there are some pretty darn big, heavy parts in that mod. KJR makes launches and burns MUCH more smooth. Thank you for this mod!

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...