Universal Storage 1.4.0.0 (For KSP 1.4.x) 13th March 2018

[Advacar]

Is there anything in US that could be tweaked to increase the strength of the connection between the 1.25m hub and parts stacked above and below it? My problem is likely due to combining US with other mods, particularly FAR and Kerbal Joint Reinforcement, but my orbiters are regularly disassembling on reentry with the failure stemming from the 1.25m hub shearing away from the stack parts above or below it under circumstances in which the orbiter holds up if I leave out the hub. Any advice would be appreciated!

I had problems with that too, but my only fix was to spam some ugly struts.

[Gaugeforever]

Did you remember to stow an empty water tank?

That being said, can we get a SAS wedge? Doesn't have to be powerful, I'm just looking to reduce the length of my craft.

Would that even make sense? Spinning weights around the circumference is what causes the S.A.S. to work right?

[CptRichardson]

Would that even make sense? Spinning weights around the circumference is what causes the S.A.S. to work right?

An SAS is a gyroscope. They don't have to be the full size of the craft in order to work. Pairs work as well.

[panarchist]

Would that even make sense? Spinning weights around the circumference is what causes the S.A.S. to work right?

SAS is what NASA calls a Control Momentum Gyroscope. They are typically in pairs and look something like this.

Mounting them in pairs allows the overall energy of the system to be preserved. (some energy has to be added periodically due to friction losses) They can be mounted in line with the centerline, but they don't have to be. If they aren't, then the axis of applied force is not in line with the axis of the vehicle - which *can* cause issues with rotation.

Older variations on SAS (Apollo days) are Reaction Wheels. Similar principles, not as efficient. One might argue that some of the SAS mechanisms are actually reaction wheels.

Edited August 19, 2014 by panarchist

[ThreePounds]

I always thought the SAS is just some mumbo jumbo that hands out free torque. Maybe there are real SAS modules, but they must behave drastically different that KSP SAS.

[OOZ662]

I always thought the SAS is just some mumbo jumbo that hands out free torque. Maybe there are real SAS modules, but they must behave drastically different that KSP SAS.

If you go to the web search of that image search, it leads you to a Wiki article explaining their function.

[Joachim]

Lol wrong thread.......................doh

Edited August 20, 2014 by Joachim

[ThreePounds]

If you go to the web search of that image search, it leads you to a Wiki article explaining their function.

I know how the real one works. I was referring to the fact, that the KSP SAS is able to continuously produce torque, which a real SAS won't be able to as it requires constant acceleration of the gyro-discs inside. At some point you will just see your ship get shredded to pieces from the shrapnel as the whole thing just bursts to pieces.

Edited August 20, 2014 by Three_Pounds

[panarchist]

I know how the real one works. I was referring to the fact, that the KSP SAS is able to continuously produce torque, which a real SAS won't be able to as it requires constant acceleration of the gyro-discs inside. At some point you will just see your ship get shredded to pieces from the shrapnel as the whole thing just bursts to pieces.

According to the Wiki article: "Some methods for unloading CMG momentum include the use of magnetic torques, reaction thrusters, and gravity gradient torque"

Note that this typically happens when CMGs are used to maintain a specific orientation throughout an orbit, and it takes a while.

Edited August 21, 2014 by panarchist
corrected original post

[dualmaster]

Hey, just started trying to use this mod and I'm having some problems. When I stack a bunch of the science bays on top of each other and then launch the vehicle the ship vibrates itself to pieces. It seems the science bays collide with each other some amount. This only happens with a large number of them stacked (8 worked fine, with 9 they self destructed).

There's also something odd going on - when I stack the science bays like this it looks like they connect smoothly in the VAB, however on the launch pad visible gaps appear between the layers. See the pictures below.

Note - this is using the 1.25m cores with science bays.

Javascript is disabled. View full album

[Daishi]

Hey, just started trying to use this mod and I'm having some problems. When I stack a bunch of the science bays on top of each other and then launch the vehicle the ship vibrates itself to pieces. It seems the science bays collide with each other some amount. This only happens with a large number of them stacked (8 worked fine, with 9 they self destructed).

There's also something odd going on - when I stack the science bays like this it looks like they connect smoothly in the VAB, however on the launch pad visible gaps appear between the layers. See the pictures below.

Note - this is using the 1.25m cores with science bays.

http://imgur.com/a/m7fv0

Weird - when the first stack of 8 work, do the seams appear?

What's happening is that the attachment points are "compressing" under Kerbin's gravity, and making the attached wedges 'pop out' (and probably intersect and explode in the process). If you took that stack into space away from surface gravity (or used the hack gravity cheat) you can see what I mean.

The weird thing is that we technically fixed that glitchy behaviour by making the quadcore physicsless in 0.75, and by the looks of it you've updated to that in order for those monoprop tanks to appear in your flight log. Could you open up your Gamedata\US_Core\Parts\US_C10_Hub_Quadcore\part.cfg file and see if PhysicsSignificance = 1 is present in the first few lines?

[dualmaster]

Weird - when the first stack of 8 work, do the seams appear?

What's happening is that the attachment points are "compressing" under Kerbin's gravity, and making the attached wedges 'pop out' (and probably intersect and explode in the process). If you took that stack into space away from surface gravity (or used the hack gravity cheat) you can see what I mean.

The weird thing is that we technically fixed that glitchy behaviour by making the quadcore physicsless in 0.75, and by the looks of it you've updated to that in order for those monoprop tanks to appear in your flight log. Could you open up your Gamedata\US_Core\Parts\US_C10_Hub_Quadcore\part.cfg file and see if PhysicsSignificance = 1 is present in the first few lines?

The seams appear even with just a couple stacked on each other. With ~8 the seams are much more noticeable but it doesn't disintegrate.

The gravity issue makes some sense - the less weight I have above it the more of them I can stack before self destruction. However, I tried using hack gravity from the debug menu as well as Hyperediting the stack into orbit immediately on launching. In both cases the larger numbers still disintegrate.

PhysicsSignificance=1 is present in the config file. And yes, I am using .75 (just downloaded today).

Also, I added a fifth picture to the album above - when I put one of the science bays in upside down, the textures don't line up the other ones. Don't know if that's just a graphical thing or what.

Edited August 21, 2014 by dualmaster

[Daishi]

The seams appear even with just a couple stacked on each other. With ~8 the seams are much more noticeable but it doesn't disintegrate.

The gravity issue makes some sense - the less weight I have above it the more of them I can stack before self destruction. However, I tried using hack gravity from the debug menu as well as Hyperediting the stack into orbit immediately on launching. In both cases the larger numbers still disintegrate.

PhysicsSignificance=1 is present in the config file. And yes, I am using .75 (just downloaded today).

Also, I added a fifth picture to the album above - when I put one of the science bays in upside down, the textures don't line up the other ones. Don't know if that's just a graphical thing or what.

Yeah that's a graphical thing - the doors are off center because the corrugation is off center so the wedges would radially "interlock" with each other.

So, on a stock install, replicated your picture, got eight quadcore stacks on the launch pad with no noticable seams (there is when you zoom out though, but that's texture mipmapping I have no control over). Physics loaded and it was pretty stable. Did one nine high, then ten, then decided to go OTT and this was the result. Similar to what you experienced?

I think its due to a breakdown in the stock game's physics engine, due to using more 'PhysicsSignificance = 1' parts than KSP can deal with. Computers with slower CPUs may freak out with fewer parts, before faster ones do. I dunno.

There's nothing much we can do, aside from turning 'PhysicsSignificance' to 0 (and dealing with springiness), or increasing the node size on the quadcores so they're massive and uncompressable (but then be impossible to attach things to). The easiest fix would probably be using less quads, sorry

Edited August 21, 2014 by Daishi

[somnambulist]

There's nothing much we can do, aside from turning 'PhysicsSignificance' to 0 (and dealing with springiness), or increasing the node size on the quadcores so they're massive and uncompressable (but then be impossible to attach things to). The easiest fix would probably be using less wedges on the quad, sorry

What's the current mass on the cores? iirc mass directly affects the strength of joints with higher mass being more stable.

[Daishi]

50kg (0.05), though from what I've seen (Scott Manley and his video on it), parts with their PhysicsSignificance set to 1 ignore any assigned mass. We did try making the quads heavier as a workaround, but nothing worked very well for something of that size.

[Paul Kingtiger]

I can't recreate it, it may be a issue caused by hardware limitations, but my PC isn't top of the range. Core5, 16gig ram (but running 32bit so that shouldn't matter).

Can you post your machine specs? It might help me to build a picture.

Edited August 21, 2014 by Paul Kingtiger

[Daishi]

I can't recreate it, it may be a issue caused by hardware limitations, but my PC isn't top of the range. Core5, 16gig ram (but running 32bit so that shouldn't matter).

Can you post your machine specs? It might help me to build a picture.

Thats odd... my PC is a little beefier and falls flat on its face. I cant get anywhere close to that amount of cores without disassembly.

---

Although I haven't updated to 0.24.2 yet... hm.

Edited August 21, 2014 by Daishi

[Tommygun]

Would that "Kerbal Joint Reinforcement" mod and its physics easing help reduce the chances of this happening?

[Daishi]

Would that "Kerbal Joint Reinforcement" mod and its physics easing help reduce the chances of this happening?

It could do. I'll update KSP and try it on another stock install.

[Patrick Kerbivan]

Would that "Kerbal Joint Reinforcement" mod and its physics easing help reduce the chances of this happening?

The physics easing may help, but there could be other unexpected consequences. As far as I've been able to discern, KJR reduces flex in joints so a particularly tall rocket is less likely to bend over one way or the other, but it doesn't seem to actually increase their strength (and may actually reduce them, especially in the case of the US quad-cores, since by default it seems to calculate joint strength based on surface area). I don't know exactly how realistic the physics simulation is, but typically when you reduce something's flexibility you increase the chance of it snapping entirely...

[tmikesecrist3]

Though your rocket should not be like trying to flying a noodle

[Daishi]

Here we see the rare PhysicsKraken, firing off high velocity shards as a defence mechanism against encroaching predators.

Any better with KJR? Seems to be perfectly rigid on the vertical nodes of the quadcores, but the radial nodes just lose the plot. I'd take it that's what was drooping and causing the seams to appear on your original post, duelmaster. Doesn't seem to be limited to the science bay either, every wedge can cause the bug.

I'd love to know what's going on in your PC to avoid all this, are your Max Physics Delta-Time sliders set to stock 0.04?

Edited August 21, 2014 by Daishi

[ThreePounds]

I must say, though, it's more than entertaining to watch.

[Tommygun]

Here we see the rare PhysicsKraken, firing off high velocity shards as a defence mechanism against encroaching predators.

Any better with KJR? Seems to be perfectly rigid on the vertical nodes of the quadcores, but the radial nodes just lose the plot. I'd take it that's what was drooping and causing the seams to appear on your original post, duelmaster. Doesn't seem to be limited to the science bay either, every wedge can cause the bug.

Well maybe you could Email BahamutoD about a new cluster bomb mod?

[Paul Kingtiger]

are your Max Physics Delta-Time sliders set to stock 0.04?

Yep, all settings are stock. The only mods I have installed are Engineer and some parts mods.