How does the breakingForce parameter work?

[CaptainArbitrary]

First off, I\'m really incredibly sorry if this is a dumb question. I\'ve scoured the wiki and searched the forum, so if I\'ve missed something, it\'s totally my fault and you should make fun of me.

The short version is that I\'m looking for some insight into how the breakingForce parameter works, so I can tune my custom parts so they handle reasonable amounts of force but will break under excessive acceleration. Like for instance, I want a payload decoupler that can support say ten tons (mass = 10.0) at up to 3g, but will break at higher accelerations than that. This seems like a job for breakingForce, but I haven\'t been able to make it work.

I\'ve done test after test, and I\'m stumped. There doesn\'t seem to be any consistent relationship, that I can see, between a part\'s breakingForce, how much mass is on top of that part, and how much that part (and the stacked mass) can be accelerated before the part breaks.

For a while, after doing some empirical experiments, I was thinking there was a straight scaling factor involved: a breakingForce of 1.0 meant the part would break when 50 kiloNewtons of force were applied to it (or whatever; I\'m calling a mass of 1.0 accelerated at 1 length unit per in-game second a 'kiloNewton'). I crunched the numbers and figured that a part with a breakingForce of 1.0 with 1 ton (i.e., mass = 1.0) stacked on it would fail at 5.098 g\'s of acceleration. I built a test rocket with that part, and sure enough, it failed at what the game reported was a max acceleration of 5.1 g\'s. Sweet! Putting an extra 100 kg on top meant — according to my spreadsheet —that the part should fail at 4.635 g\'s, and the game said it failed at a max acceleration of 4.7 g\'s. Awesome! Then I put another 100 kg on top (total of 1.2 mass units), which my spreadsheet predicted would cause it to fail at 4.249 g\'s …and the game reported that it failed at 4.3 g\'s! Everything\'s coming up Milhouse!

Then I put another 100 kg on top, for a total of 1.3 mass units, and prepared myself to see it fail at around 3.922 g\'s …and the sucker blew up on the launch pad for no reason I could figure out.

So what\'s the deal? Does anybody know how to relate the breakingForce parameter to the mass parameter and to acceleration so I can make my parts fail in a way that\'s, if not realistic, at least consistent from part to part?

[requimrar]

I don\'t know the specifics, but do you notice some lag when you launch the craft? The game actually puts it slightly above the launchpad, so as to let the terrain load. I don\'t know if it gets put or dropped, but that could be the source of the G\'s; I\'ve seen my G meter fluctuate before launch on medium-large rockets.

[CaptainArbitrary]

I think you\'re onto something there. I\'ve done more science this morning.

I built a test rig, mostly out of trusses. It looks like this:

The capsule is connected to the part above it by a small coupler which you can probably barely see. That coupler is my test part. I varied its breakingForce parameter and subjected the rig to acceleration (courtesy of the modded stock engines; each one puts out 1,750 kN of thrust). When the part breaks, the capsule will immediately stop accelerating, meaning I can then end the flight and see what the maximum acceleration was.

When I crunched the data I collected, I discovered two things. First, there appears to be a straight scaling factor between breakingForce and kiloNewtons: 50. A part with a breakingForce of 1.0 will fail when it\'s subjected to 50 kiloNewtons of force. What I mean by 'kiloNewton' here is the force that results from accelerating one mass unit by one length unit per second per second. We\'re calling length units meters; that\'s all but canonical as near as I can tell. And if the mass unit is the kilogram, then the stock parts that ship with the game are about 30% less dense than air! But if we equate the mass unit to the metric ton, then the densities of stock parts make a kind of sense (modulo a factor of two), so that makes the force unit the kiloNewton —one metric ton accelerated at one meter per second per second.

When I put in that fudge-factor of 50:1 kiloNewtons to 'breakingForce units', I started seeing margins of error between the predicted and reported failure g-loads on the order of 0.4% …then I realized the game is probably just rounding off the max g-load to one decimal place, and when I took that into account my margins of error all but vanished. (I\'ve got one anomalous result here, where the predicted failure was 4.2 g\'s but the game reported failure at 4.3 g\'s. I kind of don\'t care.)

So here\'s the apparent formula for breakingForce:

bF = m a ÷ 50

In English, to support a mass of m tons accelerated at a meters per second per second, you need a breakingForce parameter equal to m × a ÷ 50. Or if you prefer, to support m tons subjected to n g\'s of acceleration:

bF = m n g ÷ 50

where g is the acceleration due to gravity at Kerbin sea level, 9.8068 meters per second per second.

However, it appears that merely putting your rocket on the launchpad subjects it to somewhere between 2.788 and 2.815 g\'s of acceleration. In my test rig, a coupler that should have held the capsule up under 2.788 g\'s of acceleration failed as soon as I moved it to the pad, while a coupler that should\'ve held the capsule up under 2.815 g\'s of acceleration made it —and consequently failed just after liftoff when I pushed the rig through what the game reported was 2.8 g\'s of acceleration. So the math is good; it\'s physics itself that\'s wrong.

Anyway, science continues on here, so if I discover any big mistakes I\'ll correct them later. But for now at least, my guideline is to build parts that are always able to support the mass they need to support at a minimum of 3 g\'s of acceleration. Unless I miss my guess, that ought to result in parts that make it off the pad okay, then fail 'realistically' (kind of) in flight. If you push the throttle to the stops during your transfer-orbit injection burn and push your vehicle to 6 or 8 or 10 g\'s, your expensive satellite payload is gonna drop right off. At least, that\'s the hope.