furcifer said:
Avoiding the questions noted.
What questions?
furcifer said:
First you're making stuff up. Nothing is designed to utilize 99% of it's "material strength". I believe you used nominal strength before which is typically part of calculating dead loads and not active loads. Anyhow...
You said:
furcifer said:
I would strongly suggest you think about this logically, just for a second. If I build a frame designed to hold 1000lbs, and I load it with 1000lbs. HOW LONG WILL IT LAST?
If I load it with 50lbs. HOW LONG WILL IT LAST?
If I load it with 1lb HOW LONG WILL IT LAST?
If I don't load it at all HOW LONG WILL IT LAST?
Given the context of the discussion which was about the breaking strength of components (and because you didn't specify) I took "designed to hold 1000lbs" as being failure point = ~1001lbs.
I actually said the nominal failure point of a lifting component rated for 10 tonnes with a factor of safety of 5 is 50 tonnes.
furcifer said:
I'm going to presume you mean "ultimate" strength, which is the point at which a material fails. Since we're talking cars and drivetrain it's mostly metal, so this would be yield stress, the point where there's no longer elastic deformation.
Please don't, I know what UTS is. With a few exceptions most parts will already be deemed to have failed if the yield point is reached, not at UTS. Moreover I can't imagine anyone designing a drivetrain component would use limiting stress = yield. That's going to be far too high for a component subject to HCF.
furcifer said:
Stress concentrations, fatigue, etc., things that are known to happen are all part of the design. The factor of safety is an integer multiple of that design intended to account for the unknowns, weathering, damage, improper loading, variation in materials etc.
Partly correct. The factor of safety is fudge factor used because the exact loadings, resulting forces, material properties, manfacturing tolerances and environmental and useage characteristics are not precisely known. It attempts to fill the gap between how you think a part will behave and how it actually behaves in the real world. Remember, you must account for (guess[timate] at) known-unkowns as well as unknown-unknowns.
There's no rule saying safety factors must be integers and they commonly aren't. Air travel would be more interesting if they were as it'd either be very dangerous or much slower, shorter and more expensive if SF had to be 1 or 2 instead of 1.25 or 1.5.
furcifer said:
This is why fractional increases in hp don't require swapping out your Civic drivetrain for one from an F-150 if you want to boost your Honda.
That's a negative, ghost-rider. Unless the drivetain was over-specified (badly engineered from a value perspective) in the first place, you cannot increase power through it by any appreciable amount without a noticable increase in failure rate. Automotive OEMs spend a lot of money testing a lot of components to determine what is required to fulfill the design requirements of a vehicle.
You seem to think if a gearbox is rated for 200Nm, has a SF of, say 3, and designed to last 200,000 miles that you can run it at 600Nm and expect it to last at least 67,000 miles...
furcifer said:
So I don't know what you are talking about. If you could give me an example of where performance modification require a new drive train maybe I could see what you are trying to get at. I suspect maybe back in the early 1900's this may have happened but automotive engineering has advanced considerably since then.
If anything I'd expect components from yesteryear to typically have had more design margin. Such things tend to get steadily trimmed to reduce cost and weight.