MIT students build a roller coaster in one week
monograph.io
monograph.io
By using such big bolts they actually made the structure less safe because they had to drill very large holes; those holes made the wood weaker than it otherwise would have been. I don't know who at MIT insisted that they use 1" bolts but that person had no business sticking their nose in.
The Roller Coaster used to be an annual tradition but was put on hold the past few years because of EHS regulations and safety concerns. I know they really wanted to build it this year, so they obliged with the request.
If you're using desk screws you can purchase a few impact drivers which have adjustable torque settings and enjoy screwing to a specification. With lag bolts you could actually inspect everything with a torque wrench and the correct size of socket and that would work just fine.
I think the biggest advantage to bolts is that they're reusable so you could take the whole thing down and build it again next year if you wanted. Or at least the start tower, say.
But oversizing fasteners is only good up to point; beyond that and you're trading off fastener strength which could never ever be used (due to the lack of ability of the wood to sustain those kinds of loads without breaking) against reducing the ability of the wood to sustain ever smaller loads as the holes get bigger.
That sounds reasonable, and yet I'm not sure it's applicable here. It's a 1" hole in the face of a 5.5" member (2x6). In structural engineering plans that I've seen (see below) that would be allowed (for certain placements) w/o explicit permission of the engineer. From that I deduce that such sized holes do not substantially weaken the member. But perhaps there is more to it. Also consider that a larger bolt means the wood-metal interface is spread over a larger area, so lower pressure - perhaps reduces splitting?
>With lag bolts you could actually inspect everything with a torque wrench and the correct size of socket and that would work just fine.
True, but unless the lumber is pre-dried, it will shrink after construction. Whether it be lags, hex bolts, or screws, it will need rechecking after the wood settles.
All that said, I too would not have used 1" bolts. Seems like 1/2" or 5/8" would suffice. But my main objection to 1" is that it's overkill, and engineering is, after all, the science of achieving the stated objectives with the least resources. I'd love to see the calcs and assumptions of the professional engineer who mandated the 1" bolts.
From some structural engineering plans for a house:
Do not cut, bore or notch wood members expect where show in the details. Maximum holes at studs to be 0.4x width of stud. Maximum hole diameter at beam joist or rafter to be depth x 0.25, or 2 inches whichever is smaller, and are to be places within the middle third of span. Holes not permitted when depth is 4 inches or less. All other holes require engineers approval
I've never heard that definition of engineering before -- is that your personal definition? Regardless, it's moot. One of the required objectives was to get engineer signoff -- the need to get signoff from others, even others you may disagree with, is something these engineers will need to do in their careers as well.
With a factor of safety at 2+.
https://en.wikipedia.org/wiki/Factor_of_safety#Choosing_desi...
Yes, that is my personal definition, which i gained whilst studying engineering in the UK. Its similar to the one on wikipedia, except shorter and slightly broader :)
I've seen rings (like http://www.portlandbolt.com/products/others/splitrings.html) used to spread the load over a larger area of the wood, with a bolt keeping the wood members pressed against the ring.
Here's a nice reference for torque vs clamp force for a range of fasteners:
http://www.spaenaur.com/pdf/sectionD/D48.pdf
If you use a 1/2-13 fastener and torque to 75 ft-lbs it'll provide nearly 5 tons of clamp force (provided that said clamp force doesn't cause the wood to get squished). Now standard friction equation applies:
F = u * n
And wood-on-wood is between 0.25 and 0.5 so let's be conservative and use 0.25.
9000 lbs * 0.25 = 2250 lbs of frictional force keeping the one wooden member from sliding down the other one.
http://www.physlink.com/Reference/FrictionCoefficients.cfm
Here you can see the shear strength of a variety of bolts. You'll want to look at the second page under "single shear" as that's the application we're talking about.
https://nucor-fastener.com/Files/PDFs/TechDataSheets/TDS_013...
If you look here a 2x6 as a vertical column is only good to support about 900lbs so basically all of these equations and speculations are a joke. Grade 2, 1/4" bolts would have 2000lbs of shear strength each, so the only reason to go bigger is for a better interface to the lumber. But that can be largely negated so long as you have a decent amount of clamp force, for 900lbs you would need 3600lbs of clamp force at most. Which you can get with a single 5/16" fastener, and probably a pair of large washers to spread the force out to prevent crushing the wood.
https://courses.cit.cornell.edu/arch264/calculators/example7...
http://formlabs.com/en/company/blog/2014/08/27/from-3d-print...
The retention wheels are just riding on the underside of the plywood sheets. If they had a jam at a track joint, those might tear through the plywood or break off, allowing the car to derail and fall off the track. Unsupported plywood edges are weak and not good working surfaces.
Here's a standard roller coaster wheel assembly, with six wheels.
http://www.themeparkreview.com/forum/files/dsc_0207_3.jpg
Any 3 wheels can fail without serious trouble.
But at least they used 1" carriage bolts to hold it all together.
I would think that is only true if we consider that cars have more than one of these.
http://formlabs.com/en/company/blog/2014/08/15/formlabs-engi...
In January 2006 a freshman fell through a skylight and broke several bones: http://tech.mit.edu/V125/N65/65skylight.html. This seems to be what started changing attitudes.
Well, they didn't say anything about cows, did they?
Hey guys, you're MIT hackers. Involve a cow, will you? Upside-down, if at all possible...
Additionally there is probably so much friction in the system that it should not be necessary.
It was (is?) a big center of roof and tunnel hacking. One guy from my floor went out every night for a couple of years trying to collect all of Sophicles' sign-ins. And don't forget the Oddball Olympics. I remember the Master Lock picking event usually being won in less than 10 seconds.
He's one of my favorite East Campus folks to tell stories about. Awesome guy.
https://www.youtube.com/watch?v=5gcoV3MkmY8
Took a lot longer than a week, I'm sure, but it's an actual proper coaster. He later went on to work for one of the coaster companies for awhile IIRC.
Overall a neat project for a large college team to work together on but the end product doesn't appear that functional/fun.
I want to ride it. My son would love to ride it.
Or perhaps their true attitude just wasn't appropriate for this blog post.
They probably just pull it up with a rope or something.
Remember to reengage physics before starting the ride with the next rider.
Source: was one of the people pushing it back over the humps. Surprisingly a sketchy job - you have to really go for it to get it over the middle hump, and if you don't make it then you have to bail really quickly.