View Single Post
Old 01-12-22, 07:23 PM
  #37  
79pmooney
A Roadie Forever
 
79pmooney's Avatar
 
Join Date: Oct 2014
Location: Portland, OR
Posts: 10,229

Bikes: (2) ti TiCycles, 2007 w/ triple and 2011 fixed, 1979 Peter Mooney, ~1983 Trek 420 now fixed and ~1973 Raleigh Carlton Competition gravel grinder

Mentioned: 111 Post(s)
Tagged: 0 Thread(s)
Quoted: 3302 Post(s)
Liked 2,028 Times in 1,319 Posts
Originally Posted by Andy_K View Post
My degree is in physics and math, so I've always looked at engineers as a curious species. They have a strange mix of intuition and pragmatism. Having been trained in physics and math, and now working as a programmer, my mind always goes straight to setting up equations (albeit with a generous helping of simplifying assumptions). So, I'm curious, from a structural engineering perspective, does the fact that the force typically comes in at an angle modify the 1/3 2/3 rule? Granted there are cases where the impact may be perpendicular, and I guess structural engineering is all about those worst case scenarios....

Also, having started imagining force diagrams and equations in my head, I wasn't happy with @79pmooney's simplification of the force being concentrated at the end of the seat post. I wanted some integral computation with incrementally increasing force dispersed along the length of the section of post below the pivot (and some above if you've got a Cannondale!) -- which led me to the picture above with several kids sitting along the length of a see saw. But when I thought about why he'd make that assumption I decided that it must be something along the lines of "if you make a calculation based on this simplified assumption, it won't fail and the math is easier."

At some point, there's a trade off to be made. For a race bike, you might want the shortest seatpost that won't definitely cause a failure. For a touring bike, you probably want one long enough to be confident that it will never cause a failure. Many of us here are buying bikes and components that were designed for racing, but we really want to get touring-like reliability out of them.




That was the first thing that popped into my head, even before the hypothetical force equations. I've watched multiple cyclocross races where someone crossed the finish line standing to pedal with part of a carbon seatpost sticking out of their frame. That's an unlikely result with a metal seatpost, I guess, but I could imagine the seat cluster coming unclustered and having a similar effect.
Andy, you are looking for an exact solution. No issues. Just model up the seat tube, lug, TT & seat stays plus post, seat and rider in a FE program, add the load and solve. You now have spent many hours and have one solution to a problem that may never play out just like that. Huge waste of time for any engineer tasked with getting plans and products that work out the door. My simplified approach is, yes, very crude but if you can run across one or two bikes with bulged seat tubes and many without, it probably won't be hard to see that a seatpost sticking down this far and a seat tube this thin bulges but all these more conservative bikes and seatposts are fine. And what is the weight cost of a seatpost that is inserted a cm or two more than needed? 10?, 20? grams. (For an aluminum post built like a tank.)

It comes down to: what is the cost to the ride of too safe a system; in weight, race results, etc. vs the risk and consequences of a post snapping or a seat tube being bulged (metal) or cracked (CF). The answer to that won't come from theory.
79pmooney is offline