Frame Stress Analysis

[David]

Hey, as some of you may be aware I am building a frame for my A2 project. The biggest problem I am having is that to gain the marks I need to prove/show why my design is better then others. As in, for example, why joining the seatstays 20mm further away might make a stronger structure than say different angles/measurements. The only way I can see to test this is to model it in CAD and do some stress analysis. Unfortunetly my CAD skills arent the best and im not sure how accurate it would be. Is there anything else I could do?

Thanks,

Dave

[Walleee]

Well you could show some specs for material choices saying steel is strong blah blah but it's very much a matter of opinion as to which is better (mainly aluminium/steel)

What are other's making and how do they quantify 'betterness'??

[David]

Well you could show some specs for material choices saying steel is strong blah blah but it's very much a matter of opinion as to which is better (mainly aluminium/steel)

What are other's making and how do they quantify 'betterness'??

Ive got information on materials but im more looking for information on actual frame designs. As in specific details on why certain structures are stronger then others (not just because they use stronger materials or lots of triangles etc)

[Walleee]

I have some books at home im readin and they are like bicycle design bible's.

I'll throw up the author's names and such later when I get home, if i don't PM me or hassle me as I really think they could benefit you.

Best of luck.

[RobinJI]

Were doing framework internal stress analysis at college at the moment, at me on msn if you want I could probably help you with some of it, its ringwams@hotmail.com

[tank_rider]

The analysis of frames using FEA is a very sketchy business at best, due to the extreme difficulties in the modelling of welded joints etc. I guess the easiest way to do a rudamentry study would be to construct a solid model and show where the areas of highest stress are, then address those regions. That way you avoid the pitfalls of trying to show "real" figures which would always include far more error than actual engineering signals. A poor model can do more damage than good as it will cloud and often falsly influence decisions.

I think FEA is a bit too ambitious for an A2 project. It might be better going back to more basic physics principles in order to qualify your improved designs. I'd usually offer to knock up a frame in Unigraphics, but it kinda goes against the whole idea of you doing the work, and i don't have the time at the moment.

In terms of going back to basic physics, try isolating the section(s) you've changed and create a boundary around them, working out the inputs through the boundary and go from there to do your calculations etc. Thats the best approach to complex problems, when hand calcs are the only option.

[Walleee]

The analysis of frames using FEA is a very sketchy business at best, due to the extreme difficulties in the modelling of welded joints etc. I guess the easiest way to do a rudamentry study would be to construct a solid model and show where the areas of highest stress are, then address those regions. That way you avoid the pitfalls of trying to show "real" figures which would always include far more error than actual engineering signals. A poor model can do more damage than good as it will cloud and often falsly influence decisions.

I think FEA is a bit too ambitious for an A2 project. It might be better going back to more basic physics principles in order to qualify your improved designs. I'd usually offer to knock up a frame in Unigraphics, but it kinda goes against the whole idea of you doing the work, and i don't have the time at the moment.

In terms of going back to basic physics, try isolating the section(s) you've changed and create a boundary around them, working out the inputs through the boundary and go from there to do your calculations etc. Thats the best approach to complex problems, when hand calcs are the only option.

Good advice there kinda what i was getting at regarding the book's if you don't know where to start with the actual drawing the FEA will not help matter's. I find it hard enough and i've been working with solidworks for almost 3 years now.

anyway, the books im using at the moment;

Bicycle Design - Mike Burrows, Edited by Tony Hadland, published by company of cyclists??

High - Tech Cycling, Second Edition Edmung R Burke, PhD, Published by Human Kinetics (this is serious, gives rating for the drag caused by lycra!!!!!!!)

Bicycling Science, Second Edition - Frank Rowland Whitt & David Gordon Wilson, Mitpress.mit.edu

They are mainly concerning road bike's (certainley the second one anyway!) but there is some information in there and even if you don't read them, throw them in the bibliography of your write-up and it'lll look good

EDIT: you could also consider making small scaled frame's out of welding wire using solder to stick it together, then simply apply the load's with your finger's until destruction, find out where they brake and say this is where need's attention. I'd also be making larger example's of certain junction's, seatstay like you said, head tube, bottom bracket blah blah.......

Edited November 2, 2006 by jake1516

[2sixstreet]

I agree with tank_rider, get back to basics. The abuse a trials bike gets from a rigid frame and dynamic loading could be quite ambitious so i'd suggest a static analysis. You could say apply a load through the dropouts and isolate each section ie triangle and determine the stress in each member (tube). Then perhaps compare this with a standard road bike or MTB frame geometry and say why you need to make a trials frame stronger. eg gussets, CNC'd sections. As you said, by moving the a connection 20mm should show some good results once you have the initial model set up.

Then you could discuss frame advances over the years like fluid forming and tube shapes etc to reduce problems associated with welding zones.