Rim strength - engineering analysis

[tba]

Hey all

I've done a structural analysis of a typical trials rim to see how much holes affect the strength of them, and whether or not it's actually worth drilling holes in them.

If anyone's interested, give us your email address and i'll send you a pdf.

Anyone know of some way I can post it up on the forum.

It's a bit long... however there is a bit of crapping on in there, explaining a fair bit of stuff. Any feedback would be accepted. If you think i should leave out half the stuff at the start and just get to the number at the end... say so.

Well, thought some people might find it interesting...

Cheers

 

[tba]

Just realised I can attach it to a post... see above

 

[L.P.]

thats awesome.

can i ask what cad and stress analysis package you were using?

 

[bipyjamas]

Some might say, leave your rim in one piece, and take a crap before you ride....

Good point

Also, until you mentioned it, I didn't stop to think that the hubs were actually hanging from the top spokes. Learn something new everyday.

Overall fun read, thanks for the lunch time entertainment.

 

[tu plang]

What type of element did you use to model the spokes?

While a rule of thumb with FEA is that half of a symmetrical model can be used in place of the whole model, i think it would require very careful consideration your fixed degrees of freedom on certain nodes. While it is true that the upper most spokes carry the majority of the load, you also need to consider that under deflection the wheel wants to squash and as such all of the spokes except in the case of a perfectly vertical spoke will have some component of tension.

Definitely a good start though and chances are you've probably done more FEA work than i have but i just thought i'd throw in a few things you might have overlooked so that we can build a nice model.

Any chance that i can have a copy of the strand file to play around with?

 

[tba]

Hey - thanks for the feedback guys

Tu Plang - for the section i'd just cut in half, i deliberated for a bit as to how to restrain those nodes.
I ended up restraining only in the theta direction. i.e. assuming that the end part of the rim can move toward the hub (radial direction) sideways (z-direction), but it can't actually rotate (relative to itself). ie. it can't go through the other half of the rim that would usually be there.

I totally agree with you that modelling the whole rim would be accurate. I might do it later. I did it this way as it allowed me to convince myself that I was only wasting half as much time as i could be, when I should actually be working on my honours thesis.

Re element types - 4 node tetrahedrons.
From memory, the number of brick elements was in the order of ~35,000.
(My honours project consists of structural analyses of the human jaw bone to determine how the wisdom teeth affect it's strength. I'm using models containing bricks in the order of 1-1.5million... just for some perspective... however, aluminium is very simple compared to bone. Isotropic and uniform...)

Re hanging from the spokes - thanks for also clarifying what I had probably oversimplified. Thats also the reason why i deleted only as many spokes at the bottom as i thought necessary. (upon looking at each spoke tension, i probably could have deleted another one, as it showed up with a negative force each time, meaning it was in compression)

-Another note, i'm sure it would be possible to simuate the spokes as providing only tension, not compression... but i'd have to play around a bit. Might be worth looking into if I do a full wheel.

Re spokes - i just modelled them as 2mm diameter beams made of stainless steel. I pre-tensioned them to 20N. In all honesty that was pretty much a guess. They could probably be tighter, it's hard to know though.

 

[tba]

One More thing

I've also done a thermal/structural analysis of a disc brake rotor.

I did last year as a uni project. Can't remember exactly what it was like, but i've learned a fair bit more about FEA since then.... not that that means i know everything, still have plently to learn i'm sure.

By the way, if any of you know any bike manufacturers wanting an engineer to do this stuff, i'm still looking for a job next year!! haha
(Can't hurt to ask can it!)

Cheers.

 

[tu plang]

haha i wish. engineering bike shit would be a dream job (but probably a grossly underpayed one.)

I'm a good year and a half behind you but i was just curious to see how you'd done it all because i'm just starting to get more and more into FEA this semester.

 

[MasterOfReality]

Interesting.

Can this model incorporate strain softening, or are you just interested in what happens right up until the rim goes plastic?

I have been doing Finite Element and Distinct Element for the past 4 years, and have just started on equivalent continuum models using the Cosserat theory.

Always interesting to see it in another perspective.

 

[tu plang]

strand does have property settings for elastic and plastic deformation so it would be workable i think.

 

[jda]

Most of that went over my head but as a trials rider & someone who works in Materials Science I enjoyed reading it. I'd just like to add that rims come pre-drilled these days and people haven't had to drill their own for about 5 years now.

JD.

 

[alpinestar12]

That is some good work there, were you actually assessed on this or did you just do it for fun? Nice work anyway.

 

[MasterOfReality]

strand does have property settings for elastic and plastic deformation so it would be workable i think.

Ah ok.

The programs I use allow me to set the softening as a function of the plastic strain, so I can pretty much replicate a stress strain profile, but given that nobody knows what the exact values are, its all subjective anyway.

 

[luke.b]

Woah, that is cool. Never realised how much it could effect it. I'm about halfway through reading it, and need some time to wrap my head around it. I'd be very interested to read that work you did on disc brakes. Would you post it up?

 

[gbowen444]

Very interesting. Thanks for posting. I read the whole thing and it is good to see this sort of analysis even if just from an interest point of view. What about drilled sidewalls? Or different shaped (sqare, hexagon) holes?

Along the lines of what JDA said though (I don't mean to take anything away from your work), the widest non-drilled rim that I know of is 39mm, most trials riders use pre-drilled rims which come up to 49mm wide.

 

[Macbeth]

Nice work man...... Must've taken ages.....

I'd actually be curious to see the difference in strength between eyeletted and non-eyeletted rims and, as Gary said, drilling of sidewalls.....

Adam

 

[tba]

yeah.. it took a little while, not too long though.
I just did it out of interest. Wasn't marked or anything. However the disc brake one i did last year was.. Actually topped the class for it.

MasterOfReality.. it seems you know more about dealing with plastic than I do. I haven't dealt much with plastic in FEA software.

Ive heard my supervisor say a few things about using the Nonlinear solver in Strand for plastic. Which would make sense.

With my rim analysis, it was purely a comparative analysis so I wasn't real concerned..

When i looked a the discs.. i looked at a few things like, how much the expand, whether the holes drilled in them are actually beneficial (i think). Another thing i looked at was... you know how discs are always put on so as the spokes rake forward? (that makes sense?).. anyhow, i looked into whether that was optimal/detrimental/fashion. I modeled my hayes rotors, and found that if anything, they should probably be the other way around. But it probabkly doesn't matter heaps.. i'll have a dig around my computer and see if i can find my report for it

 

[Macbeth]

I'd be interested to have a read of the discs one....

I've been running rear disc for six months or something, and at first found that with the rotors in the traditional direction, they were colapsing and BADLY warping under reverse loads........ Would be interested in anything to do with making disc work as well as they can

 

[tba]

Disc Rotor Analysis

Heya

Here's the disc one i did last year.

It probably misses a bit, but anyhow.
I think it's a bit longer than the last one.... ~4000 words