Additive manufactured titanium failure (Olympics!)

[tjvirden]

There's a good thread on Cyclingtips about this, but post your thoughts here too!

I'll sum up mine by saying that, within reason, the material doesn't matter - it's the implementation that counts........

[Yoln]

What are you trying to discuss here exactly ???

[tjvirden]

Ah, perhaps I should have added some more info:

Australian Olympic track team member Alex Porter had a very nasty crash when his bars suffered a catastrophic failure. Fortunately it seems that he hasn't suffered any serious injury.

It appears to have been the titanium basebar that failed - it has been confirmed that it was a 3D printed unit from Bastion. It failed just when you might expect - as he entered the [banked] turn. That puts about 1.6g through the rider and bike (that's in the 'vertical' plane of the rider/bike) at about 62kph.

My guess is that the designers seem to have neglected the additional leverage that the aerobars have when applying pressure via the hands to the ends of the extensions - it's hard to model the forces, but 10Kgf down through the hands puts a very high force into the 'stem' area. Worst of all, they haven't conducted a "what happens if this part fails suddenly" analysis.

This type of failure is bad enough on a track but on public roads doesn't bear thinking about......

[mikemelbrooks]

The GCN commentators were saying that the stem was overtightened, For what that's worth.

[tjvirden]

GCN

Even if that was correct (have a look at where the failure was, and it then doesn't seem likely), with critical things like forks, stems and bars, nothing should get onto a bike until the "what happens if we overload this" analysis has been done.

[wheelsONfire]

I asked tech team at Rotor why they didn't do 3-D printed crankarms. They told me, they aren't durable enough.
I just guess now, but perhaps this can be applied for handlebars to. Just a thought, i have no clue!

[AJS914]

There's a good thread on Cyclingtips about this, but post your thoughts here too!

I'll sum up mine by saying that, within reason, the material doesn't matter - it's the implementation that counts........

I read the article on cyclingtips. There is no conclusion as to why it broke.

One thing I was thinking is that we have no idea of the life lived by this part since it was manufactured. Was it over-tightened? Was it dropped, crashed in training? Or, it could be a manfacturing defect or a combination of multiple things.

[Butcher]

Or that the fact it's a relatively new technology and they have not got all the bugs worked out. Crap happens when you are on the bleeding edge.

[blaugrana]

I asked tech team at Rotor why they didn't do 3-D printed crankarms. They told me, they aren't durable enough.
I just guess now, but perhaps this can be applied for handlebars to. Just a thought, i have no clue!

But for a company like Rotor, price would surely a major concern. Even if they could make a slightly lighter and/or stiffer crank than their existing alu options, it would hardly be worth it.

After all, the only 3D printed titanium components in cycling are obscenely expensive, even by cycling standards. If you are doing custom stuff like the national track teams, though, things change, because the lower upfront cost of 3D printing (you don't have to build moulds) can easily offset the higher cost per unit produced.

[tjvirden]

One thing we do know is that Bastion Cycles has been making their own "printed" titanium parts for quite a while and there hasn't been any hint of problems with what they've made. I definitely don't want this to be a thread bashing all they've done - that would be totally unfair.

From the outside, they really seem to have invested a lot of time, effort & money into their business and I don't think I'd have an issue riding a frame by them - the tricky bit with many other parts is that they *cannot* suffer instant total failure; I don't want to be a beta tester for new ideas in that realm.

wheelsONfire's comment about Rotor's view on printed cranks is an interesting one - it indicates they've done some significant testing. Rotor have had their problems in the past too (which cycling company hasn't?), pushing weight down on machined aluminum parts, including stems.

I do think there's some substantial risk in using additive manufacturing for certain parts - it comes back round to chasing weight reduction. Perhaps using it for other reasons (customisation/aero optimisation) is the best strategy, but WW is alive and well.

[ghostinthemachine]

with critical things like forks, stems and bars, nothing should get onto a bike until the "what happens if we overload this" analysis has been done.

Oh, it has, the result is "you crash". The challenge is to work out the actual peak and cyclic loads so it never fails. They still haven't managed.

I asked tech team at Rotor why they didn't do 3-D printed crankarms. They told me, they aren't durable enough.

Most 3D printing processes give you slightly worse bulk materail properties and slightly (or a lot) more inclusions and voids.
So you could make a 3D printed crank that would be as durable as a machined one, but it'd be heavier, and a whole load more expensive.

Unless you use real bleeding edge 3D processes and materials, in which case it might be lighter, but probably an order of magnitude more expensive.

[tjvirden]

I don't think "never fails" is necessary - how something fails is the really important bit. For most of the history of the bicycle, steel forks have failed - even good ones, after a lot of hard use; however, if they're made properly the failure is never catastrophic. Routine inspection is needed of course.

I wonder if the inclusions and voids in printed metal fundamentally change the failure properties. Surely all this is known, as additive manufacturing has been around a long, long time.

Edit: saying "surely it's known about", I suppose what I mean is that somebody, somewhere understands the detail of failure mechanisms in printed metals - whether that's more broadly understood and embedded in appropriate academic syllabuses, I've no idea.

[mikemelbrooks]

3d printed cranks https://road.cc/content/tech-news/sram- ... 283527?amp

[Pierre86]

I asked tech team at Rotor why they didn't do 3-D printed crankarms. They told me, they aren't durable enough.
I just guess now, but perhaps this can be applied for handlebars to. Just a thought, i have no clue!

The Aus team are all on 3D printed cranks, even the sprinters.
They've even got SRM's hidden in them.

[ghostinthemachine]

I don't think "never fails" is necessary - how something fails is the really important bit. For most of the history of the bicycle, steel forks have failed - even good ones, after a lot of hard use; however, if they're made properly the failure is never catastrophic. Routine inspection is needed of course.

no, that was an over simplification. They just try to make it statistically unlikely to fail in reasonable use. (Be that time/fatigue, over load, whatever)

I wonder if the inclusions and voids in printed metal fundamentally change the failure properties. Surely all this is known, as additive manufacturing has been around a long, long time.

yes, same as composites, they are generally allowed for/compensated for in the design. The problem with one offs, is you don't know what changes/glitches the unique features (or limited producing run) will throw up.

Edit: saying "surely it's known about", I suppose what I mean is that somebody, somewhere understands the detail of failure mechanisms in printed metals - whether that's more broadly understood and embedded in appropriate academic syllabuses, I've no idea.

Yes, but it's very difficult to apply, especially in one offs. Once you have stable production, you can do statistical sampling (as per composites manufacture).