What's new
What's new

Part Failure: 3D Printed Titanium Handlebar at the Tokyo Olympics

Orange Vise

Titanium
Joined
Feb 10, 2012
Location
California
I'm sure someone much smarter than me can answer this.

What does the grain structure of a 3D printed part look like? How does it compare to cast/forged/drawn material?
 
I'm sure someone much smarter than me can answer this.

What does the grain structure of a 3D printed part look like? How does it compare to cast/forged/drawn material?

Done correctly, it's slightly better than bar or plate stock. At my last job we had titanium parts 3D printed by LayerWise in Belgium, and along with the parts we had a bunch of tensile test coupons printed. The printed material had tensile strength results slightly higher than material cut from plate stock.
 
Done correctly, it's slightly better than bar or plate stock. At my last job we had titanium parts 3D printed by LayerWise in Belgium, and along with the parts we had a bunch of tensile test coupons printed. The printed material had tensile strength results slightly higher than material cut from plate stock.

Perhaps because it was harder than the wrought ti. Exposure to a small amount of oxygen can increase the strength of Ti, but at the expense of ductility. So the printed parts may have gotten some excess O2 exposure, but may have wound up a little brittle?

Origin of dramatic oxygen solute strengthening effect in titanium | Science

High oxygen-content titanium and titanium alloys made from powder - ScienceDirect

Darn glad that rider wasn't hurt significantly. Could have been a lot worse...
 
Surely....that's fatigue then?

Probably crack propagation starting with a small internal defect, and possibly embrittlement during sintering.

Solid titanium bars and forgings usually start from a cast billet that's vacuum arc remelted multiple times to eliminate internal defects.

With 3D printing, you have no idea what's inside. NDT is difficult, if not impossible, to perform on a finished, hollow part with lots of curved surfaces.

For mission critical titanium parts that require high structural integrity, ultrasonic NDT is done on the raw material due to its uniform shape, e.g. a round bar or disc. The resolution of the scan is limited to the surface roughness of the material, so you can't effectively scan a rough blank, but it's relatively easy and inexpensive to finish turn a simple cylindrical blank to a low RA prior to ultrasonic NDT. After NDT, the blank can be milled, and any potential external defects can be detected through mag particle inspection. Internal defects are unlikely to have been introduced during machining.

I think titanium 3D printing still has its place, since the material is inert and has high heat resistance. More care needs to be taken for structural parts though.
 
I think titanium 3D printing still has its place, since the material is inert and has high heat resistance. More care needs to be taken for structural parts though.

My impression is that atmosphere control is very important with Ti printing, as the huge surface area of the Ti powder gives a lot of opportunity for undesirable O or N uptake if there's any leaks in the enclosure or poor gas mixture control.
 
It seems to me that the AMA banned aluminum handlebars on race bikes for this very reason years and years ago. Some parts you just don't want to take a risk on.

Axles was another.

p.s. why would they be running ti anyway ? If there's no rules against it, aluminum worked on 300 lb 120 mph flat trackers. Al is lighter, so why ti ?

(Obviously the correct answer is not "because it doesn't break" :D)
 
The international cycling union, UCI, has specified 6.8 kilograms (15 pounds) as the minimum weight for a bike. Today, bike manufacturers can make bikes much lighter than that, so why is the rule still in effect when large parts of the cycling world would be happy to see it gone?

Many of the ways that a bike rider can try to bend or break the rules that have been put in place by the UCI require advanced equipment to detect, but there is one that requires little more than a bathroom scale.

Since 2000 the minimum weight for bikes used for competitions has been 6.8kg. In essence this means that if you show up ready to ride on something that weighs less, you’re not going to get past the race commissionaires.

The weight rule has been a sore point for riders, teams and bike manufacturers for years. Although the rule was instigated to protect the riders, advances in technology mean that this protection is at best limited, while it in some cases might actually be lowered by the rule.
 
It seems to me that the AMA banned aluminum handlebars on race bikes for this very reason years and years ago. Some parts you just don't want to take a risk on.

Axles was another.

p.s. why would they be running ti anyway ? If there's no rules against it, aluminum worked on 300 lb 120 mph flat trackers. Al is lighter, so why ti ?

(Obviously the correct answer is not "because it doesn't break" :D)

Every dirtbike has aluminum bars.
 
My impression is that atmosphere control is very important with Ti printing, as the huge surface area of the Ti powder gives a lot of opportunity for undesirable O or N uptake if there's any leaks in the enclosure or poor gas mixture control.

Hydrogen embrittlement is a big deal too, from any moisture in the air.
 
I still have my velodrome bikes, as well as one antique olympic bike. It's been a LONG time since I've been on the boards of a velodrome, but velodrome cycling is a cycling sport unlike most others in that, for others, a light, wiry rider has a marked performance advantage over a big, muscular rider. I was in the latter form, so on the boards, my muscle mass allowed me to outperform riders who were much lighter, and the reasons are mostly simple:

Velodrome racing is done on a fixed-gear cycle. They cannot 'coast', and you don't get to use a variety of gear ratios to get up to a given speed, or account for changes in load. You start in the only ratio you have, and it is a rather high ratio. To accellerate, it takes substantially more torque than you'd put into a road or criterium-race frame.

To further complicate matters, velodrome tracks have a very, very steep bank. IF you're on that bank, and not going fast enough, you WILL strike your uphill-side pedal. To prevent pedal-strikes, we run shorter crankarms. The added effect here, is that big heavy leg muscles are running a smaller circle, thus the accelleration from TDC to BDC and back, occurs at a lower G-force, so it takes less energy to 'spin faster', which is the essence of upper velocity limit for a velodrome cycle.

The problem, however, is that the amount of force necessary to do all this, is considerably higher than what's necessary to function in a road race, time trial, or criterium race environment.

When I was velodrome racing, we had aluminum handlebars and stems on EVERY road, criterium, and time-trial frame, but on the boards, it was steel bars and stem, or nothing. Why? Because many of the velodrome sprinters were pushing 260lbs or more... over 100lbs heavier than a criterium/road/time trialer. The forces in sprinting are 80% through handlebar reaction- When pushing down with right leg, you're pulling up hard with right hand, and pushing down with left. This puts phenominal force on the stem, and flexes the bottom bracket HARD.

I've broken steel bars and stems. I've broken crankarms, and bottom bracket axles and cups. I've cracked steel frames right at the bottom bracket, at the head tube, and bent two dozen axles.

The only technological difference between what I did 35 years ago, and what they do today, is aerodynamics, and materials technology. They've re-shaped handlebars to get the frontal area way down... that wasn't allowed when I raced, nor was running disc wheels... and carbon-fiber frames didn't exist. (Read about Graeme Obree for details there)

Handlebars, though, still flex under load. Bottom brackets flex, albeit not as much... but crankarms and pedal axles still break, bottom bracket spindles and rear axles still break, but not as often.

The fortunate part here, is that when that stem broke, he didn't land on the sharp edges with his throat. Velodrome crashes are bad, in that it is very easy to fall onto something very sharp, and one never falls 'pleasantly'.
 
With an aviation background, I would not wish to qualify a material as "suitable for purpose" based solely on a tensile tested coupon.

I would like to see the entire stress strain profile as well as fatigue lifetime. An Izod test would be good also, as the abrupt loading on a track bike is very similar to an impact.

Live and LEARN!
 
Ti alloy is more fatigue sensitive, and more notch sensitive, than steel. Just one tiny imperfection in the print and you'll get a crack starting to propagate.
 
Ti alloy is more fatigue sensitive, and more notch sensitive, than steel. Just one tiny imperfection in the print and you'll get a crack starting to propagate.

I believe that's the reason climbing gear is never made out of titanium - aluminum and steel are the typical materials there.
 
... as the abrupt loading on a track bike is very similar to an impact.

AMEN! over, and over again...

My team didn't have today's tech or high budget, so we had to rely on empirical proof. We had a 900lb steel workbench with a piece of tubing .933" and another 22.2mm ID bored in it. These were common steerer tube IDs for the '70's 80's and early '90's era frames... we would pick out an appropriate stem length and bar width for a given rider, assemble them, and clamp them into the bench, then get the rider standing at the bench, at proper height, wipe a permanent marker all over the bars and stem, and then try to physically break the bars and stem from a standing position. If they broke, we DIDN'T use them. If they didn't, we took a magnifying glass and looked for incipient fracture indications in the black marking. If no marks, we raced it. If you look at many old photos from that era, you'll see that most track-racers had black stems and bars, it was so they would have visible failure indications

Some time not long after, manufacturers started spray-painting and powder-coating bicycle stems of all types, in black. BMX racers started anodizing their aluminum ones in red, yellow, blue, and black.

They did it because they liked the contrast. WE did it because it just might save our lives.

Aside note- the worst injury I ever got whilst track racing, was a broken pedal spindle and crankarm. Left crankarm broke at the bottom bracket spindle, right pedal spindle broke at the inner bearing. Left leg ripped open from the exposed spindle, struck the boards, and wound up in the back wheel spokes, against the chainstay. Right spindle carved a notch in my right leg, but missed the tendons. That foot hit the high-side, and I rolled about four-five times down the boards, tumbling other riders like bowling pins. Splinters and bruises, and a trail of blood for a finish-line... Ahh, to be young, strong, and crazy again...
 
Ti alloy is more fatigue sensitive, and more notch sensitive, than steel. Just one tiny imperfection in the print and you'll get a crack starting to propagate.

Yup, hence the need for triple-VAR in aerospace applications. The smallest imperfection will propagate over time. United Airlines Flight 232 - Wikipedia

Titanium's best qualities are high heat strength and inertness. Strength to weight is secondary. It's really the consumer market that has glamorized Titanium as the end-all-be-all in materials, along with carbon fiber.

Titanium has its place in 3D printing, but people are going to have think more and more about how to get around the material's shortcomings. Lattice structures could help.

I would guess that the manufacturer of those handlebars wasn't diligent with running FEA on their designs, but they probably are now.
 
With an aviation background, I would not wish to qualify a material as "suitable for purpose" based solely on a tensile tested coupon.

I would like to see the entire stress strain profile as well as fatigue lifetime. An Izod test would be good also, as the abrupt loading on a track bike is very similar to an impact.

Live and LEARN!

Toss a laue x ray defraction into there as well lol
 








 
Back
Top