Would you use a 3D printed stem?

https://robotbike.co/

Someone sends me the cad of a stem the same as a traditionally made stem and I will try and get one built in Ti64 or 316L steel and test both the traditional stem and 3d stem to destruction….

BUT, even so…..3-D printing is slow for production purposes.
How long would it take to manufacture a batch of 500 stems using a 3-D printer, compared to traditional techniques…..?

Depends upon what they’re made from. Aluminium?…you’re probably right, titanium? I think you could make a decent business case for AM.

EXAMPLE – A current fashion stem fills a volumetic box of say 50*50*50mm assume 2mm of facing/machining stock on each face. so 54*54*54mm which gives a total mass for the machining blank of 0.7kg. Titanium is around €70/kg (€50 of material per part) for decent stuff. The stem weight after machining is probably around 0.125kg so a buy to ride ratio of 5.5:1 Titanium powder is around €150/kg BUT to build each 0.125kg part, you’d only need 0.15kg of material, so €22.5 per part. Machining each stem at the low cutting rate of Ti would required around 1.25 hours per part. Each 3D printed part would also require 25mins of machining (15 for setup and 10 for machining) for bolts, threads, flanges etc. Making 1 stem in AM would require a build that takes 15 hours…BUT, I can get 25 of them in a single build which takes only 20 hours. So let’s do the numbers… I can make 25 AM stems in 30 hours (20 for build and 10 for combined machining) and 25 conventionally machined parts in 31 hours (all machining), assuming machine €/hour are similar between CNC and 3DP at €50/h, then each 3DP stem costs €72.5 and each Ti machined stem would cost €99.

Further more, if you consider that in an ARCAM EBM Machine, I can nest the build to allow for the simultaneous manufacture of 315 stems, with a total build time of around 110 hours. I can make 600+ stems in 10 days.

There are obviously simplifications here, but the basics are routed in fact.

I’m hugely impressed with Daffy’s contributions to this thread – especially the point about those complex castings and machining jobs involved in building the undercarriage of an airliner.

I think it’s only a matter of time before 3d printed parts on bikes become commonplace, eliminating the need for complex moulds and machining. Would I ride with a 3d printed stem? Heck yes, if it proves to be stronger, cheaper and lighter than the machined stems I have on my bikes right now.

Probably the main point of such everyday items (stems ect) being 3D printed is to showcase the technology. An internal aerospace component doesn’t really mean much to most people , however fascinating it might be to engineers. For me unless it’s the magic bullet, strong, light and cheap I won’t be buying.
P.S That stem above is awfull 😯

Daffy – Member

Depends upon what they’re made from. Aluminium?…you’re probably right, titanium? I think you could make a decent business case for AM.

EXAMPLE – A current fashion stem fills a volumetic box of say 50*50*50mm assume 2mm of facing/machining stock on each face. so 54*54*54mm which gives a total mass for the machining blank of 0.7kg.

So you’re basing your material cost for the traditionally machined ti stem on a solid billet of material that you machine down to the shape you want…..?
But the majority of the volume would just be a titanium tube, cut to length with forms cut on each end to accept the end pieces.
So, the material requirements & presumably costs are way off…?

And a Ti stem is still a high end niche product…..

This is the point I’m trying to make – use the technique that fits the application and cost requirements.

You want to sell some high-end, supposed high tech stems to some mugs people on Kickstarter – knock yourself out; fire up the 3-D printer, get the powdered titanium or whatever out and sell them for £250 a piece.
You want to make some stems that are just as strong (as near as makes no difference) and perhaps weigh a bit more, for 99% of the population to use – then just use a tube of aluminium with some machined bits welded on the ends and sell them for £30-50 a piece…..

If 3-D printing was a viable method of making a bike stem, at a suitable price point people would be doing it.

So you’re basing your material cost for the traditionally machined ti stem on a solid billet of material that you machine down to the shape you want…..?
But the majority of the volume would just be a titanium tube, cut to length with forms cut on each end to accept the end pieces.
So, the material requirements & presumably costs are way off…?

True, most stems are either machined or cast aluminium with the welded Ti stems done purely for cost. Still a welded Ti stem is £250 at retail for a Lynskey.

1. A 3DP equivalent would still be cheaper
2. When a Ti component fails, it fails at the weld…There’s a reason we don’t do it in Aerospace.

You want to make some stems that are just as strong (as near as makes no difference) and perhaps weigh a bit more, for 99% of the population to use – then just use a tube of aluminium with some machined bits welded on the ends and sell them for £30-50 a piece….

In the same way that an aluminium frame is, by some measures, identical to a Ti one…it doesn’t mean there isn’t a market for for the latter, just because 99% of the population would buy the former, now, does it?

I’m not saying that it’s something I want to do, I’m just pointing out that it IS a viable technology and that when used for the correct application, It allows you to make something that can be light, cheap (relatively) and fast. Very few other technologies can make a similar claim.

i hope that when someone does try 3d printing a Ti stem, they remember to hip it…

Daffy, where are you purchasing your ti powder from? we are paying more like £450/kg

A question for those that do — @Daffy etc. How much annealing is needed on these made-from-powdered-metal 3D printed components? I would have thought quite a lot. Does the time and space within which to do that have a big effect on the total production costs?

Shirly considering only the stem is the old fashioned approach? 3D printing gives the ability to think it terms of overall function and not assemblies of easy to manufacture parts. That has already been said but in this case wouldn’t the ‘future’ be a combined bars and stem part which is uniquely fitted and shaped to each rider? 🙂

ahwiles – Member
i hope that when someone does try 3d printing a Ti stem, they remember to hip it…

Hiping a Stem would make very little difference due to the wall thickness; it would likely make it worse.

hexhamstu – Member
Daffy, where are you purchasing your ti powder from? we are paying more like £450/kg

LPW is one, but the quality can vary, so you need to have it independently verified. Sandvic, Alcoa, etc. Chinese powder can be had for as little as €95/k…but you get what you pay for…to a point.

A question for those that do — @Daffy etc. How much annealing is needed on these made-from-powdered-metal 3D printed components? I would have thought quite a lot. Does the time and space within which to do that have a big effect on the total production costs?

We use a stress relieving heat treatment on the laser parts which require it, but no other heat treatment as standard for Ti parts. We do several types of heat treatment of Al and INCo. For larger parts (walls thicker than 3mm) we do tend to HIP the part to close any porosity which may be detrimental to performance. HIP of a single part in a bespoke cycle can be pricy, but cost per part in a packed cycle is vew low 10s>100s of €, nothing more.

We maintain our own heat treatment ovens (they’re cheap to buy, maintain and certify), but use external for HIP due to the NADCAP certification requirements.

Thanks Daffy, interesting.

I suppose the worry with cheap 3D printed stuff (if/when it happens) would be whether the manufacturer skimpled on the heat-treatment. Much like you might not completely trust the lay-up inside cheap carbon frames.

Cost-wise, the overall process doesn’t sound like it ought to be wildly different from machined metsl, if it was done on a mass-production basis.

@hexamstu metalysis in Sheffield speak to them you may be pleasantly surprised

Go big or go home

Daffy – Member

Hiping a Stem would make very little difference due to the wall thickness; it would likely make it worse.

so not hip-ing would be the safer option? – yikes!

HIP closes internal pores, but opens pores close to the surfaces, creating a greater notch-fatigue effect…not what you want on a stem. You’d be better laser polishing and then peening the surface on a thin walled structure.

Thread is delivering.

deanfbm – Member

Thread is delivering. Agreed

Daffy – Member
You’d be better laser polishing and then peening the surface on a thin walled structure.

interesting, thanks.

(how much does that cost? and can you do it to internal surfaces?)

ahwiles – Member
interesting, thanks.

(how much does that cost? and can you do it to internal surfaces?)

MUCH more difficult on internal surfaces, but if light can enter and be projected to the surface at an angle less than 45 degrees to the surface…yes.

For internal surfaces, you’d be better with a chemietch or plasma based approach, especially on a complex surface such as the geodesic stems above.

The 3D stem above is a classic case of being designed rather than engineered.

Why is every structural intersection a sharp point on a part which is subject to fatigue loads and torsion? “Because it looks ace…”

*got meself a new mancrush*

Always thought the most obvious and beneficial area 3D printing could become prolific would be medical prosthetics and the like.

Can only imagine the weight savings for aeronautical applications will usher in some very interesting/different/innovative designs in the future.

What about this one?

The problem with both the stem (which is gorgeous) and the approach in Abaqus (via ntopology) is that you’re giving it mechanical properties from billet material. The problem is that when you get to really fine structures(<1mm), especially in AM/3DP, the mechanical properties may be as little as 10% of what you believe they are.

For that one, I’d really like to see it physically tested to see if it matches the computational predicted behaviour.

As there is an expert here, what would the expected cost be for printing a custom watch bezel?

And are there places offering this service?

Are you wanting a bezel designed and printed or have you already designed it?

If your serious of printing stem i’ve got a conservitive/ traditional model of a 50mm stem I printed in plastic.
Done a bit of designing parts for printing on my bike project and design as always seems to be a key point, the process & function still dictates design to a point.

Couple of questions.
1) in all these hollow designs how is the powder removed from the middle/ cavity? I assume a hole is left some where.

2) how do you get a job doing 3d printing/ design and validation. It’s something i’d seriously consider moving to.

Couple of questions.
1) in all these hollow designs how is the powder removed from the middle/ cavity? I assume a hole is left some where.

Correct.

2) how do you get a job doing 3d printing/ design and validation. It’s something i’d seriously consider moving to.

It depends upon which aspect you want to get into.
A manufacturing/mechanical engineering background for operations, mechanical engineering/numerical Methods for analysis and design,
material Science for materials. There are quite a few aspect/roles.

Would I buy one? Not how they look now – I can’t think I’d ever get one clean. Until they can come out nice and shiny and polished like my CNC’d & anodised alloy version I’m not going near it. At 106g I don’t see there’s much to lose from my Renthal stem – certainly not on a cost/reward basis anyway.

the range of STW experts on this thread is amazing.

the range of STW experts on this thread is amazing.

I’m glad I asked about this now 🙂

Are you wanting a bezel designed and printed or have you already designed it?

I have 3d files from a 6th generation prototype. Which is a little meaningless to me tbh.

3D printing definitely has its uses. We have complex templates made at work that perfectly car bodywork contour’s.

We have been using 3D printed exhaust manifolds on turbo engined race car for a while now, they are running around 1000 degrees and making a lot of power