3d printed stem spacers???

[Queen of Skulls]

I hear ya.  If you're set on using a consumer FDM printer, you can design in some holes for metal pins.  Then you'd have some more solid material to take up the preload vs the plastic.

Best of luck!

I mean you just do solid infill. It becomes like a solid piece of plastic. I also have a resin printer but I would have to see about what resin is the best if there is one. for  this job.

[Sakizashi]

I dont know if you will get right properties from a resin printer. Even 100% infill might not be enough. Might be worth looking at your material and see how samples lab test and make sure you have a good margin of safety. Or you can tighten a sample up to 4nm or something and see if it creeps enough that you need to retighten it after a few weeks.

My $.02: Once you have the fit worked out using the home printer, get the functional ones made by someone with a nice high end machine. MJF parts are cheap and while glass filled nylons are 2x the price of standard Nylon12, I believe MJF Nylon12GB (also Pa12gb) are what the 3d printed headset parts used on pro bikes use.

[Queen of Skulls]

I dont know if you will get right properties from a resin printer. Even 100% infill might not be enough. Might be worth looking at your material and see how samples lab test and make sure you have a good margin of safety. Or you can tighten a sample up to 4nm or something and see if it creeps enough that you need to retighten it after a few weeks.

My $.02: Once you have the fit worked out using the home printer, get the functional ones made by someone with a nice high end machine. MJF parts are cheap and while glass filled nylons are 2x the price of standard Nylon12, I believe MJF Nylon12GB (also Pa12gb) are what the 3d printed headset parts used on pro bikes use.

mmm... I smell an experiment. trying out different materals.  can do cf molds too. but I think the pa12 is the best best longterm. or maybe asa. as its very uv resitstant.

also I could get some machined out of al too. maybe.

[gmh]

also I could get some machined out of al too. maybe.

I feel like at that point you might as well just buy another set from Velobuild.  The spacers I got from them were ~$30 usd shipped.

[Queen of Skulls]

But this is custom spacers. Like I am going to do a 50mm- 70mm.. also making it only one layer increases rigidity and reduce the likely hood of an issue.

[strayan_rice_farmer]

I've ran some 3d printed stem spacers on my bike for ~1000km after I built it up until the white color irked me enough to order some cheap carbon ones.

It was printed out of 100% PLA, didn't have any issues with creep in that time.
Since the load is compressive and fairly low, there will be hardly any creep especially if you print it 100% infill.

That being said, use PETG or something that won't deform in the heat

[joegal]

I've ran some 3d printed stem spacers on my bike for ~1000km after I built it up until the white color irked me enough to order some cheap carbon ones.

It was printed out of 100% PLA, didn't have any issues with creep in that time.
Since the load is compressive and fairly low, there will be hardly any creep especially if you print it 100% infill.

That being said, use PETG or something that won't deform in the heat

3D-printed spacers above the stem are a whole different story. They are not loaded and therefore not critical at all. They are only loaded while you tighten the headset. Theoretical (I ofc do not advise) you could remove the spacers after you tightened the stem bolts.

However, a lot of Allez Sprints use 3D printed spacer adapter for the SL7 stem, as a loaded part...

[Queen of Skulls]

3D-printed spacers above the stem are a whole different story. They are not loaded and therefore not critical at all. They are only loaded while you tighten the headset. Theoretical (I ofc do not advise) you could remove the spacers after you tightened the stem bolts.

However, a lot of Allez Sprints use 3D printed spacer adapter for the SL7 stem, as a loaded part...

Well from what I can tell the spacers used are pretty flimsy and are likely abs. As nylon is vulnerable to uv. It's very likely the load is being handled in the front and back. So as long as they are beef and I use a plastic like abs or Asa... Petg would be fine too. Worst case scenario is I keep an eye on it. And change it out if there are signs of wear or compression. However when you are riding. Your weight pushes the head tube down on to the lower bearing of the head set. And releasing it on the top side where the spacers are. These return when you put axial force on the stem. But no where as great. As it would snap the fork long before then. Worst case scenario I get them printed in pa12 or something similar. Also al machining would be easy and cheap

Most of problems is that parts aren't 100 percent infill.

[Sakizashi]

I am really sorry if this comes off as rude, but I think it might be worth taking a step back on this project and clarifying your assumptions about the current design and limitations before proceeding. There are a lot of assumptions about how this will perform relative to the original design that could have serious safety implications.

1. Spacer stack height. 50mm is a lot of spacers to put under a carbon stem. 70mm is playing with fire IMO. Depending on manufacturer that's right at the upper limit of what would be recommended. Most like Trek and Specialized are in the 40mm range for the max amount of spacers. Cannondale is the exception where I think on some models they have said no more than 55mm of spacers. An extra long compression plug like the 90mm ones sold by Origin 8 might help, but the idea of that many spacers is already pushing the limit on designs in an industry that arguably is already playing with thin safety factors.
2. Design. Having a single, rigid, piece will help limit the steerer from bending but this is going to be dependent on getting a great fit to the steerer tube and the cone spacer under it. A split spacer will make that much harder to achieve.
3. Production method. Given the 50mm-70mm of spacers and the use of FDM maintaining preload may be a challenge. Depending on your layer thickness and assuming you have the orientation controlled to prevent shearing, even a 100% infill part is likely 80% as strong as the equivalent injection molded part. It may even be less as it will be hard to get a part that size made without voids.
4. Load on the fork. I don't think there is an issue here with the overall load on the fork with that many spacers, but that will depend on how close to the limit you are pushing things with total system weight, fit, and stem length. Certainly worth checking.
5. Material. I don't know what Velobuild uses for their spacers, but most of this type of stem spacer claim to be nylon or nylon/carbon. UV isn't an issue as most use stabilizers and the end product is considered UV stable. This is also true for many MJF and SLS PA12 based nylons.

The price of failure here can be high as the resulting bending moment can cause the fork to snap above the headset bearing.

[Queen of Skulls]

I am really sorry if this comes off as rude, but I think it might be worth taking a step back on this project and clarifying your assumptions about the current design and limitations before proceeding. There are a lot of assumptions about how this will perform relative to the original design that could have serious safety implications.

1. Spacer stack height. 50mm is a lot of spacers to put under a carbon stem. 70mm is playing with fire IMO. Depending on manufacturer that's right at the upper limit of what would be recommended. Most like Trek and Specialized are in the 40mm range for the max amount of spacers. Cannondale is the exception where I think on some models they have said no more than 55mm of spacers. An extra long compression plug like the 90mm ones sold by Origin 8 might help, but the idea of that many spacers is already pushing the limit on designs in an industry that arguably is already playing with thin safety factors.
2. Design. Having a single, rigid, piece will help limit the steerer from bending but this is going to be dependent on getting a great fit to the steerer tube and the cone spacer under it. A split spacer will make that much harder to achieve.
3. Production method. Given the 50mm-70mm of spacers and the use of FDM maintaining preload may be a challenge. Depending on your layer thickness and assuming you have the orientation controlled to prevent shearing, even a 100% infill part is likely 80% as strong as the equivalent injection molded part. It may even be less as it will be hard to get a part that size made without voids.
4. Load on the fork. I don't think there is an issue here with the overall load on the fork with that many spacers, but that will depend on how close to the limit you are pushing things with total system weight, fit, and stem length. Certainly worth checking.
5. Material. I don't know what Velobuild uses for their spacers, but most of this type of stem spacer claim to be nylon or nylon/carbon. UV isn't an issue as most use stabilizers and the end product is considered UV stable. This is also true for many MJF and SLS PA12 based nylons.

The price of failure here can be high as the resulting bending moment can cause the fork to snap above the headset bearing.

I got the compression tube 80mm, I am shooting for 60 to 70mm. I don't go hard anyway. And the plastic that they use is probably nylon so something like Asa is going to be more stiff and r Ridgid van nylon also the use of One Versus several. The use of several is going to be more flexible than one. Right now, I'm testing in pla because it's cheap and then when I'm ready to do it for real it will be probably Asa as I'm thinking that pet G a little bit more brittle.

[Queen of Skulls]

3. Production method. Given the 50mm-70mm of spacers and the use of FDM maintaining preload may be a challenge. Depending on your layer thickness and assuming you have the orientation controlled to prevent shearing, even a 100% infill part is likely 80% as strong as the equivalent injection molded part. It may even be less as it will be hard to get a part that size made without voids.
4. Load on the fork. I don't think there is an issue here with the overall load on the fork with that many spacers, but that will depend on how close to the limit you are pushing things with total system weight, fit, and stem length. Certainly worth checking.

The price of failure here can be high as the resulting bending moment can cause the fork to snap above the headset bearing.

I came across this... https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7041383/

The compression test results are shown in Figure 9 and indicate improvements of only 4% in the injected samples in comparison with the 3D printed ones (100% infill). It is worth noting that Young’s Modulus is higher in the printed parts (50% improvement with respect to the injected samples). This suggests that when compressive loads are applied, the 3D printing process leads the parts to show higher stiffness than in the IM process. In compression tests, this is due to the direction of the force being opposite to that in the tensile test and the separation of consecutive layers of the printed parts is more difficult. Thus, using high infill density values avoids an early breakage between consecutive layers. Qualitatively, it can be observed that the behaviour of the 3D printed part decreases as of a determined strain. Furthermore, the decline appears just when the compressive stresses of both samples are equal. This decrease is a consequence of the separation taking place between consecutive layers. This conclusion is supported by the enormous separation observed in Figure 10B corresponding to 60%-filled parts, while this occurred incrementally in completely filled samples. In the 60%-filled non-unidirectional samples (Figure 10C,D), polymeric hardening took place under compression stresses (Figure 11). This can be an advantage for parts that work under compressive stresses with no deformability constraints. Inversely, the unidirectional pattern broke down due to the presence of gaps between consecutive layers. Improvements up to 73% in yield strength and 33% in Young’s Modulus were reached when 100% instead of 60% infill density was used, as seen in Figure 9.

So I dont think there is a problem with compression. Its more of the shearing load which I doubt will be that great.  Seems like 3d printing adds stiffness compared to injection as stated in the paper. Which I can tell that the pla part is more stiffer than 2x the Hight  abs injection part.  So as long as I use 100% infill for compression it should be as good as IM. As I said I wont be doing any crazy racing or pushing it.

[Sakizashi]

A couple of thoughts.

This paper is comparing short fiber reinforced polymers, not the materials you have proposed using, there is some discussion in other papers about whether or not 3d printing can actually help you gain directional strength because of the ability to control fiber orientation.

The caution i am expressing regarding FDM vs. production quality injection molded parts is also what you will hear if you contact production experts at places like Xometry that work on rapid prototyping with all of these technologies and is what most engineers who work with these technologies will tell you. For example, when HP says "MJF produces a part with anisotropic differences that vary between 95% and 90% of an injection molded part" they are, like these papers, assuming the same formulation of material. Unfortunately even for something like "standard" PA12 there are fairly significant differences between the materials themselves that would be used for different production processes including FDM, SLS, MFJ and injection molding. You can pull the data sheets from Stratasys, HP, etc and see for yourself.

Of course this stuff continues to change and today we see 3d printed parts being used for small production parts and on race bikes at highest levels. That would have not been the case only a few years ago.

Obviously the risk is yours to take and it very well may work fine, but this project is pushing the limits of both the materials and the design envelope for most carbon bikes in multiple ways. Good luck!

[propje]

I designed and printed my spacers on my 3d printer.

The main problem was that I have a Velobuild frame, but the integrated handlebar with the shortest stem was too long. So I bought a Kosevlobike handlebar with a shorter stem. Since the shape of the velobuild frame is not the same as the Kosevlobike handlebar, I designed a part to make this transition.

After several months dailing in my position on the indoor trainer, I got the correct spacer height. For me it was 45cm. Quite high, but I am not as athletic as I want to be. Since I don't think I will change the height anymore, I didn't design a 2 piece part. I believe this might decrease the rigidity and the impact strength.

For the materials, i used PETG.
I had bought a spool of carbon fiber enforced PETG, but I have read that the carbon fiber envorced PETG has a lower impact strenght in the Z-Axis ( https://blog.bambulab.com/bambu-carbon-fiber-filaments/ ). And for as far as i think, this Z-Axis impact strength is the most important one. I live in Flanders, and the roads and cobbles give quite some battering, which translates to impact on the Z-Axis.
And, since the impact on the Z-Axis is on compression, I don't think that layer-adhesion is a big issue.
For the creep, I have left some milimeters to retighten afterwards. If I have to tighten to frequent, I will change the material or reprint.

I used a 100% infill

I have ridden it now only for 60km. Including a downhill cobblestone section. The spacers held well.

If the spacers don't hold, I will update my findings here. So, unless if I had a very catastrophic failure, when there is no repost, the PETG spacer is holding on.