Customizable 3D Printed Lift/Cant wedges for F2 bindings

[jim_s]

Yeah, just always be sure to take a good look at the thin edge of the shim, to be sure that both the edge and the ridge/groove wall thicknesses are sufficiently thick for your purposes. (And confirm this before you send it to print!  Changes to lift and/or cant will often require a little experimentation and adjustment of the base depth value, to be sure that things are sufficiently thick.

[jburk]

Might be beneficial to include a non-printable object in the scene located near the edges of the shim whose height could be compared to the thinnest part of the shim for a visual reference for minimum thickness.  

"Shim's thinnest edge must be thicker than this cylinder is tall" sort of thing.

[jim_s]

If you're working directly in OpenSCAD, you can turn on the axes and scale markers to see ruler-like marks on the axes ( icons in OpenSCAD), but you have to be zoomed in pretty far for them to get down to the unit (mm) level, and you have to be looking directly from the side view, and in Orthogonal view ( are the left and right side view icons, is the Orthogonal view icon) to get an accurate view, and even then, some squinting is required.

I'll play with the numbers a bit this weekend - I should be able to calculate the base depth needed to achieve a minimum edge thickness and/or groove/ridge wall thickness, and then either output that in the console window, or as you suggest, include some reference in the "Preview" layout.

[JRAZZ]

23 hours ago, jim_s said:

it was around $550

I wonder if that's because it's "solid". I bet if I make a version that's open on the bottom the price would drop.

Still, that's as expensive as I thought it would be 

[jim_s]

Yeah, I think its all about material volume. (I don't know if maybe print time plays into it, as well, or not.) The F2 shims are pretty low volume, but even then, $60 for a full set (ie, both feet, in the nice SLS Nylon) is pretty expensive - and that's for a 'reasonable' set - something in the 3-4 degree range for lift and cant, and minimal base depth. The following set would be double the price (I checked! ), though I'd probably pay it...

 

 

Edited April 21, 2018 by jim_s

[jburk]

This is from the other thread on this topic:

@Donek: did you ever get a chance to try out printing in some shims in PETG?  

 

[Beckmann AG]

10 hours ago, jim_s said:

The F2 shims are pretty low volume, but even then, $60 for a full set (ie, both feet, in the nice SLS Nylon) is pretty expensive - and that's for a 'reasonable' set - something in the 3-4 degree range for lift and cant, and minimal base depth.

In 'real world' terms, that's seriously cheap for a functional one-off. 

[jim_s]

3 hours ago, Beckmann AG said:

In 'real world' terms, that's seriously cheap for a functional one-off. 

Agreed, and given that a set of stock wedges (ie, buying both lift and cant stock sets for one foot) will run about $24 per foot, $30 isn't far off of that for exactly the settings you want, down to whatever nth of a degree you desire. But, one would hope that it could be even cheaper than that. (And, it is if you go with an FDM service - you're then talking about $10 per foot, but the SLS is likely to be far more of a worry-free, long-term/permanent solution, I suspect.)

Truth be told, despite the fact that I have a perfectly good FDM machine at home, I'm inclined to order a set printed in nylon via SLS, just because I think it would be a better long-term result.

[jburk]

I’m planning on experimenting with cant/lift combos in ABS on a $200-300 home printer to keep the costs down, then getting a final set printed in something like SLS nylon by a service.  Should be close to break even with 5 to 7 different variants, and much quicker turn around. 

[Donek]

15 hours ago, jburk said:

This is from the other thread on this topic:

@Donek: did you ever get a chance to try out printing in some shims in PETG?  

Not yet.  Hopefully this weekend.

 

 

[MarkJeangerard]

Jim, this tool is awesome. I am not a programmer. Nor am I an engineer. It would have taken me a month to figure out how to create a single cant lift shim in SketchUP. I probably never would have found success in OpenSCAD. Now I'll have my first prototype in a week or so.

Your adjustable wedge template in OpenSCAD is astonishingly easy to use and you and others explanations in this thread and elsewhere on the forum  give all of us desperately needed solutions. And I don't think that is over stated. The implications of uncommon parts never being out of stock alone! 

@Jack Michaud I think this type of effort is going to revolutionize binding setup, among other things. Some sort of sticky will probably become appropriate, guiding those who seek towards threads focused on self generated solutions for their particular boots and bindings and whatnot. 

[MarkJeangerard]

Ok. So.

3DHubs. Uploaded the STL files, chose materials (ABS Black), searched by closest printer to me (https://www.3dhubs.com/service/saiteik3d in Henderson, NV). 18 bucks, I'll pick them up. 

He responded that he didn't have enough black today. I said red is fine, chances are I'll ride these one day only and then want to make changes. 

Print was ready in 2 hours. In my hand in three. (Las Vegas,  NV) I told him I didn't expect the labeling to come out so good. He said it was a matter of prepping the bed. But no details. He said he used 15% infill. 

If you're wondering how I came up with the angles they are very close to the last setup I rode. I took notes on the day. 

I'm posting pics. I would love know what anyone thinks about the quality. 

---

I just installed them on the bindings. I'm glad all the local ski areas are closed. I want to ride right now. But there are two problems. I just reread Jim's Thingiverse page. He says to use a high infill level. I did not specify and the guy who printed them guessed at 15%. What would be considered "high" infill?

Also, it has become very apparent that the 22mm, 26mm, and 36mm bolts that come with the bindings do not provide enough options. One of the bolts goes right to flush with the T-nut and two others only have 1 or 2 threads holding. (I did not tighten them. I was just excited to feel the stance.) Depending on how stupid I go with revisions, I might want a box with 22, 24, 26, 28, 31, 34, 36, and 39mm bolts in it. I would say don't even start without 31mm bolts in hand. It's not the short and long extents that get ya. It's the intermediate sizes. 

They fit the bindings as well or better than the factory shims. The tongue and groove wiggle a little on both these and factory. A little less on these depending on the individual shim. 

Boy, oh boy do the angles feel natural. I need to ride them, of course, but it sure is a lot better than just throwing 4s and 6s at the problem. I can tell that on the carpet. 

 

 

Edited April 26, 2018 by MarkJeangerard
I just installed them.

[yamifumi]

Wow, this is awesome. Now I need to figure out if the shims from F2 fits SG bindings or not.....

[jburk]

@MarkJeangerard: What material did you have these printed in?

[jim_s]

@Mark Jeangerard - Overall, I think the print quality on these looks pretty good. (I like the red, too! )

The banding on the underside is expected - the printer can only print in flat planes, so it has to approximate a sloped surface by making a series of stepped planes. A low layer height will result in smoother steps, but some studies have been done that show lower layer heights can actually result in weaker materials, as regards elongation, bending, etc. (Now, in our case, the part is primarily in compression, so a denser/lower layer height might be more beneficial than if the part were being stretched and bent, but I haven't done any comparative tests on that...) If you look at the attached photos, the layer height on that print was .1575mm - so, smoother sloped surface (my print has 25 bands across the bottom, whereas his has 10 bands), but I can't say whether that has any real world positive effect, beyond aesthetics.

I'm impressed with the detail he achieved on the labeling (I'd love to know his surface prep secrets!), but on the other hand, I tend to run my extruder and bed on the hot side, and I tend to over-extrude (ie, the rate at which the filament is fed through the hot end that melts and applies it), and run a low layer height - I joke that I like it hot, wet, and sloppy!  - as a result, though, I tend to get little gaps and spaces filled in quite solidly (by contrast, you can see the individual filament beads on his print, and there are some areas that don't fill all the way in, such as the little void between the bolt hole and the edge of the shim). I tend to believe that the 'hot, wet and sloppy' approach will result in a more solid part, but again, I've not done any comparative testing to see if the end result is physically stronger, or not.

As regards infill percentage, I'd be wary of something as low as 15% - for a decorative print, that would be fine, but for something structural, I'd definitely recommend making the infill about as high as you can. (I've read various articles that seem to indicate there are diminishing returns above around 60-70% infill, though 100% will almost certainly still be stronger than 70% - just likely not by as much as the difference between 70% and 40%, for instance.) I also adjust my 'shells' setting, so that the walls of the shim come out 100% solid ('shells' dictates how many top, bottom and side layers are made solid, before infill starts being used - I adjust my shells so that the vertical walls don't end up using any infill, but are instead just solid material).

All these various settings are a means of managing trade-offs in the final print and the time and material it takes to achieve the final print. You're basically shuffling around various quantities such as time, material usage, print quality (aesthetic) and strength. If you're willing to sacrifice time and material (ie, cost goes up), your quality and strength can go up. If you're more interested in cutting time and/or material, then your quality and strength will likely go down. For something structural like this, I think the trade-off should definitely be maximized to prefer strength.

There are a few informative articles and discussion on this out there. Ones I've found particularly informative include:

http://my3dmatter.com/influence-infill-layer-height-pattern/

https://revistas.unal.edu.co/index.php/ingeinv/article/view/56610/58902

 

[jburk]

@jim_s, can you clarify a concept for me? Are the number of shells related to the "wall_thickness" parameter in your published .scad file, or is the shell value solely for printer control and wall_thickness affects the geometry in the exported STL?

[JRAZZ]

Regarding infill, the shape of infill is just as important in here as is the percentage.

As a rule of thumb plastic can carry approx. 3kg for each sq. mm in compression. Assuming that the infill he used is "vertical" then the ultimate failure load for your shims would be the area[mm^2] x 15% x 3.  A more accurate calculation would include the walls but even if you run the simple case you'd find that the load is in the hundreds or  thousands of KGs, e.i. tons. If the infill is a cubic or something different then the value is lower but honestly I can't see these failing unless you hit a really really bad landing.

 

That being said I try to run my prints with 1.2-1.6mm wall thickness and 40% infill (triangles). I found that this gives a very good compromise between ultimate strength, weight, and print time. Anything above that is probably not going to add anything significant.

 

@MarkJeangerard I wouldn't be concerned with the prints. They look fine to me. Even if they were to fail they would compress which seems to me a safe mode of failure.

[jim_s]

@jburk - yes, you've got that correct. The wall_thickness parameter in the OpenSCAD file controls how thick the wall is rendered in the virtual 3D model (the STL file generated from OpenSCAD). The shells value is something you specify in the 'slicer' software that converts the virtual 3D model in the STL file into a series of layers that the 3D printer prints. (So, the slicer and its settings are separate from the OpenSCAD file and its settings - except as noted below.)

FDM 3D printers have a nozzle with a fixed diameter (0.4mm is very common) through which the melted filament is fed, so a bead of filament's height and width will generally match the nozzle width (not taking into account any 'squishing' associated with the layer height). The slicer, in addition to slicing the 3D model up into layers (where each layer corresponds to a slice that is the thickness of the chosen layer height), controls things like how thick to make external walls (shell thickness), how much infill to use, etc. (Some slicers have the shell thickness specified in mm, and some specify the shell thickness as the number of shell layers, where each layer is assumed to be the thickness of the nozzle diameter.)  There are usually settings for both vertical shells (sides of a print) and horizontal shells (top and bottom surfaces of a print). A typical shell thickness setting is 0.8mm (or 2 shell layers, for a 0.4mm nozzle), which is fine for decorative items. For something structural, thicker shells will be stronger. The area between the shells in a print will be filled with infill. (Infill being a way to fill larger volumes using a matrix of filament, with gaps in the matrix, to save on filament use and time to print.)  So, if you have a wall_thickness in the F2 Shim model of 5mm, and you were to specify a shell thickness of 0.8mm (not recommending this, just as an example), that would leave a gap of 3.4mm between the shell walls, which would then be filled with a matrix pattern of infill material, the density of which is indicated by the infill percentage. (This is an area where your slicer shells setting is related to your OpenSCAD model's wall_thickness setting...) If you specified an infill percentage of 100%, then you'd get a solid wall in your print, no matter how thick your wall is specified in OpenSCAD. Alternatively, if you don't want to use 100% infill, but still want the walls to be solid (in these shims, the walls are what are bearing the vast majority of the forces...), then you could instead just set your shells thickness to 2.5mm (each shell - the one on the outside, and the one between the outside and the hollow pocket underneath) will be 1/2 the width of the wall. Or, if specifying shell thickness in layers, 6 layers at 0.4mm each would give you 2.4mm thick shells, which would technically leave a 0.1mm gap in between, but at least with my hot, wet and sloppy printing method, that gap gets filled in. If you're daintier with your print technique, then you'd want to up the shell layer count to 7, to be sure you got a solid wall.

The top and bottom shells will be whatever thickness you specify (in the slicer) for the horizontal shells (different slicers call it different things), but as the shims are typically lifted and/or canted, one side or end is going to be thicker than the other side or end, so its more difficult to force a completely solid top and bottom based on the number of horizontal shells - what works for the thin end/side won't work for the thicker end/side. So, you can revert back to using 100% infill (definitely strongest!), or you can accept that there is going to be some infill in the 'roof' of your shims. (I personally don't worry too much about infill in the roof - I think the walls are what are really bearing the forces.)

The following article goes further into explaining some common slicer settings - its worth a quick read:

https://www.3dhubs.com/knowledge-base/selecting-optimal-shell-and-infill-parameters-fdm-3d-printing

 

[jim_s]

11 minutes ago, erazz said:

As a rule of thumb plastic can carry approx. 3kg for each sq. mm in compression. Assuming that the infill he used is "vertical" then the ultimate failure load for your shims would be the area[mm^2] x 15% x 3.  A more accurate calculation would include the walls but even if you run the simple case you'd find that the load is in the hundreds or  thousands of KGs, e.i. tons. If the infill is a cubic or something different then the value is lower but honestly I can't see these failing unless you hit a really really bad landing.

I freaking love having you involved in all of this, @erazz!!

[Donek]

I had warping and bed adhesion issues with the PETG.  On the 3rd attempt, I just gave up and printed on a raft.  That fixed the warping, but the lettering looks crappy.  My riding this Friday will be in softies, so no testing unfortunately.  Maybe next Friday.

[JRAZZ]

I hate PETG

Still, lettering aside, looks pretty good!

[Donek]

3 hours ago, erazz said:

I hate PETG

Still, lettering aside, looks pretty good!

It can be a bit stringy., but it's stronger than PLA and a better low temp performer than ABS.  It claims to be as easy to print as PLA, but I think it's somewhere between PLA and ABS. I was surprised by the warping as I had not seen any information on that being a problem.

[MarkJeangerard]

16 hours ago, jburk said:

What material did you have these printed in?

Standard ABS is what the web form said. Saiteik only mentioned that it was ABS. (I went back and edited the post, and put the printers' 3DHubs link, as well.)

12 hours ago, erazz said:

I wouldn't be concerned with the prints. They look fine to me. Even if they were to fail they would compress which seems to me a safe mode of failure.

Would that kind of failure be fairly obvious? I intend to check the bolt tightness, inside of the rubber T-nut 'cap', and snowboard top sheet regularly.

15 hours ago, jim_s said:

The banding on the underside is expected - the printer can only print in flat planes, so it has to approximate a sloped surface by making a series of stepped planes. A low layer height will result in smoother steps, but some studies have been done that show lower layer heights can actually result in weaker materials, as regards elongation, bending, etc. <snip>

The 3DHubs order form specifically asked if I wanted a 300μm or 200μm layer height I did a quick google search and came to the same conclusion so, went with 300. I'm glad to have that point reinforced here. 

[JRAZZ]

3 hours ago, MarkJeangerard said:

Would that kind of failure be fairly obvious? I intend to check the bolt tightness, inside of the rubber T-nut 'cap', and snowboard top sheet regularly.

Checking the bolts for tightness is definitely the right thing to do. I expect there to be some relaxation after the first day or so.

I think that a compression failure would result in a loose feeling. The bolts should still hold you in place but the boot would feel loose. 

[jburk]

On 4/24/2018 at 11:30 PM, MarkJeangerard said:

Also, it has become very apparent that the 22mm, 26mm, and 36mm bolts that come with the bindings do not provide enough options. One of the bolts goes right to flush with the T-nut and two others only have 1 or 2 threads holding. (I did not tighten them. I was just excited to feel the stance.) Depending on how stupid I go with revisions, I might want a box with 22, 24, 26, 28, 31, 34, 36, and 39mm bolts in it. I would say don't even start without 31mm bolts in hand. It's not the short and long extents that get ya. It's the intermediate sizes. 

I'm wondering if the way to go would be buy 40mm bolts and a rethreading (thread repair) die.  Mark it to length, thread on the die, another 2 nuts to act as a guide, and cut the bolt with a dremel.  Can anyone out there tell me if the die is overkill, and will just filing a chamfer and taking the nuts back off clean things up enough?

These bolts look like they'll be a good replacement for the stock hardware, the head height is a bit lower than stock (3.3mm vs ~3.6mm) and the flange width looks to be the same at 13.6mm with my meagre skills with a $5 caliper.