Poker Chip Racks (3D printed) (1 Viewer)

[Nex]

The effects of being banned to home office during Corona...

I previously said I didn't see any reason to explore this area.
But now that I was repurposing all the spinners I got with my HCE CPC chips as a small separate set, my view has changed.

Making just three big racks in some nice colored material wouldn't be using that much filament that I'd have to buy additional spools, so I went for it. I still am unwilling to print so many racks that I could store one of my original CPC sets in whole in them (~3300/1600/1700 chips). Just for the tighter fit, it's definitely not worth it to me. For the looks though, the wood filament printed ones are awesome when stained.

My new Prusa Mini's build volume isn't big enough to spit out 5x20 racks, so I've been printing those on my MK3S. Instead, I made some 3x20s on the Mini.

These smaller racks are intended to be used by players at the table. Initially they would hold their starting stacks or standard buy-in, then once they are seated and placed their chips on the table, the rack could be used to put top-off/rebuy chips in them in a cash game to keep them separate from the live chips. (You buy these in advance, but they aren't live as long as they're in the rack. You can move chips from the rack to the table between hands to top off your live stack, and of course only up to the max allowed buy-in. Effect: less work with bank handling)

I built optimized custom support structures right into the models for the overhangs on the underside of the racks. Printing them is still tricky and requires the material to be used to have somewhat decent bridging performance, but you can break the supports away fairly easily with minimal residue, and they use much less material than autogenerated ones. The small wings on the underside of each barrel are there to keep a rack seated properly on top of another rack of chips.

Models are parametric, so I can generate them with any number of barrels and any barrel capacity. Barrel diameter adjustable as well of course. Even multiple rows of barrels are possible. Only limited by build volume of the 3D printer.

A 5x20 rack takes 7 1/2 hours to print and consumes about 80 grams.
The smaller 3x20 ones take 4 3/4 hours and 53 grams each.

This goes for a 0.4mm nozzle. Printing with a 0.6 obviously is faster, although they still take a lot of time (5 hours for 5x20).

Only took me two prototype prints until I ended up with a solid, working design. Even the prototypes worked, but on the first the chip fit was so tight that you'd have to pull out the barrels with force, and on the other, there was a rounding/detail error with the supports which led to some drooping filament on an otherwise successful print.

Had some failed 5x20 prints due to warping. The results also work, but don't look 100% good. Made some adjustments to placement on the print bed and now it works reliably in this aspect. A few more failed due to nozzle clogging from a filament spool that was tainted with foreign particles. New spool, fresh nozzle, all fine.

 

[RainmanTrail]

Very cool. I think I'm going to have to get some of that wood filament next.

 

[dutch_whaley]

Nice work, boy I wish I had a 3D printer for these quarantined times!

 

[One Eyed Dollar]

Very cool! The wood looks awesome. What software do you use for the parametric modeling?

 

[Nex]

What software do you use for the parametric modeling?

Autodesk Inventor. Sadly the results aren't portable/readable/editable by other software as far as I know, but I did look at free CAD software and was everything but convinced. OpenSCAD looks like a PITA to create complex models to me. It's already challenging enough to make parametric models using all the GUI tools and aids that Inventor offers... and Fusion 360 forces you to save your designs in the cloud.

 

[Amish Rabbi]

Small ones look very handy

I’m still surprised everyone with clay chips is ok with storing them with the ridges from the built up 3D printing layers. I figured that would be a deal breaker for everyone

 

[Nex]

The wood filament prints can be sanded to reduce the effect.

Or you could use a nozzle with even thinner diameter and lower layer height to print, resulting in much less visible layer lines at the cost of printing speed.

They're most visible in the recesses for the chips, so when you have chips in the rack you don't see that anymore. The texture at the top layer isn't that bad, especially if you increase the number of perimeters, then it's all straight parallel lines and no more zigzag pattern.

In any way, if you look at these (0.4mm nozzle, 0.2mm layer height) from maybe 3/4 of a meter distance, it's barely visible even when not doing all those tweaks.

 

[One Eyed Dollar]

Small ones look very handy

I’m still surprised everyone with clay chips is ok with storing them with the ridges from the built up 3D printing layers. I figured that would be a deal breaker for everyone

Yeah, I've been experimenting with trying to make them as smooth as possible, and I can still see some wear on the chips. Sanding the wood racks should take care of it I would think. I print my little shuffle stack racks with a 0.6mm nozzle and 0.15mm layer height, with ironing, and there is still some roughness. But I figure shuffle stacks are getting worn down anyway, so it's okay for those.

 

[ekricket]

Yeah, I've been experimenting with trying to make them as smooth as possible, and I can still see some wear on the chips. Sanding the wood racks should take care of it I would think. I print my little shuffle stack racks with a 0.6mm nozzle and 0.15mm layer height, with ironing, and there is still some roughness. But I figure shuffle stacks are getting worn down anyway, so it's okay for those.

if you printed these sideways would that change your ability to make rounded chip cavity’s? Or at least make rounded ridges?

 

[One Eyed Dollar]

if you printed these sideways would that change your ability to make rounded chip cavity’s? Or at least make rounded ridges?

It definitely makes the slots for the chips smoother, but one of the sides becomes an overhang so you have to add a lot of support, and when you break the supports off, that side is usually rough and has to be cleaned up. It's tricky.

 

[Nex]

Not to mention dimensional stability would be the lowest in the most stressed direction if printed that way.

 

[Amish Rabbi]

The wood filament prints can be sanded to reduce the effect.

Or you could use a nozzle with even thinner diameter and lower layer height to print, resulting in much less visible layer lines at the cost of printing speed.

They're most visible in the recesses for the chips, so when you have chips in the rack you don't see that anymore. The texture at the top layer isn't that bad, especially if you increase the number of perimeters, then it's all straight parallel lines and no more zigzag pattern.

In any way, if you look at these (0.4mm nozzle, 0.2mm layer height) from maybe 3/4 of a meter distance, it's barely visible even when not doing all those tweaks.

Oh I didn’t mean from visibility, I ment from a wear on the chips perspective. And considering the weight of chips, with stacked 3D printed racks I would be wary of the layers being imprinted on the chips

mine are just for plastic chips but I do them vertical with separately made end plates screwed in. Keeps it smooth and have no issues with strength so far on a 5x25 chip rack full of chipco.

have you tried the variable layer heights with the wood filament? I have seen some cool results with that to make it more wood looking

 

[Nex]

I wouldn't be worried in that regard, at least for my CPCs.

PLA plastic may be stiff and brittle overall, but it still has a tiny amount of malleability (inb4 OOC thread) on the surface at room temperature. It's incredibly easy to scratch and cut it. I wouldn't be surprised if it was rather the chips that made small imprints on the racks instead.

Variable layer height tends to be used to optimize print time/quality balance on bigger models with small sections of roundings. Given the chip recess spans over 4/5th of the total object height, it wouldn't optimize much in comparison to simply printing the whole object with a different layer height. Using variable layer height in terms of random changes likely would cause issues with the roundings of the chip recesses. Also not convinced it'd actually look more organic that way. After all, you'd still see 100% perfectly straight parallel lines despite the variation in thickness.

 

[OutlawECHO]

Nice design Nex. I am printing a prototype tray currently using my Prusa Mini. Since it is just a prototype, I went with just one 10 chip sample. I am printing it on its side, but both ends will be pinned and glued. This should solve the overhang issue, give a better surface to the chip channel, and allow it to print slightly faster. I will post a picture of the design if it works out.

Autodesk Inventor. Sadly the results aren't portable/readable/editable by other software as far as I know, but I did look at free CAD software and was everything but convinced. OpenSCAD looks like a PITA to create complex models to me. It's already challenging enough to make parametric models using all the GUI tools and aids that Inventor offers... and Fusion 360 forces you to save your designs in the cloud.

You should have the option to save your 3d model as either a STEP or IGS file. Either of those can be opened by almost any 3d modeling software (however, the feature tree will not be complete and it will just import as a dumb solid). Also, Fusion 360 is a free software that should feel similar to Inventor. Like any software, it takes some time getting used to where all the tools are.

 

[Nex]

I am printing a prototype tray currently using my Prusa Mini. Since it is just a prototype, I went with just one 10 chip sample. I am printing it on its side, but both ends will be pinned and glued.

You will make your own experiences, but I've found that single-piece designs are usually better in most aspects I care about for the kind of models I usually create (functional parts) than multi-part designs - particularly those that have to be glued, rather than slotting them together. Manual postprocessing work like gluing introduces room for error and worsens the tolerances.

There are limits to it too, yes, and having to remove supports does also impact tolerances. However, it is a lot better with intelligently placed custom supports (baked into the model) rather than rolling with auto-generated supports, and in a lot of cases you can print the object in an orientation that the little residue of the cut off supports won't affect anything of importance.

OED's two-part rack design may be superior to my single-part design for people concerned that the rack would be abrasive to their chips; on the other hand, he seems to be wasting a lot of material on super thick double walls that are necessary to integrate the slotting. But it seems to be fairly reproducible with only little critical manual work (like cutting off brim) impacting tolerances, and even that could be avoided by using a different build plate.

Either of those can be opened by almost any 3d modeling software (however, the feature tree will not be complete and it will just import as a dumb solid).

So, not really better than just supplying a standard STL file that you not only can import as a dumb solid, but also drop into many slicers as-is, unlike STEP (afaik).
None of the options available allow someone else who does not have the proprietary software to change the parameters of the design.

Also, Fusion 360 is a free software that should feel similar to Inventor.

Like I said in the paragraph you quoted, Fusion 360 (at least when I last checked) doesn't allow to save your stuff locally. It forces you to share your designs with the company. "Cloud" is just a fancy term for someone else's computer you do not have complete control over, and where who knows who may be looking through your files any time.

I prefer to decide myself who I share my own data with... Not just personal stuff, but also stuff I invested significant work into that could be exploited commercially.

 

[OutlawECHO]

I agree that you can't beat an one-piece design for tolerancing. I ended up with decent success with my first prototype, but you can see the seam line which may bother some people. With my machine, I can still keep the tolerancing pretty tight, easily within .2mm.

The alignment pins on this one were a little weak. I also didn't need to have separate pieces for both ends, just one. I updated some dimensions for better tolerancing and made a prototype for starting tournament stack. It may be an uncommon stack at 8-25, 8-100, 2-500, 3-1000, but it is what we have been playing with.

Extra information if interested:
Filament: Hatchbox PETG ($21/1kg spool)
Material Usage: 60.25g (~$1.27 not including electric)
Print Time: 6h10m (for both the main piece and the end cap printed at the same time)

I could probably speed up the machine and shave off maybe an hour.

So, not really better than just supplying a standard STL file that you not only can import as a dumb solid, but also drop into many slicers as-is, unlike STEP (afaik).
None of the options available allow someone else who does not have the proprietary software to change the parameters of the design.

Correct. The only real benefit over .STL (at least in my experience) is that with a .stp file, it knows faces, so you can you push/pull/resize/delete commands. They are limited in redesign, but are useful if you just needed to adjust a hole or extrude to work with your machine's tolerances.

Like I said in the paragraph you quoted, Fusion 360 (at least when I last checked) doesn't allow to save your stuff locally. It forces you to share your designs with the company.

You can export your files as a .f3d file locally. You are still tied to using F360 though (like pretty much any 3d software). You are correct that 'save' will only save to the cloud.

I understand your concern about your files and not wanting to give access and such. In my case, most of my 3d modeling stuff isn't confidentially and I don't expect it to be marketable. It sounds like you do a lot more design work and inventing with potentially intent to sell and in that case, I would definitely be paying for software and storing my designs locally as well as backing them up.

Lastly, I would like to thank you for sharing your design as it inspired me to play around with a chip design of my own.

 

[BEANO52]

I agree that you can't beat an one-piece design for tolerancing. I ended up with decent success with my first prototype, but you can see the seam line which may bother some people. With my machine, I can still keep the tolerancing pretty tight, easily within .2mm.
View attachment 608841 View attachment 608842

The alignment pins on this one were a little weak. I also didn't need to have separate pieces for both ends, just one. I updated some dimensions for better tolerancing and made a prototype for starting tournament stack. It may be an uncommon stack at 8-25, 8-100, 2-500, 3-1000, but it is what we have been playing with.
View attachment 608849 View attachment 608851

Extra information if interested:
Filament: Hatchbox PETG ($21/1kg spool)
Material Usage: 60.25g (~$1.27 not including electric)
Print Time: 6h10m (for both the main piece and the end cap printed at the same time)

I could probably speed up the machine and shave off maybe an hour.



Correct. The only real benefit over .STL (at least in my experience) is that with a .stp file, it knows faces, so you can you push/pull/resize/delete commands. They are limited in redesign, but are useful if you just needed to adjust a hole or extrude to work with your machine's tolerances.


You can export your files as a .f3d file locally. You are still tied to using F360 though (like pretty much any 3d software). You are correct that 'save' will only save to the cloud.

I understand your concern about your files and not wanting to give access and such. In my case, most of my 3d modeling stuff isn't confidentially and I don't expect it to be marketable. It sounds like you do a lot more design work and inventing with potentially intent to sell and in that case, I would definitely be paying for software and storing my designs locally as well as backing them up.

Lastly, I would like to thank you for sharing your design as it inspired me to play around with a chip design of my own.

Wow, just found this, very cool design!

 

[splice42]

This now has me thinking about making my own... I've tried printing some racks before but the designs I got a hold of didn't print very well because of some warping issues. I've since learned how to use Fusion 360 for some design work and done some basic designs of my own for some stuff in the house, designing a chip rack won't be that hard...

Wood filament looks good but it took me so much work to print things cleanly without clogs. Might give it another try because those racks look pretty damn nice!

 

[Nex]

The bigger the nozzle diameter, the lower the chance for clogs is my personal experience. I have one spool of wood filament that would constantly clog a 0.4mm nozzle, but works like a charm when printing with a 0.6mm nozzle. My chip rack design doesn't have any super fine details, so it works out.

 

[splice42]

Yeah the 0.6 nozzle change is def. part of my settings for printing wood filament. I did end up getting through a couple of barrel cases for Glencairn whisky glasses, it's a real nice look with the wooden filament. Just need to find a stain that works well with it too!

 

[Nex]

Any stain designed for real wood should work, just that with wood filament you a) need to sand it reeeeaaally well beforehand and b) it'll still soak up much less than real wood will, you are going to need more passes if you want a bold color. I've stained a bunch of prints already and it is a challenge, but definitely doable. Biggest danger is applying too much stain at once which then will not be fully soaked in and leave ugly drip marks on the surface - not evenly spread out.

Edit: At least that's true for the brand of wood filament that I used, which sadly is no longer in production. Other manufacturers might mix in a different % of wood dust so results can vary. But anyway that's the experiences I've made.

 

[davislane]

what the pcf equivalent for 3D printing? I need to check out some more project that people are printing.

 

[Nex]

what the pcf equivalent for 3D printing? I need to check out some more project that people are printing.

Check out @One Eyed Dollar's store. His modular poker game die is a nifty design in particular.

 

[davislane]

Check out @One Eyed Dollar's store. His modular poker game die is a nifty design in particular.

yeah im familiar with his work. Have bought a few items from there.

 

[RainmanTrail]

what the pcf equivalent for 3D printing? I need to check out some more project that people are printing.

Reddit has a pretty good 3D printing community.

 

[splice42]

I did end up knocking out a quick chip tube design and printed it in translucent filament in vase mode. Took a few tries in the design phase to size it just right for my chips (Venerati) and 10k starting stack composition (12/12/5/6) but they look pretty good and ended up working great. Cap is a simple pressure fit and actually holds surprisingly well with no extra features needed, just a tiny lip. The tubes sure sped up the rebuys and made distributing starting stacks real convenient. I'll definitely be using them for home tourneys going forward.

 

[RainmanTrail]

I did end up knocking out a quick chip tube design and printed it in translucent filament in vase mode. Took a few tries in the design phase to size it just right for my chips (Venerati) and 10k starting stack composition (12/12/5/6) but they look pretty good and ended up working great. Cap is a simple pressure fit and actually holds surprisingly well with no extra features needed, just a tiny lip. The tubes sure sped up the rebuys and made distributing starting stacks real convenient. I'll definitely be using them for home tourneys going forward.

View attachment 793502

Very cool. I've been wanting to do this as well, but haven't had the time to do 3d printing projects lately. If you'd be willing to share your files, I'd love to give this a try when I get more time.

 

[splice42]

Very cool. I've been wanting to do this as well, but haven't had the time to do 3d printing projects lately. If you'd be willing to share your files, I'd love to give this a try when I get more time.

I can share my files (both the Fusion 360 design and the STLs I exported from them) but can't attach them here. I'll try to find a good place for them.

 

[splice42]

You can find both the STLs and Fusion 360 design files for my simple Venerati chip tube and lid here: https://www.thingiverse.com/thing:5021873

 

[markleteenie]

I did end up knocking out a quick chip tube design and printed it in translucent filament in vase mode. Took a few tries in the design phase to size it just right for my chips (Venerati) and 10k starting stack composition (12/12/5/6) but they look pretty good and ended up working great. Cap is a simple pressure fit and actually holds surprisingly well with no extra features needed, just a tiny lip. The tubes sure sped up the rebuys and made distributing starting stacks real convenient. I'll definitely be using them for home tourneys going forward.

View attachment 793502

Cool print! What translucent filament did you use?