Online since 2002. Over 3300 puzzles, 2600 worldwide members, and 270,000 messages.

TwistyPuzzles.com Forum

It is currently Sat Apr 19, 2014 6:06 pm

All times are UTC - 5 hours



Post new topic Reply to topic  [ 15 posts ] 
Author Message
 Post subject: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 12:05 am 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Dear designers, I love what you do! Without you I wouldn't have so many amazing puzzles to play with and admire.

That being said, there are a few things you could do to make my life (and perhaps the lives of others in the future) easier...


Alright so maybe this isn't really a love letter. There really are a few small things that could really help in the polishing of parts.

First, I should recap how parts get polished.

Shapeways' Process:
Shapeways uses diagonal-cut cylindrical ceramic media similar to this:
Attachment:
ceramic_media.jpg
ceramic_media.jpg [ 64.64 KiB | Viewed 997 times ]

Because of the diagonal-cut to this media, it can partially reach recessed areas of your parts. Also, the media wears and leaves a powder which does an okay job of wearing parts too. This is a great media for Shapeways to use because it's fast and only one step. It does a good job of smoothing parts and shining them up at the same time.

My Process:
I polish in three stages. The first stage is very coarse grit (think volcanic sand) and the second stage is extremely fine grit (think powdered sugar). Both of these stages need good flow of grit through the parts and they do a great job of reaching all of the areas of a part. For the third stage I use porcelain spheres which do a great job of wearing / smashing down the external surfaces until they shine.

There are three big problems with my process that require a lot of manual labor or planning:

1) Grit, especially the coarse grit, can get trapped in pieces or just not flow well through the pieces which causes them to wear very unevenly.
2) The porcelain spheres can get jammed or stuck in pieces so badly that it takes upwards of 15 minutes PER PIECE to remove the spheres and even then, sometimes they have to be cut or drilled out.
3) Very large pieces with sharp points, or large, highly non-spherical pieces can wear very unevenly. A "large" piece is a piece where any dimension of the piece measures 30mm or more.

I believe these issues can be addressed by designers without much additional work.

Dealing with very large pieces:
If your puzzle has large pieces with sharp points, it really helps to re-enforce the piece. One option would be to shell the piece to 1mm or greater. Preferably about 1.5mm. This can drive up the cost of the part though so another option is simply to reinforce the tip(s). For example:
Attachment:
mlc_reenforce_tip.png
mlc_reenforce_tip.png [ 46.56 KiB | Viewed 997 times ]

The red portion really should be re-enforced. The three corners around that tip and then the other three below those around it are also good candidates for re-enforcing.
The two tips on the big edge piece in the back of the image are also great candidates for re-enforcing.

Enabling great powder flow:
The more powder drain holes your pieces have the better. This is true when Shapeways cleans the nylon powder media out of your parts and it's especially true when I tumble your parts. I have run into parts with a single powder drain hole that have been horrible nightmares to tumble. Imagine waiting for a giant hour glass to drain grit out of it, one grain at a time. Now imagine pieces that only drain when shaken violently. Now imagine me standing at my kitchen sink violently shaking pieces for hours to drain the grit out. Now imagine a T-Rex standing at my kitchen sink with arms so short he can't even grasp the parts. Okay so maybe I made up the T-Rex thing...

Parts that fill out with powder that don't drain well tumble very poorly. They get weighed down and wear unevenly. Here are a few examples of powder-drain nightmares:
Attachment:
powder_trouble.png
powder_trouble.png [ 1.56 MiB | Viewed 997 times ]
These are large parts with large chambers that fill out with powder and only one (or two) drain holes. The hole size on these is great because they don't allow porcelain spheres in, but they also don't really allow powder out very fast. For the part on the left, the hole is also at the apex of the curve. Powder doesn't want to drain out of the top of an inverted curve. If the top of the curved surface is the only (good) option, it's better to make the hole a long rectangle or to add a few other holes where you can.

Designing with small spheres in mind:

Porcelain spheres are, by far, the best stage for quality results in the polishing process. Unfortunately they're also, by far, the most work to clean up after. I use 3mm porcelain spheres. They actually vary from about 2mm to 3.5mm but the majority are in the 2.8-3.0mm range.

There are two options for making parts porcelain-sphere friendly:

1) Don't let ANY spheres into the parts
2) Make holes so big that spheres flow really freely in and out.

This means holes need to either be smaller than 2mm or larger than 5mm. Anything in the 2-5mm range causes horrible nightmares. Here are a few nightmare examples:
Attachment:
sphere_trouble.png
sphere_trouble.png [ 3.54 MiB | Viewed 997 times ]


A) This is a 3mm powder drain hole. Now the part is packed with spheres and the only way to get them out is to drill / cut a larger hole.
B) This part has the same problem as A.
C) There is a sizable chamber in the corner of this part and the opening is an irregular shape roughly large enough to allow a 4mm sphere in. Now the corners of this part are packed full of spheres and it takes a lot of manual dental-pick and air-compressor work to get the spheres out.
D)The grating on this part is roughly 3mm in width which allows hundreds of spheres in but getting them out is a huge amount of work. The only reasonable way to get all the spheres out is to cut into the part.

The best way to allow powder flow without allowing sphere flow is to make drain holes that are between 1mm and 1.75mm in size. If you use rectangular drain holes (my favorite) then you can make rectangles that are Xmm in length but only up to 1.75mm in width. If in doubt, ask yourself: can a 2mm sphere make it through this hole? If yes, the hole is too big.

If you can't make small holes (and there are lots of legit reasons why you can't) then try to make sure holes are big enough to allow 5mm or larger spheres through. This makes getting packed spheres out very easy.


Ideal parts:
Here are some example parts that are perfect for both powder and spheres:
Attachment:
powder_sphere_perfect.png
powder_sphere_perfect.png [ 1.62 MiB | Viewed 997 times ]

These parts allow great powder flow but don't allow spheres to get into the parts.


Example interaction with a designer:
Ben (of benpuzzles fame) was kind enough to work with me on his Master Little Chop design. Allow me to quote a bit from our interaction:

Quote:
There are basically three guiding principles for reviewing a design to try to make it tumble as well as it can. Ask yourself:

* Are there any highly non-spherical parts with sharp points that will wear too quickly (you don't want this)

* Will a bunch of 3mm spheres get packed into the pieces (you don't want this)

* Will powder flow through and drain out of the pieces easily (you want this)

For the 3mm spheres, you either want no spheres to be able to get into the pieces, or you want the spheres to get in and come out very easily.

The 3mm spheres actually vary in size from about 2mm to about 4mm and they can get packed very dense. I see the holes in the bottom of the pieces are somewhat big, definitely big enough for lots of spheres to get in:

Attachment:
mlc_bottom_holes.png
mlc_bottom_holes.png [ 59.91 KiB | Viewed 997 times ]


Quote:
Unfortunately the pieces have big enclosed chambers and the spheres will get into every crevice and be very, very hard to get out. The corners and the edges will be especially troublesome. I'd suggest making the bottom holes only 1.75mm in diameter to prevent spheres from getting in them at all. The other options would be to make the holes even bigger and to make other holes on the parts that spheres could get in and out of. That way the spheres would get in but it would be easy to get them all out. Unfortunately there isn't a lot of room for additional 5mm + holes elsewhere on the pieces. I think a better strategy is to avoid getting the spheres in them in the first place.

For the holes in the parts, since they are going to be made smaller, it's important that powder can get in and out of the parts other ways. I'd suggest adding rectangular drains like so:


Attachment:
mlc_big_piece_slots.png
mlc_big_piece_slots.png [ 63.66 KiB | Viewed 997 times ]


Quote:
The skinny part of those rectangles should be between 1.0 and 1.75 mm wide and they can be however long the part can support. dimensions like 1mm x 6mm or 1mm x 4mm would be fine. They'd be too small for spheres to get in but big enough that powder could get in and out easily.

Here are all of the places I'd suggest adding these drains:

Attachment:
mlc_center_drains.png
mlc_center_drains.png [ 54.94 KiB | Viewed 997 times ]

Attachment:
mlc_small_triangle_drains.png
mlc_small_triangle_drains.png [ 50.01 KiB | Viewed 997 times ]

Attachment:
mlc_house_drains.png
mlc_house_drains.png [ 43.72 KiB | Viewed 997 times ]



Finally, if you want to make your design polish-friendly and you aren't sure what to do, I'd be happy to consult. Feel free to PM me with questions!

_________________
Prior to using my real name I posted under the account named bmenrigh.


Last edited by Brandon Enright on Tue Dec 03, 2013 7:59 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 2:38 am 
Offline
User avatar

Joined: Sat Apr 21, 2007 11:21 pm
Location: Marin, CA
Fantastic post!

Am I correct in assuming that the spheres that get trapped between shell layers are easily cleaned? I see some on your radiolarian part, but not circled in red. I guess they can just be popped out?

I'll be applying these notes as I go. Thanks.

_________________
Jason Smith posted here as 'io' through 2012.
Visit Jason Smith's PuzzleForge on Shapeways!
Jason Smith's Puzzles - YouTube Channel.


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 2:46 am 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Yeah the spheres in the rail grooves are really trivial to get out.

The spheres in the screw holes are horrible but it's my own darn fault for forgetting to put screws in the holes to fill them before tumbling in the porcelain spheres.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 4:02 am 
Online
User avatar

Joined: Sun Mar 15, 2009 12:00 am
Location: Jarrow, England
Regarding hole sizes, don't Shapeways require at least two 2mm (min) diameter holes or one 4mm (min) hole for powder removal? This seems to be right in the middle of your not recommended range. :(

Also, on a few parts I have had printed, I used holes with a diameter of 0.02mm to "fool" the checking software into seeing the parts as correct, but the hole disappears during manufacture, trapping powder inside. However, they have got wise to this.

If the hole(s) could be filled before tumbling, would that not be better?

_________________
My Shapeways Shop: http://www.shapeways.com/shops/gus_shop

Image


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 7:30 am 
Offline
User avatar

Joined: Sat Sep 07, 2013 3:53 pm
I got an error message from shapeways regarding the printability of my Super OctoCube in Polished Materials. Here is what is said:

Quote:
Here is why we cannot print your model:

The minimum bounding box per shell for this material must be at least equal to x+y+z >25 mm. Please enlarge the model to ensure that it can be cleaned and shipped effectively. For more information about the design specifications for this material please visit: http://www.shapeways.com/materials/stro ... guidelines


Then this image was provided:


Attachments:
photo.png
photo.png [ 96.48 KiB | Viewed 900 times ]

_________________
Image

http://www.shapeways.com/shops/ZemPuzzles
Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 10:24 am 
Online
User avatar

Joined: Sun Mar 15, 2009 12:00 am
Location: Jarrow, England
Yes this would be an error. They can print this, but because it is so small it would be easily lost during the polishing process. You will have to remove this part(s) from the model in order to get all of the other ones polished, and then print them separately (or with another puzzle) in WSF and polish them yourself :(

_________________
My Shapeways Shop: http://www.shapeways.com/shops/gus_shop

Image


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 1:02 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Gus wrote:
Regarding hole sizes, don't Shapeways require at least two 2mm (min) diameter holes or one 4mm (min) hole for powder removal? This seems to be right in the middle of your not recommended range. :(
I'm not aware of any such restriction. I've had to do a lot of powder and grit removal though which is why I like the long rectangle drain holes so much. They make powder removal much easier.

Gus wrote:
Also, on a few parts I have had printed, I used holes with a diameter of 0.02mm to "fool" the checking software into seeing the parts as correct, but the hole disappears during manufacture, trapping powder inside. However, they have got wise to this. If the hole(s) could be filled before tumbling, would that not be better?
I really dislike when large chambers are closed like this. During tumbling the density of the part is wrong and it doesn't churn in the grit properly causing it to wear poorly. For the dyeing process, I can't use the pressure cooker or I'll collapse the part. Then I have to boil the part for even longer which allows a lot of moisture (and worse, salt) to penetrate the walls and get into the inner chamber. For weeks or even months the part slowly leaches salt, creating a ring of white salt crystals. I usually have to drill a hole myself and drain the powder before I even start the tumbling.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 2:14 pm 
Offline
User avatar

Joined: Thu Dec 02, 2004 12:09 pm
Location: Missouri
Brandon Enright wrote:
I'm not aware of any such restriction.
Neither am I. And if there is I've broken it several times myself.
Brandon Enright wrote:
I've had to do a lot of powder and grit removal though which is why I like the long rectangle drain holes so much. They make powder removal much easier.
First I see several of the offending parts are mine and I've likely just shipped you many more. Sorry about that. I can add more holes. That is easy enough. That Bubbloid corner though had its inner geometry simply determined by a shell operation. To fix that I'd have to seal off that geomentry and then add some smaller holes. That is doable I think. Just not something I thought about while designing the part. Though long narrow rectangles make me nervous. I would think they'd remove alot of the strength from a part. Granted SW&F is very strong so maybe its a false concern but I'd be happier with a row of circular holes.

Oh and one other thing I do think about as a designer. I personally don't like seeing holes on a finished puzzle so I try to hide them away on the bottom, or near the bottom, of the parts. I generally try to minimize them altogether... but I will make an effort to change. It's just a personal preference I need to over come.

Carl

_________________
-
Image

Image


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 3:27 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
wwwmwww wrote:
First I see several of the offending parts are mine and I've likely just shipped you many more. Sorry about that. I can add more holes. That is easy enough. That Bubbloid corner though had its inner geometry simply determined by a shell operation. To fix that I'd have to seal off that geomentry and then add some smaller holes. That is doable I think. Just not something I thought about while designing the part. Though long narrow rectangles make me nervous. I would think they'd remove alot of the strength from a part. Granted SW&F is very strong so maybe its a false concern but I'd be happier with a row of circular holes.
Your parts aren't that bad but they're easy to pick on because I had them handy and I knew you wouldn't mind :D

Your Bubbloid corner really is an example where it's hard to really fix the issue. You could add a cover with small holes in it, but I think that's a bit much to ask of designers.

When looking through your Realx5x5x5 parts I spotted some areas that will be troublesome for tumbling. As long as you're okay with it, I'd be happy to post some screenshots of the parts I think will be trouble with a discussion of possible options for avoiding those problems. I'd really mean this to be educational rather than to suggest you should go back and adjust existing designs.

wwwmwww wrote:
Oh and one other thing I do think about as a designer. I personally don't like seeing holes on a finished puzzle so I try to hide them away on the bottom, or near the bottom, of the parts. I generally try to minimize them altogether... but I will make an effort to change. It's just a personal preference I need to over come.
More holes is also great. 3 holes instead of 1 means powder is 3x more likely to drain and will generally drain at least 3x as fast. WSF really is so strong though that I can't imagine rectangle cutouts having any meaningful impact on part strength.

Also, rectangle (or square) holes are a lot friendlier to the triangle count :)

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 5:31 pm 
Online
User avatar

Joined: Sun Mar 15, 2009 12:00 am
Location: Jarrow, England
wwwmwww wrote:
Brandon Enright wrote:
Gus wrote:
Regarding hole sizes, don't Shapeways require at least two 2mm (min) diameter holes or one 4mm (min) hole for powder removal? This seems to be right in the middle of your not recommended range. :(
I'm not aware of any such restriction.
Neither am I. And if there is I've broken it several times myself.
See http://www.shapeways.com/materials/strong-flexible-design-guidelines

_________________
My Shapeways Shop: http://www.shapeways.com/shops/gus_shop

Image


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 5:35 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Gus wrote:


There they have:

Min Escape Hole: 4mm (1 hole) · 2mm (2 holes) · Big models require multiple holes

But I think all they mean by that is that it's really hard to reliably get all of the powder out with smaller holes. It's certainly not a restriction for printing parts. Sometimes they're really good about removing powder and sometimes it looks like they don't even try to remove the powder, even with much bigger holes.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 6:50 pm 
Offline
User avatar

Joined: Thu Dec 02, 2004 12:09 pm
Location: Missouri
Brandon Enright wrote:
When looking through your Realx5x5x5 parts I spotted some areas that will be troublesome for tumbling. As long as you're okay with it, I'd be happy to post some screenshots of the parts I think will be trouble with a discussion of possible options for avoiding those problems. I'd really mean this to be educational rather than to suggest you should go back and adjust existing designs.
Sure... go ahead. And its easy enough for me to make changes. Its just too late for the first print.

Carl

_________________
-
Image

Image


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 7:55 pm 
Offline
User avatar

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Thanks a lot of Carl and Ben for letting me use their models as examples :D

Here are my concerns for the Real5x5x5 pieces. I don't actually know the sizes of the pieces because Meshlab is not an engineering / CAD tool and doesn't provide a good (or any from what I can tell) way to measure pieces. So some of the problem I point out might not actually be problems because I just don't know the sizes involved. These examples are meant to illustrate the kinds of problems that I look for.

Attachment:
real555_edge_sphere_trouble.png
real555_edge_sphere_trouble.png [ 80.65 KiB | Viewed 729 times ]
If spheres could get up into these chambers it will be very hard to get them out. The spheres have a tendency to pack themselves into self-supporting bridge / arch structures and to break the arch you often have to cause the part to expand (I use an air compressor).


Attachment:
real555_edge_side_sphere_trouble.png
real555_edge_side_sphere_trouble.png [ 64.45 KiB | Viewed 729 times ]
If spheres can get through this hole into the chamber I will likely have to cut them out.


Attachment:
real555_555edge_sphere_trouble.png
real555_555edge_sphere_trouble.png [ 143.62 KiB | Viewed 729 times ]
If spheres can get up into this chamber through the neck they will be very hard to get out. Fortunately the neck gap is long enough that various picks will likely give me enough room to work with to get spheres out.


Attachment:
real555_555corner_wear_trouble.png
real555_555corner_wear_trouble.png [ 57.33 KiB | Viewed 729 times ]
This corner is very big and will wear unevenly. There is nothing that can be done about that. A few things will help a lot though. Powder drain holes on the sides to allow powder to flow through will help prevent the piece from filling completely up with powder and "sinking" to the bottom where it will wear even faster. The more powder can flow through without being trapped the more evenly it will wear. No matter how many drains are added though, the tips I have circled probably need re-enforcing. I think the three side corners are likely to wear more than the corner at the tip but all 4 would really benefit from more material.


Attachment:
real555_555xcenter_powder_trouble.png
real555_555xcenter_powder_trouble.png [ 28.1 KiB | Viewed 729 times ]
This neck looks too thin to be printed. It will probably be printed partially or completely closed. This will probably cause annoying dye and salt and moisture trapping when the part is dyed. I'd suggest adding an upper drain hole to let some trapped powder out. If the neck isn't too think to print, it could still be so think that coarse grit gets stuck in it. When coarse grit gets stuck it slowly leaks out in later stages. If coarse grit leaks in the porcelain sphere stage it will prevent the pieces from developing a shine.


Attachment:
real555_core_powder_trap.png
real555_core_powder_trap.png [ 60.92 KiB | Viewed 729 times ]
This corner will probably trap lots of nylon powder in the printing process. It will be hard to get out and the powder will slowly leak out when the puzzle is assembled. If you dye the core the chambers will trap dye water and it'll take a long time to dry. The geometry of the chambers just make it hard for stuff to get out. I'd suggest adding 8 big holes, one in each spot where a "corner" on the cube would go.

_________________
Prior to using my real name I posted under the account named bmenrigh.


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Tue Dec 03, 2013 11:13 pm 
Offline
User avatar

Joined: Thu Dec 02, 2004 12:09 pm
Location: Missouri
Brandon Enright wrote:
I think the three side corners are likely to wear more than the corner at the tip but all 4 would really benefit from more material.
Thanks Brandon. I don't like trapped power either. I should have cross-sections more parts to look at the insides. I was just pushing things to get the $50 discount. I can probably go back and address all these issues for version 2.0. However those 4 corners you point out are already 1.5mm thick. I make all my inner walls 0.75mm think and all the walls that are on the puzzle's surface are 1.5mm thick. Before I go tweak thinks I want to see how this first one turns though.

Carl

_________________
-
Image

Image


Top
 Profile  
 
 Post subject: Re: Designing puzzles with polishing in mind: A Love Letter
PostPosted: Wed Dec 04, 2013 3:24 am 
Online
User avatar

Joined: Sun Mar 15, 2009 12:00 am
Location: Jarrow, England
Brandon Enright wrote:
Gus wrote:


There they have:

Min Escape Hole: 4mm (1 hole) · 2mm (2 holes) · Big models require multiple holes

But I think all they mean by that is that it's really hard to reliably get all of the powder out with smaller holes. It's certainly not a restriction for printing parts. Sometimes they're really good about removing powder and sometimes it looks like they don't even try to remove the powder, even with much bigger holes.
When I first printed the Sword in the Soma pieces I cheated by using 0.02mm holes but Shapeways picked up on it. So I added multiple 2mm holes as required, and they did a very poor job at removing the unused powder:
Attachment:
File comment: Unused powder
Soma Sword pieces.jpg
Soma Sword pieces.jpg [ 693.87 KiB | Viewed 663 times ]
In fact, most of this power came from the small inner parts and the sword. For the large pieces, I couldn't be bother to try to get out the powder so I just sealed up the holes with a drop of superglue after I dyed them.

_________________
My Shapeways Shop: http://www.shapeways.com/shops/gus_shop

Image


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 15 posts ] 

All times are UTC - 5 hours


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  

Forum powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group