I love gridfinity, especially for drawers. It feels like 70% of the time when I want to store something there’s already a nicely molded box out there from someone else.
Bento3D is a web tool that allows
you to create 3D printable dividers
and toolboxes with millimeter precision.
Is the "millimeter" precision noteworthy? This is kind of a CAD program, right? Why would it have trouble with any precision? Isn't it just juggling numbers?
it's noteworthy because a lot of similar solutions using OpenSCAD (progmatic on-demand creation of boxes/containers) are written poorly, and in many cases they only allow an arbitrary precision as set by the authors.
>This is kind of a CAD program, right? Why would it have trouble with any precision? Isn't it just juggling numbers?
this one kind of made me laugh only because of my familiarity with the history of CAD programs and what poor tasting dog-food has been served.
There is a long history of conversion errors, process errors, arbitrary formats and unit types, whatever. A CAD program that truly 'juggles numbers' is the holy grail, and the high tech solutions nowadays are getting pretty close to getting it right -- but it's been a journey and they're still not really there; every CAD suite has a list of no-nos that must always be kept in mind, and they're not engineering/science no-nos, they're "it'll break the software when I try to create a chamfer around this type of edge." kind of no-nos that are quirky and specific.
To be honest I'd also assume to be able to create arbitrarily sized boxes.
Last time I used OpenSCAD I didn't notice any arbitrary precision limitations — maybe the ones who expose interfaces to the web have them, but even there I don't seen how anything but millimeters would make sense (then again I live in Europe).
I think they wanted to highlight the unit they chose for their exported drawings.
> Isn't it just juggling numbers?
it is, until you have to hand your plans off to someone else to work with in their toolchain
every time I want my local laser cutting shop to cut my piece correctly, I have to export my drawing from my CAD suite as a DXF, open Inkscape, set the page size to something physical like A4, import the DXF, align it on the physical page, then print the result as a PDF. This is the only way that the laser shop reliably receives a properly scaled drawing. Any deviation from this procedure has led to wasted time and material.
It's to draw contrast to "design systems" like Gridfinity, which is based on boxes that are multiples of 42mm on each side.
There are tradeoffs between the two approaches, with no real "right" answer. Gridfinity trades some flexibility for compatibility, the 42mm squares mean you can move the gridfinity boxes on a base without reprinting anything. This gives you much more flexibility, but if you want to change the size of one part you'll have to reprint a bunch of things.
The precision should be no trouble, but you should understand your printers tolerance capabilities.
If you print a 30mm hole and a 30mm box to go in it, it might not fit, usually you would oversize the hole or undersize the box. By how much depends on your printers characteristics. Easy to test by printing said box then measuring how close to 30mm it was.
Organizers are so essential in my workshop and homelab. The mill alone has hundreds of items to keep track of. You can only go so far with misc cardboard boxes. I've been using SolveSpace MCAD to design simple just right size boxes, but these are so much nicer.
It seems like one could do this design in Solvespace with several parameters for the overall box. The interior compartments/partitions could be done in a single 2D sketch.
The dividers look to be full height of the toolbox (not layered trays), so why are they printed as a separate part from the toolbox?
Not knowing 3D printing, my first guesses are that dividers can be done in a different color (to have high-contrast with the contents), or different material/process (for different outside/inside requirements), or to facilitate reconfiguring an existing box for different contents (lower cost to print just the divider insert, than to print an entire box and add metal hardware)?
Wow. This is amazing. I spent ages trying to find boxes so I could neatly store batches of parts for assembly. Now I can generate boxes with exact dimensions. This is great!
Really nice workflow for a really common 3d print use case. I'd like to see the lid re-worked to print without support (the tall standing orientation is also asking for it to get knocked over). Not sure why they insist on 0.1mm layer height. That makes prints take forever.
It looks like it could be printed flat and they are forcing everything in one print. That also increases the chance of failure and maximizes the resulting waste. I have wasted enough material with long multipart prints that I limit myself to ~8h if I can.
Readit News
https://www.reddit.com/r/gridfinity/comments/1ami7f4/i_desig...
>This is kind of a CAD program, right? Why would it have trouble with any precision? Isn't it just juggling numbers?
this one kind of made me laugh only because of my familiarity with the history of CAD programs and what poor tasting dog-food has been served.
There is a long history of conversion errors, process errors, arbitrary formats and unit types, whatever. A CAD program that truly 'juggles numbers' is the holy grail, and the high tech solutions nowadays are getting pretty close to getting it right -- but it's been a journey and they're still not really there; every CAD suite has a list of no-nos that must always be kept in mind, and they're not engineering/science no-nos, they're "it'll break the software when I try to create a chamfer around this type of edge." kind of no-nos that are quirky and specific.
Last time I used OpenSCAD I didn't notice any arbitrary precision limitations — maybe the ones who expose interfaces to the web have them, but even there I don't seen how anything but millimeters would make sense (then again I live in Europe).
I think they wanted to highlight the unit they chose for their exported drawings.
> Isn't it just juggling numbers?
it is, until you have to hand your plans off to someone else to work with in their toolchain
every time I want my local laser cutting shop to cut my piece correctly, I have to export my drawing from my CAD suite as a DXF, open Inkscape, set the page size to something physical like A4, import the DXF, align it on the physical page, then print the result as a PDF. This is the only way that the laser shop reliably receives a properly scaled drawing. Any deviation from this procedure has led to wasted time and material.
There are tradeoffs between the two approaches, with no real "right" answer. Gridfinity trades some flexibility for compatibility, the 42mm squares mean you can move the gridfinity boxes on a base without reprinting anything. This gives you much more flexibility, but if you want to change the size of one part you'll have to reprint a bunch of things.
CAD programs are generally unitless. You can design a box with OpenSCAD "down to a micrometer" if you want.
Designing a joint that works well also will poorly 3d-printed parts is usually what matters, but it's not like we're lacking customizable solutions:
https://www.printables.com/model/757297
(not detracting from the website.. just commenting on the headline)
If you print a 30mm hole and a 30mm box to go in it, it might not fit, usually you would oversize the hole or undersize the box. By how much depends on your printers characteristics. Easy to test by printing said box then measuring how close to 30mm it was.
0.2mm is usually the clearance I go with if I want a loose fit.
I rather write my own parametric CAD from scratch.
Not knowing 3D printing, my first guesses are that dividers can be done in a different color (to have high-contrast with the contents), or different material/process (for different outside/inside requirements), or to facilitate reconfiguring an existing box for different contents (lower cost to print just the divider insert, than to print an entire box and add metal hardware)?