Have any of you sucessfully used SolidWorks for carton design, package design, or any type of cardboard folding/unfolding? We're considering using it for that among other things, and I'd like to hear how it works for that. I'd use the sheetmetal tools, of course, but there are some differences from sheetmetal design that may or may not be difficult to do. Let me know...thanks!
Brian, since nobody has answered your query I'll give you my $.02. Although I haven't used SolidWorks for carton design I would think that it would work just fine for that. There could be inconveniences caused by the sheet metal portion's "Auto-Relief" function, or the built-in fail-safe that prevents folds that come too close to interfering with other folds. With a little experimentation and experience you should be able to do a lot with very little trouble. Turning off "Auto-Relief" and being just a little careful will get you a long way.
There was a thread on this subject quite a while ago (I'm guessing 99 or
2000, but I'm really bad on dates). You might try the google newsgroup search to find it. All I can remember about it was that someone suggested using a K-factor very close to 1, as the inner layers of the cardboard buckle instead of compressing and the outer layer doesn't stretch much.
Well I've come pretty close to being successful. The biggest obstacle that's going to stand in your way is an old bug that I've been trying to get the SolidWorks to fix for the longest time now.
Basically, there is a black dot (or sketch point if you use the old- skool method) that represents the "Fixed Face" of your sheetmetal (or cardboard) box.
It likes to move around if you change dimensions. The bug comes from the fact that you can't add a relation to the dot so that it would always stay put when you resize things.
You can add a relation to the sketch point, but it goes away when you exit the sketch (can we fix this already SoldWorks!!)
If you just plan on having one size with no configurations or design- tables, then you should be fine. If you plan on doing what the software is supposed to do (that is use configs for different sizes), your part will fail miserably.
The fun part is spending hours trying to figure out why all of your bend sketches are failing ;^)
If you are clever enough, you will finally realize that when the dot moves, your fixed faces will change, thus causing a chain reaction of errors.
Brian I have used SW to create packing cartons and die cuts since 1999 without any major problems. As Mike stated, the floating point on the fixed face is the only small hiccup but you soon get wise to it. I don't create different sizes of cartons or die cuts via configurations (mulipart/size configs are a pain in the rear in a Smarteam/PDM enviroment), For cartons I create a single file with 3 configs, Developement, Top Open and Top closed, for die cuts it's sometimes 2 configs, development and folded or somtimes it has to be 3 configs developement, folded and assembled. The cartons are easily modified to create new sizes, I use save as and then add or subtract on 5 dims to create the new size carton. I can give you some examples of what I have done if you wish, send me a email (remove the dot) so we can arrane delivery.
I see exactly what you mean. This is shady and should not be falling apart. Personailly I have used sketched bends only once (maybe twice) so far.
I have been able to make many boxes that were parametric and did not fall apart, at least a way exists (model-> sheet metal OR baseflange + edgeflanges)
Generally I associate bending from a flat sheet with sketched bends with self flagellation for reasons of mystic ecstacy (grin) - Viva La Difference! It's too much like "virtual fabrication" - like having to start all of ones models with a cube and remove material.
As proven by soapbox Paul's example (and yours as well) this CAN work and does work but as you illustrated it's alot like trying to sprint while on stilts (you will fall hard eventually).
I vaguely remember seeing this in 1998 and saying "wow, this #%^K&!* stinks - let's not use this funtion". Shamefully this is still a problem. That's the stuff that drives users mad - that's the stuff that needs fixing.
Thanks for showing the problem,
By the way, if I did design packaging, this stuff would serve nicely for modeling RSCs and many cardboard inserts and would also kick-booty on multi layer foam. Some tuning & learning required of course.
Can I ask why you all seem to start sheetmetal with a flat pattern, why not design the part bent to the finished parameters and then unfold to reveal the flat pattern? I have had great success using this method, the only thing that you have to bare in mind is that no 2 features can touch or overlap. I had intended to put some examples up on http:\\
\sw\ over the weekend but a blown power supply wrecked my home PC. I should have something up there by 10.00pm GMT tonight if anybody is interested.
I agree, I would not normally design sheetmetal parts from the flat. The soapbox example is only modeled from the flat because I choose to unfolded the original soapbox carton, scanned it, sketched over it and folded it. The soapbox was a test part I did years ago.
Most people do it your way most of the time. There are a few times where starting in the flat and adding bends works much better. The one I am familiar with is modelling a part, say a flexible printed circuit or a ribbon cable, where the bends are not perpendicular to the edges of the part. I don't believe it can be modeled properly without starting in the flat, at least not without a great deal of tricky work. Works great when you do it the "right" way, generating the twisted surfaces on the sides automatically.