Resizing a design question

Once you go solidworks. You never go back. :wink: just saying. I have a design controlled by a excel sheet. Change the cell named ā€œslotsā€ and everything works. (within reason)

As a disclaimer, my job owns my license. I donā€™t know if I would buy one myself. Especially not as a hobbiest. So in that regard I definitely understand the challenge. And have worried what I will do if I lose access to solidworks somehow.

4 Likes

Unfortunately this is Very Hard. Your choices (as far as I know) are:

  1. Resize everything. As noted, now your design is the wrong size.
  2. Manually resize all the tabs and slots. Donā€™t forget: some you want to left-justify, some right-justify, some center-justifyā€¦
  3. Use a parametric design tool.

I usually pick one of these approaches for my own project depending on the relative importance of simplicity, size, etc.

Unfortunately material variation usually exceeds what you can cheat with glue, in my experience.

10 Likes

If this doesnā€™t get implemented the store has potential to become a real mess.

I like @Jules suggestion about making parameterized lines pink (or whatever). Iā€™d also add a second color, say cyan, that denotes which lines are also moved when parameterized lines are extended/truncated.

(Assuming the Glowforge software is capable of detecting which lines are connectedā€¦) The ends of pink lines attached to black lines would remain fixed and the other endpoint would be moved. Cyan lines attached to endpoints of pink lines would move. The endpoints of pink lines that arenā€™t attached to black lines would expand equally in both directions.
Hmmmā€¦ endpoints of pink lines attached to non-cyan lines would also probably need to remain fixed.

Of course, weā€™d probably need at least two parameterized colors and at least two ā€œalong for the rideā€ colors for the two main kinds of action a lasercutter can take: cut and engrave. Would we then use even more colors for scoring and different engrave depths?

Maybe the Glowforge software could detect gradients and use those for resizing slots/tabs.

7 Likes

Standards, standards, standards. The same things should be done the same wayā€¦ unless you find a better way, then do that.

2 Likes

What do you mean?

Do Illustrator/Inkscape have parametric features? Using colors is a kludge to allow non-parameteric software to be used to make designs that would benefit greatly by dynamic adjustability.

ā€“editā€”
Shoot, I realized that I may have misinterpreted the original question.

@polarbrainfreeze, were you asking about what you could do to make a design work with the Glowforge software or were you asking to find an immediate solution to this problem? I assumed you were asking the question so that you could preemptively modify your designs now so that transitioning them to the Glowforge store would go smoothly.

3 Likes

Maybe itā€™s because Iā€™m still learning Fusion. But Iā€™ve built things using user parameters to solve this exact issue but when the model gets very complex itā€™s still a pain to manage in my experience.

I like the idea of http://openjscad.org however I donā€™t like using it very much (even as a computer scientist). We need an easy way of sharing and modifying parameterised models.

1 Like

The problem is that the svg does not contain the constraints that are obvious to you. A small thing like the red and cyan line being perpendicular to each other is not information contained in the svg. Now an intelligent piece of software might be able to infer some of the constraints but imagine that you want a curve to be relational to material thickness of similar. Gets complicated very quickly.

Also common ruleset for lasers acting as printers (to a computer) is that color defines laser power and thickness of line defines vector cutting vs raster engraving. This is a hack on an existing protocol but gets the jobs done and ensures compatibility with a lot of software/hardware.

2 Likes

I used to use Actions in AI, which is a type of macro, (push one button to run it, and it prompts you for the necessary input, then makes the changes), but itā€™s been a while since I wrote one. Refresher work is gonna be needed.

And Iā€™m running an old version of AI (CS5)ā€¦weā€™re going to have a problem with people running all different versions of the software as well as totally different software. So there might not be an easy answer for this one.

2 Likes

I agree that the store has the potential to become a mess if itā€™s not implemented. If we do not have a method for adjusting designs to material width, we at least need clear guidelines on how to post our designs to handle this (perhaps have several copies of the designs for several different widths we support).

I like your ideas of using color for lines that need to be resized. But I do have reservations about how practical it is in real life. I know when I do a design in InkScape, one shape has one line color. I canā€™t easily say ā€œthis line is this colorā€. I can always break that shape up into line segments, but itā€™s a pain in the neck to do so, and it has other negative consequences.

Iā€™m wondering if thereā€™s not an easier way to denote which lines need to move when resizing. Perhaps a layer above the cut layer that has lines that intersect the lines to resize (with 3 specific colors, one for right/bottom justify, one for left/top justify and one for center justify). Something like:

In essence, we are defining a standard for ā€œparametizingā€ an svg file. I was really hoping that something like this already existed, and we did not have to define it ourselves.

I know a lot of people are saying: ā€œjust use Fusion 360 or some other tool that already has parametric designsā€. Thatā€™s fine for making my own designs. I already do that.

But if Iā€™m selling a design on the design store, I canā€™t assume that the person who is buying it has that same tool installed, knows how to use it, or has the time to figure it out. We want the store to be as easy as possible to use. I donā€™t think itā€™s reasonable to ask customers to download and install Fusion 360, open the design, change the parameter, save the proper sketch as a dxf, open that sketch in Inkscape, resize it, color the lines we want to cut vs the lines we want to raster, etc. etc. etc. There are a LOT of steps between Fusion 360 and having a cut file that is usable.

I would much prefer to just put up an SVG file on the store, and have a quick GUI interface the user can use to adjusts the design for material width (or something that does it automatically if they are using proofgrade material).

An SVG file is just an XML file, so it should be possible to come up with a program that does this, given the right standards. Iā€™d be happy with a program that will work for 99% of designs. Someone will always be able to make a design that will just not resize properly given our standards, but those are the edge cases.

Sorry for the long post.

6 Likes

Some things are made harder than they should be, and this seems like one of them.

At a certain point, I think one needs to say, this plan has been tested and is designed for this proof grade material at this thickness (since proofgrade materials will undergo QC). Measure your material before attempting to cut.

The impetus should be on the designer to not only design for different materials/weights but also test for different materials/weights if they want variations available. Further, I imagine compatibility will be guaranteed (or should be) for only proof grade materials.

The goal should be for a professional catalog - where every single design has been thoroughly tested by a designer before being offered for sale. If you want to offer a plywood, mdf, and acrylic version at 1/8" and 1/4", each variation (all 6!) should be tested by the designer on proofgrade materials before being offered for sale.

Different materials cut differently, kerf differs between materials and material thickness and all of those factors should be tested before offering a design for sale.

Just my 0.02.

6 Likes

Perhaps that is the best solution. As someone who wants to design for the design store, Iā€™m exploring different avenues to offer the most flexibility to someone who buys my designs.

Imagine I post a design that I tested with just Plywood and I know it works with that material (like the chess set I posted here: Laser Cut Board Games).

Someone buys the designs, emails me, and asks ā€œWill this work with Acrylic?ā€. What do I say?

  • no, I only tested it with Plywood. You can cut it in Acrylic, but it probably wonā€™t work as well, since the acrylic is wider than the plywood.
  • yes, but youā€™ll have to use Inkscape to adjust the width of the holes here, here and here so they are slightly wider. Hereā€™s a tutorial on using Inkscape.
  • probably, but Iā€™ll have to purchase the acrylic, have it shipped to me, adjust my design, and send you a new file.

None of those answers strike me as ā€œuser-friendlyā€. And I donā€™t think itā€™s reasonable to ask designers to test it with every single feasible material (unless they are on the GF payroll).

I can probably guess which materials are going to be popular, design for those, and offer different files for each of those materials. And thatā€™s most likely what Iā€™ll do at the start.

Iā€™m just trying to figure out if there are tools out there that could make my life easier, and offer a better experience for users.

5 Likes

I get that - user friendly. I really do.

And user friendly, to me, is having a design that has been thoroughly tested on the mediums you intend to sell for.

In my mind, Glowforge is basically designing a stock library for laser cutters and as an artist, youā€™ll receive royalties. Think Getty, Shutterstock, or Adobe Stock. Not Etsy, where buyers can contact the seller and inquire about modifications, commissioned pieces, etc. Why compete with that huge platform?

Again, I surmise, but I feel like the goal here is world domination (hehe) - to have a Glowforge in a lab at every school and then in every home. Not just ā€œmakersā€ as we think of them now. People that want to buy a design that has been tested and with proofgrade material, will just work. So, for reputation stake, itā€™s important that designs are thoroughly tested and just work.

I feel like GF is investing very heavily in the material and proofgrade business because thatā€™s the huge money maker of the business down the road (imo). Get us early adopters on board to start making and designing, and then go for every home and say hey, you donā€™t even have to design. We have the worlds largest catalogue of laser cut designs that are guaranteed to work on our material.

Likely, the hardware will be even cheaper in the future. Its hard to consistently make money on hardware when itā€™s already being cloned, or attempting to be cloned. The future is in the value of the design catalog and guaranteed proofgrade materials - I think.

Sorry, rambling a bit. :slight_smile:

3 Likes

I agree with what you say. I have a very similar vision of the GF store and proof-grade material. But I hesitate a little with this sentence:

because:

  1. Thereā€™s a cost to testing on all the different mediums. Especially if we get into different types of wood. Each wood species behaves differently, and will have different settings on the GF. I will not have the ressources or time, as a designer, to buy each type and try it. I can certainly try it on a couple, not all of them.
  2. I might not predict all the mediums someone intends to use for a design.

I think when a design us put on the store, the designer should say:

  • It was tested with material A, B and C.
  • It should also work with material D, E, and F (because they have the same widths as A, B, or C).
  • It should also work with any material that has one of those widths.
2 Likes

Well, you guys let me know what final standards are once everyone has decided on something.

If I were going to consider designing for the store, I would try to keep it very simple.

  1. Make one version in whatever material you choose.
  2. Post a picture in that material.
  3. List the nominal size of whatever the material actually was, in both inches and mm.
    ( If itā€™s proofgrade, and that turns out to be fairly standardized, we might just be able to get away with saying 1/4" ply proof.)

Then just make the file available as is. Those who have a minimum of knowledge will be able to re-work the file for their own use in other materials. (If you want to be nice about it we could decide on some standard like pink fill or stroke for parts that need to be targeted or something like that.)

In addition:

  1. Offer to do the redesign in another material as a service for a nominal additional fee. But make it optional, not expected. (You would not believe how much time you are going to spend re-working your files for people who do not know how.) They give you the nominal measurement, you adjust the file and send it to them directly via email.

Once you see that there is interest in a particular file, you can do up several versions for the proofgrade materials to have them on hand if someone requests it down the road. But you will continue to recoup the additional fee for having to rework the file. (Itā€™s not being greedy, itā€™s being smart. Your time is worth something, and designs tend to have a short shelf life.)

5 Likes

Unless the design has a mechanical element that depends on certain properties of a material (friction, water resistance, flexibility, strength) I cannot see what can possibly be gained by ā€œtestingā€ a design in a multitude of different materials.

ā€œCool acrylic coaster! What is that, 3mm? Would it work in 4mm?ā€

A picture frame designed with 1/8" acrylic in mind will probably function just fine as a picture frame if itā€™s cut out of 1/8" plywood. Asking a designer to ā€œtestā€ a ā€œno solicitationā€ plaque in multiple thicknesses of acrylic, grades of leather, solid and ply wood, laserable vinyl, and stencil material will do little except waste designersā€™ time and discourage people from contributing.

4 Likes

@Hirudin : Do all designs need to be tested on all materials? Absolutely not. When tolerances are an issue - yes. If Iā€™m paying money for a design and using proofgrade materials, I want it to work when I hit print. Therefore, designs with potential tolerance issues should be tested and variations of the file made available for different materials.

I have no idea how the catalog will be set up. None of us do. Maybe my thoughts on files being thoroughly tested before inclusion into the catalog is my ultimate vision of what the catalog will be: a stock library of laser designsā€¦ once we get past the early adopter stage, people are not going to want to finagle with designs. They are going to want to hit print and make their dinosaur head.

If your design is of a nature that it will/could fail due to a difference in material specs, it should only be sold for the materials it has been tested for - and then the proper file variant could be supplied that has already been tested.

Maybe I misunderstood the nature of this threadā€¦ if so, disregard.

5 Likes

I did the same thing with a design I did for box puzzles. I made something similar to these: https://www.puzzlemaster.ca/browse/novelty/foam/

Once I change a variable, it resizes all of the parts and then I output a SVG file from Solidowrks.

I know that does not work for people without that software, but just scaling didnā€™t work for me.

1 Like

I still remember the day that @jacobturner schooled me on parametric design. I had just not done anything like that, especially in 3D modeling. I understand the facility of programs that have this function. As I have been working with my kitchen matchbox, I have thought about putting it into OnShape or Fusion360 but just am unwilling to start down that path yet. I just fiddle with it in Sketchup and put it into Inkscape. But I definitely need to design two different matchboxes, one to fit the Diamond Brand, which has a bigger box, and one to fit the generic Best Value matches that are smaller. Then I need to design the tabs to deal with different thicknesses of material. I think acrylic, baltic birch and walnut will all be great materials for this design. In the end I am better off learning the parametric program. But I can get Inkscape to resize the tabs as needed by changing the grid pattern size underneath and snapping to the new dimension to be able to accommodate different thicknesses of material.

This is a good discussion and is spurring me on to do some more design fiddling.

5 Likes

The Color Kludge to make SVG semi-parametric is intriguing, and could work out.

SVG are allowed to have text in the file, and that is ignored by the cutter (possibly not the same for Glowforge though, as being able to use text in an engrave file is more user friendly). In this case you would be able to use text to dictate if a color should be resized, and what type of justification to use.

Howeverā€¦ maybe we can stop wondering if the shopper in the Catalog has the skill to make a cut file from a parametric model, and instead ask for the store itself to be able to do that.

This version of the catalog would work like this:

  1. We can upload a fully parametric file from OnShape or other, and then specify what user inputs are allowed.
  2. When you buy a model from the catalog your account is flagged as owning that model (much like the Apple App store), and you can come back to that catalog entry any time to re-download the model with different parameters.
  3. After buying a parametric model, you are able to modify the default values for the various parameters, and then click a button to download the various corrected cut files.
8 Likes

ask for the store itself to be able to do that

Yes please.
Instead of loading the Glowforge preview/cut app with features do it in the store!

It is still a very hard problem to solve, but at least it means that the results could be saved on an individual basis so the software does not need to calculate the parametric values each time.

3 Likes