We covered a lot of ground in Part 1 and some of the fundamentals of laser craft - the laser, materials and prep, cutting, air flow, exhaust, etc. This post will spend a little time discussing designing and manufacturing a good puzzle.
Material Selection
A quick search on laser cut puzzles will reveal a host of media that can be used to make a puzzle, ranging from conventional commercial paperboard, to various types of veneered plywood, to medium density fiberboard (MDF), to cast acrylic. The use of these different materials can not only change the aesthetics of your puzzle, but influence the manufacture, packaging, and pricing of your work, if you're looking to sell them.
Common puzzles available at toy stores, bookstores, big box retail, etc. are made of a material called Paperboard. While we do use a lot of recycled chipboard (think soda cartons and cereal boxes) in our shop, we don't work much with paperboard given the lower quality nature of the material and its availability as puzzle products in massive quantities. If you want paperboard puzzles, there are plenty out there. We're wanting to make something special that will last a lifetime, perhaps as an heirloom product.
Plywood
Plywood is the go-to material for most laser cut puzzle makers. Plywood offers a durable surface that if cared for properly will ensure that your puzzles last for generations. Crafting plywood comes in a range of thicknesses from 1/8" (~3mm) to 1/4" (~6mm). We've used it in 4mm and 5mm thicknesses as well, however these are rare in the United States. There are a variety of cores in different crafting plywood including birch, poplar, and MDF to name a few. The layers of inner plies are composed of different veneers, range in thickness, are cross-banded, and laminated with adhesive. Another reason to use plywood designed for laser cutting is that adhesive layer and any gasses and residue that can be emitted during cutting and engraving. You don't want to be dealing with toxins that you're unaware of in a product. Another aspect of plywood to consider is the veneer, with birch and maple being poplar choices, but they also comes in a wide variety of specialty hardwood veneers. Keep in mind that unless you're making something a little different, most puzzles will have an image as part of their construction, so using a specialty wood might not be a great option for cost reasons.
Plywood also cuts quite well on a laser but it can be a bit finicky at times. Voids in the interim layers and glue spots that present challenges for lasers to cut through are not uncommon with plywood. As we mentioned in our previous post, selecting a higher quality plywood that is engineered for laser cutting is a better option over plywood from big-box stores. There's nothing worse than a puzzle cut that fails 12 minutes into the process due to a void. It's also worth noting that plywood produces a fair amount of smoke, soot, and resin during cutting that leave a residue and possibly discoloration without proper airflow. It's a balancing act.
All of that said - plywood is the material of choice for puzzle making and it's what we use.
You'll find a few of the manufacturers that we use in our shop on our materials page.
Medium Density Fiberboard
We do work quite a bit with MDF in both 1/8" (~3mm) and 1/4" (~6mm) thicknesses and as a material, MDF responds quite well on a laser cutter. In fact it's one of the materials that laser crafters generally have a great experience with. It cuts well under the right settings (speed, power, focus, airflow) and is easy to paint, coat, and glue for popular layered products. It also works well for puzzle making. Typically MDF has two finished sides, as opposed to hardboard with one, and that smooth surface is great when it comes to adhering an image, which we'll discuss in another post. If you're wanting to just do a one-off project and happen to have MDF available, it's great for experimenting in puzzle making and relatively cheap when compared to other materials at about 1/2 to 1/3 the cost of plywood per sheet. In short, MDF is a great way to get started.
Acrylic
Acrylic isn't as popular of a material when it comes to puzzles, but we do run across an acrylic puzzle from time to time. It comes in a range of thicknesses, is available in a wide variety of colors, patterns, and finishes, and can be an alternative for a challenging style of puzzles with or without images.
The Puzzle Cut
The art that goes into puzzle making isn't just about the image that is used to provide clues into reassembling the pieces, but is as much, if not more about the cut of the puzzle itself. Most of us are well familiar with the conventional puzzle cuts, with the classic knob and pocket style of interlocking pieces that are typically flag or square in shape with some variance. Beyond this common style, there are wide variety of options when it comes to puzzle cuts, and many elements of a cut to consider when planning a new puzzle.
To get a sense of the array of options, one of the best sources of information on the history and variety of puzzles is Bob Armstrong's methods for puzzle classification. Bob does a great job of cataloging and classifying elements of puzzles through analysis of his historic puzzle collection. On his site, he details different knob styles, line styles, patterns, strip cutting techniques, whimseys and more. It's notable that Bob's material centers around classically crafted puzzles, i.e. puzzles that are cut with an actual jigsaw as opposed to a laser. While we're going to use a laser, the different styles, shapes and methods still apply. Then, given the precision of a laser, there are new opportunities to explore, e.g. continuous puzzles, where tessellated pieces and an appropriate image come together to create a puzzle that has no real beginning and end. Pretty cool concept.
Conventional Cuts
To get started, I'd highly recommend using a conventional puzzle cut. This allows spending more time on the construction and cutting process without having to focus on the elements that make good puzzle cut, which can be tricky. There are a number of puzzle cut generators that are available on the internet and I use this one for conventional cuts. It currently generates two different types of puzzle shapes in a conventional knob and pocket style, with the ability to tune some parameters of the puzzle generation. Then you can download the cut as an Scalable Vector Graphics (SVG) file for use with your design and laser cutting software.
Performing the Cut
Once you've used the tool to generate an SVG, it can then be imported into Adobe Illustrator, CorelDRAW, or Inkscape for further manipulation. Most often we take it directly into Lightburn and just work with it there.
Depending on your workflow, you may be like us and have decided to lift your work piece from the bed to prevent flashback. You can find more on that in the previous article. Lifting a piece from the bed has a lot of benefits in terms of airflow and mitigating flashback from the cutting bed. There are however some things to think about with this approach. Lifting the piece from the bed means that it will fall a short distance to the bed as it's cut. This means that the craftsperson (that's you) must take into consideration a number of factors including
The order in which pieces are cut from the material
The flexibility of the materials being cut
The airflow pushing on the piece
The effect of our good old friend, gravity
The point at which the laser starts and ends in a cutting pass
In the image above you'll note that there are three layers in the puzzle; red, green, and black. Our puzzle cutting optimization settings (not shown) typically follow a sequence of "Order by Priority" and then "Order by Layer".
With "Order by Priority", Lightburn will first honor a priority setting for cuts. Each cut in the image above is either a red line (11), a green line (17), or the black outline (1), for a total of 29 cuts, and we've gone in and assigned a unique priority value to each one. Following cut prioritization, a cut with a lower value is performed before ones with a higher value setting, e.g. 10 goes before 20, goes before 30, etc. We typically assign values in increments of 10s to each cut, 110, 120, 130, etc. and start a layer (color) at a number evenly divisible by 100, e.g. 100, 200, 300, etc. That's because in a piece with a lot of cuts, one sometimes misses something and you need to insert a cut in-between others. Incrementing by 10 leaves plenty of room to do that.
Then with "Order by Layer", if there are several cuts that happen to have the same priority value, say 50, cutting is prioritized by the layers as they appear in the upper right "Cuts / Layers" box, and are honored top down by color. For this puzzle the red lines with the same value would go first, then the green lines, then the black lines last. Now you might be thinking, "if you're giving each cut a different priority, why bother with the layers then?" Well, sometimes for a complex puzzle it easier to deal with groups of cuts, and using layers allow us to show/hide, output/exclude them en-masse. So layers aren't just useful in determining cutting or engraving order, but for other operations as well during the design process. It may seem complicated, but if you tinker for a bit, you'll become more comfortable with it.
Order Matters
There are a few reasons to assign a discreet value to each cut for priority. First, with laser cutters, order of operations matters, and with puzzles they matter a great deal. If you've been using a laser for any amount of time, you'll know that sometimes you're engraving before cutting, sometimes cutting before engraving, depending on what you're making.
With puzzles we're typically just cutting, but some of the same general rules of cutting apply, e.g. cutting from the inside out, or starting from a specific direction. Further, we lift or pieces off the bed, and as you might imagine, cutting the edge of the puzzle before cutting the inner pieces would result in the puzzle falling to the bed prematurely, and then all the pieces being cut while the beam was now out of focus. That's a piece destined for the trash.
Additionally, as the lifted puzzle is cut, depending on the material and its weight, as well as the size of the cut, gravity will pull and airflow will push on the piece. In the video above, you might imagine that as long strips are cut from the material, gravity and airflow will cause them to travel towards the cutting bed, taking them out of reach of the beam's focal point. Subsequent cuts of these bed resting pieces may be successful, but the cuts are often noticeably wider that the rest of the piece because they are well past the focal point, or perhaps worse, the out of focus cuts will only be partially successful towards the bottom of the material and will remained attached. When this happens, the puzzle is yet again destined for the bin. While a few improperly cut pieces are not unusual under ideal circumstances, no-one wants to take a hobby-knife to 100 pieces of a poorly cut puzzle. We want it to cut correctly, every time.
In our video, you'll also see that we make the vertical cuts first. Again, this has to due with the weight of the resulting strips. As the first horizontal cuts occur, we start with the middle cut. This first cut splits the vertical strips into equal parts, distributing weight evenly across them. Remember, gravity is pulling and our air assist is pushing on these strips already, causing them to want to drop below the focal point where we want them. Distributing the weight evenly with the initial cut, and then having subsequent cuts trim out the puzzle pieces from the ends first, cause the weight of strips to decrease as pieces fall to the bed, and then the strips spring back a little with each cut. This makes for a better puzzle cut. Once initial weight concerns are mitigated, subsequent cuts occur in sequence towards the edge to save time, yet another factor in manufacturing at scale.
Finally, sometimes the starting and ending points of a cut line matter as well. In our experience this is most common when a 'T' intersection happens between two cuts. If the precise intersection of the cut lines is just a little off, say due to vibration and subsequent material movement or perhaps mechanical backlash, then the puzzle may have noticeable points at which two cuts do not exactly align. Regardless, by starting one cut first at the top of the T junction and drawing away from the 'I', and then crossing it with the other cut that forms the top of the 'T', the effects of these anomalies can sometimes be mitigated.
Practice, Practice, Practice
All of that said, a puzzle cut is very subjective. Sure there are general rules to follow, and you'll be able to quickly create conventional puzzle cuts, however as puzzle cuts become more complex, each requires careful consideration and planning to execute successfully time and time again. This is all just another part of the craft and if you're wanting to produce puzzles even at a modest scale, you'll want to learn to create a method for each puzzle that is repeatable with a high rate of success. When we say modest scale, we do small batches of 3-4 for new concepts that we're taking to market and larger batches of 10-12 for successful models with repeat sales. If you're a small maker like us, your time is precious, and since the cut is close to the end of the process, you're already invested in a piece both from a materials and a time perspective. Cutting is the point at which failure is mostly costly during manufacture. It's the point of no return. The final thing to repeat is that designing a cut isn't a one-and-done exercise. The old joke, "How do you get to Carnegie Hall? Practice, Practice, Practice." rings true here. We do many test cuts before we take something to market. We definitely scale down the cutting to save on materials, but it's not unusual for us to do 8-10 test cuts before we make a puzzle that can be manufactured and is one that we like, considering the aesthetics of the cut and image. We're planning on getting into the puzzle image and printing in our next article. Thanks again for reading and reach out on Facebook if you have any questions.
Commentaires