Lasersaur That Cuts 1/2″ Plywood And 10mm Acrylic (Pew-Pew)

Remember when building your own 3D printer was a big deal? We’re starting to think that building your own laser cutter might be the next hot topic.

Boasting a 16,000 square-foot facility, the Dallas Makerspace is an impressive collaboration of local artists, engineers, makers, and thinkers. Recently they embarked on building a serious laser cutting machine. They chose to go with the an open-hardware design rather than buying an off-the-shelf unit. What they built is based on the Lasersaur plans. (Another popular open-source build is the buildlog.net unit.)

They ended up with a huge 24″ by 48″ cutting bed and with a laser tube rated for 100 watts continuous output. It can cut 1/2″ plywood and 10mm acrylic with ease. The entire machine is built from 20mm Misumi aluminum t-slot extrusions, making more like a giant erector set then a commercial built machine. We hadn’t seen too many of the Lasersaur builds out in the wild, so we thought you might like to see one too.

Now, before you start ordering parts to build your own, you should know that a top of the line build like this will run you about $7-10k. But by comparison if you were to go with something with the same cutting area and power, you’d be looking at something like the “Epilog Fusion 40” at a whopping $40k. With that said, we expect to see more budget laser cutter builds. Cost can be cut dramatically when you go for a smaller machine, with less cutting area, and less power. With that, you can use less expensive steppers, drivers, and frame. We suspect a little as $700 for a smart shopper could yield a very respectable laser cutter.

If you’re interested in learning more about the Dallas Makerspace, we took a video tour back in early 2014.

54 thoughts on “Lasersaur That Cuts 1/2″ Plywood And 10mm Acrylic (Pew-Pew)

  1. So…. I’m assuming it’s a 100W CO2 laser, liquid cooled, not a solid state device? Some details would be nice, cut speed, smoke recovery etc… maybe a video of it working?

    1. Yes. Any laser of that size will be water cooled. Smoke abatement is an interesting topic I want to cover in a post soon in a very DIY sort of way. As far as details – the links should give you some good specs. But you’re right – we should have posted a video of it cutting (but honestly, once you’ve see one laser cutter, you’ve kinda seen them all)

      http://www.youtube.com/watch?v=d3HGPoVQvKk

        1. Soooo, whacha building?
          And have you found a way to place the targets in a way that wastes less plywood, (I mean less space between them and swapping one of the targets end for end?)

          1. it’s a bigish UAV.
            https://www.youtube.com/watch?v=MHt6Tu-DhnY
            Since this video there was a major “hard landing” and I have to rethink the full flying rudder and elevator, which worked fine in tiny scale, but vibrated like a bitch when scaled up.

            I wasted heaps of balsa cutting these, having 30 people wanting to use the laser cutter was a pain!
            It was the second to last day of Zoz Brookes “Digital fabrication master class”, which was great!

  2. Well we bought a G.Weike machine for our hackerspace- it’s 100W (measured 98.9 W), 900x600mm + Z axis + all the possible upgrades + a spare tube= cost us around 7k Eur (8-9k USD for that time) with all the VAT + Import + crane.

    Works great, stunningly high build quality given that it’s made by chinese and we have had no major problems with it.

    I understand the thrill of building something yourself, but I totally don’t get it in regard to a 3 axis (or is it just 2 for the laser saur) laser machine.

    1. G-Weike’s aren’t all that well built though, which is too bad. They look impressive but you look at the wiring and you start questioning the purchase decision. Our maker space has one and they’ve done OK, but it’s needed a lot of help. They replaced the door hinges, laser lenses and power supply in only the first year of operation. I’m sure they had some other problem I’ve forgotten.

      I’m not saying Lasersaur is a better idea. Though its working area is a third wider (1200mm instead of 900mm).

      But even then, one should look for the working area they need, not one they want just because it’s big, because there’s a lot of hassle in focusing the giant machines.

      1. The software is crap, but the wiring is not that bad. The only weird thing is that the power supplies for non-high voltage are split in to multiple ones, with the smallest being a cheap looking AC->DC power adapter.

        We put up a ghetto POE injected RPi + Arduino mini pro + AN522 RFID with a screen so that the access cards can be used to turn it on/off + count the lamp impulses between the controller and the high power supply.

    1. Waterjets turn out to be a LOT harder than lasers. Laser tubes are cheap now thanks to Chinese labor. Until we find a new way to generate tens of thousands of PSI at significant flow rates, building intensifiers and such will likely remain pretty costly. Not to mention that the basic horsepower requirements are very high. Commercial machines run north of 25 HP, and even then are comparatively slow. A consumer scale unit capable of fitting in a normal garage will be SLOOOW.

      There’s also the issue of supporting infrastructure like water tables, abrasive recovery, nozzles, etc. thst will remain even if you figure out how to generate pressure cheaply. Unless you want to cut stone or glass, plasma, a router table, or a machining center offer more speed, versatility, or both at much lower cost.

  3. For the typical HaD reader who wants a laser cutter, I’d recommend getting a DC-K40 from eBay and fixing (i.e. replacing) the dreadful electronics. I did that thinking that I’d play for a bit and then just use it as a donor machine for the tube and PSU, but to be honest it does what I need for now so I’ve stuck with it.

    1. yep i did the same thing took a punt on it for home, i even have new gantrys but ended up keeping it mostly as is.

      i added a motorised ztable to mine, changed the the optics and managed to tame moshidraw enough to make it useable. for $400 can’t really go wrong. there are a couple of versions of the k40 i got the one with the digital readout. its an ok platform for small scale, and if you outgrow it, donor it.

    2. Oxfred, got any links on doing just that? I’ve been considering the purchase of a ‘guaranteed broken’ commercial laser cutter for $500, but your suggestion looks promising. What does it take to turn one of these into a usable tool?

        1. I have a K40 and enlarged the default cut area (to get your 200 x 300) .. adjustable lens mount to adjust for material thickness (building a mechanism to raise and lower the bed is a hassle) .. but I got the version that uses the board and plugin for Corel Draw. It works surprisingly well.

          The ONLY thing holding me back from building a machine a bit more capable is the lack of electronics and software. The 3d printer electronics and software obviously has matured greatly. There is no great, stand out, open source electronics or software tool chain for lasers.

          Sure .. there are one or two .. but while notable .. are old and less than capable especially when talking about raster engraving.

          I anxiously await a maturation of laser electronics and software. As commented earlier .. laser tubes, lenses, and structural components are very affordable these days.

          1. How did you do your enlargement? New housing or just squeeze a bit more out of the factory case? I was hoping to get about 10% more without changing housings and I see a few options, but I was hoping to see how others have done it.

  4. Is the lasersaur software is in usable state already?
    Last time I checked it was not suitable to any serious CAD work, the basic workflow was like,
    import the svg and cut it. You could not even measure a known distance in the drawing,
    if it imported fine, more importantly the zoom was not available either.

    From time to time I checked their github account, but the activity was next to nothing. Then I stopped bothering it.

    Did the state changed, or is it still the same? Cutting nice decoration is certainly a usecase. But cutting dxf (with tolerances) is a question totally different.

    1. My local space had it’s 80w CO2 electronics die and we opted to replace them with the lasersaur two years ago. After a year of broken promises, a hundred hours of troubleshooting, and eventually seeing that it can’t, and won’t ever (without a complete overhaul) be capable of doing what it claims, we threw it out and purchased the replacement chinese electronics.

      I’d strongly recommend against anyone using the lasersaur project. Which is a shame, it’s one of the original kickstarters and it SOUNDS great. It’s just, not a good solution is all there is to it. (Neither is the proprietary chinese crap, dongle, and terribad software we went with instead… but… that should give you an idea of just how bad Lasersaur is that we *preferred* that to it.).

  5. I’ve been looking to put together a fiber laser for metal engraving… Any plans/suppliers people have been successful with? Laser source, columnator, galvo head, lenses? Stage automation optional at this point. Commercials for 20-70w run 30k-80k :-(

    1. Fiber lasers themselves are ridiculously expensive at this time, you are looking at several k$ just for the laser alone…

      A much better alternative for DIY would be a flashlamp driven Nd:YAG, which while horribly inefficient should be doable for a couple of hundred $$$ from ebay junk.
      If you add a zero to this, you could possibly make it into a diode-driven one.
      Galvos could be had cheap, even the dielectric mirrors (metal mirrors will probably not survive even the first pulse) for them can he had cheap, but the servo drivers and f-theta lens will cost you unless you get ebay-lucky…

  6. just purely for laser kit golfing

    i think this is all you’d need including software , just add mirrors and a focusing lense, extractron fan, cutting surface, panels for the sides of the frames and a PC. 100W with 1200x900mm hogsheads, there is a 1600x1200mm which i have, but there is no frame kit for that size frame.

    http://www.lightobject.com/XLE-1200900-XY-Stages-complete-kit-for-DIY-CO2-Laser-2-yrs-warranty-P737.aspx $995
    http://www.lightobject.com/Aluminum-frame-kit-1200×900-for-CO2-laser-machine-P868.aspx $850
    http://www.lightobject.com/High-Quality-Reci-100W-Z4-CO2-Sealed-Laser-Tube-10000hrs-P119.aspx reci Z4 100W $995
    http://www.lightobject.com/3MD560-Leadshine-compatible-3-Phase-84A-1-axis-Stepping-Motor-Driver-P836.aspx $62 ea. need two
    http://www.lightobject.com/100W180W-PWM-CO2-Laser-Power-Supply-AC110V-P74.aspx 100W PSUI for Laser $385
    http://www.lightobject.com/AWC608-Commercial-DSP-CO2-Laser-Engraving-Cutter-Controller-P321.aspx controller $385
    http://www.lightobject.com/DC-24V-15A-Switching-Power-Supply-Good-for-CNC-Stepping-Motor-Drive-P447.aspx 24V 15A $48
    http://www.lightobject.com/CW5000-800W-Water-chiller-for-CO2-laser-machine-AC110V-60Hz-P823.aspx chiller $600

    about $4K – $4.5K for all the parts sourced locally. i maybe have missed a few parts, as i just threw this together, but these are the high ticket items, you can find them cheaper.

    compare that to a smaller complete, shipped from china 900x600mm 100W RECI from gweike which about 4K + shipping and has a motorised Z table and extraction fan.

  7. This is a reply to a few things, we are in a new building now so it’s 16,000 square feet. :)

    The software for the Lasersaur, LaserApp, is pretty solid at this point. Raster is the one elephant in the room for the software. As far as cutting accurate designs from CAD files and such it’s fine for that. It expects SVG or DXF with millimeter units (1220mm x 610mm cutting area).

    We went with an open source build so we could learn from it and explore more options in the future, like a significantly higher wattage tube, integration with our RFID access system, larger bed sizes, etc. With a completely open source design we have access to everything we need to make that happen for a future build. There are better deals out there in terms of specifications for the dollar but that wasn’t our primary goal.

    If you have questions about the build or Lasersaur stuff feel free to email me, luke@dallasmakerspace.org, cheers!

    1. Another reason was the fact that we got tired of having to source $300 small parts. Which in this kind of environment you have to do on almost a weekly basis. (Can someone here PLEASE make a Open Source CO2 Laser PS?)

      Our FS unit is great for rastering, but we have people who want to do larger vector projects. While we did consider an Epilog, the power to size ratio is not optimal. Yes, an Epilog is faster… but you also have to figure in the people who don’t necessarily need to do engraving on a daily basis.

      On the account of people wanting to use this for hard-core CAD work, I have made the back to my portable iPad monitor. So… it is possible, and if you don’t like to feed it a SVG or DXF you can input GCODE and cut your file that way.

      Hackability was also a factor. We do plan on integrating access controls into the unit, which is a royal pain with other units like our FS unit that uses mains voltage directly on the stepper drivers. (or an Epilog which uses servos)

      This being said, we have a mini-engraver in the works. We would also like to build another Lasersaur, perhaps with a Nd:Yag laser. (which would be a fair hack in itself)

    2. Anyone who has built a 3d printer can fully understand the mechanics of a laser. Parts are super reasonable now. Electronics and especially a good open source software package. That is the gap. Nothing seems in the works or progressing. Smoothieboard was to support laser cutters, but I see no builds. Same with software.. Nothing really moving along. It is sad my K40 craptacular electronics and CorelLaser plug in is the best low cost package out there. Is there any community working on this?

  8. We have a couple of these 100watt lasersaurs – we use them in full scale manufacturing, 12 hours/day, 365 days a year. Awesome machines for the price…just need a ton of sweat-equity to put them together. But having a machine where you understand 100% of the hardware, software, firmware, etc is well worth the effort!

  9. 100W and only 1/2″ acrylic? I have cut 1.25″ thick on a 80 watt. Something is wrong if thats as much as they are getting.

    You just cant compare a machine like this to an Epilog Fusion, the Laser in the epilog alone costs more than the whole lasersaur, the laser is a much better RF driven unit, I think by coherent. Also servos on the axis with a linear encoder on the raster axis. Also the software and control are on a whole another level.

    1. Well, they didn’t say it was the limit, they said it was easy to cut.

      And yeah, I get annoyed when people compare their project to some super high dollar product when it’s a poor comparison and there’s more realistic comparisons to be made.

  10. After having cut inch think plate steel on one of these babies http://www.bystronic.com/en/Products/laser-cutting-systems/Bystar.php every time I see one of these 100w builds all I can think of is “aaaawww look how cute it is”

    With such a low power laser over such a large cutting bed, unless you have your optics dialled in 110% you are going to have a bad time. In a makerspace this probably isn’t a big deal, after all you can file the edges, etc. but even on a pro machine when you’re talking 10 – 15mm thick stainless steel cut 90% of the way through, bubbling up molten stainless then having the cutting head smash into it all because there is a little bit of gunk on a mirror, it’s something you really need to keep an eye on.

    Don’t go for a large bed unless you absolutely need to use it and often, the optics will go out of focus and will cause no end of problems, it’s not worth it.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.