BAMF2011: Lasersaur Is One BIG Laser Cutter!

Psst…wanna buy a laser cutter, but not ready to sell your internal organs? Nortd Labs’ Lasersaur project aims to create an open source large-format laser cutter/engraver that undercuts (har har!) the cost of commercial models by an order of magnitude.

A Lasersaur built strictly using componentry from the globally-available bill of materials will set you back about $5,000. But if you can source some of the parts locally…or better yet, if you have a knack for scrounging (the stepper motors, for example, are common in junked inkjet printers)…it may be possible to knock that down by a grand or more. That’s still a big chunk of change for the lone garage tinkerer…but for schools or hackerspaces, or anyone who can pool some labor and funding, this could make laser work a lot more practical. Entry-level commercial models can be had for similar cost, but these are tiny in comparison — Lasersaur’s cutting bed is an enormous 140 by 72 centimeters!

Aside from bringing down the cost, another goal of the project is to break from the Windows/Corel hegemony that’s been the rule with most commercial laser platforms. Lasersaur’s “brain” is Arduino-based and cross-platform by nature. The hardware is interfaced through a custom motor control shield, easily assembled with all through-hole components.

The Lasersaur project is currently in a closed alpha stage, with the aim of a fully documented 1.0 open source release this summer. Project contributors have early access to the design documents and software, if you’re eager to get crackin’.

38 thoughts on “BAMF2011: Lasersaur Is One BIG Laser Cutter!

  1. its large but i dont think its “one BIG lasercutter” seeing some of the professional plasma cutters XD

    in theory you make one small one making a big one would not be to much more expensive right?

    the laser is the pricey part XD

  2. Arduino? Meh. Arduinos are generally considered too slow for any serious stepper work. Plus adding a trajectory planner. Maybe for cutting, but I cant see it for engraving.

  3. @barry99705
    “two entirely different things”
    are you high?
    the only difference is the head
    a good plasma cutter CNC head will run you $1200 about the same as a good 80-100w CO2 laser and driver (tho i prefer the firestar-f100)

    @BiOzZ true but sometimes a larger drive system can run a pretty penny depending on how you go about doing it

    @macona
    running at top baud a USB connection in general is to slow for live driving
    using cashed driving it can be just as fast but im not sure that an arduino can store enough data to do a proper job

    I prefer using dual parallel ports for any CNCs i design
    its cheap, its super fast and can drive up to 8 steppers, 2 PWMs, 2 indicators, 8 bumper switchs, 2 general inputs (wonderful for probing) and with some HAL driver modifications you can stick in an IR receiver and use a remote to do manual adjustments

  4. $5k plus HOW MUCH time and frustration to put it together / tweak it / fix it / upgrade it / get it working *just right* / taunt it / have it taunt you back / dial it in again / replace the laser tube frequently because it isn’t a high quality air cooled tube / etc.

    I own an Epilog that cost 4 times as much. But I didn’t have to put it together. There are downsides of course – harder to truly “fix” it as it is closed source, more expensive to maintain and purchase but you buy it and it works. Nothing to install or assemble and honestly, my time is worth something.

    I really hate it when people quote the parts only and omit labor because it paints an inaccurate portrayal of the *ACTUAL* costs of building something.

  5. Again, not trying to torpedo this project. I think the approach is great and I fully understand that things take time to be made, perfected, remade, perfected again and then finalized (for now). I would love to be able to make a unit like this for a few grand.

    That said – this is essentially a Chinese laser in terms of the optics (at least the ones that are currently being used). It uses a water cooled laser tube. The good news is that they are somewhat cheap. This project’s BOM has them at $200 each. They can and do die more often than commercial air cooled tubes however. That may not be much of a problem with this project – or it may – I really don’t know.

    Just keep in mind however that the chiller being sources (which is $400) is Chinese.

    http://www.cncoletech.cn/Laser%20Water%20Chiller.html

    It boasts such amazing features as “Professioanl radiator with great volume of heat diapation that forces the wind cool.”

    Needless to say, I am a bit skeptical of this chiller. You can buy nice ones on ebay for the same price or less however (Neslab makes some nice ones) but the one they have picked (mostly to keep costs down) is… sketchy at best. I disclaim that I have not done a teardown on this particular chiller myself but I would personally avoid this one.

    Perhaps the beta version manages to integrate an air cooled laser and maybe later also allows you to swap out to a YAG laser to mark/engrave metal? One can always dream.

  6. Lastly, the BOM has roughly 2000 parts. This thing will probably take at LEAST 40 – 60 hours to put everything together (and this assumes your tools are at the ready and you have ample space to spead out – not to mention time needed to tweak it and dial it in). Even if you can hire somebody at minimum wage (in the US) to assemble this (assuming the instructions are that good), it is going to “cost” an extra $300 – $465 in labor. More realistically, this thing would take a few weekends or about $1000 or so in labor.

    If you don’t mind it or can get free labor – fantastic. But again, I encourage people to not assume labor is “free” when they publish numbers on what things cost when it comes in the form of a project like this.

  7. TANSTAAFL?

    There Ain’t No Such Thing As A Free Laser?

    Being serious about the real cost factor differences between a commercial “Product” laser unit and a Hacker’s Prototype? The differences make cost factors go from under stated to simply not comparable. There’s an inherent set of parameters for a commercial product sold as plug in/begin using that are worlds apart from this excellent ProtoHack.

    BtW, air Vs Fluid cooling for Lasers is not a binary one good/one bad scene either. For example- in one set of designs the coolant should be DI water. Non-conductive, Not toxic or flammable etc. BUT- DI’s often a risk of becoming “hungry” and dissolving metals it contacts. Air Cooling’s a risk of airborne particle entrainment plus a quicker curve from flow stops>component DOOM. Properly set up fluid cooler systems “can” have some fail safer modes where thermosyphon is a backup. Air path systems often do not act so graceful. Cost Vs Fail Gracefully. If the Lasersaur costs enough less as cash out of pocket, we’ll see more of them

    Who knows, this may be to Lasercutting what Cupcake was for Hobby FDM gear.

    This overall is a worthy project.

  8. At the risk of offending dozens of people (or more), hobby FDM is still a joke and always will be. The concept of using a small ABS glue gun isn’t even very good in commercial units.

    This project however seeks to bring actual commercial level lasering into the realm of serious tinkerers. THAT is useful, nitpicking aside.

    This is basically an open sourced Chinese laser. Which can and do work fairly well. Generally with a bit of hacking and tinkering but they do work and they are less expensive.

  9. I’m always amazed at the price differentials based on size. I know it is mostly extracting extra profit from customers that can afford it to subsidize the base models and partly because larger beds require tighter tolerances to reduce waste(much like ICs).

    But really, if the commercial stuff doesn’t come down in price for basic 2/3 axis platforms, the hobby market is going to grow up to eat their lunch. And they probably can’t take refuge in 4/5 axis for much longer after that.

  10. Yeah, I think @hackerspacer has missed the point as well. So people LIKE building stuff.

    It is nice to see that it’s mostly of the shelf parts, unlike the Buildlog one the popped up a few days ago that used custom made rails & bearings – awesome – critical parts that doesn’t yet exist, will probably only have a single production run, are only available from one person, and can’t really be duplicated by anyone else. The fanboi crowd thought that was perfectly ok.

    Linear rails are fairly cheap, go buy them!

    I’d ditch the Arduino. If you don’t like PC/Windows, then use Linux/EMC2. With an old PC it’s essentially free.

    And @Hackerspace, the laser is only 40W, most people run them without a chiller, usually cooling is a bucket of water with a fishpond pump in it. Mind you, those lasers usually have smaller beds.

    Also, the optics don’t appear to be Chinese – they’d be much cheaper if the were. Just the laser is.

  11. I like building stuff too – bear in mind that my comments are directed at using things to make things to make money to buy new things :)

    I am perfectly comfortable with the concept of making things “for fun” but not profit. That just isn’t my perspective is all.

    And yes, these optics are solid. I also would prefer to see linux over arduino only because the raw power of a stock arduino may be inappropriate for this (but then again it may not – I have not tested it). Can anybody more educated than myself chime in? What about some of the newer arduino compatible boards that have more horsepower that we have seen in the last day or two here?

  12. @Tony

    The Makerslide linear bearing system on buildlog.net is being licensed as cc-sa attribution with no commercial restrictions.

    If it works well, it will have a future beyond Barton Dring, it’s creator.

    Having spent some time reading buildlog lately, I’ve been very impressed by the amount of time and expertise he freely shares.

    Queeg

  13. @Hackerspacer, the Arduino gives you a USB interface. Most PC-based controllers require the PC to have a parallel port.

    The problem with the Arduino (or rather the shield they are using) is firmware. CAM programs like EMC, Mach3 & TurboCNC are well tested, cheap or free, so why develop something new? Other USB controllers exist as well.

    These are odd sort of projects. Too expensive for the average hobbyist, too time consuming for the business types.

    They’re really aimed at start-ups, cash poor but plenty of time, this can make the difference between starting a business or not. Once you do get started though, you may find you simply don’t have the time to build a second machine to keep up with demand, so you simply buy a new one (eg Epilog).

    See http://www.mechmate.com for a similar project.

  14. @Queeg, I have seen that. At 30khz its slower than any of my other cnc machines. Might be fine for a low res machine. And compared to a smoothstepper that will do up to 4mhz pulses I cant see spending the time and money on a dual arduino kludge. Instead of sticking with atmega they should do it once correctly and use something with more oomph.

  15. @Queeg, open source means squat if you can’t make it.

    To make the buildlog.net wheels you need a lathe, which I have.

    To make the v-rail extrusion I need the die and an aluminium foundry, which I certainly don’t have, and I doubt you do either. Anyone have an aluminium extruder handy? Oh good, run me up a die while you at it, ta.

    NASA could open source the Space Shuttle – so what?

    You will need to register, but here – http://www.mechmate.com/forums/forumdisplay.php?f=6 is the way it should have been done.

    And that’s in steel, by comparison alloy is a piece of cake.

    As Barton says the off-the-shelf v-rails he was using were overkill for the laser (quite true), but going the custom extrusion route isn’t that bright an idea either.

    Sometimes wheels don’t need to be re-invented (poorly in this case), and that goes for the custom shield this project uses too.

  16. NASA could open source the Space Shuttle – so what?

    They did, actually.

    Space Shuttle Operator’s Manual, Revised Edition [Paperback]

    Ok so maybe not *exactly* but much of what NASA learns is technically available to the public.

    CNC plasma is nice but fails miserably at cutting details. Great for cutting large hunks of metal out of slightly larger hunks though.

    Even the most high def plasma with the best commercial controllers and the right software still fails. At that price, you might as well buy a 2000+ watt laser (better performance in that application) or waterjet (about the same price, more accurate although slower and more expensive to run).

    The arduino isn’t the only thing that can give you a USB interface though. I don’t think it is appropriate for this application given the clock speed. Can it *work* – yes. Should it? No.

    Maybe with this? BAMF2011: chipKIT is Arduino to the power of 32

  17. @hackerspacer
    There are enough companies making plasma equipment so somebody must be buying them.
    But it’s good to know your Epilog is the one and only tool for all your shop needs (what do you do with all the extra shop space? I heard you don’t even own a screwdriver).

  18. I am not saying plasma is useless. I am saying it, like any tool, has its place. And that place is not replacing a 40 watt laser cutter.

    We own other tools as well when the need arises for them.

    I own a standard def (kit based) plasma table, a high def plasma table and a waterjet, among other tools. We don’t own a high powered laser for a number of reasons. Mostly because we do mostly aluminum and cutting aluminum on a laser is tricky at best unless you have a 4000+ watt laser (depending on thickness of course).

    140 by 72 centimeters = 55.11 inches by 28.34 inches. The largest Epilog is 24″ x 36″.

    This really is a decent sized laser. How quick and how accurate is it is the question. Epilogs are mighty slow but can do 1200+ dpi raster engravings.

  19. Yeah, having plans for the Space Shuttle and actually building one are two totally different things. I doubt even NASA could build a Shuttle these days, there’s only so much you can scrounge on eBay.

    Adding ‘open source’ to a bad idea doesn’t help, you can say those magic words over & over and it’s still a bad idea.

    It’s occurred to me that this laser doesn’t have any exhaust venting. I guess they’ll figure that out eventually.

  20. I love comments like : why spend $5.000 and lots of time for something, when you can spend $50.000 and have something which works immediatly.

    Some people don’t understand why this website is called Hack a Day ?

  21. I’m not a fan of the lasersaur project, mainly because buildlog did it cheaper, simpler and first.

    Tony has consistently proved himself to be misinformed when it comes to open source hardware but he is spot on in terms of the controller.

    An arduino is literally the worst choice of hardware to power any kind of cnc machine. It has none of the capabilities you need and is ridiculously underpowered (high speed cnc requires pulses faster than an arduinos instruction cycle).

    Then they reinvent the entire CAM process but worse and you don’t even save any money for your trouble.

  22. Ah, @Ryan, but the Adruino is open source, so it should be perfect, right?

    So fanboi, educate me. Show me where I’m going wrong, and get me back on the true ‘open source’ path. Do I just say “Hail Open Source” ten times and we’re good?

    Show me how the buildlog rail system (custom manufactured, single source and currently 50% vaporware) is better than the Lasersaur setup.

    Note that both systems are open source.

    Note that we are talking lasers which are a lightweight system, similar to a pen plotter.

    I’m sitting in Sydney Australia, which one can I easily build?

    Over to you, fanboi.

    The Lasersaur people do admit they didn’t go out of their way to get the lowest prices, and it shows. $275 for a lens! Plenty of people will sell you the full optics setup (head, lens, mirrors & mounts) for less than that. My last buy was just over $200.

  23. Why pay one-fourth the cost for a DIY/kit version of something which might take some assembly, when you can pay full price for a commercial unit which works right away?

    Who would, with any seriousness and following of this site, ask that?

    Most folks on here like assembling their own stuff, afterall half the fun is getting there. And we like saving money, and we don’t mind if it’s DIY quality (which sometimes is as good as commercial)

    Aside from that, one of the most famous lines of personal computers started as a PCB hobby in a garage, which then became a basic kit sold to folks who then had to buy the I/O devices and the monitor.

    I’m sure lots of folks said things like “Sure, but it’s no Altair”.

    Hate to think of what our present would be like if there were no DIYers back then.

  24. I love comments like : why spend $5.000 and lots of time for something, when you can spend $50.000 and have something which works immediatly (sic).

    The thing though isn’t that it is $5000 + lots of time vs $50,000. A pre built, top of the line commercial unit is closer to $10000 – $20000, so it really isn’t an “order of magnitude”. You can go all out and spend more but not $50,000 worth.

  25. @Hackerspacer: who needs nuances? it’s just a laser xD

    Things like a z axis, auto focus, exhaust, red dot homing, variable dpi engraving, bottom up raster engraving, air assist, aluminum honeycomb cutting surfaces and the like are all, in total, very nice things to have when operating a laser. Commercial units have them and again, I reiterate this project is fantastic but I would not consider one until it is as feature rich as a commercial unit.

    Building something and then struggling with it at length over simple, time consuming hassles is not the hallmark of a solid build, in my opinion.

    That all said,this project will get there. I give it about a year to a year and a half to solve most of those issues.

  26. Ugh, such class privilege. Here’s a thought, I can’t afford a commercial laser cutter. They tend to cost more than I make in an entire year.

    So the “how much is my time worth formula” is pretty easy to compute. I simply can’t buy one, see how easy that was?

    However, as a student who has the time to maybe scrounge around for parts, these open source projects are great. Think of the little people, the ones who can’t just throw money at commercial systems.

    It’s also great cred too when you can show your professors that you can build DIY solutions to otherwise commercial systems.

    In fact, my DIY camera controller systems and sensors are really popular with my friends and profs and it’s definitely helping me with grad school portfolios, they tend to like that sort of thing.

    It wouldn’t exactly be special if I just said “hey look at this cool thing I bought.”

    So stop speaking for everyone or repeating the same crap we already know. Yes, there are busy people who don’t have the time. Get over yourself, we aren’t all like you.

  27. I suppose I missed it somewhere, the portion of the project which said it was out to upstage commercial units and provide the exact same range of features and industrial-capable-service.

    I checked the project goal section which mentioned wanting to provide an open source readily-repeatable solution to personal laser cutting.

    I’ve seen the light now.

    Why have Hack a Day when you can just go to Thinkgeeks and buy cool commercial gadgets.

    Shoot, why have a commercial laser cutter when you can just buy the parts custom-made commercially?

    Why use anything Google, when you can just pay for a search engine, or pay for GPS navigation?

    Why spend my time building something for fun which comes in a kit form when I can just buy it?

    Why build a house, there’s plenty of caves which came already available from nature.

  28. Why buy a commercial unit when eventually an open source one at least meets if not surpasses the commercial offerings? THAT is what I am personally waiting for. I don’t speak for everybody here and I never said I do. For me, it makes more sense to purchase a commercial unit at this point in time. The pendulum is swinging the other way however and before too long (1 – 1.5 years?), my Epilog will be (somewhat) matched by DIY efforts like this.

    That is, in my opinion, awesome.

    But I can’t operate a commercial venture with promises and what will come. But when it does, I will embrace it openly and warmly. I would rather my machines be open source rather than closed sourced, over priced and proprietary.

    Which is, why I reiterate once again, why I am very encouraged by this project.

    I am also following this guy very closely: http://3dhomemade.blogspot.com/

    He is basically making an open sourced high def 3d printer. That can be made for under $5000 that literally competes with $40,000 – $120,000+ units in terms of speed, quality and final output materials. In some ways it could even surpass it.

    But he too is a good 1 – 1.5 years out (or more) from being completed.

    So again, I am not arguing this project is crap because it can’t run with the “big dogs” of commercial or industrial units. I only stated that I am waiting for it to match the capacity of commercial units before it becomes viable for me and my usage requirements. Your mileage may vary – and that doesn’t make me all high and mighty and I never intended to come across as if it did.

  29. Sometimes ‘buy vs build’ comes down to ‘just buy it’.

    Not just in the commercial sphere as @Hackerspacer is referring to, but for the rest of us as well.

    The small 40W lasers cost about $1,000 these days and require minimal setup, and if you price the components, well, you don’t save much money. You do gain the joy and aggravation of DIY though…

    It’s not the first time that’s happened, people used to build radios & TV’s too. I don’t think it’ll happened with large bed lasers though.

  30. I’m buying into this.

    I very much want to be able to mod/fix/hack my own laser cnc. And yes, the arduino may be a little underpowered for driving this badboy, but there are other, very affordable, multi-axis drivers that can be put in place of the arduino such as to be compatible with Mach3 etc.

    All in all, the lasersaur team did a bunch of work, and their design is sound, but still in progress. I applaud them for their efforts and would rather my hard-earned laser dollars go to some upstart hackers than to epilog or universal. I want my maintenance contract to be with myself, thanks, not with a for-profit company.

  31. I understand where Hackerspacer is coming from. Just like a CNC router table. Yes, you can build one yourself for pretty cheap. About on the order of what it costs to build a laser cutter vs buy a commercial one. But for a place like that needs to have the machine operating like a school, business, or techshop like entity it is not a viable option. In these locations you need to be able to get the machine up and running in minimal time and if the machine goes down get troubleshooting help on a 9-5 basis or even get a tech to come out when it breaks down. That is the limitation of home made machines. Just like in the days before mass production many of the parts in a home made laser cutter are one offs or limited supply parts from one or two vendors.

    But for someone who wants to build a machine for themselves there is usually little issues here. I am slowly, I mean slowly, building my own laser cutter as well. It is a low priority project.

    But the lasersaur justs costs too much for what it is, even as a kit. You can buy a 1000mm x 1200mm bed , 80 watt chinese made machine for $8k, 600mm x 900mm bed, 80 watt for $6300. And this has a real control on it (chinese DSP)and comes with the chiller, exhaust fan, and after alignment it is ready to go. The software is kind of “meh”, but it works pretty well. The hardware itself seems to be really good and has a couple features that i would like to have seen on the Epilog.

    It comes down to this, the lasersaur should at least compete with a chinese machine of equal bed size and power, at this point it does not.

  32. @Tony- actually a 40w on ebay cost around $350-400 not 1,000$

    @macona- i actually don’t think the lasersaur competes with the chinese already. I saw it at MF in SF last weekend, and I have to say it competes much more with my 30k universal then the crappy chinese systems I have played with. Its a super solid system.

    I do not think/get the impression they were trying to make the cheapest system, I think they were trying to make a system which can be built worldwide, that is super solid and build a system which can grow with its community. Being on the alpha list, I can tell you there are tons of people actively building the lasersaur as we speak and tons of people actively working on it behind the scenes..

    Seriously, give it 1 year or so, with the community it has behind it, I have no doubt it will surpass/beat all the commercial systems in both performance, software and price point.

Leave a Reply to andreaCancel 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.