Aleph Objects, Inc. is a Free Software, Libre Innovation, and Open Source Hardware company. We currently manufacture the LulzBot line of desktop 3D printers, and sell parts and plastics. What do you think we should make next?
A laser cutter platform. I believe that glowforge took a great approach to the laser cutter but I would love to see an opensource project that can accomplish the same thing. https://glowforge.com/
More diverse line of 3D printers, or more customizations, straight from factory. Offer all-metal framing. Offer OpenBuilds linear bearings. Offer laser/microscope attachments. Offer milling spindles. Offer inkjet heads. Offer different depth probes. Generic CNC platforms, like cars, seem to have a lot of “accessories” that can suit an individual driver’s preferences. At least for metal brackets and OpenBuilds hardware, there seems to be demand on these forums.
Electronics assembly automation, specifically pick-in-place, also seems to have underserved demand.
Of course, I am a bit biased, having been attempting these things myself, but nonetheless, I submit the basic premise for consideration.
They could make one of these pretty easily once I upload the STL’s somewhere…
https://forum.lulzbot.com/t/project-laserscythe-or-what-exactly-do-you-do-with-leftover-taz-rods/3011/1 Part cost for that was around $230 for a one-off. (laser was $100, Controll circuitry was $50, Motors and belts $30) In bulk from different suppliers you could probably get it down further, though the enclosure would probably cost a bit more if you don’t get lucky and find a fleet of surplus metal printer cabinet boxes for $5 each like I did heh.
I’d like to see a “pro” version of the Taz, maybe with a smothieboard or duet board core. There is a small but growing market for a completely solid, built and no adjustment required printer that can outperform the Stratsys machines at the $3-$5k pricepoint. There would also be a market for a larger machine, but again that would be somewhat limited.
You could look at CNC manufacturing, but thats a pretty crowded market. That would give you theoretically an in house milling capability though which might come in handy.
There’s always the ROV market!
There is definitly room for improvement in the 3d scanner market too. That reminds me I need to finish the BQ scanner frame redesign sometime in here.
There is also a market for small specialty machine tools. If you look at a micromark catalog there are all sorts of small bandsaws and sanders and whatnot selling for $300. An open source version of any of those tools, or even a desktop small version of something modular and upgradeable along the lines of a shopsmith Mk 7 (http://www.shopsmith.com/markvsite/index.htm) would be of interest.
A small (size of the TAZ), inexpensive, open source CNC mill capable of working wood, plastic, aluminum, etc. Extra points for being able to work steel. Add-ons might include continuous liquid cooling and a forth axis.
TAZ pro, openbuilds rails, linear bearings,sensors for filament width, auto leveling, three point bed leveling, drop in beds, strengthen extruder mount, other extruder options(Cyclops) …
+1 to piercet’s suggestion regarding Duet electronics
OpenBuilds hardware does seem popular. Would also like to see Y-axis twisting eliminated though, as it is difficult to adjust.
Disagree with drop-in beds, at least as a standard/default feature. Keeping these things aligned is tough enough already.
@geburgess See http://www.taigtools.com/cmill.html .
I’d say a CNC is the next natural progression for a maker-enabler ;/ CAM company. Unfortunately, the market is pretty saturated with kits and even open source designs.
The ideas of “enhancement add-ons” is interesting also. I wouldn’t mind a kit for the X-axis openbeam and offically supported capacitance auto-leveling. Of course both these diverge from the original TAZ design.
I’d like to see a CNC Plasma table in the $4-6k range I’m considering building one soon anyway but that would be nice.
I submit that the next step in plastic extrusion 3-D printing should be a small hot-end that uses pellets. Plastic extrusion industry runs on myriad varieties of pelletized materials. Just imagine what we could build with all the those cool types of polymers out there!
Lulzbot…build us a Pellet Extruder!!!
Tim in D
(I can help…I own a machine shop)
Well, since AO had officially announced a few years back (one source is Hackaday, saw it somewhere else too) that they would be planning on selling a version of a Lyman Filament Extruder i vote for them to continue development on that. I dont know if there is any interest in that anymore or if it’s really all that economical. But AO said they would make one, so they should keep their word.
Personally i think you should experiment in the CNC mill / lazer engraver market. Sure it’s crowded, but you already have some of the expertise with such a device already. Not sure what other hardware based things you should get into, but there is definitely room for expansion in several areas. Perhaps partner even more with SparkFun electronics where they make your electronics (like your own Rambo boards) and you make hardware that they want. Better and cheaper 3D scanners perhaps. Perhaps also experiment with larger and other types of 3d printers (like resin).
from a open source software perspective someone pumping huge funding and/or hiring lots of programmers to develop open source CAD is desperately needed. The only viable and user friendly CAD in my opinion is Solidworks. But it is expensive, not open, and tied to a non-open operating system (not cross platform). Despite this it is still my software of choice even though i am a native linux user, and even AO has used it extensively to design the lulzbot mini. Freecad has potential, but it sucks. It’s interface is clunky and does not have an intuitive interface like solidworks. Blender is beautiful 3d software, but it’s way to complex and is not designed with CAD use in mind. If someone could take the beauty and elegance of blender, mix in a little bit of freecad and then top it off with the best features of Solidworks we would have a winner.
With the recent news that RepRapPro is closing down what do you think about Aleph Objects offering the RepRap Fisher Delta printer? I think that would be awesome, especially because i want one to play with. It seems so new that i haven’t seen anyone in the U.S. offering any.
Have you guys contemplated making an actual opensource replacement for thingiverse? It would be a massive investment in infrastructure and knowledge, with an uncertain payoff, but I think the community would trust an open source company not to abuse their models more than certain other ones. You would have to commit to doing it right though. If it isn’t as polished as thingiverse and as easy to use at launch, it won’t gain market share.
I’m not sure if it would monitize well except as a product advertisement page, but it would offer a huge amount of visibility and give you a public perception of being as large as makerbot in the eye of the general public. None of the other alternatives out there really get the model down right. There are either too many steps, or the pages are slow or glitchy. Mind you thingiverse has huge room for improvement itself, but even with their licensing sillyness, they remain the most usable option.
A website with a drag and drop part upload interface and better support for model instruction formatting that didn’t crash and had a useable search feature that actually returns the items you are searching for the first time when you search for an exact item by name that had a rock solid garuntee of continued opensource content and no moves to grab control of uploaded parts would eventually take over. I just don’t know if the intangable benifit would be worth the capital to make it work. It is for stratsys and makerbot, but it would be hard to show an actual correlation on profit before and after it was implemented due to the whole history of how that came about.
Hmm, that wouldn’t be too hard to implement in a smaller form. I am intrigued about the Z axis though. does a plasma cutter need a Z axis? The lower frame would almost have to be metal. I’m guessing the temperatures around a plasma cutter nozzle would vaporize ABS pretty quickly
Real-time control of torch height helps maintain cut quality and efficiency, analogous to precisely focusing a laser. Required travel range is small though.
The USB link needs to be better.
There is no reason a little bit of noise on the line needs to result in a dead printer. It should just ask for a resend of the last packet. Resulting in a little slower transmission speeds.
And this one is more important to machining (mills and lathes etc.) then 3D printers.
Much more real time feed hold, feed rate override, and single step.
When you push feed hole the machine should stop it should not finish the current block before stopping.
What market will you serve?
“industrial fabrication / high training threshold”
“entry to mid level / easy to use”
The mid-level moderate learning curve market Lulzbot serves currently will get eaten by the less expensive, easy to use market.
Building in Biolumo and piercet’s comments on software-
If you want to serve the “entry to mid level / easy to use”, build a better software / workflow stack:
File library alternative to thingiverse (make an “Import My files” tool )
Modeling software (we based is pretty great… who want’s to download/ install?? will it work on my OS??) Tinkercad is great for a broad range of skills.
Slicing / management / TROUBLESHOOTING software (My print failed, what do I do now??)
Make a web enabled print controller so the printer can be operated from the web tools
Integrate the printer mechanisms with the software to help users troubleshoot the issues
Make the software in a manner that allows installation on a LAN so the whole system can run without an internet connection, if desired. (Tinkercad down for maintenance stinks!)
A competitor WILL make this improved software / hardware stack. Make it open source so all cnc fabrication devices can potentially benefit.
How about an open source “hardware operating system”? Is http://www.ros.org/ enough of a starting point?
Host a Cura hackathon…
I can’t speak for everyone, but I very much prefer downloading and installing software locally. Web apps are limited by their nature and lack of direct hardware access. They also break for every user if the server goes down. Even on recent hardware, tinkercad is stupidly slow. When a slower machine running 123D works perfectly fine. I need a quad i7 to use tinkercad without pulling my hair out. A core2duo is enough to use 123D for basic design work. I don’t have any in between hardware, so I’m sure the low sample size skews my results some, but it’s a reasonable point. On the design front, improving FreeCAD is an interesting idea and helps an existing open source project.
Many of the other suggestions are interesting. I personally have never had issues with the USB transfers, even for longish prints. I don’t think I’ve ever gone more than 12 hours though. I do think error correction is a good idea anyway. Data transfer issues shouldn’t be able to kill the job if the host is still running. Just resend the bad data.
The original question seems to be related to hardware… I’m not sure what I would like to see there as I’m somewhat new to the world of 3D printing. I don’t see much point in another filament extruder. I own a filastruder and it’s great, but has some limitations inherent in the process. I think that market is served well by them. If it were possible to do a reasonably priced grinder to recycle failed prints, that would be excellent. However, that’s a tough nut to crack. Same thing with a pellet based print head. It’s been tried, and there are a number of issues.
I can think of a few upgrades that might be good for existing products.
Filament width and feed rate sensor. Automatically compensating for diameter changes and detecting stripped filament would be nice. Even better if doing so would allow one to fix the filament and resume the print.
Removable print surfaces, if they can be done well. I’m playing with a zebra plate and it’s working quite well. If this were combined with a more accurate auto leveling sensor that could compensate for the more irregular surface better, it would be great.
There seemed to be some interest at AO for scanning. Is that no longer a goal? To be fair, it seems to more of a software problem at this point.