Tom, great to finally hear the flip-side to the 3d printing industry. Thank you for keeping the conversation Realistic, most TH-cam personalities won't swim these waters. Happy Extruding....
Especially closed loop control. As soon as we figure out a cheap reliable way of autotuning PID loops with cheap encoders, brushless (or even cheaper brushed) motors can be used instead of the decently expensive steppers we currently need. Even if we still use the NEMA17 steppers, closed loop encoders would be nice to verify accuracy of microsteps, change acceleration settings on the fly based on gcode geometry response, and detect faults.
Tom, I think this is my favorite of yours so far. I teach a class on building a 3D printer and this topic is something I stress each class. You hit the nail on the head. Great job!
Congrats on the move, but I hope you still plan to have a hangprinter room! Besides that, I am currently going through what you did with that first printer as I am currently upgrading my school's BitsFromBytes Rapman as the old board and proprietary gcode rendered it useless for printing with any sort of currently relevant slicing software. I had to get a new board and nozzle, so due to my limited allowance from the department, I used a RAMPs and an E3D Lite6. I then needed to buy a new block and nozzle as my teacher managed to shear the nozzle in half after trying to disassemble it after a minor plastic leak. Thankfully I now have it working almost perfectly except for the cheap dull feeder gear which I intend to replace, and the lack of part cooling fan, but apart from that, it is nearly there!
There is actually a printer manufacturer that is trying to make a hassle less 3D printer. You should check out Nectar 3D. They are based in the Netherlands so depending on how close you live to the border it shouldn't be too difficult to visit them! ;)
90% of what I have printed since having my new 3D printer has been design by me. Without a small engineering and 3D design background I would be annoyed with my purchase. Your video is spot on.
the problem is simple. Everyone know how to type because its an essential knowledge in society. But not everyone know how to sculpt, because its not essential knowledge in society. since typing is essential, then 2D printer is essential. and since sculpting is not essential knowledge, then 3D printer is not essential. 3D printer will only become essential tool for consumer market if the skill for sculpting/3D modeling become essential. lets say, if in the future, everything is digitized meaning we dont have to go to store to buy things, we can just download the items and 3D print it, then yes. 3D printer is become consumer market.
and it won't be a 3d printer really, it would be some atomizer that can make things out of random things you don't want. you can download a new shirt, throw in some old clothes that are ripped or don't fit and press "Make". shits going get real in 100-200 years. then you can just make everything.
but it is still typing though. lol. and i also think that, 2D printer is essential because it is basically an information carrier. you print texts, images, graph, news, contracts, etc. its an information carrier. since information is what society is driven of, then its essential. 3D printing for now is just for prototyping and novelty. there is no information you can carry in that thing. the market is very niche.
I'm not sure about that "buying it in a store instead" idea. If I print enough stuff, I offset the cost of a 3D printer. For example, 1 spool of filament might print a dozen or even 50+ of a specific item before running out. For example, something as simple as a soap holder (customized to fit a specific area) might cost you $5 at the store. Buying 4 of them would cost you the same as a roll of filament. So you could very much print 4 items and the filament already breaks even in paying for itself. Another example: A stringed instrument capo. You can find them from $5 to $20 online. Or you could print one. Now of course if you get a $1000+ printer, it would be harder to break even on the cost, but a few dozen prints on a $200-$400 printer can offset the cost of the 3D printer and filament, if you would have otherwise bought those items in a store. There are things you can buy in a store, but sometimes even small modifications to a self printed item can make a big difference in its usability, or make life easier. For example, I have a tiny bathroom sink, split faucets for hot and cold. I can't even use the edges of the sink to store anything. But I can print something that will fit it to give me room to store things or to mount things. Or print something to combine the hot/cold into one stream. Small things add up. a few dollars here, a few dollars there, can turn into hundreds of dollars over time. Sometimes things may be cheap, but soo limited in their usage you might not bother to buy it or to even look for it in a store, its use might even be soo trivial a store might not even carry it. 3D printing can allow you to have that thing at a whim without much expense or effort. While it helps to be able to use programs to make your own models, I think there's a lot of models out there for people who want to print stuff for their home, garden, etc. Pricing is shooting up these days even on relatively cheap things. You can't buy much for a dollar anymore. But if you can print 20 of something with filament, you've essentially paid $1 to print each item (not including the cost of the printer).
Here's a list of features off the top of my head that I would like to see as a standard in consumer 3d printers (I know it would be expensive, so for now it's just a dream) - All the things Tom already mentioned in this video - Rpm control of fans like in PCs (the part cooling fan in my Prusa doesn't start spinning below 14%, and yes, I ruined some prints until I found that out) - Ambient humidity sensor - Absolute x/y/z-axis mesurement and control like in cnc machines - Precise and dynamic flow control of the extruder without having to tinker with the extrusion miltiplier and/or checking the exact filament diameter - Heatbeds that only need one initial calibration and don't flex/warp on a wide temperature range
Thank you for the excellent and much needed long-term perspective on consumer-grade 3D printing. For now I would be happy to have a self-leveling bed, a filament feeder to make loading and unloading easy and a reliable long term build platform that actually works with ABS.
you're so right !!! Friends that bought a 3Dprinter in the same time than me ... techs continue to use it and involve ... normal user ... let it in the dust somewhere when the fun stops or when they had the first "clog" ! Others .. like me ... owns 4-5 printers now ..
@9:50 These are great great points that should be tackled. I wonder if one could make the inside of a printer a single color, and have something take a webcam image, and compare it to something like the build image and use that to compare, and if there's too much of the base color, then pause, home everything and continue, or just pause/kill the print and set off an alert....
One easy thing is to have an easy interface to add your name to pre-selected items, in set areas. People with the cash will be willing to drop the $200+ just to have the opportunity to personalize their trinkets.
Adding a sensor to the back of any stepper to detect the the rotational shift of the shaft and monitoring that to detect if the axis motor is where it should be is not hard to do. Look at the mechanism of the old ball type mouse - it detects small movements of the ball (albeit with a highly inaccurate friction coupling to the ball) and a couple of LED's and light sensors gets the job done and that is considered "old tech" today. Take an ordinary car windscreen wiper motor for example - yes it is big BUT a small modification to the brush assembly (one is permanently grounded which has to be removed from earth and wired out) and you have a motor that CANNOT be "forced" to turn if it does not want to due to the worm reduction gearbox and you need the mechanism similar to that from an old school mouse (or a modern lazer type) to detect rotational motion and direction and you have replaced the "normal" stepper with something that will be able to "self correct" (in conjunction with software of course) which today's stepper motors cannot. And they are WAY more powerful !! In batches of thousands they can be produced far more cost effectively than stepper motors which have SO MANY coils and other stuff - this motor has 1 large armature with 12 or 14 coils - thick wire and a pair of magnets in a casing with a gear set and containing a commentator and a brush assembly. Controlling it is simple PWM and rotational sensor reading - in effect it is a large powerful servo motor. Have you ever tried to stop a wiper motor - you injur yourself if you are not careful !!! Control circuitry (the high power side) is a single h-bridge and power mos-fets or whatever, not some nonsense about quarter or eighth step manipulation which is often not a reliable way to get positional accuracy. The manufacturer who comes up with a robust, "agricultural" (strong) design in a 50x50x100mm package will make a killing because it will be a better, more robust, more cost effective and stronger method of driving printers, mills, even robots. So what if the drive shaft comes out at 90 degrees to the 100mm axis of the package, designers will design around it - that is what creative designers do ! If you cannot "stand it up vertically" due to clearance constraints, lay it over 90 degrees on its side. SO WHAT? Most think that conventional steppers are the way to go. Some years back most people thought that the earth was flat. One group has already been proven wrong and the other soon will be when people realize that the CNC universe does not revolve around the stepper motor - the stepper is just the "cool" thing to use - there are more cost effective and efficient ways of moving a tool or a bed around.
I recently discovered that the Eris is a good place to start for beginners. Sure it is not cheap compared to a bunch of low end 300 dollar machines but so far it has been 100 percent hassle free from the unboxing to the first print. Only hard part was inserting the filament and installing the software because I never used Matter control but after that all great prints. The only thing that I can not speak to is the reliability of the machine over time or the maintenance involved because it is too early to tell. As far as the rest it is a very smart printer it auto levels and if anything the prints stick almost too well to the build plate even with cheap filament. It is also portable which means you can take it to school or other places for demos which is a plus and the quality of the prints are quite impressive for such a small printer.
I'm not acctually a consumer but rather a child with an interest for tinkering. I ordered the tevo tarantula and most likely I will print a few cool things then get bored and realiez I don't need it but I believe that a 3D printer build kit for 200 dollars is much better than a 300 dollar summer camp. It's cheaper, much more fun and you get a very cool finished product that can acctually be used.
...would it make sense to put shaft encoders with every stepper motor? I definitely think it is a good idea for the filament path, so we don't need to tune extruder steps/mm. It would also detect jams and the like. Also, automatic print removal from the bed would be a neat feature. I've found that most parts come off easily from my glass printing surface once it has cooled down, but they stick nicely while it is hot. But they only stick in the first place because I use hair spray... I'm excited for the future if this is the way it's going, in any case!
Great analysis, Tom. When you put it in perspective with 2D printers... Yeah, there's no disputing there's a path yet to track to get 3D printers there. Thanks and congrats!
I think the consumer market is not a good place to aim for right now. The market isn't big enough, so it's hard to make parts cheap enough. A sub-$400 printer with all the sensors needed for reliable operation would require hundreds of thousands of machines to be made in a batch. 2D printers are heavily engineered and optimized pieces of equipment, and if you tried to make one in the unit count of 3d printers (a few thousand max), they would not be less than $400. Engineering time, making molds/tooling, getting pcbs manufactured. It's all scales dramatically with unit count.
@Thomas Sanladerer I hope your getting settled in. Thank you for the lecture, I found it informative. I've been following 3D printing for about 5 years, but I have yet to commit to a printer. My problems/concerns are price, print quality and technology and the vicious cycle that currently exists between these three categories. I just haven't found "the" printer to get my feet wet. I'm currently looking at the Creality CR-10. It certainly covers price and print quality and to date I've had a hard time finding reasons NOT to buy it. But advances are happening so quickly and prices continues to come down while print quality goes up. Other printers I've been looking at are the Wanhao models from Monoprice V2 Plus or Ultimate. For me as a hobbyist/maker print quality is important. SLA fits this bill, but the high cost of ownership (up front and operating costs) plus the mess of SLA keeps me from going that route. Now we have SLS approaching "affordable" for serious hobbyists/makers. It's maddening which also fed into some of my frustrations the other day. Anyway I continue to watch and learn from you sir. Thank you again.
Good points Tom, and great talk. Each of your points is perfectly achievable, but each one drives up costs. Autoleveling is getting there, position/feedback sensors can be added, but we need better firmware to handle the feedback, and ambient temps, etc. This would most likely appear in the next wave, of 32bit controllers. Software is a sticky wicket, you want something that nearly anyone can pick up, but experts will soon want more. I think 2 key points would be better use of developer time - 1) compatible 3d source files, so one can take the beginner tinkercad file and load it into FreeCAD or a high end CAD for better refinement or whatever. and 2) BETTER libraries, thingiverse is a drag, it slow, its clunky, of the 100 things I downloaded only things that are knick-naks work, things that are supposed to FIT together, are always wonky, and don't fit. I got a 3d printer when the prices for a CBOP (chinese box of parts) got below $200. Price gets people in, a consumer level friendly box will be more expensive. Nice house. Now get to work !
I am actually working on the "next step" in 3D printing...but can't disclose any details...still in the very early stages. Edit: maybe that was too vague. I'm challenging the existing workflow and making hardware also
I waited until this year to buy one because I hate fixing things. I think now is the perfect time if you've been eying them. I bought a Monoprice Select Mini, and while I've had minor hiccups, I've not had to replace or fix anything. The only shortcoming I see is that it doesn't print metal, and when they have an affordable desktop that can do that.. I just can't imagine.
Huge congrats on your new place Tom. Looking forward to seeing videos from there. Fantastic insight, I too think manufactures should focus more on quality than being cheap. I will gladly pay a bit more for something that has quality.
Excellent analysis Thomas. I think it is useful to compare the emergence of domestic 3D FDM printing with that of the domestic sewing machine. Both technologies are essentially very simple, yet to really get value from them you need to have a good grasp of 3D modelling, whether that be in the form of paper patterns or CAD. Consider the success of Burda Style and Butterick.Furthermore, although a sewing machine can easily be used to re-attach a single button (is it really worth it?) advanced users need a deep understanding of how different fabrics, needles and threads interact and when they should be used. I think a fundamental difference between the two technologies is the almost immediate feedback that a sewing machine operator receives as they work the materials. A sewing machine is still just a complex hand tool. Automation of "computerised" domestic sewing machines is exemplified by elaborate embroidery, but would we ever see a sewing machine that you load up and walk away from and then come back to find a made-to-measure shirt?
so why does an extruder motor skip on a specific part of the print every time ? mine is doing this , freash clean nozzel so idk , the print is still pretty ok though plz fill me in on this
I've wondered as well. When will filament sensors be standard for anti spaghetti or like you suggest temperature. It's fun to explain to folks how the extruder works and it doesn't care if there's anything underneath it. It just assumes the last layer is there. Some just give me a funny look?
Paint the ceilings white. I did and it looks very nice and unique. You might even go for a white wash look. Beats all that work and expense. I can relate. I too built my first machine back in 2010 Prusa MK1.
Yes. You HAVE TO use a good primer as the first coat. If you don't use a primer, the paint will peel off. Wait a couple of hours and then paint with a good latex paint and roller. If you use a good paint, you'll need one coat if you want your ceiling to be white (recommended), since the primer is usually white. If you use any colors besides white, you'll need 2 coats or more. If you use a cheap paint, you'll need 2 or more coats. Cheap paints are not labor friendly.
hey Thomas, I just finished my very own 3D printer, no kit tho I build it all myself from scratch up. It works better than I expected but there is some work to do because the prints aren't that nice as a delta printer or as a prusa, do you have some tips to help stick the layers better together?
I've noticed a flood onto the forums of people who don't know their new machines, haven't taken the time to do even basic research and familiarisation with the concepts and just seem to have picked up a printer on a whim. I think it's too early for the larger consumer market by far.
Good points Thomas. I have a lot of experience in AutoCAD starting on an IBM 286, and manufacturing using CNC lasers, turret presses, Mills and Lathes. Decades of experience. I've found the design process the most difficult part to 3D printing. Fusion 360 is great, somewhat flawed, but without the mindset that comes from conceptualizing and producing real parts, I would have been lost. 40 years of experience helps, but the process is still not easy, and will frustrate the average newby user. I have a Rostock Max V3 that I built and it is chugging along with only occasional calibration needed. After designing and printing over 90 custom things, the learning curve is still there but not that difficult to master.
I couldn't have said it better. You should see the 3D Printing group on facebook. Its a train wreck. All day its just "why doesnt it work" "what am i doing wrong" "my printer sucks", and then theres the idiots saying "my 150$ Anet owns 2000$ printers"... I cant even anymore. And most of the guys that have a at least one working settings sheet, think they now know better than anyone else, and will argue it to the end even if you present data to why theyre wrong, then they throw a tantrum.... Thanks a lot mainstream media outlets....
Well, in my job I work with state of the art Stratasys FDM 3D printer, and in my home I have an Anet A2 (among other printers) and in terms of quality there is almost no difference (after upgrading some not electronics parts), so the people claims are not so far from reality.
Simon Silva Lemarchand for basic materials I agree, pla especially but I know that on my anet due to its lack of an all metal hot end there are many materials that I can't print, also it's build volume is much smaller then most higher priced printers
Toby Cowles but that is an easy upgrade to do, and also relativitly cheap. In my case, I have all my printers with all metal hot ends, but that is not the most important thing when you want to print a complex material, I think a hotbed is the most crucial item for that kind of task. Anyways, even if you have a state of the artículo 3D printers, There's always something that you are unable to do with it.
What about those that know they are no 3D engineering god of l33t destruction pwnz0r, and need a 3D printer for making many examples, for cheaper than commissioning companies for a similar bulk (costs more)? Just wondering :)
Hi Tom. I heard that if you grit the surface of the glass bed with 90 grit sand paper using a bit water (wet sanding in diagonal direction, only one direction), nothing will lift off from the surface during printing, even Nylon. Is that true?
I tried it and it works with ABS + ABS slurry. Very cool. God..my parts are so straight. It doesn't work with Nylon + abs slurry. For Nylon I need to find slurry based on Nylon, but nylon is chemically resistant. Just few chemicals can dissolve it. Actually I posted it on facebook Anet A8 forum and the link here , but you guys erased it for some reason. Anyway the video is still there on the forum.
Honestly? We are living in 21st century with amazing access to any kind of knowledge. If someone doesn't have any background knowledge on 3d printers, 3d modeling, CAD software, why shouldn't they learn this stuff? You have google, forums, youtube channels like Makers Muse, 3d Printing nerd, and ofc Tom's Channel. You can easily learn stuff without getting out of your home. There are plenty of tutorials, guides, reviews etc. and if someone wants to start 3d printing, why shouldn't they? In my opinion, if someone is so lazy that he thinks he can buy 3d printer and press 1 button to print everything he wants without any reading(or watching in this case) it's his/her business, why bother? In 21st century, with such access to data, lack of knowledge is a choice.
Would love to see you make a video on how to make your printer "smarter". I know filament runout sensors are easy to implement. What else could be done?
Tom, good video! My thought on this topic - to many adjusting options in the software (slicer) and mechanical flaws of 3D printers (vibrations etc. ) give a lot of room for error. The printing results are also affected by the bazillion options of available filaments. I do not really think that one company will be able to solve this by bringing a consumer friendly solution to the market, since it requires custom software + very accurate mechanics which results in a very high cost of the printer. You can clearly see this with today’s printers that print very good. Considering the above, machine learning (AI) could be the way to go. In my opinion, the best solution would be to include as many sensors as possible in any type of printer and collect all the printing data in a cloud. This would give enough data points for a cloud based AI to remove the errors and adjust your print settings while printing. Basically, if you are connecting you printer to the AI, the AI provides adjustments to your print settings based on the information it knows and your printing results. I assume you and other smart people from the 3D printing world could definitely agree on the minimum data points required that should be collected. I see it more like an open software project - anyone could (manufacturer or DIY) put needed sensors on their printers and connect to the cloud based AI. I think that such a system could make any printer (with bad mechanics, made out of wood etc.) to perform very well, since the printing could be adjusted in real time and considering the mechanical flaws of the system (mechanical issues have always a pattern so the AI should be able to counteract). As a result, you could make a printer out of Lego and print very, very high quality prints and the software cost would be 0 :)
I agree… I have been complaining about this from the beginning. Direct solid modeling with a nice interface and simple workflow is the way to go if we hope to advance 3D printing past the realm of the user-nerds with any kind of market staying power. Procedural and parametric or history based CADs will never be the way for consumers (and are even clunky for prototyping on the fly or modifying existing designs), and 3d sculpting/animation programs like blender (which I also use occasionally) or ZBrush (while great for art/figured/etc) are not at all user friendly, and just plain not any good for technical CAD. Tinkercad is too simplistic, but Autodesk was onto something with 123D Design - it just needed a few more features and a little shove toward 3D printing and they would have had a perfect middle-ground consumer product. Unfortunately they killed it off and rolled the features into a crashy, clunky, overcomplicated, swiss-army-knife, cloud-teathered abomination that I begrudgingly use since it is unfortunately among the best (if not THE best) comprehensive options out there. I also think that the next HUGE step in hardware is closed feedback loops and modern processing. Mesh leveling and multi-extrusion are great and all, but we have to patch up and modernize the foundation before we add more floors to the house. When I first started poking into 3D printing, my mind was blown that 99% of the products out there were open-loop systems with zero error correction, running on 8-bit boards. The Reprap project was a springboard, but it became a platform because it was just too easy to take their work and roll it into a good-enough system. If we want consumer products… take a page from the consumer 2D printer books and use cheap, lightweight, DC motors with encoders, and cheap boards with enough processing grunt to run floating point operations, realtime feedback data, and a decent display with a thought-out UI. I pull apart dozens of trashed printers for parts, and I have not seen an open-loop, stepper-controlled printer in a good few years now.
I been messing around with 123Design for a while. What other program is close to it, but have more features? I dont like online cad. I want a program that I could install on my computer.
+TheDarkstar826 any "normal" CAD is a logical next step from 123D Design. If you're looking for free options, that would be Fusion360 (cloud-based), Onshape (web only) or FreeCAD (total pain to use)
Hi Tom, thanks for the video. This is somewhat relevant to me as in the summer I'll be working in an arts summer camp in the US teaching 9-14 year-olds 3D modelling and printing using Makerbots. I was wondering what kind of software you think I should use to teach them modelling. TinkerCAD, 123Design? They are all assumed to have no prior knowledge. I have a Vive and was hoping to do some VR clay modelling, but aside from that are true any others you might recommend? Thanks!
Hi Tom, I have Anet A8 3d printer. I like it. I just think it's kinda small. Is it possible using Anet's electronic main board and screen to build larger 3d printer, let say , with the printing bed size 500mm x 500mm or even larger. I have master’s degree in mechanical engineering and I have a mini machine shop at garage with machines and tools. I can model anything using SolidWorks. I just need a bigger printer to print my Big ideas :) I can make a frame from aluminum extrusion and I can find any polished shafts with bearings and screws with a bit bigger step motors. Some parts I can print with my Anet. Can I alter the extruder and put larger nozzle to speed up the printing? I mean with larger print size I will need to extrude more plastic in a time, so I will need higher printing speeds Is it going to work? Again, will Anet's A8 main board support that size bed? I know you are the first guy who can have some idea about this.
CoolStreamT Most likely You can keep all electronic components, you just need to change the dimensions in your software; easier with some boards than others but usually relatively simple with lots of information online. Also just buy a larger size extruded, also then change settings. As a side note this is a good example of Tom's point. Statistically, not personally, an engineer is of average to above average IQ and if an engineer is unsure how to modify their 3D printer it shows how far we have to go before this technology becomes common place in our homes for us average to below humans. This is not a criticism, it is just a very good example of why 3D printers are looked at as novelty items by a large portion of society.
For bigger size bed, I need bigger step motors to move more heavier beds and extruders. So bigger motors need more power ( different voltage). Can I plug these bigger step motors into the Anet A8 board and start printing? Could you tell me what printer brand board I can use to do that for sure? If I challenge you, maybe Tom can answer the question.
Hello Tom, I can't find anyone who owns a Zeus 3D printer to get help from... My icon for Zeus_UI is there but it won't load the application to launch the program so I am unable to print. please help if you can. Thanks Kells
These are some fantastic points and used a great analogy with the 2D desktop Printer. I honestly don't think that a 3D Printer in every home is really something that will be seen anytime soon. Possibly whenever a future product comes along that allows people to make practical objects in a sub 10 minute time frame. I believe that once it gets to this point, we will have a TON of companies that are charging for replacement parts that you can just print out at home. I fear for this day, but I can see this becoming a trend IF we have a printer with a 98% success rate in every home.
Microsoft 3D builder is really quite nice as a basic tool I think. Combine this with a simple "ready to go" compact enclosed printer like a daVinci jr and I think you are in better shape for a beginner. Yes, I know it makes consumables more expensive, and is not open source, but also more usable - my DiVinci rarely has a failed print, and I always know how much filament remains and if a print will print. Unlike my new "mechatronic opensource super flexible lego kit" - which is super fun but as reliable as an Italian sports car. It has parts made in the daVinci because I know they be durable and dimensionally correct. Just some thoughts based on six months of learning curve so far :)
I've been saying for years it won't catch on until you 'dumb it down' to the level of using an iPad while making it as reliable as one. I believe we will eventually get there just like we did with computers which were in the same state of disarray in the 80's (maybe even worse since communication was so limited) and the Internet was in a similar state of disarray (But not nearly as much due to immediately implementing open standards) in the early to mid-90s. My first introduction to the Internet involved compiling a TCP/IP stack because windows 3.1 never had one, using a 28K modem (56K hadn't come out although 33.8k were but were expensive for an extra few k of bandwidth) and you phoned into an ISP and used a Unix shell account to actually hook into the Internet via the command line (the horror) and the only browsers available were Mosaic and an early beta version of Netscape. My first website was coded by hand in Notepad and was 5 pages with mostly text and a few graphics but man was I proud of it. And all my friends and family said it was too complex and would never catch on .... but 10 years later every single one of them had it
What about a company sends a file which has a license for 1 part to print then it doesn't work and they charge you for it. Hope I made sense. Great video Thomas
Great Speech!!! I totally agree! A lot of people say things like: "people have to be patient" or "you can't just instant be a pro-inventor/designer/maker" or "There's always a learning curve" or "It's not just for everyone" Well that's the exact problem! Something that IS "for everyone": - is intuitive - knows what a consumer next choice/step is - fixes problems/errors silently in the background (doesn't bother consumers with problems the answer could be guessed or logically made) - works almost out of the box for everyone! - etc etc.. That's what a 3d printer is NOT! Great comparison with MAC computer, that's basically just a Linux machine fully (fucked up) and packed in a more manageable jacket that satisfies things @Thomas Sanladerer sad or the few things i just listed! Those type of "skills" do sound like magic on this exact moment considering 3d printing and manufacturing/creating but trust me!!.... There was a time long long ago, tech-nurds also tough the same about computers. Well look at computers (and such) right now! Maybe the real (rooky/not-expert) consumer only scratches the survive of what a computer is really capable of and happens behind the scenes or even know of all the hundreds/thousands events taking place by just a simple click! (for example: going thrue the OSI-layers down to zeros and ones, trust me that's a long long way of instructions!!!!)
so true, I agrees, though it's not going to be easy to make a consumer grade 3D printer that actually is user friendly, but seeing as it is, it's going to be a while to get there, always great videos, I enjoyed all of your videos...keep it up!
Delta printers just need either a very accurate distance sensor for each carriage on every axis, or a gaming mice sensor with strong LED scanning movement on a belt on each axis to detect irregular movement. Semi-Closed loop is really nothing special. Just the right pick of smart components. Augmented reality technologies from the recent Google I/O could also play a part in detecting errors with machine learning algorithms. This means just stick in an ARM smartphone chip with camera already just like self-driving cars. Simple implementation would be a depth sensor like the Tango Project and new Asus AR phone which helps the machine learning algorithms monitor the 3D space of the print bed at every minute. Seriously we should all learn from past history from how so many developers underestimate capabilities of smartphone cameras that they don't even realize that an app can measure a humans' heartbeat when coupled with the right LED light.
Hey Tom! What are your thoughts about Microsoft's Paint 3D? Everyone know how to use the good old paint, so do you think that it's 3D counterpart will kick off 3D Printing?
I wonder if there is a solution in the foreseeable future. There's a lot of things that have to go right for a 3D part to come out right. We don't see this same kind of push, at least to the same degree and fervor with other CNC-type machines. There's a lot of complexity going on, so many things have to go right. Even if you do all the right things there's something that can be a roadblock.
The more closely I look at 3D printer parts on eBay, the more worried I become. There are ratings issues, protection issues, clearance issues, layout issues, interference issues, and the list goes on. We are flooded with cheap rubbish electronics that should come in a swan vesta box and that require serious rework to make safe.
After watching a lot of youtube and gaining a better understanding of 3d printers it would seem to me that the 'Leapfrog Bolt' printer ticks all the requirements of a reliable quality printer, reflected in its price. Would like your opinion of this product.
I don't expect any manufacturer to take the lead on this, they aren't willing to share their improvements. I expect this to be the result of community development. The auto-bed leveling/detection is a good example, The community developed this feature as a free add-on and now it's a standard feature in all new designs, not because the manufacturer is doing anything, but because the free software they base things off of includes it by default. As far as additional sensors go, It seems that we are running out of I/O pins on the default arduino boards taht are the standard basis for things. Until we make a jump to something with more I/O and more processing available, I think we are at a plateau. However, with things like the ESP-32 chip hitting good availability, I can see this breaking free fairly soon. A simple microswitch in the filament path to detect that there is no filament there would be a trivial thing to add. We are seeing a handful of closed-loop designs start to show up in the CNC world, and I would expect to see some of this start to show up in 3d printers once things stabilize a bit. One problem is getting people to back away from the "because I can" and get to "it's good enough", the o=drive project for example can use hobby brushless DC motors and shaft encoders to make _really_ fast machines that know exactly where they are, but it's designed to drive motors at 100A @24v, and as a result is fairly pricy. We need to get a closed-loop equivalent of a ramps board out there, something that's just good enough (with flexibility and upgradability enough to be used in many systems) and let the far east clone manufacturers start mass producing them to drive the cost down. This is going to be a bad thing for whoever tries to make money selling such a board, but a great thing for the community/industry. This is another reason why I think we're at a plateau until the community makes something 'good enough' to get everyone to catch on.
Sensors? SENSORS? .... Filament sensors are know to work until the cheapo plastic the cover of the microswitch erodes (abs/pla can do that with time and pressure) .... and if they are integrated on the extruder it depends on the program . Usually it means "Welp, print stops right whenever this is 0". What if it malfunction? No dev ever leaves a bypass option for those more daring or with a dying printer awaiting replacement. But closing loops and setting more sensors is a nice track. Just do your FMEA's , proper development and have a plan when everything fails. Because of time, manufacturing or simply the user mishandling your machine IT WILL FAIL.
Didn't come here for the blog at the start. If you want to do that make it separate and clearly marked, or at the end of the video so we don't have to search for the real start of the video.
Totally agreed. Asides from smarter hardware, the entire software tool chain of 3d printing still need to be a lot smarter. I think slicers are mostly too dumb. there are a lot of things that you can do to make a slicer smarter. Cura 2.x is heading that direction but there are a lot more that can be done.
Mega Making software development might take less time than hardware (but debugs forever) probably needs to go simultaeniously to work out the kinks, in any case I'd do it...but then again who'd pay me to do it? so open hardware/software might not be the way to go
I am an university researcher, and I would think that research into closed loop control for open-source machines should be taking place at universities. To that end we have actually started a post graduate research group focusing solely on additive manufacturing and specifically for research on desktop 3d printers (non professional). I believe this is where making will be moved into the mainstream consumer market, could not agree more with you on the current barriers to entry!
I plan on getting a 3D printer because I would love to be able to learn about it how it work everything and I'd also love to be able to design my own stuff too
I revoke my last comment about steps. You are right in your cal video. Use the prusa calculator. I also agree with this one but less emphasis on creating your own models and more on making the printer smarter. Even just as simple as the software and hardware working together better. In opinion the printers wont take off and be consumer grade until something equivalent of the computer able to watch the print being printed and make its own adjustments. (Cameras and recognition software). Skipping sensors. Etc etc.
Hi Thomas! Your video series is excellent and is very informative. I was wondering what your thoughts on the 3DprintClean Air Scrubbers/Enclosures to reduce the amount of Ultrafine Particles (UFPs) released into the air with burning plastic and the long-term health consequences. I have not found a competitor for their product and wonder why not everyone is using enclosures for their printers (not only for ABS smelly fumes)Thanks!
+Infosec Val it's a touchy subject, both as to how you can even remove UFP and whether the concentrations 3DP produces are actually harmful in every case. I'll leave th arguments to scientific studies, as I do not have the tools or expertise to judge the subject.
REALLY great points you're making. At present, 3D printing is a bit of a "fiddly" hobby and not ready for the masses. More reliable and "bulletproof" hardware is important, and I think progress is happening on that front, but the lack of user-friendly software is the biggest issue, IMHO. Though I will say that sites like Thingiverse will keep the lack of firsthand creation power as a less important issue. You'll have hobbyists finding lots of great pre-created models that will keep them happy for a long time. In the process, you'll create a market for people who CAN and WILL create custom 3D models for people. The same way there is still a market for professional graphic design services (shameless plug), you'll still have a need for people to make good models.
I think I am going to send you the 3d printer that I am building, tell me if it is close to what you are looking for. Shipping might be really expensive from the US, so I'll probably disassemble it into a kit that will reduce shipping costs and that you will not have to tinker with, similar to the way that Lulzbot ships their printers.
I think the best place for massive innovation is the $4k-10k prosumer/small business printer market. The only real contender in this market is Ultimaker, but their printers are not significantly better than a Prusa i3. It's all the same basic components. I'm working on some new printer designs in this price range with these goals in mind: precision, super high speed, support material, reliability. I think that's the quad-fecta that would make this price range of printer exceedingly useful.
Around 6:40 you emphasize the usability of 3D modelling but also hint at what is a more fundamental problem: a lack of design literacy. That is "design" in the sense of understanding a need and connecting it to a solution by working out all the steps in between. Imagine having this same discussion about paper printers in a world where "normal" people don't know how to write... Calligraphy is only a part of writing, regardless of how fancy your pencil is (and that's what a 2D printer ultimately is, isn't it? A really fancy pencil...). Similarly, the 3D model-print toolchain, while tremendously valuable, is only part of a design process. The solution to that moment when the initial excitement wears off is to better communicate the idea that 3D printers are a gateway to broader abilities. That gateway aspect might be the critical difference for this third class of printer (non-enthusiast, non-industrial). Designing for your own needs can involve a huge skill set and it is not encouraged by a consumption-focused culture. People have to start somewhere and a great way to learn any complex skill is to start by making copies, tweaking them more and more until you're making your own thing. In that sense, just making copies of 3D models is an essential starting point as long as there are immediately accessible ways to tinker that don't break the process. That's the critical point you make about more accessible printers: more important than 100% reliability is that they have enough sensors to provide meaningful feedback to get someone back on course if something's going pear-shaped. Printers made for improving design literacy must be built for robust "tinkerability"*. Still, like writing, it would be nice if there was a broader emphasis on the value of design skills, just like there is for reading and writing which it is no longer "normal" not have. P.S.: the other thing that makes this combination important in this use case over the other two classes of printers is that skills might be more perishable for this group. If you're only a casual, non-technical user, you might forget how one bit or the other works or how all the variables fit together. Building the toolchain for robustness means that it will just work when needed. Design literacy is served by having feedback built into all the operating steps so that as they work their way up and down the learning curve, it's never too challenging.
You touched on safety only briefly, and skipped the part of hazardous materials, fumes and particle emissions. For a true consumer level 3D Printing, it needs to be safe beyond just fire hazards. There are only a few manufactures today that have taken any steps on the emissions issues. Consumers aren't going to place their printers in the garage or open the windows is the dead of winter to avoid toxic gases put off my materials such as ABS, or clear out PM2.5 particles above national safety standard. HEPA filters, Charcoal filters and air quality monitors in enclosed temperature controlled cases should be standard. Otherwise lawsuits from adverse health effects will ensue...
Today, finding an image that you want is a search engine away. At least close to that. In a day and age where you can assuredly find several high-quality printable 3D models for basically everything, then even people who can't make their own models will want a 3D printer. It also has to do with word of mouth. The day a colleague tells people about how great a 3D printer is and how it changed his life, others will want to get one. Hearing about basic issues or "it costed $3K and I'm not using it too much" is obviously discouraging.
3D printers will never be for everyone just the same way as high end custom PCs and woodworking scroll saws will never be for everyone. The average person has no need for a 3D printer. Perhaps in the future when 3D printers can print "anything" there will be one in every home. People will pay for the files for an object rather than the object itself. 3D printers and manufacturing tools and so there only for makers and designers at the moment. Great video as always Thomas!
Servos are just brushed motors with integrated gearbox and closed loop control circuitry. At the torque requirements of 3d printers, I doubt you can find a servo cheaper than just taking your own brushed motor, gearbox, encoder, and having the already existent control board (commonly an ATmega1280) do the closed loop math.
I would guess that a company like Lulzbot would be the perfect place to improve the printers and share theyr findings with all of the open source community. The way i see it, thats really almost the only way an open source company can stay successful with theyr business, if they innovate faster than the closed source competition can. However i actually think that there are too many patented processes revolving around 3D printing to prevent the innovations made in the professional printers years ago tickling down into the broad consumer and maker space. If some old patents on the fused filament production process had not expired we would not have most of the printers we all know today...
Tom, great to finally hear the flip-side to the 3d printing industry. Thank you for keeping the conversation Realistic, most TH-cam personalities won't swim these waters. Happy Extruding....
"Happy Extruding...." lmao
Yes! Closed loop control, fault detection and correction is definitely the way forward!
Especially closed loop control. As soon as we figure out a cheap reliable way of autotuning PID loops with cheap encoders, brushless (or even cheaper brushed) motors can be used instead of the decently expensive steppers we currently need. Even if we still use the NEMA17 steppers, closed loop encoders would be nice to verify accuracy of microsteps, change acceleration settings on the fly based on gcode geometry response, and detect faults.
Tom, I think this is my favorite of yours so far. I teach a class on building a 3D printer and this topic is something I stress each class. You hit the nail on the head. Great job!
You have exactly the same colour-pallette as your new apartament. :D
You've even put around some blue stuff matching your eyes - bottles, clips, folders... :P
XD I'm dead. XD
Congrats on the move, but I hope you still plan to have a hangprinter room! Besides that, I am currently going through what you did with that first printer as I am currently upgrading my school's BitsFromBytes Rapman as the old board and proprietary gcode rendered it useless for printing with any sort of currently relevant slicing software. I had to get a new board and nozzle, so due to my limited allowance from the department, I used a RAMPs and an E3D Lite6. I then needed to buy a new block and nozzle as my teacher managed to shear the nozzle in half after trying to disassemble it after a minor plastic leak. Thankfully I now have it working almost perfectly except for the cheap dull feeder gear which I intend to replace, and the lack of part cooling fan, but apart from that, it is nearly there!
There is actually a printer manufacturer that is trying to make a hassle less 3D printer. You should check out Nectar 3D. They are based in the Netherlands so depending on how close you live to the border it shouldn't be too difficult to visit them! ;)
90% of what I have printed since having my new 3D printer has been design by me. Without a small engineering and 3D design background I would be annoyed with my purchase. Your video is spot on.
the problem is simple. Everyone know how to type because its an essential knowledge in society. But not everyone know how to sculpt, because its not essential knowledge in society. since typing is essential, then 2D printer is essential. and since sculpting is not essential knowledge, then 3D printer is not essential. 3D printer will only become essential tool for consumer market if the skill for sculpting/3D modeling become essential. lets say, if in the future, everything is digitized meaning we dont have to go to store to buy things, we can just download the items and 3D print it, then yes. 3D printer is become consumer market.
and it won't be a 3d printer really, it would be some atomizer that can make things out of random things you don't want. you can download a new shirt, throw in some old clothes that are ripped or don't fit and press "Make". shits going get real in 100-200 years. then you can just make everything.
that atomizer is technically a 3D printer
Hardly anyone knows how to type. 2 fingers isn't typing!
hardly anyone knows how to use a 2D printer either... work tech support. people are complete morons.
but it is still typing though. lol. and i also think that, 2D printer is essential because it is basically an information carrier. you print texts, images, graph, news, contracts, etc. its an information carrier. since information is what society is driven of, then its essential. 3D printing for now is just for prototyping and novelty. there is no information you can carry in that thing. the market is very niche.
I'm not sure about that "buying it in a store instead" idea. If I print enough stuff, I offset the cost of a 3D printer. For example, 1 spool of filament might print a dozen or even 50+ of a specific item before running out. For example, something as simple as a soap holder (customized to fit a specific area) might cost you $5 at the store. Buying 4 of them would cost you the same as a roll of filament. So you could very much print 4 items and the filament already breaks even in paying for itself. Another example: A stringed instrument capo. You can find them from $5 to $20 online. Or you could print one.
Now of course if you get a $1000+ printer, it would be harder to break even on the cost, but a few dozen prints on a $200-$400 printer can offset the cost of the 3D printer and filament, if you would have otherwise bought those items in a store.
There are things you can buy in a store, but sometimes even small modifications to a self printed item can make a big difference in its usability, or make life easier. For example, I have a tiny bathroom sink, split faucets for hot and cold. I can't even use the edges of the sink to store anything. But I can print something that will fit it to give me room to store things or to mount things. Or print something to combine the hot/cold into one stream.
Small things add up. a few dollars here, a few dollars there, can turn into hundreds of dollars over time. Sometimes things may be cheap, but soo limited in their usage you might not bother to buy it or to even look for it in a store, its use might even be soo trivial a store might not even carry it. 3D printing can allow you to have that thing at a whim without much expense or effort.
While it helps to be able to use programs to make your own models, I think there's a lot of models out there for people who want to print stuff for their home, garden, etc. Pricing is shooting up these days even on relatively cheap things. You can't buy much for a dollar anymore. But if you can print 20 of something with filament, you've essentially paid $1 to print each item (not including the cost of the printer).
Here's a list of features off the top of my head that I would like to see as a standard in consumer 3d printers (I know it would be expensive, so for now it's just a dream)
- All the things Tom already mentioned in this video
- Rpm control of fans like in PCs (the part cooling fan in my Prusa doesn't start spinning below 14%, and yes, I ruined some prints until I found that out)
- Ambient humidity sensor
- Absolute x/y/z-axis mesurement and control like in cnc machines
- Precise and dynamic flow control of the extruder without having to tinker with the extrusion miltiplier and/or checking the exact filament diameter
- Heatbeds that only need one initial calibration and don't flex/warp on a wide temperature range
Thank you for the excellent and much needed long-term perspective on consumer-grade 3D printing. For now I would be happy to have a self-leveling bed, a filament feeder to make loading and unloading easy and a reliable long term build platform that actually works with ABS.
you're so right !!! Friends that bought a 3Dprinter in the same time than me ... techs continue to use it and involve ... normal user ... let it in the dust somewhere when the fun stops or when they had the first "clog" ! Others .. like me ... owns 4-5 printers now ..
@9:50
These are great great points that should be tackled.
I wonder if one could make the inside of a printer a single color, and have something take a webcam image, and compare it to something like the build image and use that to compare, and if there's too much of the base color, then pause, home everything and continue, or just pause/kill the print and set off an alert....
Congratulations on the move. Huge fan of your videos. Planning to design 3d printer myself and would love your comments on it when its ready.
One easy thing is to have an easy interface to add your name to pre-selected items, in set areas. People with the cash will be willing to drop the $200+ just to have the opportunity to personalize their trinkets.
Gotta say this is one of your best videos and definately still applies in 2017.
Adding a sensor to the back of any stepper to detect the the rotational shift of the shaft and monitoring that to detect if the axis motor is where it should be is not hard to do.
Look at the mechanism of the old ball type mouse - it detects small movements of the ball (albeit with a highly inaccurate friction coupling to the ball) and a couple of LED's and light sensors gets the job done and that is considered "old tech" today.
Take an ordinary car windscreen wiper motor for example - yes it is big BUT a small modification to the brush assembly (one is permanently grounded which has to be removed from earth and wired out) and you have a motor that CANNOT be "forced" to turn if it does not want to due to the worm reduction gearbox and you need the mechanism similar to that from an old school mouse (or a modern lazer type) to detect rotational motion and direction and you have replaced the "normal" stepper with something that will be able to "self correct" (in conjunction with software of course) which today's stepper motors cannot.
And they are WAY more powerful !!
In batches of thousands they can be produced far more cost effectively than stepper motors which have SO MANY coils and other stuff - this motor has 1 large armature with 12 or 14 coils - thick wire and a pair of magnets in a casing with a gear set and containing a commentator and a brush assembly.
Controlling it is simple PWM and rotational sensor reading - in effect it is a large powerful servo motor.
Have you ever tried to stop a wiper motor - you injur yourself if you are not careful !!!
Control circuitry (the high power side) is a single h-bridge and power mos-fets or whatever, not some nonsense about quarter or eighth step manipulation which is often not a reliable way to get positional accuracy.
The manufacturer who comes up with a robust, "agricultural" (strong) design in a 50x50x100mm package will make a killing because it will be a better, more robust, more cost effective and stronger method of driving printers, mills, even robots.
So what if the drive shaft comes out at 90 degrees to the 100mm axis of the package, designers will design around it - that is what creative designers do !
If you cannot "stand it up vertically" due to clearance constraints, lay it over 90 degrees on its side. SO WHAT?
Most think that conventional steppers are the way to go.
Some years back most people thought that the earth was flat.
One group has already been proven wrong and the other soon will be when people realize that the CNC universe does not revolve around the stepper motor - the stepper is just the "cool" thing to use - there are more cost effective and efficient ways of moving a tool or a bed around.
I recently discovered that the Eris is a good place to start for beginners. Sure it is not cheap compared to a bunch of low end 300 dollar machines but so far it has been 100 percent hassle free from the unboxing to the first print. Only hard part was inserting the filament and installing the software because I never used Matter control but after that all great prints. The only thing that I can not speak to is the reliability of the machine over time or the maintenance involved because it is too early to tell. As far as the rest it is a very smart printer it auto levels and if anything the prints stick almost too well to the build plate even with cheap filament. It is also portable which means you can take it to school or other places for demos which is a plus and the quality of the prints are quite impressive for such a small printer.
I'm not acctually a consumer but rather a child with an interest for tinkering. I ordered the tevo tarantula and most likely I will print a few cool things then get bored and realiez I don't need it but I believe that a 3D printer build kit for 200 dollars is much better than a 300 dollar summer camp. It's cheaper, much more fun and you get a very cool finished product that can acctually be used.
How about servos instead of steppers? They provide feedback (closed loop) and are used in other cnc applications such as plasma cutters
defactobeertv Closed loop servos add more cost and complexity, but of course they can be made.
...would it make sense to put shaft encoders with every stepper motor? I definitely think it is a good idea for the filament path, so we don't need to tune extruder steps/mm. It would also detect jams and the like.
Also, automatic print removal from the bed would be a neat feature. I've found that most parts come off easily from my glass printing surface once it has cooled down, but they stick nicely while it is hot. But they only stick in the first place because I use hair spray...
I'm excited for the future if this is the way it's going, in any case!
Great analysis, Tom. When you put it in perspective with 2D printers... Yeah, there's no disputing there's a path yet to track to get 3D printers there. Thanks and congrats!
I think the consumer market is not a good place to aim for right now. The market isn't big enough, so it's hard to make parts cheap enough. A sub-$400 printer with all the sensors needed for reliable operation would require hundreds of thousands of machines to be made in a batch. 2D printers are heavily engineered and optimized pieces of equipment, and if you tried to make one in the unit count of 3d printers (a few thousand max), they would not be less than $400. Engineering time, making molds/tooling, getting pcbs manufactured. It's all scales dramatically with unit count.
@Thomas Sanladerer I hope your getting settled in. Thank you for the lecture, I found it informative. I've been following 3D printing for about 5 years, but I have yet to commit to a printer. My problems/concerns are price, print quality and technology and the vicious cycle that currently exists between these three categories. I just haven't found "the" printer to get my feet wet. I'm currently looking at the Creality CR-10. It certainly covers price and print quality and to date I've had a hard time finding reasons NOT to buy it. But advances are happening so quickly and prices continues to come down while print quality goes up. Other printers I've been looking at are the Wanhao models from Monoprice V2 Plus or Ultimate. For me as a hobbyist/maker print quality is important. SLA fits this bill, but the high cost of ownership (up front and operating costs) plus the mess of SLA keeps me from going that route. Now we have SLS approaching "affordable" for serious hobbyists/makers. It's maddening which also fed into some of my frustrations the other day. Anyway I continue to watch and learn from you sir. Thank you again.
Good points Tom, and great talk. Each of your points is perfectly achievable, but each one drives up costs. Autoleveling is getting there, position/feedback sensors can be added, but we need better firmware to handle the feedback, and ambient temps, etc. This would most likely appear in the next wave, of 32bit controllers.
Software is a sticky wicket, you want something that nearly anyone can pick up, but experts will soon want more. I think 2 key points would be better use of developer time - 1) compatible 3d source files, so one can take the beginner tinkercad file and load it into FreeCAD or a high end CAD for better refinement or whatever. and 2) BETTER libraries, thingiverse is a drag, it slow, its clunky, of the 100 things I downloaded only things that are knick-naks work, things that are supposed to FIT together, are always wonky, and don't fit.
I got a 3d printer when the prices for a CBOP (chinese box of parts) got below $200. Price gets people in, a consumer level friendly box will be more expensive.
Nice house. Now get to work !
I am actually working on the "next step" in 3D printing...but can't disclose any details...still in the very early stages. Edit: maybe that was too vague. I'm challenging the existing workflow and making hardware also
I waited until this year to buy one because I hate fixing things. I think now is the perfect time if you've been eying them. I bought a Monoprice Select Mini, and while I've had minor hiccups, I've not had to replace or fix anything. The only shortcoming I see is that it doesn't print metal, and when they have an affordable desktop that can do that.. I just can't imagine.
Huge congrats on your new place Tom. Looking forward to seeing videos from there.
Fantastic insight, I too think manufactures should focus more on quality than being cheap. I will gladly pay a bit more for something that has quality.
Excellent analysis Thomas. I think it is useful to compare the emergence of domestic 3D FDM printing with that of the domestic sewing machine. Both technologies are essentially very simple, yet to really get value from them you need to have a good grasp of 3D modelling, whether that be in the form of paper patterns or CAD. Consider the success of Burda Style and Butterick.Furthermore, although a sewing machine can easily be used to re-attach a single button (is it really worth it?) advanced users need a deep understanding of how different fabrics, needles and threads interact and when they should be used. I think a fundamental difference between the two technologies is the almost immediate feedback that a sewing machine operator receives as they work the materials. A sewing machine is still just a complex hand tool. Automation of "computerised" domestic sewing machines is exemplified by elaborate embroidery, but would we ever see a sewing machine that you load up and walk away from and then come back to find a made-to-measure shirt?
so why does an extruder motor skip on a specific part of the print every time ?
mine is doing this , freash clean nozzel
so idk , the print is still pretty ok though
plz fill me in on this
I've wondered as well. When will filament sensors be standard for anti spaghetti or like you suggest temperature. It's fun to explain to folks how the extruder works and it doesn't care if there's anything underneath it. It just assumes the last layer is there. Some just give me a funny look?
Paint the ceilings white. I did and it looks very nice and unique. You might even go for a white wash look. Beats all that work and expense. I can relate. I too built my first machine back in 2010 Prusa MK1.
+UFOhunter you're saying just smack a coat of paint over the laminated ceiling panels?
Yes. You HAVE TO use a good primer as the first coat. If you don't use a primer, the paint will peel off.
Wait a couple of hours and then paint with a good latex paint and roller.
If you use a good paint, you'll need one coat if you want your ceiling to be white (recommended), since the primer is usually white. If you use any colors besides white, you'll need 2 coats or more.
If you use a cheap paint, you'll need 2 or more coats. Cheap paints are not labor friendly.
Good video Tom, thanks. Wishing you well with your house move.
hey Thomas, I just finished my very own 3D printer, no kit tho I build it all myself from scratch up. It works better than I expected but there is some work to do because the prints aren't that nice as a delta printer or as a prusa, do you have some tips to help stick the layers better together?
I've noticed a flood onto the forums of people who don't know their new machines, haven't taken the time to do even basic research and familiarisation with the concepts and just seem to have picked up a printer on a whim.
I think it's too early for the larger consumer market by far.
Do you have a schematic or instructions about the filament diameter compensator that you mentioned?
Good points Thomas. I have a lot of experience in AutoCAD starting on an IBM 286, and manufacturing using CNC lasers, turret presses, Mills and Lathes. Decades of experience. I've found the design process the most difficult part to 3D printing. Fusion 360 is great, somewhat flawed, but without the mindset that comes from conceptualizing and producing real parts, I would have been lost. 40 years of experience helps, but the process is still not easy, and will frustrate the average newby user. I have a Rostock Max V3 that I built and it is chugging along with only occasional calibration needed. After designing and printing over 90 custom things, the learning curve is still there but not that difficult to master.
Do you have a link to the filament diameter sensor? Is it an open project?
I couldn't have said it better. You should see the 3D Printing group on facebook. Its a train wreck. All day its just "why doesnt it work" "what am i doing wrong" "my printer sucks", and then theres the idiots saying "my 150$ Anet owns 2000$ printers"... I cant even anymore. And most of the guys that have a at least one working settings sheet, think they now know better than anyone else, and will argue it to the end even if you present data to why theyre wrong, then they throw a tantrum.... Thanks a lot mainstream media outlets....
Well, in my job I work with state of the art Stratasys FDM 3D printer, and in my home I have an Anet A2 (among other printers) and in terms of quality there is almost no difference (after upgrading some not electronics parts), so the people claims are not so far from reality.
Simon Silva Lemarchand for basic materials I agree, pla especially but I know that on my anet due to its lack of an all metal hot end there are many materials that I can't print, also it's build volume is much smaller then most higher priced printers
Toby Cowles but that is an easy upgrade to do, and also relativitly cheap. In my case, I have all my printers with all metal hot ends, but that is not the most important thing when you want to print a complex material, I think a hotbed is the most crucial item for that kind of task. Anyways, even if you have a state of the artículo 3D printers, There's always something that you are unable to do with it.
Simon Silva Lemarchand yes, I probably will upgrade to an all metal at some point, mine already came with an aluminum hot bed standard,
What about those that know they are no 3D engineering god of l33t destruction pwnz0r, and need a 3D printer for making many examples, for cheaper than commissioning companies for a similar bulk (costs more)?
Just wondering :)
Hi Tom. I heard that if you grit the surface of the glass bed with 90 grit sand paper using a bit water (wet sanding in diagonal direction, only one direction), nothing will lift off from the surface during printing, even Nylon. Is that true?
+CoolStreamT haven't tried that yet, but it might work.
I tried it and it works with ABS + ABS slurry. Very cool. God..my parts are so straight. It doesn't work with Nylon + abs slurry. For Nylon I need to find slurry based on Nylon, but nylon is chemically resistant. Just few chemicals can dissolve it. Actually I posted it on facebook Anet A8 forum and the link here , but you guys erased it for some reason. Anyway the video is still there on the forum.
Honestly? We are living in 21st century with amazing access to any kind of knowledge. If someone doesn't have any background knowledge on 3d printers, 3d modeling, CAD software, why shouldn't they learn this stuff? You have google, forums, youtube channels like Makers Muse, 3d Printing nerd, and ofc Tom's Channel. You can easily learn stuff without getting out of your home. There are plenty of tutorials, guides, reviews etc. and if someone wants to start 3d printing, why shouldn't they? In my opinion, if someone is so lazy that he thinks he can buy 3d printer and press 1 button to print everything he wants without any reading(or watching in this case) it's his/her business, why bother? In 21st century, with such access to data, lack of knowledge is a choice.
Is closed loop control or some sort of encoder ever going to become a popular upgrade?
Been working in this direction for a couple of years. We are getting close...
Tom, do you have two cents to throw in on closed loop controller board upgrades for stepper motors?
Got a site for that filament sensor? Is it on the market for sale?
Cracked up at "Plunge down, yeah! You can do it!".
Would love to see you make a video on how to make your printer "smarter". I know filament runout sensors are easy to implement. What else could be done?
Tom, good video! My thought on this topic - to many adjusting options in the software (slicer) and mechanical flaws of 3D printers (vibrations etc. ) give a lot of room for error. The printing results are also affected by the bazillion options of available filaments. I do not really think that one company will be able to solve this by bringing a consumer friendly solution to the market, since it requires custom software + very accurate mechanics which results in a very high cost of the printer. You can clearly see this with today’s printers that print very good.
Considering the above, machine learning (AI) could be the way to go. In my opinion, the best solution would be to include as many sensors as possible in any type of printer and collect all the printing data in a cloud. This would give enough data points for a cloud based AI to remove the errors and adjust your print settings while printing. Basically, if you are connecting you printer to the AI, the AI provides adjustments to your print settings based on the information it knows and your printing results.
I assume you and other smart people from the 3D printing world could definitely agree on the minimum data points required that should be collected. I see it more like an open software project - anyone could (manufacturer or DIY) put needed sensors on their printers and connect to the cloud based AI.
I think that such a system could make any printer (with bad mechanics, made out of wood etc.) to perform very well, since the printing could be adjusted in real time and considering the mechanical flaws of the system (mechanical issues have always a pattern so the AI should be able to counteract).
As a result, you could make a printer out of Lego and print very, very high quality prints and the software cost would be 0 :)
I agree… I have been complaining about this from the beginning. Direct solid modeling with a nice interface and simple workflow is the way to go if we hope to advance 3D printing past the realm of the user-nerds with any kind of market staying power. Procedural and parametric or history based CADs will never be the way for consumers (and are even clunky for prototyping on the fly or modifying existing designs), and 3d sculpting/animation programs like blender (which I also use occasionally) or ZBrush (while great for art/figured/etc) are not at all user friendly, and just plain not any good for technical CAD. Tinkercad is too simplistic, but Autodesk was onto something with 123D Design - it just needed a few more features and a little shove toward 3D printing and they would have had a perfect middle-ground consumer product. Unfortunately they killed it off and rolled the features into a crashy, clunky, overcomplicated, swiss-army-knife, cloud-teathered abomination that I begrudgingly use since it is unfortunately among the best (if not THE best) comprehensive options out there.
I also think that the next HUGE step in hardware is closed feedback loops and modern processing. Mesh leveling and multi-extrusion are great and all, but we have to patch up and modernize the foundation before we add more floors to the house.
When I first started poking into 3D printing, my mind was blown that 99% of the products out there were open-loop systems with zero error correction, running on 8-bit boards. The Reprap project was a springboard, but it became a platform because it was just too easy to take their work and roll it into a good-enough system.
If we want consumer products… take a page from the consumer 2D printer books and use cheap, lightweight, DC motors with encoders, and cheap boards with enough processing grunt to run floating point operations, realtime feedback data, and a decent display with a thought-out UI. I pull apart dozens of trashed printers for parts, and I have not seen an open-loop, stepper-controlled printer in a good few years now.
I been messing around with 123Design for a while. What other program is close to it, but have more features? I dont like online cad. I want a program that I could install on my computer.
+TheDarkstar826 any "normal" CAD is a logical next step from 123D Design. If you're looking for free options, that would be Fusion360 (cloud-based), Onshape (web only) or FreeCAD (total pain to use)
Hi Tom, thanks for the video.
This is somewhat relevant to me as in the summer I'll be working in an arts summer camp in the US teaching 9-14 year-olds 3D modelling and printing using Makerbots. I was wondering what kind of software you think I should use to teach them modelling. TinkerCAD, 123Design? They are all assumed to have no prior knowledge. I have a Vive and was hoping to do some VR clay modelling, but aside from that are true any others you might recommend?
Thanks!
Hi Tom, I have Anet A8 3d printer. I like it. I just think it's kinda small. Is it possible using Anet's electronic main board and screen to build larger 3d printer, let say , with the printing bed size 500mm x 500mm or even larger. I have master’s degree in mechanical engineering and I have a mini machine shop at garage with machines and tools. I can model anything using SolidWorks. I just need a bigger printer to print my Big ideas :) I can make a frame from aluminum extrusion and I can find any polished shafts with bearings and screws with a bit bigger step motors. Some parts I can print with my Anet. Can I alter the extruder and put larger nozzle to speed up the printing? I mean with larger print size I will need to extrude more plastic in a time, so I will need higher printing speeds Is it going to work? Again, will Anet's A8 main board support that size bed? I know you are the first guy who can have some idea about this.
CoolStreamT Most likely You can keep all electronic components, you just need to change the dimensions in your software; easier with some boards than others but usually relatively simple with lots of information online. Also just buy a larger size extruded, also then change settings.
As a side note this is a good example of Tom's point. Statistically, not personally, an engineer is of average to above average IQ and if an engineer is unsure how to modify their 3D printer it shows how far we have to go before this technology becomes common place in our homes for us average to below humans.
This is not a criticism, it is just a very good example of why 3D printers are looked at as novelty items by a large portion of society.
For bigger size bed, I need bigger step motors to move more heavier beds and extruders. So bigger motors need more power ( different voltage). Can I plug these bigger step motors into the Anet A8 board and start printing? Could you tell me what printer brand board I can use to do that for sure? If I challenge you, maybe Tom can answer the question.
Hello Tom, I can't find anyone who owns a Zeus 3D printer to get help from... My icon for Zeus_UI is there but it won't load the application to launch the program so I am unable to print. please help if you can. Thanks Kells
+ONTHEGO PHLEBO get in touch with AIO, there should be a relatively easy way to reset the entire machine
What were those random numbers?
it is counting the "as you might be able to tell........"
o
Wooo, I like how you sit on the table.
These are some fantastic points and used a great analogy with the 2D desktop Printer. I honestly don't think that a 3D Printer in every home is really something that will be seen anytime soon. Possibly whenever a future product comes along that allows people to make practical objects in a sub 10 minute time frame. I believe that once it gets to this point, we will have a TON of companies that are charging for replacement parts that you can just print out at home. I fear for this day, but I can see this becoming a trend IF we have a printer with a 98% success rate in every home.
Microsoft 3D builder is really quite nice as a basic tool I think. Combine this with a simple "ready to go" compact enclosed printer like a daVinci jr and I think you are in better shape for a beginner. Yes, I know it makes consumables more expensive, and is not open source, but also more usable - my DiVinci rarely has a failed print, and I always know how much filament remains and if a print will print. Unlike my new "mechatronic opensource super flexible lego kit" - which is super fun but as reliable as an Italian sports car. It has parts made in the daVinci because I know they be durable and dimensionally correct. Just some thoughts based on six months of learning curve so far :)
I've been saying for years it won't catch on until you 'dumb it down' to the level of using an iPad while making it as reliable as one.
I believe we will eventually get there just like we did with computers which were in the same state of disarray in the 80's (maybe even worse since communication was so limited) and the Internet was in a similar state of disarray (But not nearly as much due to immediately implementing open standards) in the early to mid-90s. My first introduction to the Internet involved compiling a TCP/IP stack because windows 3.1 never had one, using a 28K modem (56K hadn't come out although 33.8k were but were expensive for an extra few k of bandwidth) and you phoned into an ISP and used a Unix shell account to actually hook into the Internet via the command line (the horror) and the only browsers available were Mosaic and an early beta version of Netscape. My first website was coded by hand in Notepad and was 5 pages with mostly text and a few graphics but man was I proud of it.
And all my friends and family said it was too complex and would never catch on .... but 10 years later every single one of them had it
What about a company sends a file which has a license for 1 part to print then it doesn't work and they charge you for it. Hope I made sense. Great video Thomas
Hey Thomas!
Really like your videos!
Good luck finishing moving!
Keep up the good work :-)
Alread ycan't wait for the second season of fillaween.
PS: Where could I find your filament diameter adjustment sensor?
Great Speech!!! I totally agree!
A lot of people say things like: "people have to be patient" or "you can't just instant be a pro-inventor/designer/maker" or "There's always a learning curve" or "It's not just for everyone"
Well that's the exact problem! Something that IS "for everyone":
- is intuitive
- knows what a consumer next choice/step is
- fixes problems/errors silently in the background (doesn't bother consumers with problems the answer could be guessed or logically made)
- works almost out of the box for everyone!
- etc etc..
That's what a 3d printer is NOT! Great comparison with MAC computer, that's basically just a Linux machine fully (fucked up) and packed in a more manageable jacket that satisfies things @Thomas Sanladerer sad or the few things i just listed!
Those type of "skills" do sound like magic on this exact moment considering 3d printing and manufacturing/creating but trust me!!....
There was a time long long ago, tech-nurds also tough the same about computers.
Well look at computers (and such) right now! Maybe the real (rooky/not-expert) consumer only scratches the survive of what a computer is really capable of and happens behind the scenes or even know of all the hundreds/thousands events taking place by just a simple click! (for example: going thrue the OSI-layers down to zeros and ones, trust me that's a long long way of instructions!!!!)
so true, I agrees, though it's not going to be easy to make a consumer grade 3D printer that actually is user friendly, but seeing as it is, it's going to be a while to get there, always great videos, I enjoyed all of your videos...keep it up!
hi Thom can you do a video on how to update slic3r in repetier host please Regards Rick
Delta printers just need either a very accurate distance sensor for each carriage on every axis, or a gaming mice sensor with strong LED scanning movement on a belt on each axis to detect irregular movement. Semi-Closed loop is really nothing special. Just the right pick of smart components. Augmented reality technologies from the recent Google I/O could also play a part in detecting errors with machine learning algorithms. This means just stick in an ARM smartphone chip with camera already just like self-driving cars. Simple implementation would be a depth sensor like the Tango Project and new Asus AR phone which helps the machine learning algorithms monitor the 3D space of the print bed at every minute. Seriously we should all learn from past history from how so many developers underestimate capabilities of smartphone cameras that they don't even realize that an app can measure a humans' heartbeat when coupled with the right LED light.
but if the average person could easily get skilled in 3d designing+printing then we all (the community) would consequentely starve to death!
Hey Tom!
What are your thoughts about Microsoft's Paint 3D? Everyone know how to use the good old paint, so do you think that it's 3D counterpart will kick off 3D Printing?
+Pierre LF used it, couldn't see how it would be useful.
I wonder if there is a solution in the foreseeable future. There's a lot of things that have to go right for a 3D part to come out right.
We don't see this same kind of push, at least to the same degree and fervor with other CNC-type machines. There's a lot of complexity going on, so many things have to go right. Even if you do all the right things there's something that can be a roadblock.
amazing tour!
The more closely I look at 3D printer parts on eBay, the more worried I become. There are ratings issues, protection issues, clearance issues, layout issues, interference issues, and the list goes on. We are flooded with cheap rubbish electronics that should come in a swan vesta box and that require serious rework to make safe.
After watching a lot of youtube and gaining a better understanding of 3d printers it would seem to me that the 'Leapfrog Bolt' printer ticks all the requirements of a reliable quality printer, reflected in its price. Would like your opinion of this product.
lot's of good points here. also, i just ordered 100+ euros worth of filament from dasfilament. i sure hope it's as good as you say :-)
I don't expect any manufacturer to take the lead on this, they aren't willing to share their improvements.
I expect this to be the result of community development. The auto-bed leveling/detection is a good example, The community developed this feature as a free add-on and now it's a standard feature in all new designs, not because the manufacturer is doing anything, but because the free software they base things off of includes it by default.
As far as additional sensors go, It seems that we are running out of I/O pins on the default arduino boards taht are the standard basis for things. Until we make a jump to something with more I/O and more processing available, I think we are at a plateau. However, with things like the ESP-32 chip hitting good availability, I can see this breaking free fairly soon. A simple microswitch in the filament path to detect that there is no filament there would be a trivial thing to add.
We are seeing a handful of closed-loop designs start to show up in the CNC world, and I would expect to see some of this start to show up in 3d printers once things stabilize a bit.
One problem is getting people to back away from the "because I can" and get to "it's good enough", the o=drive project for example can use hobby brushless DC motors and shaft encoders to make _really_ fast machines that know exactly where they are, but it's designed to drive motors at 100A @24v, and as a result is fairly pricy.
We need to get a closed-loop equivalent of a ramps board out there, something that's just good enough (with flexibility and upgradability enough to be used in many systems) and let the far east clone manufacturers start mass producing them to drive the cost down. This is going to be a bad thing for whoever tries to make money selling such a board, but a great thing for the community/industry.
This is another reason why I think we're at a plateau until the community makes something 'good enough' to get everyone to catch on.
Sensors? SENSORS? .... Filament sensors are know to work until the cheapo plastic the cover of the microswitch erodes (abs/pla can do that with time and pressure) .... and if they are integrated on the extruder it depends on the program . Usually it means "Welp, print stops right whenever this is 0". What if it malfunction? No dev ever leaves a bypass option for those more daring or with a dying printer awaiting replacement.
But closing loops and setting more sensors is a nice track. Just do your FMEA's , proper development and have a plan when everything fails. Because of time, manufacturing or simply the user mishandling your machine IT WILL FAIL.
Didn't come here for the blog at the start. If you want to do that make it separate and clearly marked, or at the end of the video so we don't have to search for the real start of the video.
Totally agreed. Asides from smarter hardware, the entire software tool chain of 3d printing still need to be a lot smarter. I think slicers are mostly too dumb. there are a lot of things that you can do to make a slicer smarter. Cura 2.x is heading that direction but there are a lot more that can be done.
Mega Making software development might take less time than hardware (but debugs forever) probably needs to go simultaeniously to work out the kinks, in any case I'd do it...but then again who'd pay me to do it? so open hardware/software might not be the way to go
I am an university researcher, and I would think that research into closed loop control for open-source machines should be taking place at universities. To that end we have actually started a post graduate research group focusing solely on additive manufacturing and specifically for research on desktop 3d printers (non professional). I believe this is where making will be moved into the mainstream consumer market, could not agree more with you on the current barriers to entry!
I plan on getting a 3D printer because I would love to be able to learn about it how it work everything and I'd also love to be able to design my own stuff too
I revoke my last comment about steps. You are right in your cal video. Use the prusa calculator. I also agree with this one but less emphasis on creating your own models and more on making the printer smarter. Even just as simple as the software and hardware working together better. In opinion the printers wont take off and be consumer grade until something equivalent of the computer able to watch the print being printed and make its own adjustments. (Cameras and recognition software). Skipping sensors. Etc etc.
Hi Thomas! Your video series is excellent and is very informative. I was wondering what your thoughts on the 3DprintClean Air Scrubbers/Enclosures to reduce the amount of Ultrafine Particles (UFPs) released into the air with burning plastic and the long-term health consequences. I have not found a competitor for their product and wonder why not everyone is using enclosures for their printers (not only for ABS smelly fumes)Thanks!
+Infosec Val it's a touchy subject, both as to how you can even remove UFP and whether the concentrations 3DP produces are actually harmful in every case. I'll leave th arguments to scientific studies, as I do not have the tools or expertise to judge the subject.
Thanks for your response, Tom!
REALLY great points you're making. At present, 3D printing is a bit of a "fiddly" hobby and not ready for the masses. More reliable and "bulletproof" hardware is important, and I think progress is happening on that front, but the lack of user-friendly software is the biggest issue, IMHO.
Though I will say that sites like Thingiverse will keep the lack of firsthand creation power as a less important issue. You'll have hobbyists finding lots of great pre-created models that will keep them happy for a long time.
In the process, you'll create a market for people who CAN and WILL create custom 3D models for people. The same way there is still a market for professional graphic design services (shameless plug), you'll still have a need for people to make good models.
This video definitely fits into the category of "Do you actually need a 3d printer or just the number to someone who has one?"
I think I am going to send you the 3d printer that I am building, tell me if it is close to what you are looking for. Shipping might be really expensive from the US, so I'll probably disassemble it into a kit that will reduce shipping costs and that you will not have to tinker with, similar to the way that Lulzbot ships their printers.
So are you moving closer to any actual civilization (like Munich or Nuremberg) or further away?
Have you tried the Robox? it seems to have a lot of these features,
Hoping for the usability to achieve 2D printer level. These "consumer" level FDM printers are so far away from "Click & Collect".
I make pretty complex devices with tinkercad. Well, that's accessible enough.
I think the best place for massive innovation is the $4k-10k prosumer/small business printer market. The only real contender in this market is Ultimaker, but their printers are not significantly better than a Prusa i3. It's all the same basic components. I'm working on some new printer designs in this price range with these goals in mind: precision, super high speed, support material, reliability. I think that's the quad-fecta that would make this price range of printer exceedingly useful.
Around 6:40 you emphasize the usability of 3D modelling but also hint at what is a more fundamental problem: a lack of design literacy. That is "design" in the sense of understanding a need and connecting it to a solution by working out all the steps in between. Imagine having this same discussion about paper printers in a world where "normal" people don't know how to write... Calligraphy is only a part of writing, regardless of how fancy your pencil is (and that's what a 2D printer ultimately is, isn't it? A really fancy pencil...). Similarly, the 3D model-print toolchain, while tremendously valuable, is only part of a design process. The solution to that moment when the initial excitement wears off is to better communicate the idea that 3D printers are a gateway to broader abilities.
That gateway aspect might be the critical difference for this third class of printer (non-enthusiast, non-industrial). Designing for your own needs can involve a huge skill set and it is not encouraged by a consumption-focused culture. People have to start somewhere and a great way to learn any complex skill is to start by making copies, tweaking them more and more until you're making your own thing. In that sense, just making copies of 3D models is an essential starting point as long as there are immediately accessible ways to tinker that don't break the process. That's the critical point you make about more accessible printers: more important than 100% reliability is that they have enough sensors to provide meaningful feedback to get someone back on course if something's going pear-shaped. Printers made for improving design literacy must be built for robust "tinkerability"*.
Still, like writing, it would be nice if there was a broader emphasis on the value of design skills, just like there is for reading and writing which it is no longer "normal" not have.
P.S.: the other thing that makes this combination important in this use case over the other two classes of printers is that skills might be more perishable for this group. If you're only a casual, non-technical user, you might forget how one bit or the other works or how all the variables fit together. Building the toolchain for robustness means that it will just work when needed. Design literacy is served by having feedback built into all the operating steps so that as they work their way up and down the learning curve, it's never too challenging.
You touched on safety only briefly, and skipped the part of hazardous materials, fumes and particle emissions. For a true consumer level 3D Printing, it needs to be safe beyond just fire hazards. There are only a few manufactures today that have taken any steps on the emissions issues. Consumers aren't going to place their printers in the garage or open the windows is the dead of winter to avoid toxic gases put off my materials such as ABS, or clear out PM2.5 particles above national safety standard. HEPA filters, Charcoal filters and air quality monitors in enclosed temperature controlled cases should be standard. Otherwise lawsuits from adverse health effects will ensue...
Today, finding an image that you want is a search engine away. At least close to that. In a day and age where you can assuredly find several high-quality printable 3D models for basically everything, then even people who can't make their own models will want a 3D printer.
It also has to do with word of mouth. The day a colleague tells people about how great a 3D printer is and how it changed his life, others will want to get one. Hearing about basic issues or "it costed $3K and I'm not using it too much" is obviously discouraging.
i think its already other, it's just to expensive to make for the common guy
that is huge... my workshop... 1/4 of one of your room...
Well said. Thanks Tom.
dude... excellent video. thanks!!!
3D printers will never be for everyone just the same way as high end custom PCs and woodworking scroll saws will never be for everyone. The average person has no need for a 3D printer. Perhaps in the future when 3D printers can print "anything" there will be one in every home. People will pay for the files for an object rather than the object itself. 3D printers and manufacturing tools and so there only for makers and designers at the moment. Great video as always Thomas!
Firstly we should use Servo Motors instead of stepper motors^^
Whats better about Servo motors rather then steppers?
Isn't it then just as easy to use an encoder?
What do you guys think about "Clearpath Servo's" ? Are they worth their money?^^
Servos are just brushed motors with integrated gearbox and closed loop control circuitry. At the torque requirements of 3d printers, I doubt you can find a servo cheaper than just taking your own brushed motor, gearbox, encoder, and having the already existent control board (commonly an ATmega1280) do the closed loop math.
I would guess that a company like Lulzbot would be the perfect place to improve the printers and share theyr findings with all of the open source community. The way i see it, thats really almost the only way an open source company can stay successful with theyr business, if they innovate faster than the closed source competition can.
However i actually think that there are too many patented processes revolving around 3D printing to prevent the innovations made in the professional printers years ago tickling down into the broad consumer and maker space. If some old patents on the fused filament production process had not expired we would not have most of the printers we all know today...
I think we are finally reaching this point.
No music.
cool good luck with the move