Same experience for me. I bought a cheap printer from eBay and slowly started re-designing and replacing existing parts by printing them out on the printer itself. Nearly 100% of the printer has been replaced, and I’m finalizing the design for a second printer. I’ve started making contributions to the firmware, slicers, and host software to get features I want, learned to model in Blender, learned about programming micro-controllers, started researching thermoplastics, material science, and the list goes on.
3D printing is an incredible intersection of software, electronics, machinery, chemistry, and an open, community-driven R&D environment. It is the most fun I’ve had since being a child building RC cars.
I want to offer additional context for other people who read this and may not be familiar with how 3D printers work. The core components of a 3D printer are not currently printable. These components are: power supply, control board, steppers, rails, bearings, belts, hotend, and the bed. When people talk about printing parts for their printer, they are almost always referring to the secondary parts like the carriage assembly, extruder gears, fan shrouds, and frame components that can be made from a large variety of cheap materials (acrylic, MDF, ABS, etc.). We are a very long way from a printer being able to actually replicate itself.
Someone questioning your use for operating 50 printers reminded me about a recent raid on a Dutch drugs crime organization. They used 3D printers to custom print fake Nintendo game cases, ink cartridges and fake make-up compacts and then used those to hide the drugs.
How far away are we from 3D printer + CNC mill/laser cutter + robot arm for assembly?
Incremental deposition may not yield a working 3D printer, but couldn't a small ensemble of machines construct all their parts? (Minus the chips, for now)
Great question, and the answer is probably "it depends". I've got all of those except the robot arm/part picker and even if you had really expensive tools like metal sintering printers and a 5-axis CNC mill, you would still have to buy a lot of the components that are produced with specialized machinery. I'm going to go out on a limb and say that maybe in 20 years we'll be able to self-replicate machinery with raw materials inputs and a lot of work. It wouldn't be even close to economically viable to do so, but it might be possible.
It's important to remember that the majority of the core technology in 3D printing today actually dates back to the late 1980s. We're starting to see some interesting developments in materials and capabilities, but there are still plenty of limitations that need to be overcome.
I highly doubt your estimate is even in the ball(bearing) park.
Spindles, motors, circuit boards with components on them and bearings, slides and so on are all multi material or very complex processes usually only doable if you produce a lot of something in one go.
Just try to think about what it would take to print something as trivial as lacquered copper wire for stepper motor windings or a circuit board with a reasonable level of integration.
And the biggest issue with that prediction is that there is no gain from it: printing the non-commodity parts is the whole trick to efficient 3D printing, mass produced parts will have incredible accuracy and very low pricing so use them when you can and 3D print the remainder.
I don't think you're arguing against what I actually said. I said in 20 years it might be possible to replicate a printer using nothing but raw materials and a lot of work, but it wouldn't make sense economically. That's very different than saying you'll be able to print all the parts you need in one go, or that you would want to, which obviously isn't viable without molecular-level assembly. That's a holy grail in the distant future. But making PCBs is already possible without specialized equipment, metal sintering gets you pretty far on components needed for things like steppers and threaded rods, etc. If you follow the research being done (ex. [1,2]) it's pretty clear that the boundaries of what's currently possible are being pushed in interesting directions.
To me the biggest hurdle is the electronics, which currently do require special tooling to even produce basic components. You're probably right that we're more than 20 years from self-replication ability (again, not practicality), but I'd be surprised if it's more than 50 years out.
If you operate 50 printers you're presumably doing some sort of business. Nobody operates 50 printers for a laugh.
It sounds interesting. Are you printing arbitrary parts that other people send you, printing parts of your own design for your own products, or something else?
Yes to all of those. It's a micro manufacturing shop (we also have laser cutters, a CNC mill, woodworking equipment, casting materials, etc.). We mostly sell products that we design and build on Etsy & Amazon, but also do small batch runs for people who need 50+ pieces of something and the economics of injection molding don't make sense. It actually started as a makerspace, but the economics of that model were not sustainable, so we scaled into small-run product manufacturing.
1. Basic electricity
2. PWM
3. Battery chemistry and care
4. Transmissions/gear ratios
5. How DC motors work. Winds and turns and brushes, etc.
6. Suspension basics
7. Basic RF (crystals and channels and avoiding interference)
8. Soldering
And all sorts of hard-to-list mechanics and know-how.
I didn't realize it at the time (and I don't think my parents did, either), but I learned so much more from that hobby than I did in school, for those areas.
3D printing feels the same way. I only recently got the Monoprice 120mm^2 unit, and it's working great for what I've thrown at it so far.
What are you using for all the non-plastic parts? Do you print slides and bracing as well?
The board is an MKS Sbase 1.2 with an ARM chip running Snoothie, and I’m using the integrated stepper drivers. I switched from using the typical Arduino Mega + Ramps setup because working with Marlin was really difficult, and it felt like they were fighting a losing battle trying to cram more features into an 8 bit architecture. The motors are varying sizes of Nema 17s, but I don’t remember the manufacturer.
I have started experimenting with printed linear rails, and likely V2 of the printer will have printed rails on the z axis. The bushings are printed nylon that thread directly into the bed, x carriage, etc. I would also like to start experimenting with printed wiring, but most conductive filament is more like resistor wire, so that would take some figuring out.
As far as the frame and all that; yes it’s all printed. You’d be surprised how much the part count starts to fall when you no longer have to attach different parts together and you instead start integrating them into a single, printed part.
Agreed. One of the reasons I've enjoyed robotics so much is that it presses on all three of mechanics, software, and electronics. Since there is rarely someone who is good enough at all three it gets a huge boost by community involvement by folks with complementary skill sets. The problem with robotics was that once you did the basics, obstacle avoidance, line following, balancing perhaps. You Were sort of stuck on the next thing to work on. But 3D printers are robots that can make more interesting things so they are useful in their own right.
My current favorite application is to make holding fixtures for things like breakout boards so that I can fabricate systems out of a bunch of separate boards easily.
3D printing is an incredible intersection of software, electronics, machinery, chemistry, and an open, community-driven R&D environment. It is the most fun I’ve had since being a child building RC cars.