How to Design a 3D Printer




Introduction: How to Design a 3D Printer

I recently built my own 3D printer with a friend. The process was both challenging and rewarding, and it's awesome to have a 3D printer of my own design. However, I found very little information out there on how to make a 3D printer from scratch on my own.

The goal of this instructable is to help those interested in 3D printing to have a chance to design and build something that is truly their own and providing resources to help organize the process of making a 3D printer. This process mirrors the engineering process very closely, but will also provide useful resources along the way.

Step 1: Decide What You Want

Before you even begin to think about your 3D printer's design. You have to decide what you want out of your 3D printer. Many people who build their own printer simply want the lowest price possible. Others want to be able to print more advanced materials, such as ABS or dissolvable filaments. For our project, we wanted a large print volume and a triple extruder. This is where you make your wish list, and browse around the internet for ideas. This is the phase where you look up 3D printing on instructables (gotcha) and poke around 3D printing forums.

If you're looking for ideas, there are quite a few good places to look. I lurked around YouTube, Reddit (r/3Dprinting), instructables, and most of all RepRap (this just goes to their homepage) where you can find a plethora of 3D printer ideas and designs. Perhaps you will find the perfect design and just follow their design or only make slight tweaks. If so, great! But if you're like me. You decided to take aspects of several designs and decided to make your own from scratch.

Be warned, this phase is addictive. Many people just imagine how cool it would be to build a 3D printer and read about others who have. If that's what you enjoy, great. If you want to actually make a printer, you'll eventually have to move on.

Step 2: Sketch Out Ideas

Now you're starting to dream big. In this phase, you get to draw out different designs and play with different ideas. Nothing is set in stone and you can tweak the idea until it's perfect (or so you think). Nothing is set in stone, so play with different ideas. Decide what on your wish list is possible and what has to go in order for the printer to be more realistic. Ask friends and family, post questions to 3D printing forums, puzzle it over in your head. This is where you get to swing from idea to idea without having to worry about changing documentation or physical parts. Again, the internet is your friend. Go out there and look at what other people are doing and apply it to your design.

Again, this phase can be a trap. Don't just skip past it, but remember that eventually, you'll have to settle on a design. No idea is perfect, and yours won't be. Just find something that you think you can do and go for it.

Step 3: Model Your Design

If you have access to the resources, DO THIS. It can be a pain at times, and sometimes it feels like simply finalizing your sketches, but you catch many obvious mistakes that you simply overlooked before. Many times, this will mean you have to go back to the drawing board and change radically, other times you'll simply find interfering parts or impossible shapes that only need a small tweak to work. Redesign will happen, but meticulously going over every aspect of your design will save you time, money and heartache. This also gives you a chance to look at what's out there in terms of parts and suppliers.

You should put a lot of time and effort into this phase, but you don't have to go through it alone. McMaster Carr is a relatively expensive part supplier (I don't usually buy from them for personal projects), but they have free CAD files for virtually everything in most file formats. GrabCAD is also another great site for getting professionally done files for use in your design. Finally, many part suppliers will have 3D files available if you do a bit of digging.

If you don't have access to 3D modeling for whatever reason, do your best. Build a cardboard mock-up, draft it too scale, do something and take it seriously. However, you probably have access to some kind of modeling software. Google SketchUp, 123D Design, and even Blender are all free modeling programs that each have their quirks, but will get the job done if you're willing to learn. If you're a student, you can most likely get access to any Autodesk Program or SolidEdge for free. I had access to Solidworks through my school, but they sadly seem to be lagging behind in this respect.

You also want to look into the electronics design in this phase too. You'll have to forgive my lack of knowledge about software or design considerations, but if you have questions about that side of things, let me know in the comments and I can ask my friend who worked with me.

You can get trapped in this phase too, but by this point, you've gotten pretty serious. As long as your making consistent progress, the longer you stay in this phase, the easier things generally get. From here on out, you're no longer just dreaming. The new threat is discouragement. Look back periodically on how far you've come from those napkin sketches, look forward to what it will become. Keep going and you'll get it done eventually.

Step 4: Ask for Advice and Edit

Now that you have an awesome model, you need to look at it critically. What will work awesome, what won't? Do you have any concerns? If don't know if something will work, do some research. You'll probably have to make even more changes, but it'll pay off in the long term, trust me.

Also, ask others about the design. I sent my files to friends who did woodworking, old engineering teachers, posted my designs and concerns to forums to let people nitpick (again, Reddit is great for this). You don't want to follow every scrap of advice thrown your way, but seriously challenge your assumptions about the design. You're still not married to it and nothing is set in stone. Tweak and change it until you think it'll work. Again, don't lose hope. You're making progress and without this phase, your 3D printer will be a piece of garbage.

Step 5: Set Up Parts List

Now it's finally time for the FUN part!!! Haha, just kidding. You probably already have a rough list of what you need from the previous steps, but it's time to formalize it. I set up a spreadsheet of everything I needed, where I was going to get it, and a link to the site where I would get that part. This is where you get your exact budget and decide on manufacturers and distributors, and generally make all the grown up decisions about your printer. This should take a few days, because being through helps you later on when going back over what you need and where you got x or y component. You'll probably forget things, or add stuff while you troubleshoot on the fly, but that's life.

This is where you decide on quality. There are some things that you can't skimp on, but other parts give you more wiggle room. Think about what you need, what each part does, and how much you're willing to spend on your printer. I spent a lot of time on RepRap during this phase just figuring out where people got their parts, what brands they used and where quality mattered the most. Reddit is again very helpful, and I highly recommend asking people things.

Step 6: Start Making the Printer!

Yay! You get to make it now! My favorite part. You will laugh, you will cry, you might swear uncontrollably or dance around the room in joy, but now nobody can deny that you're building your very own 3D printer! This is where you buy parts and start putting them together.

If you did everything up until now perfectly, you will fly through this phase without a hitch, everything will fit together just right and you'll have it running in no time. If you're human however, there will be snags. It will be a good idea to revisit your models, tweak them, figure out what needs to be changed and change it. There will be problems through every step of the assembly, but if you put a proper amount of work into the previous steps, you shouldn't have any crippling design flaws or expensive issues. At this point, it comes down to elbow grease and tinkering with things. If you know anyone who's savvy with tools and willing to help, it'd be smart to bring them on board. I was the mechanical wiz and my friend was the electronics guru, so we were able to cover all the bases pretty well, but still needed outside expertise and, you guessed it, had to go on forums for help occasionally. Programming needed an especially large amount of internet help, but once it's done, you'll hopefully have a usable chunk of bug free code regardless of prior programming knowledge. I had to make some parts over and over again, or "modify" them to make things work, but eventually everything fits and you'll have a working 3D printer before long.

Remember, this phase has it's low points too. Don't give up! Just because you've made it this far doesn't mean that it never gets discouraging. I love this phase of the project and find it incredibly rewarding, but there are times when nothing works and you don't know why. Fixing those problems is one of the best parts of making anything.

Step 7: Print Stuff!

You did it! The printer isn't what you imagined int he beginning, but it's here and it's working. Now it's time to enjoy your creation and talk about in incessantly (I do at least). 3D printing is not easy either, but it's a labor of love. Besides, you get to make cool stuff and it's an awesome feeling. No matter what you want your printer for, it's yours. Make it!!! I'm rather fond of a a website called Thingiverse that allows you to upload or download 3D printable designs for you to make. You can also use that software from earlier to design 3D printable parts for your own personal use.

To make the actual printing commands, you need a slicing software to convert from models to g-code. I personally use Simplify3D for all of my stuff, but it costs a decent amount of money and you may want to look into one of the many free options such as pronterface, Slic3r, or KISSlicer. There will be a learning curve, but 3D printing is all about tinkering with the settings.

This isn't so much a step as a reward. Go out there and enjoy your new creation. It'll be frustrating at times, but it's yours and wonderful. Make things!

3 People Made This Project!


  • Fix It! Contest

    Fix It! Contest
  • Tiny Home Contest

    Tiny Home Contest
  • Creative Misuse Contest

    Creative Misuse Contest

22 Discussions

can u email me each and everything pls

Great! Thanks. I understand now and will try follow these steps.
Great job!
Keep it up

Guys, this is great .

I understand you hooked up steppers and limit switches to the Rumba board. What do you use to actually print - what software ? thanks

3 replies

We used a Megatronics v2.0 from sainsmart with modified marlin firmware. We use Simplify 3D for the slicing. Does that answer your question?

I see so you have the hot head that can run 3 fillaments? How do use that? What software lets you controll multiple extruders? And how do you design the differnt colors?

I'm pretty sure it's very similar programatically as having two extruders (I'll have my friend fact check me in the morning). This end also allows color mixing, which is a little more complicated, but it's made by reprap, so they have tons of documentation. You just have to have a board which allows 6 servos total (3 movement, 3 feeders). There is also a facebook group for the diamond hotend, and they've been more than happy to help people out.

In simplify3D, you just add another tool in the process settings, so it's pretty easy. I haven't done anything in multiple colors yet on this printer, but the process I use with my flashforge dreamer is:

1. Design an assembly in Solidworks

2. Save assembly as an STL (make sure you export them as separate files)

3. Import all the component files into my slicer (simplify 3D)

4. Control click center and arrange or make sure that they all have the same origin

5. Create a custom process for each part and only have each process apply to the specific colored part

6. Print using all of the processes

The important thing is that you make parts that are separate files with a common origin that lets them line up. I'm not sure how this works in other programs, but I'm sure you can find the proper way to make assemblies in whatever program you use. Does that help?

I love that you went through your thought process and resources - instead of just handing us the plans and parts. That's more valuable for a tinkerer than just the BOM. Good job, and keep it up. That said, what would your software / programming buddy say about increasing the resolution - what hardware would need to be upgraded, and software-wise, too?

I also like your design where instead of the print heads moving up and down the platform does that instead.

2 replies

There are a couple of limitations that limit resolution. The FDM method is limited to a certain accuracy that we cannot get around (hence the development of other printing techniques such as SLA, DLP, SLS, etc) The actual nozzle of an FDM printer has a minimum size (ours is a fairly standard 0.4 mm). Printers cannot accurately print smaller than the diameter of their nozell. That being said there are several hardware factors as well as a few software settings that affect resolution.

Electronics wise the type of stepper motors and stepper drivers used will affect your resolution. Stepper motors have a limited accuracy (their step size), the stepper drivers have different microstepping settings (typically 1/16 or 1/32 in 3D printers and other CNC applications).

Software wise (and firmware) there are a couple of settings that will reduce the resolution of the printer. Generally the slower you print the better quality you will get (up to a certain point - reducing speed then becomes superfluous or even detrimental). Reducing the acceleration and jerk settings will have similar effects. The print speed, acceleration, and jerk settings are included in both the slicing software and the firmware (or some combination thereof). Those settings will be accessable in any decent slicing program. I personally love simplify3D but that is a fairly expensive solution. Slic3r (included with a lot of the common printer interface software packages - Reptier Host, Pronterface, etc) is a decent introduction to the world of slicing settings - I would definately spend some time tinkering with settings in those programs before spending money on simplify3D.

From a mechanical standpoint you want to have as little play as possible in your moving parts. This means you want to have very high quality linear bearings, tight drive belts, and to have as light a print head as possible (reducing the print head mass will also let you print the same quality at a higher speed or vise versa). Sorry for the book, this is only a not so brief tl;dr - there is a ton more info if you dig around on the Reprap website (its a bit difficult to find stuff I will admit). You can also take a look on the simplify3D (the slicer we use) website.

Thanks! I wasn't sure how well it would go over since I haven't actually made an instructable before, but I'm glad people are finding it useful! There's enough specific guides out there for other designs, a designing guide can be nice for the tinkerer's out there.

Honestly the resolution is more of a mechanical and settings issue than a hardware one. I bought cheap linear bearings instead of shelling out the extra dollar or two for nicer ones. Since taking those pictures, we've slowed the printer down to 2000 mm/min, played with extrusion and layer settings and tightened the hotend heat-sinks to stop a leaking issue. I'll have my buddy fact check me on this, but I think the only hardware upgrade that would make much of a difference would be the stepper drivers. For my slicing software, I've had very good luck with simplify3D (even though it's kinda expensive) since it give you a lot of options, but its pretty idiot-proof to use (which I need sometimes).

That design just seemed more space efficient, simpler, and easier to tinker with to me. I know having the hotend move up and down is common with reprap stuff, but I just don't like it to be honest.

Finally! Ive wanted a 3d printer so bad, but they are just so expensive. Good job! I love it!

NEMA 17 bipolar or unipolar? And how much torque of your stepper motor b/c NEMA 17 is a series. It don't specifying about it's torque.

Hii. I am also working on 3d printer for my final year project. But I am confused about the torque of stepper motor and the type of driver and ramp. Can u please tell me the specification of these things that u use in your project?
Please help me buddy I have only 1 month dead line to submit my project.

2 replies

We used pololu DRV8825 based stepper drivers @ 1/32 microstepping ( The motors were NEMA 17 ( and NEMA 23 ( form factors. By ramp I am guessing you are talking about a RAMPS board. We used a Megatronics v2.0 because it can support 6 individual stepper motors (so we could handle a triple extruder). Lemme know if you need more info and good luck!

My friend is the guru for those, I'll have him answer your question when he gets back. We used Nema 17 servos for the xy axis and a Nema 23 for the z axis (I'm pretty sure they all came from openbuilds). I don't know much about the drivers. We based what we bought off of the drivers that came with the diamond hotend, they were a bit overpriced though, so we just bought similar ones off of amazon.

Sorry I can't help more, again, my friend will be able to answer those later today.

Hii. I am also working on 3d printer for my final year project. But I am confused about the torque of stepper motor and the type of driver and ramp. Can u please tell me the specification of these things that u use in your project?
Please help me buddy I have only 1 month dead line to submit my project.

D raj

2 years ago

Can u provide an instructable for ur printer if yes plz rply

1 reply

I plan to post an in depth instructable for my design, but our documentation just isn't where I want it to be yet, and there's a bit that I still need to take care of. The total cost was somewhere around $1500. That bought us a diamond hotend triple extruder and a 1'x1'x3' build volume. I can maybe help you get started if you're interested and want to start sooner rather than later.

good job guys! it's kind of funny, as a parent I read this and just grinned at how proud your parents must be to have young man that they raised with that much fortitude! that tickles me more than your project matter even though the geek in me would like to have one of those myself! I can't imagine either of you being anything less then very successful in your life! once again, good job

1 reply

Thanks, it's been a long road and both our sets of parents are awesome. They've actually been VERY supportive of our projects (I was raised around tools, so it was kinda my dad's fault). College has taught me just how nice supportive parents and family are. The jury is still out on how well our careers will go, but engineering is awesome and pays well, so I'm feeling positive.

Seriously though, if you wish you had one, you can get them for under $300. It can be a time consuming hobby (they're just not easy to use yet) but I personally love it. Most people who spend time with me know that I talk about it WAY too much. Eventually I'll make a step by step for our specific design, but it's a bit on the expensive side, we still want to tinker with some stuff on it, and the documentation is a bit of a train wreck right now (keeping CAD files organized is a nightmare once you're building).