Modding Project: Prometheus (DIY 3D Printer)-August 21, 2011- Replication!

So after years of trying to build things with only a meager selection of tools, I have finally decided to start gathering (and building) the tools necessary for my own workshop. I am also in the process of getting all of my stuff out of storage and trying to finish up a few other projects, so I can't promise super frequent updates, but I will try to keep everyone up to date on major progress.

Some of you may have noticed that there are a few open source projects going around that involve rapid prototyping. I have always had an interest in rapid prototyping and the various ways to go about it. You can categorize just about everything into 2 styles: Additive prototyping, and subtractive prototyping. This project is going to be based on the former.

Some of the 3D printer groups out there are:
RepRap and its cousin project MakerBot
Fab@Home

I have finally jumped on board and will be participating in building myself an FDM (Fused Deposition Modeling) 3D printer based on the work of the RepRap group. The latest design is referred to as "Mendel" and is the second generation of their 3d printer.

What a completed Mendel looks like


One of the unique aspects about the RepRap group's efforts is that the machines are designed to print out their own components. They ultimately hope to be able to build a complete self replicating system but that is still a ways off. However the current design makes extensive use of components that are meant to be created using a rapid prototype machine. Now this does leave us with a chicken and egg situation, there are a couple ways to go about solving this problem.

1. Talk to a friend with a 3D printer or other rapid prototyping system and have them make the parts for you
2. Make your own through some other means (mill, Dremel, whittle, hammer and chisel, etc)
3. Purchase or otherwise acquire a set of parts made by one of the many small businesses that are growing around these communities.

Lucky for me I am in the fortunate position of situation 1 so this makes my life much easier. The rest of the parts are pretty much off the shelf or easily sourced through online retailers.

And so the build begins...

Index of Updates:
July 29, 2010 - Starting to Rapid Prototype the parts
July 30, 2010 - Making progress with the printing and an explanation of some of the mechanics
August 2, 2010 - Assembly of the X-axis
August 19, 2010 - Assembly of portions of the Y-axis and Z-axis
September 5, 2010 - Assembly of the frame
September 10, 2010 - The Extruder
September 20, 2010 - Finishing the Y-axis and adding the drive belts
February 4, 2011 - Wiring up the electronics, and keeping them cool.
February 5, 2011 - Wiring the extruder, and running the first few prints.
February 12, 2011 - Building an enclosure for the printer.
April 22, 2011 - Working on the enclosure some more, along with a few more prints
August 7, 2011 - Back at it again.
August 21, 2011 - Replication!

 

Starting to Rapid Prototype the parts

July 29, 2010

Since this little build uses lots of rather complex parts, the best way to get them is to have them formed using a rapid prototyping process. In my case a commercial 3D Printer. So I have started printing off the parts while I gather up the rest of the parts needed for the printer.

He we are starting the first batch of parts that need to be rapid prototyped


About 4 hours later


Starting up the 2nd batch

 

nice!

FDM is ok, but surface finish isn't that great.

 

I read the issue of Make (I think it was the issue-before-last) which was all about 3D printing with much jealousy. I think I'd better finish my CNC router first, though

 

Yeah it was a hard decision for me between a router and printer, but I hope to make some of the components for my router with this printer. Though it would be nice to have a CNC router to make some the parts for this printer *sigh* the never ending run around of our hobbies.

 

Machine self replication and 3D printers are fascinating! This is a nice start to a project with a lot of potential - I'm gonna follow this one

 

nice, i really have to think about building a printer head for my cnc

 

Making progress with the printing and an explanation of some of the mechanics

It would not be that hard to do, in fact there are quite a few people who have done so. When someone fits a print head to an existing setup it is usually referred to as a RepStrap by the RepRap community (in case you want to look it up).

Anyways, it is time for a bit of an update:

July 30, 2010

I am still in the process of printing out parts for my printer, but I can now say that I am about 50% complete with the printing and hope to have them all done by Sunday or Monday. Which brings me to the subject of the mechanics of the printer.

The system is built with 3 axes of movement: X, Y, and Z. The first generation of the RepRap group 3D printer (referred to as Darwin) also used a 3 axes implementation, but differed in how the system handled the movement of the axes. Darwin moved the print head on the X and Y axis, while the whole print platform traveled up and down for the Z. The Mendel (what I am building) moves the print head on the X and Z axes, and moves the bed for the Y travel. There are several advantages to this:

1. Now there is much less strain on the motors that will be driving the system. It is much easier to move just the print head up and down as opposed to the whole print platform. And the linear strain on the Y axis isn't that much since now we are just moving the bed along rails while sitting on top of a ball bearing assembly.

2. The potential for future upgrades: The Mendel was designed to be a bit more upgradeable and flexible in its design, in comparison to Darwin. It was created so that in future revisions you could add an auto tool changer that could swap out the print heads on the fly or switch to a router or any other attachment you can come up with and code in. There has even been some experimentation in printing circuit boards with modified extruders that lay down low temperature alloys. Though the boards were messy and nothing fancy, it was still impressive to see a home made machine print out a crude circuit board and then lay down "traces" of metal.

3. The new design is much more simple and requires less parts, which is always a plus. Also, it is stronger and takes up less space while still retaining nearly the same capacity of the first generation machine.

Ok, enough reading on with the pictures.

I opened up the printer this morning and found my 4th batch of parts completed


These two pictures are all of the parts I have printed so far



These are the rods that I will be using for my slides for my X,Y and Z axes. They are 5/16" BZP (Bright Zinc Plate) rods. Unfortunately in a streak of absentminded measuring I cut one of the rods 1cm too short. But upon closer examination of the designs I should still be able to assemble my printer. Besides, I am planning on building a 2nd printer to accompany the first and at that time I will put in A2 steel rods in both of them (smoother finish and all around a better choice of material)


Here are the bearings that will be used all over the machine, yes there are a lot, and yes I will use just about all of them.

 

Wait till it get AI and start replicating killer robots, then it won't be so "fascinating"





Fo reals, this project is really really cool, I will be following this one

 

These rule. I'd have to pass on Lenne's idea of a print head for the CNC. I don't have big enough balls to modify my machine yet. -Even if it is just a parallel cable chop...
I'll be watching this closely too.

 

Assembly of the X-axis

August 2, 2010

Well this post is going to be mostly pictures so here we go! Feel free to check out the album for even more pictures.

There was a slight hiccup in the process of printing out the parts for my 3D printer and as such they will not be complete for another day or two as we had to rebuild a full print tray, but it is a minor problem and has only cost me a little time. Luckily I have enough parts now to begin the assembly of the printer.

At least I got to pick up another tray's worth of components.


In other news I purchased the fasteners that will be used to assemble my printer (actually that is only 1/2 of them)


Here are the parts to build the carriage assembly for the X-axis


The bearings are sandwiched between washers to keep gunk out and also to keep them from binding on the frame.


Nearing completion of the X-axis carriage


Alot of the frame was printed to trap nuts for easier assembly


6 bearings on the right, 4 on the left, constrain the carriage


It is remarkable how tight of a fit it is, there is almost no play at all, the carriage will be further tightened after the axis is assembled


The beginning of one of the idlers that will hold the belt that drives the carriage and also acts as an endpoint to the X-axis assembly


That idler almost completed


Part of the idler assembly, also this contains the nut trap that transfers the rotational motion of the threaded rod into linear motion of the Z-axis


These parts will make up a portion of the other idler assembly and also contain the majority of the bearings that will constrain the movement of the Z-axis. Along with another nut trap to transfer motion.


The parts of the previous picture assembled (check my album for intermediate pictures)


Here is the rest of the 2nd idler assembly, it contains the motor that will be driving the X-axis.


The large mudguard/fender washers function as guides for the X-axis drive belt that will be running over the bearings (under the washers)


Putting together all of the assemblies that make up the X-axis


Almost there


The completed X-axis assembly (for intermediate pictures check my album)


The belt and drive gear will be fitted when the other main assemblies are complete and the printer is assembled as a whole.

 

EEEK! My head is spinning just looking at that. I'd hate to be the one reading the assembly instructions.

 

Assembly of portions of the Y-axis and Z-axis

August 19, 2010

Hello, time for another update. This time I have been working on some of the assemblies that make up the Y-axis, as well as some of the Z-axis. Comments and questions are always welcome, and here are the pictures:

This will be the motor bracket for the Y-axis


The large washers cover the bearings and act as guides for the belt, which will be installed later


This will be the other bracket for the other end of the Y-axis


The bearings are sandwiched between smaller washers as well


These parts will make 2 of the 4 assemblies for the chassis for the Y-axis


Two completed assemblies


Three done


Fourth and final assembly for Y-axis chassis


These will be the base of the Z-axis


I need to find some better washers for this particular part


For more pictures see the album.

 

Assembly of the frame

September 5, 2010

I finally took shipment of my PTFE (Teflon) stock, now I can finish the mechanical construction of the printer. So a bit about the frame, this printer has been designed to be built from as many of the shelf parts as possible. There is extensive use of BZP (bright zinc plate) threaded rod for the structure of the printer. It is also used for a pair of leadscrews to drive the Z-axis up and down.

Since I was going to be working around this printer a lot after it is completed I have taken the time to remove any sharp edges that might cut or scrape anyone. So after cutting all of my threaded rod and regular rod, I chucked it in my lathe and slightly rounded the ends of every rod. Now there are no sharp edges to cut me (or anyone else).

Before


Trimmed down to size


The frame of the printer is built around 2 equilateral triangles. The framing of these two triangles serves as the base mounting point for all of the other hardware that makes up the printer. Once fine tuned and adjusted, the frame is extremely rigid and strong. The threaded rod and the smooth rod are the only imperial parts in the whole build being 5/16". Every other component is metric, and future child printers will be entirely metric as well.



You can see how it is starting to take shape and how all of the motors will function in the printer.


At this point I pulled out my caliper and triple checked all of my spacing and tightened it all with liberal use of a thread locking compound (Loctite blue). Then I checked the squaring of the build. It really pays off to take your time in the beginning and make sure that everything is accurate as possible, so that you don't have to try to backtrack afterwards and try to figure out why your printer has a bad case of vertigo.


Then it was time to insert the guide rails, the leadscrews for the Z-axis, and attach the X-axis assembly (I accidentally had it upside down, it is fixed in later pictures).


So a bit about the motors:

These motors are Nema 17 Bi-Polar stepper motors with a step increment of 1.8 degrees. Stepper motors differ from your run of the mill DC motor in that you don't just apply power to the lines to get them turning. A stepper motor has a series of coils that must be energized in the proper polarity and sequence to get them to rotate. These motors will be driven using Allegro’s A4983 chip, which is capable of 16th micro-stepping and delivers up to 2A of current.

The motors


One of four stepper controllers


The motors have smooth shafts and the gears are held on by a bolt through a trapped nut in the gear itself. So it is necessary to cut small flats into the shaft to give the bolt something to bite into.

 

The Extruder

September 10, 2010

As some of you may have already noticed, the pictures in my album usually precede my posts to this thread, so keep an eye on the photo album if you want to get a sneak peak at what is coming up. Now on with the update.

The extruder is probably the most interesting part of the whole mechanical structure. It is the device that takes in a 3mm filament of plastic and extrudes a .6mm bead of molten plastic; Which is hot enough to fuse to the layer of plastic below and bond to form the next layer of the model. This requires a fair bit of heat and quite a bit of torque to feed the plastic through the print head. But this is where there is a conundrum as the printer is designed to print most of its own components. So the question is, how do you build an extruder capable of melting plastic to print parts without melting the already printed components?

The solution is to use a machined part that can withstand the temperatures. There are many styles of extruders that people have come up with to achieve this, but I am going to be going with PTFE (Teflon). Due to its extremely low coefficient of friction, any back-flow that may occur as the print head idles doesn't stick in the barrel, and just gets forced back into the nozzle as the feed resumes. The extruder is fed by a system that "pinches" the filament to force it through the heater assembly. The pictures will give you a better picture of what is going on. If it is still not clear or you want more pictures of a specific part of the printer just ask and I will see what I can do.

Here are most of the parts before I received the Teflon.


The assembly is geared to offer more torque.


The springs and bearing that help "pinch" the filament and keep a constant pressure so that the opposite side will have something to grip.


I ground a small channel into a M8 bolt and then used my tap to add the markings that will grip and advance the print material.


Bearings that will support the bolt and allow it to rotate freely.


With the bolt inserted.


Now you can get an idea of how this thing works.


The bolts thread into the plastic.


Since I do not have the ability to drill a small enough hole for the nozzle, I made my own design that can use a copper welding tip.


Test fitting the barrel with some filament.


It fits!

 

Awesome project, what's the advantage of the additive vs subtractive prototyping machine?

 

Thanks, each fulfills a specific purpose. Additive rapid prototyping can create hollow parts and very complex designs out of a single part; Where as subtractive rapid prototyping would require multiple parts to achieve a hollow workpiece, and may not be capable of the same complexity.

Subtractive generally creates a better surface finish and can use a wider variety of materials. It can also be more precise depending on your equipment. the finished product generally tends to be stronger due to the part originating from a piece of solid stock. There are many other differentiations between them but that is most of the differences in a nut shell.

 

Thanks for explaining, can't wait to see this finished.

 

Finishing the Y-axis and adding the drive belts

September 20, 2010

This post will consist of the completion of the Y-axis that was started on August 19, 2010. The Y-axis is probably the most complex part of the whole machine. And since I was unable to gain access to a CNC router I ended up just printing out a template and cutting/drilling it by hand. I must say that I am pretty happy with the results considering it was all with simple tools. I only seem to have botched a couple of the holes and it seems that it will still work (though I plan on putting in a better plate eventually.)

From DXF to Masonite.


The assemblies from awhile back and the fasteners needed for finishing it up.


Assembling the y-carriage


The small rectangular blocks will pinch the drive belt, holding it to the carriage.


And now to mount it to the frame.


If you look closely you will notice that the mounting blocks for the Y-axis rails are actually two parts, and they are identical; 8 pieces in total for all of the mounts.


Now all of the belts are mounted in place.


This is the top of the carriage, and will serve as the print platform (another layer of Masonite will be placed on top held with bulldog clips when printing).


The springs and bolts will allow me to level out the platform after it is assembled.


Mounted, but not leveled, that will happen after the build is complete. Also you have a good view of the Z-axis belt tensioner. The Z-axis is the only one that requires a continuous belt and thus needs a tensioner to compensate for different belt sizes.


Edit: I just realized that I had forgotten a few pictures of the belt system. Hopefully these will give you a better understanding of how it works.

 

WOW....this is so cool, good job......my biggest question is how much?