[3D Printing] 3D Printing Superthread!

[Max McGrumpy]

I've been a member off and on here over the years (George knows, I'm not skirting a ban or anything)... Over the last few I've been increasingly invested in 3D printing and what it can do for me and for society in general. I've put forth a lot of energy to create a Facebook technical group and a budding website that does independent reviews of FFF hardware, machines and material. On top of that, as I coach and teach people I learn new things all the time.

I noticed there are at least a handful of us on here that are dabbling, I remember a fellow from last year I helped him get a Robo R1+ setup and printing. Since I've been back I noticed two separate conversations buried in other threads... useful information but it won't be easy to find it again.

So here's what I propose: Those of us looking to ask specific questions on 3D printing and share our experiences please do so in this thread. If you bring it up in another thread, post a link to it here so we can reference it later!

I've tagged this thread so it should come up in search later on.

We're not quite to the point where we need a Homeshop 3D Printing Forum (or magazine! ) and we may never get there but this will help us consolidate the information and make it slightly easier to find as we go forward.

[George Bulliss]

Okay, I’ve got a question. I’ve been gathering pieces with plans to build my own printer. Probably won’t happen this winter, but I will be getting started on the design soon.

Originally I had planned on building a gantry style system, but have been seeing more printers with the X and Y on top and the Z handled by raising or lowering the table. Seems like it would make sense to keep the hot end movement all in one plane, but I’m not sure if there are other reasons for going this route.

Thoughts on the merits of either design?

[Max McGrumpy]

A lot of people prefer the Z-drop design with the table that raises and lowers. You are only accelerating/decelerating in one direction, usually down, and only a few steps at a time. This is in contrast to a moving table that has to accelerate/decelerate the entire mass of the table in multiple directions rapidly. Z-drops tend to be more accurate because there is less chance of losing steps and drifting out of position. I've tuned each of my moving table machines to the point where that rarely happens but it still does. The Z-drop ones, I've never had one miss a step. Well one time and that was because the driver on the board overheated. I replaced the heatsink and it has been fine since.

I have a representative sample of each, and a Delta-style system with a fixed table. The only reason I don't have more Z-drop machines is the sample units I've been testing have been all moving tables. I have two, and they do very good work once dialed in.


Now to confuse it even more, you'll want to look at Bowden vs Direct Drive extrusion. Bowden puts the extruder motor in a remote location and your head is just the hot-end/fan assembly. Direct drive you feed the material into a feed cog and it pushes directly into the hot-end. Bowden has the advantage of being a lighter head, but the downside is the backlash in the system makes getting really good retract (dribble control) challenging. And you have a harder time running soft flex-materials like TPU. Direct-drive you have a heavier head, but better retract and most setups can be tweeked to run flex materials with very little effort.

[DEVILHUNTER]

I don't think there is any problem with either ways, just make it the easiest way for you. In my opinion, small extruded aluminium profiles is the way to go.

This christmas I mounted one with my brother. It was only 190€ with shipping from Germany to Spain, incredible cheap. Even came with a roll of filament. The power supply didn't last long, but an ATX PSU and working again. The structure is very weak but even with that I were extremelly surprised on how precise we were able to get the parts.

Funny story is that, instead of a hot plate, we were using just a glass with some laca. The point is that everybody recomends a specific brand of laca, Nelly. We tried other and didn't worked, then we used the Nelly one and we had to make some efforts to take the pieces out off the glass. Is there anything similar over the rest of the world?

By the way, wood fiber filament is awsome. The look on the parts is very reallistic and even smells like burned wood when you are printing.

[George Bulliss]

Thanks, I'll probably be headed in the Z-drop direction, when I get around to it. The direct drive sounds like the way to go too. It shouldn't be too hard to handle the weight if it's only moving in X and Y.

Next question, I have a box of NEMA 17s and 23s. I know most printers are done with 17s, but my 23s are better quality. Any reason not to use the larger motor?

[Max McGrumpy]

I've found that Aquanet Extra Hold hair-spray works as a bed coating very well for printing most novelty items, toys and gimmies. The Aquanet brand has a poly-vinyl adhesive that grabs really well, yet releases the part. And it washes off with water.

For other more challenging prints I use gluestick of a certain formulation, just common every-day gluestick. And when I can't use either I put down blue painter's tape. The 3M brand has a specific surface finish that transfers to the bottom of the part and the adhesion to the bed is excellent. Other brands work too, it's a question of trial and error.

There are other bed fixing options like mirror-glass, PEI sheet, kapton sheet, and many others! All have advantages and disadvantages. I'm of the mind that if it works for you it's the right way. I don't push any particular school of thought or brand when it comes to this technology -- keeping an open mind allows me to learn new stuff and keep it interesting.

[Max McGrumpy]

Thanks, I'll probably be headed in the Z-drop direction, when I get around to it. The direct drive sounds like the way to go too. It shouldn't be too hard to handle the weight if it's only moving in X and Y.

Frankly with good tuning I don't see a difference. I prefer direct drive for the simplicity.

Next question, I have a box of NEMA 17s and 23s. I know most printers are done with 17s, but my 23s are better quality. Any reason not to use the larger motor?

Most kits are sourced cheaply. That means smaller weaker components because that's what the cheaper flimsier drivers on the less-quality clone-board can handle because the ratcrap power-supply isn't capable of putting out what the plate claims.

Sourcing your own components and with the knowledge you have, you'll be able to build something better than a pre-fab kit by a huge margin.

[enginuity]

Starting this week I'm going to be building a 3D printer with a group of students for a local elementary (middle / intermediate) school as a way to introduce them to technology, and as relatively safe alternative to machining. My hope is to get them interested in robots and just making stuff.

I'll admit I was lazy and just bought a kit, primarily because while I know I could build it myself, the kit was convenient, and forced me to do the project, and I didn't have time to put all the pieces together between the time I decided to do the club and when it starts.

George - from my understanding stepper motor size and selection is a more than just pick something - although I'm not familiar with the intimate details. A 3D printer needs speed - something that requires some tuning. What I do know is 3D printers really should be using servo motors with encoder feedback to get the speed up - especially on the high motion XY portion. (also I finally subscribed to HSM, not only for a great magazine but to also show my support for this forum and its great contribution to the hobby).

Has anyone come out with a reasonably priced servo motor option for 3D printers?

[Max McGrumpy]

Not that I am aware of... not at the level I work at anyway. It is probable that the high-end professional and production machines have the technology.

I'd love a servo driven delta-printer...

[DEVILHUNTER]

Has anyone come out with a reasonably priced servo motor option for 3D printers?

There is a new open source servo driver that fits on the back of NEMA 17 motors. Can be also used in NEMA 23 but is limited to 2A. It uses a 14bit magnetic encoder and the same 48MHz microcontroller as the Arduino Zero. It's called Mecaduino:

http://tropical-labs.com/index.php/mechaduino

In fact I found it about 2 weeks ago while looking for info about the encoder and it was very usefull. Yesterday I got the encoder but I need to buy some PCBs first to test it. I have been reading all the code and is going to be very usefull for my project (BLDC servo drive). I will probably get something similar for my CNC mill.

[enginuity]

There is a new open source servo driver that fits on the back of NEMA 17 motors. Can be also used in NEMA 23 but is limited to 2A. It uses a 14bit magnetic encoder and the same 48MHz microcontroller as the Arduino Zero. It's called Mecaduino:

http://tropical-labs.com/index.php/mechaduino

In fact I found it about 2 weeks ago while looking for info about the encoder and it was very usefull. Yesterday I got the encoder but I need to buy some PCBs first to test it. I have been reading all the code and is going to be very usefull for my project (BLDC servo drive). I will probably get something similar for my CNC mill.

Wow, that's one of the more exciting projects I've seen in awhile. Thanks for posting it.

[Max McGrumpy]

I'll be sharing that with some folks in my other group -- they'll know how to make use of that.

Thank you!

[Black Forest]

I think if you want this to be a go to thread then the first thing you need is a glossary of terms.

As in what the hell is a Z-drop?

[reggie_obe]

I think if you want this to be a go to thread then the first thing you need is a glossary of terms.

As in what the hell is a Z-drop?

Latest slang term for an old social disease.

[Max McGrumpy]

A table that advances (drops) down, driven by the stepper motor on the Z-axis.

Machine prints a layer, table drops .2mm (or whatever you have it set for...) and begins printing the next layer.

Now the other method is to move the entire head assembly up and down by various means.

At this point with proper tuning and programming they are indistinguishable in terms of output, it's a matter of preference.