Designing my own new printer

Now that I'm familiar with the operation of a 3D printer through the use of my prusa, I'm looking to design my own printer. Not because of any shortcomings of current designs, but just as a challenge to myself; to understand and to solve the issues and problems that come along the way. I'm sure there are many things I'm not thinking of right now, but I look forward to encountering them and solving them.

My initial thought is to minimize the movement of the print bed. On some prints, the oscillation of the print bed is quite dramatic, and I can only think that this would only weaken the adhesion to the bed. So my initial idea is to build an XY moving print head.carriage at the top of the printer with an print bed that gradually lowers.

For the XY plane, my plan was to build out of 20mm 80/20 extrusions along with the rails and wheels from openbuilds.com. I haven't really worked out the bed lowering design yet, but I've already ordered some components so I can start experimenting.

Does anybody have any critiques to this idea?

The only bit of advice that I can give is to make a 3D model, and to design using that. Being able to work and experiment digitally is so much better than doing it physically. No material cost, easily undo mistakes, and the ability to save different iterations is the bee's knees.

My only issue here is that other than openscad, I am not conversant with anything with CAD in its name. I've tried to use freecad, and I even have a free license for autodesk as I am a mentor for my local high school's FIRST robotics team, but they baffle me. I have a technical background - I am a software engineer by profession, but something about CAD programs still proves to be elusive. If I could get over the initial learning hurdle - and believe me I want to - then I agree with modeling as the good first step.

If someone could give me a pointer to a good tutorial, I'd appreciate it.

Try inventor fusion - there are quite a few tutorials to get you started. And it's free.

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

Much the same as you, happy with my reprappro mendel, fancied designing something myself, wanted the bed Z only and XY running on extrusions.
XY on this has tested great (running a print with it piggybacked off my Mendel), ran a few hours no bother at all, there's still loads to do yet just in case anyone fingers my missing bits

Nice design - quite different from what I was thinking of. My current thinking is to have the X axis much as it is in the prusa, but instead of smooth rods, I would use extrusions with rails mounted on them. This entire assembly would then essentially make up the Y carriage. I suppose that's a lot of mass to be moving around, but my present prusa Y carriage with a 6mm x 230mm x 230mm aluminum sub bed is fairly heayy as well and it works fine. Your approach has a very light carriage, but it looks as though it requires 2 steppers for each the X and Y axes. Is this true? Not that it's a bad thing, but I'm trying to understand the trade-offs.

@ Tommo

Those wheels you are using are not suited for high precision linear motion. They are not self-centering. I suspect when you try to print something you are going to have huge positioning repeatability issues because of the slop between the wheels and the extrusions.

I was planning on using these wheels and these rails.

I'm just not sure if the wheels would be better between the rails, or the other way around. I'm presently leaning towards wheels on the inside because then the carriage could be somewhat co-planar with the track. Otherwise, the carriage would need to be above or below the track in order to mount the wheels on the outside.

!!Yea, 2 motors cross corners for X and Y, as I'm not having to move the motors so weight's not a problem, I could have used a single motor to drive both ends of the carriage but this solution seems to work well so far, keeping the moving weight down without designing something I can't actually make has been .... Interesting.

I Just use sketchup, it's not great but it's quick and easy to knock something up, that way I don't worry to much about scrapping the idea and starting again, spend to long and it's hard to let go of a bad idea, when I'm happy it's re done in Autocad (for the machined bits at least).

I think I have seen a gantry style XY like I think your suggesting, possibly somewhere on 3ders.org, I expect there's a page or two in this forum covering pro's and con's for that design, almost every question I have had iv'e answered with a search here.

@crispy1, you can't really see the runner part in the photo but it's actualy in two parts with the inner spring loaded to the outer so the wheels don't need to center, the whole thing centers between the runners, the wheels can float within the channel without it affecting the possition of the carriage, I didn't want to use the inner edges of the channel for anything other than a guide, not a bearing surface.

akhlut Wrote:
-------------------------------------------------------
> Try inventor fusion - there are quite a few
> tutorials to get you started. And it's free.


akhlut, are you sure it is free?
Just asking as I have downloaded it from their site and thought it said something like PC version expires 1st of April 2013 but the Mac version has no expiration.
I really hope I am wrong as I do like IF.
Regards
Jan

You're right - I just went to the web site too and it's only free for another month. I might be able to use it longer because I have a legitimate license to the full product, but I haven't tried yet.

jbernardis Wrote:
-------------------------------------------------------
> My only issue here is that other than openscad, I
> am not conversant with anything with CAD in its
> name. I've tried to use freecad, and I even have
> a free license for autodesk as I am a mentor for
> my local high school's FIRST robotics team, but
> they baffle me. I have a technical background - I
> am a software engineer by profession, but
> something about CAD programs still proves to be
> elusive.

It is absolutely worth the investment to become proficient in a proper MCAD package.

Yeah, I was hesitant with that too. Asked on their forums and was told by an employee that:


Community Mailer
Jan 20

to me
Hello akhlut,
You have received a private message in the Autodesk Discussion Groups.

From: schneik

Subject: Fusion after April

Body:
There will be an update to the Fusion 2013 download in late March that will remove the time out so that the software runs indefinetly like the Mac App store version.

We will are moving new development to Fusion 360 which will be a pay for product but you can continue to use the 2013 version as is for free.

-Kevin Schneider

Autodesk, inc.


So it looks like it's worth learning!

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

jbernardis Wrote:
-------------------------------------------------------
> My only issue here is that other than openscad, I
> am not conversant with anything with CAD in its
> name. I've tried to use freecad, and I even have
> a free license for autodesk as I am a mentor for
> my local high school's FIRST robotics team, but
> they baffle me. I have a technical background - I
> am a software engineer by profession, but
> something about CAD programs still proves to be
> elusive. If I could get over the initial learning
> hurdle - and believe me I want to - then I agree
> with modeling as the good first step.
>
> If someone could give me a pointer to a good
> tutorial, I'd appreciate it.

See if you can get an educational license for Rhino. I have a full seat of Rhino 3 and used it everyday for almost 8 years, but I currently work in Solid Edge and learning Solid Works. When I first came across Rhino I had tried a number of low cost 3D programs, and couldn't get anywhere with them, even though I was very proficient in Generic 3D years before. Rhino really clicked with me. I started modeling the product my boss and I were developing in Rhino and after about 3 days I was able to show him a nearly complete 3D model of something I had working on for months in 2D. That convinced us to buy a seat and never look back. There is a 30 day full trial version available and very generous educational deals. I still hang out on the newsgroup, even though I haven't used the software on a daily basis in nearly 10 years!

Gary H. Lucas

akhlut Wrote:
-------------------------------------------------------
> Yeah, I was hesitant with that too. Asked on
> their forums and was told by an employee that:
>
>
> Community Mailer
> Jan 20
>
> to me
> Hello akhlut,
> You have received a private message in the
> Autodesk Discussion Groups.
>
> From: schneik
>
> Subject: Fusion after April
>
> Body:
> There will be an update to the Fusion 2013
> download in late March that will remove the time
> out so that the software runs indefinetly like the
> Mac App store version.
>
> We will are moving new development to Fusion 360
> which will be a pay for product but you can
> continue to use the 2013 version as is for free.
>
> -Kevin Schneider
>
> Autodesk, inc.
>
>
> So it looks like it's worth learning!


Thank you akhlut!

Yeah, I was pretty happy when I got the message!

Now to pry myself away from 123D Beta9. It's hard to switch away from a tool you know so well...

---

- akhlut

Just remember - Iterate, Iterate, Iterate!

[myhomelessmind.blogspot.com]

Try creo elements - you have to register and say "I'm a student" but you don't have to give any more details. It's a bit like sketchup in operation but much much more powerful. Only downside is it only saves to a proprietary format or stl. (unless you pay monies!) It's made by the same people as pro engineer.

Look into Alibre for a cheap parametric modeler. There should be good learning resources for that. CAD is not something you pick up instantly. Do not start out trying to design something. Do all the tutorials and training exercises available before trying to get work done. After that, your first few projects will still be a little frustrating, and poorly organized. After that, you will have some proficiency.

With or without CAD, you will find that planning is greatly enhanced by doing lots of sketching of ideas. The pencil is still faster than any CAD software. Sketch on a pad of graph paper so you can maintain consistent scale when seeing if things fit in the space you visualized. You can iterate through a lot of ideas quickly before trying to set out precise geometry in CAD or spending money building ideas that don't work out.

Quote

It's made by the same people as pro engineer.

Given my experience (in a professional setting) with Pro E, this is pretty much the opposite of a glowing endorsement.

... a really interesting free program for 'redesigning' STL-files is MeshMixer: [www.meshmixer.com]

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

jbernardis Wrote:
-
> If someone could give me a pointer to a good
> tutorial, I'd appreciate it.

There are some excellent tutorials on Youtube for Cad programs like Openscad and other programs.
All you have to do is sit back and watch them. You can even download them and keep them on your computer for future reference. Most start at the very beginning level and get more complicated when you're ready. Many questions you wouldn't think to ask are answered.