So I'm young, lazy, and broke. What's the cheapest combination of parts to build a reprap. Also assuming I have no access to a lathe, drill press, or other heavy machinery, what would be the easiest to build.

Or maybe I can pay someone to print out the parts and send them to me...

Any Volunteers?

buy the kit from bitsfrombytes.co.uk
and the electronics from store.rrrf.org

Nothing but handtools needed

'sid

Course, that aint very cheap. For cheap and nothing but hand tools needed, go with the electronics from RRRF and then build a McWire cartbot on your own. Don't use acrylic for the stages, yous scrap plywood or MDF. Beg some skater friends for their old 608 skate barrings. Use uncles and grandpas for bolts and such--they always keep nuts and bolts, always. Borrow a soldering iron from someone, you may have to call around to find one.

On the other hand, this isn't really the easy way to do it. You probably can't have cheap and easy at the same time. It'll be cheap or easy.

Demented

I'm in a similar boat. For the past several weeks I've been collecting old printers from friends, businesses, and Freecycle, taking them apart for their motors, opto's, and straight metal rods. I plan to try and build as many of the plastic parts as possible from scrap 2x4 and a hand drill / dremel -- I know this won't work well for many of them (such as the pulleys), but for some of the lower-precision parts (such as the rod-tie-blocks) I think it should work just fine.

I plan on getting an Arduino / Sanguino soon, but I'm sortof holding off on the electronics because I'd like to get version 2 of the stepper motor controllers (though I haven't heard much solid information about them yet -- I'm more just holding back out of hope and ignorance than out of confidence).

I know I could build a McWire, but I'm thinking I might rather try and build it as close to the Darwin as possible, and then slowly replace the wooden components with self-fabbed plastic parts as the machine progresses. So I might be looking to purchase some of the harder-to-hand-fabricate plastic pieces from someone who has an existing RepStrap going already, since ideally I shouldn't need the full plastics kit from BfB. I very well might buy their extruder kit.

--clint

Edited 4 time(s). Last edit at 11/12/2008 05:05PM by HanClinto.

Yeah Clint, looking at some inkjet printers I've destroyed, it occurs to me you could use the print head as is - use two for the Y axis, and one for the X. They even come with their own driver boards, but good luck trying to make use of them. Too bad everything isn't open source!

Z axis is a bit trickier, but maybe you can do something like Nophead's new Z axis with 4 motors; maybe use the paper feed motors from the previously mentioned three printers, plus one more. They're a little smaller than the main stepper, but they might work. That would be only 4 inkejet printers for a Darwin. Plus a lot of work. I found a box of 8 mm printer rods at the local surplus shop, lucky for me.

The printers I've found discarded all had steppers in them; you could use the RRRF electronics with those no problem.

I think I might go with the bits from bytes kit...
anyone know what the shipping is on that to America?

@Wade: So far I've taken apart maybe half-a-dozen printers (including an old electric typewriter and a laser printer). Most of the older printers have large end brackets that brace in the horizontal rods, providing a lot of encumbrance that would get in the way of salvaging the assembly as a usable axis.

Most of the newer printers have smaller assemblies that are quite usable as a standalone axis (some of them just unscrew with a handful of screws and can be lifted out of the plastic as a still-functioning unit), but those generally don't use steppers, but rather a small hobby-sized DC motor with a linear optical encoder strip.

I'm saving a couple of the newer printers to try and make a dental-plaster z-corp style inkjet 3d printer (using the print axis as the x-axis, an old scanner tied into the paper feed as the y-axis, though haven't found a great scrap z-axis).

It seems that the new printers have small enough assemblies to use standalone, but the motors are too small (and aren't steppers). The old printers have steppers, but the assemblies are too bulky to use intact.

Ultimately, I'm just salvaging a few components and recycling the rest.

I'm all ears for cheaper and easier ways -- I'm thinking perhaps I should be writing down the model numbers of the various printers that I'm dismantling, and giving a report of the usefulness of the components inside.

--clint

Edited 2 time(s). Last edit at 11/12/2008 07:28PM by HanClinto.

I'm broke too, so I write software for reprap

Another way of getting a cheap reprap is to do what you can now for the reprap project. Then in several months when someone in your area can reprap the plastic components and is looking for two people to reprap parts for, and the remaining components are cheaper / better than they are now; ask for the parts in return for plastic. Also, since you've contributed in the meantime, they'll likely be willing to help you troubleshoot your assembly.

Enrique

Enrique speaks the truth!

Dom, shipping just the plastic parts to the US is quite reasonable, just get the heavy steel parts locally. That's what I did. Zach in New York has steppers (RRRF) as well, they work great. Be prepared to spend a bit of cash though; as Zach says, this is still experimental. If you want super cheap, wait a year or so till there's a lot of repraps cranking out parts; right now we haven't quite hit critical mass.

See I would wait, but if I can scrape together the cash, I want to be on the cutting edge and be part of the whole experimental phase. That's the most fun time when building things on your own, working until you get it right. I can only imagine that the excitement would be exponentially higher when working on a massive international project like this that could change the world.

dom Wrote:
-------------------------------------------------------
> See I would wait, but if I can scrape together the
> cash, I want to be on the cutting edge and be part
> of the whole experimental phase. That's the most
> fun time when building things on your own, working
> until you get it right. I can only imagine that
> the excitement would be exponentially higher when
> working on a massive international project like
> this that could change the world.
>
While I'm not sure that "massive" is the proper word Reprap certainly has the potential to "change the world" in a rather major way.

I think that possibly what you ought to do if you want to get into development is to sit down and work out (1) what skills you bring to the task, and (2) how fast you learn. I think that Zach might be a good role model that you might want to look at. When he got involved back in 2006, his skill set was limited, to say the least. What he has accomplished, pretty much, as I understand, by teaching himself since then is awe-inspiring.

Always remember the Classic sign on the wall in every design shop:
-----------------

1. Fast

2. Cheap

3. Good

Pick two.

-----------------

For good you can read easy in this case.

Design, documentation and software are useful now, and don't cost money to develop.

You want a path to the cutting edge of reprap development. Not every road is straight.

Enrique

I'd be willing to trade part or even a whole machine in exchange for having work done. I'm making a whole new design, new electronics, new mechanics, new firmware and new software. New everything. This will obviously be a fair amount of work, so I'd like to have some of the work done by other people.

I could split up the tasks, and if any of you wanted to you could take one of the tasks and do it and once we have a finished machine working, I'd send you everything you need to make yourself one, if you take a big enough task.

You'd have to do the task exactly as I want it. I believe in having a lead designer for a project and everyone doing things according to the lead designer's guidance. I'm not a dictator, I just believe in doing things the way that works, the way that has been proven to work by industry.

I can't afford to give out a machine for a small amount of work, but I think I'd be willing to split this into 5 majors tasks and give out 5 machines for that. It'd be required that we finish the machine and get it working. I want people on this project to be motivated to help each other with their tasks, so if we don't get the thing working we don't have machines to give out. So if you help the other people on their tasks, your more likely to get your machine.

Or maybe, if you want, you could just help me as much as you can, not having a specific task, and I'll try to make it worth you while.

I'm trying to design on the basis of making millions of these machines. To reduce the cost as much as possible, so that the people making these things won't have to spend as much. This adds extra design work, extra component sourcing work, extra work in a lot of areas. If you are ok with working, and don't mind extra work, and want to see millions of these things made, help me.

We're going to have linear encoders which resolution around 0.0001 inches, which will cost about $5 each to make. Our main drive motors cost us $1 each. We found toy motors as well that cost $0.17 each, which we'll use for many things. The motors are surplus, so we may in future have to redesign at little, but we'll probably be able to use 50,000 motors each at that price if we hurry, which means a savings for the cause of $5 million or more. I think that justifies redesign work later on. It doesn't matter if other people save the money and not us. As long as the money is saved, it is worth the work.

For programming, we might divide into two major tasks. We'll need a multiplatform (Windows, Linux, etc.) program that both controls the machine locally (sends G-codes, gives status info, views cameras) and also controls a machine via Internet (sends G-codes, gives status info, views cameras), and also interfaces the machine to the Internet to allow internet control, and also allows local limitations to be set on the machine to prevent certain things being done via Internet. The machine will hook up via USB, but it could appear to the computer as either USB or RS232. It would have to be programmed in C++ with Code::Blocks and using wxWindows for multiplatform capability.

That same program will also need to interface with an oscilloscope, frequency and waveform generator, logic analyzer, multimeters, variable power supplies, and other electronic test instruments which will be part of this machine. That could be a separate project though.

This machine I'm making is actually designed primarily as an electronics machine, but it will be accurate enough and rigid enough to do 3D printing and even I think milling with small cutters.

There's other tasks to of course. So if anyone wants to do this, let me know.

Tony

---

Creating the society of the future
[conceivia.com]

Ant/Tony, et al,

Can you give some detail on your linear encoders? They sound good -- but perhaps too good to be true. (I'd be happy to learn otherwise.) Small DC motors are widely available; in my experience, matching gearboxes (of sufficient ratio) are harder to find. Again, I'm interested in learning about your approach.

-- Larry

For the linear encoder we are likely going to use a mouse chip, which is basically an 18x18 pixel video camera. It costs $1.60. To that we add a microcontroller, $0.62 to $0.90, and some communications circuitry for about $0.30. If we need more speed, we might need to add a regular optical interrupter as well, but probably not. The mouse chip has pretty good speed, but if we try for extreme resolution it'll be slower.

We'll also need some sort of stripped tape or something. Can be done on a laser printer, except we need it to be longer than regular paper. The mouse chip would average out inaccuracies in the strips to a fair extent.

I'm thinking we might just use black and white movie film for our stripped tape. We can create a simple machine to put the strips on it very easily.

Tony

I don't know anything about programming, I am OK at 3d design so maybe I'll make designs and post them on the reprap page of download-able designs. I don't know if that will be any help but why not.
I'm better at vector art than 3d but I don't think that helps here!
So if I can help you in any way ant I'd be happy to, but I don't think I can.

Edited 2 time(s). Last edit at 11/16/2008 10:35PM by dom.

It is very important to have confidence in yourself. If you don't think you can do something, you can't do it. If you do think you can do something, you can. It is really that simple.

Take for example Machine Language programming. An 8 bit microcontroller typically has about 50 words in its language. It is much easier than learning French or Spanish or whatever, which have hundreds of thousands of words. Yet in spite of the extreme simplicity of machine language programming, there are few people who are willing to learn it. Most people think it is just to difficult.

When I learned Assembly Language, it took me a month or so before I would accept the fact that it was really that simple. I kept thinking it can't be that simple, 'cause people told me it was very difficult. For a month I kept telling myself it can't be that simple.

Once I accepted that, I was able to program. Very simply, the change from believing I can not do it, to believing I can do it, was a very abrupt change taking place in minute. Before that I could not program in Assembly Language, and after that I could.

Going from Assembly Language to Machine Language was pretty much without a thought, since the two are so similar.

We do need things to be done right by people who know how to do things. I define "knowing how" to mean people who have confidence in their ability. Not to be confused with arrogance, which is actually people who pretend to have confidence. People with arrogance are extremely lacking self confidence, and so they pretend to hide their shame in not knowing. Therefore, people with arrogance are extremely bad at what they do. People who do truly have confidence are extremely good at what they do.

Anyway, there are many other areas we need help on, and we'll be needing help not just in making the machine but also long after we have it running. Documentation, for example, is something we'll need done as we near completion or during the process. Then afterwards we'll need websites made and models designed, people to create things for people to create.

All sorts of things will need to be done. If you don't think you have a useful skill right now, you could take some time and learn a useful skill. It is always useful to learn new skills, and you can learn just about anything on your own. Many skills only take a few months to learn and get fairly good at it.

There's also promotional art work that we'll need to have done, not so much right now, but in the future.

Tony

Hi Dom,

Glad you're going to design stuff. There are several places to upload models, besides the reprap page there is thingiverse:
[www.thingiverse.com]

and reprap models:
[www.reprap-models.com]

the reprap "Things to Print" forum is at:
[forums.reprap.org]

If you want to make some models for others, I suppose besides posting an offer to model in the General forum, you could post an offer in the Things to Print forum, or email me.

We need many designs, indeed eventually we'll want designs for almost everything. You definitely will be helping.

I'd love to make 3d files, but I like using sketchup, so I looked around for a converter and found a little script that will hook me up. The problem is I'm not sure this will work right when printed on a reprap, it looks fine when I save it to stl and open it on AOI, but if someone could do a test before I start going berserk and make tons of stuff that would be great.

I want to get as much return as possible with the least amount of effort.

Some call it laziness, I call it productivity.

And also @ Ant; if by documentation you mean lots of typing, I'm definitely in!

Zach and Brie have now made the big time.

[blog.wired.com]

Geek Dad has found thingiverse. This may help in some fashion I hope.

Bob Teeter

---

Bob Teeter
"What Box?"

heres a test cube from google sketchup.

its a 5*5*5mm cube

If someone could print this and tell me if it works that would be great.

I haven't built it, but viewing it in AOI is seems to be a 0.2mm cube. A bit too small for my machine!

---

[www.hydraraptor.blogspot.com]

heres another try using a different plugin that allows you to specify units

Hi Chris,

... the "0.2"-size is simply the transfer without conversion from mm to inch - multiply with 25.4 and you have 5mm ...

Viktor

haha, stupid unit conversions! And down goes another space craft!!!

Demented

Ant, you November 17 post is legendary

very nice analogy between Spanish & assembly... & machine code...