The Guide to Not Buying a 3D Printer
chopmeister.blogspot.com
chopmeister.blogspot.com
Until you reach the stage where you are making stuff a substantial portion of your time renting is by far the better option (especialy in fields where there is still a lot of development).
If you want to get one to play with the technology on a daily basis buy one that is as open and flexible as possible so you can move up as new developments become available.
This is a pretty good option:
http://www.emachineshop.com/machine-shop/Rapid-Prototyping-S...
Might be less hassle than renting an actual machine.
Renting has the additional advantage that you're not stuck with the particular technology that you bought. If two jobs benefit from 2 printing technologies, you can have both.
On the other hand, a benefit of having your own proto shop is that if you're skilled, you can prototype and modify quicker than you can specify. And don't overlook ancient methods. You can still do a lot with woodworking tools.
When you work wood with metal tools you're going to have do a very thorough cleaning job afterwards because the acids in wood will wreck equipment quickly.
Thanks for the warning. I hadn't realized this problem exists.
I've been a member of the Austin Hackerspace (http://atxhs.org) for a while, and it's paid dividends in saved expense on tools. Plus the community of experts it has attracted has allowed me to quickly finish projects that would take months of research on my own.
I'd argue that additive manufacturing via hobbyist FDM printing is a lot cheaper for one off designs than subtractive manufacturing of any form, because the consumable cost is limited to utilized filament and a tiny amount of printer maintenance rather than huge amounts of waste combined with expensive, finicky tool bits and the same kinds of maintenance.
I do agree that for items like a Raspberry Pi case which could be mass-produced via molding to much better tolerance at a low cost, 3D printing might be silly, but for truly one-off, low accuracy consumer parts I don't think 3D printing can be beat.
I personally have also not had much problem getting square, true, dimensionally accurate prints with a crappy printer (Solidoodle 2). The most deflection I've ever seen is about .8mm across the 150x150mm build area. That isn't perfect and obviously won't work for precision equipment, but combined with ABS's flexibility and forgiving nature it's certainly good enough to make one-off enclosures, fasteners, buttons and knobs, adapters, and other "glue" parts, which is most of my use for 3D printing.
I will say that 3D printers are hard to set up, and that's probably the source of a lot of the warped cases you've seen. CNC mills and lathes are just as hard to set up, though - you've just generally paid the vendor tons of money to do it for you.
They are promising assembled printers nowadays for less than I spend on hotend, motors and electronics...let alone frame and assembly.
I am a printer vendor and have been for the last 3 1/2 years...I have worked on many opensource projects related to the reprap community. I have done my best to help people tune their printers in IRC regardless of where they sourced them...any printer can be made to print...making one print good takes time. How much time is based on how good the printer is assembled mechanically, electrically as well as tuned in the software side (firmware and slicer).
I often tell people "These things are like table saws...not toasters...just because you have a table saw does not mean you can use it well right away"
I've made various lens to body adapters (mostly old legacy manual focus lenses to my Sony A7 [E-Mount]), custom lens caps for odd-sized lenses, custom-sized extension tubes, etc. Just today I 3D printed a custom mount to physically attach a generic ("Neewer" brand) hotshoe flash that I rewired into my Fujifilm Instax 210 instant camera to replace the shitty direct-flash mechanism that the camera ships with so I can use the camera indoors and get decent exposures.
But, yeah, if you can't think of anything you would use a 3D printer to build (if you had one), it is likely to be collecting dust within a couple of weeks after you get it.
Maybe there's a startup idea in there :)
A toy soldier that costs the better part of a dollar in materials, or one that costs fractions of a penny? Which one is going to provide better margins?
I was thinking along the lines of printing kid-designed soldiers or weapons :)
Absolutely niche (though maybe not that much niche for stuff like Warhammer 4000 or tabletop war games or whatever :) ).
I'm not too familiar with 3d printers and hadn't heard anything like this about Makerbot before -- anyone know what the conflict there is about?
Funny enough they were starting to be a lot less interested in the deal when they heard I had a patent on it. I even tried the "look, you've covered your behinds, I'm covering mine" angle.
(I do have a patent on the design, but I'm choosing to not enforce it - http://robots-everywhere.com/re_wiki/index.php?n=Main.LCheap... if you want to either buy one, or download the designs for one)
It's not fast.
FWIW, I not only provide the design, I also provide the design (and a video) of the jig I use to make them.
If you're going to do the copper thing make sure to get some kind of fume extraction going, metal vaporization is an excellent way to poison yourself!
A poor mans version of this would be amazing:
http://www.thefabricator.com/blog/direct-diode-lasers-enter-...
Poor: a few thousand $ for a head.
Makerbot disputes that investors caused them to change but "For the Replicator 2, we will not share the way the physical machine is designed or our GUI" http://www.makerbot.com/blog/2012/09/24/lets-try-that-again/
Basically, they're very keen on other people releasing their stuff as open source hardware so they can use it, but refuse to release their own stuff. It also doesn't help that certain prominent, business-owning members of the community called anyone holding Bre Pettis to the standards he expected of others terrorists and claimed to be standing up for "polite discourse" by doing so.
There are certainly things I like better about the Solidoodle, such as the ability to use a sheet of glass on the bed, which allows me to use ABS/Acetone slurry to make parts stick to the bed, the ability to tweak all of the CAM settings, the mechanism for feeding the filament into the hot end is easier to work on, I've never had a clog on the Solidoodle, and there are several power users who have shared improved designs for various parts of the machine.
But if I need to make a part that has to be within tight tolerances, or have a good surface finish, I prefer to print with the Makerbot.
Does anyone know of a sub $3000 open source FDM printer that can print as well as a Makerbot?
The question of why I suggest avoiding makerbot is complex and not just about open vs. closed source. I will elaborate that in a future article in detail, since it raised quite a lot of questions.
I'm posting this only because I think it's a legitimate answer to an important question... Is there an affordable printer that prints as well as (or better than) a MakerBot?
I have been using the FlashForge Creator and Creator X to make 3D printed prosthetic devices, and I'm thrilled with the quality. All of the photos and videos I have on my blog at 3duniverse.org were printed on the Creator or Creator X printers. I'd say these printers print better than the MakerBots, which we've been testing side-by-side.
The FlashForge printers range in price from $977 for the original Creator to $1349 for the latest Creator Pro. The printers come fully assembled and include two full spools of filament.
But note that I only started selling them after falling in love with them through hundreds of hours of direct hands-on experience with them. I haven't yet found another printer on the market that has better price/performance value.
The moral is they will eagerly drop support for your $3000 purchase in just a few years.
We've solved the "hard to print" aspect first, and will be moving further into solving the "hard to calibrate", "hard to troubleshoot", "hard to share", "hard to integrate" aspects soon. I want these machines to be easy to use some day, and I think we're on the right track.
For example, I use it to print custom gaskets for the circuits that I make and sell -- this results in less installation errors, which is worth it just for the replacement units.
They have a small office in Techshop though they are in the process of getting more space. Their machines are awesome and a lot of the guys who work on it are in the bay (which means you can hit them up and ask questions easily). Also the driver code is opensource.
The other day one of the plastic hooks on the shower curtain broke. Instead of trying to glue it as I would have before the printer, I just printed a new one. Works great.
The printer: http://imgur.com/a/URcNC#0
For now i think paying someone else to print is the best option, for low volume. I guess it's been the same with many new technologies, initially at least.
If you don't use servos or steppers+encoders for feedback you may just as well assume that anything that you produce is unique. The variability in wire-feed (gripping wire is a tricky proposition) and head temperature adds to this. The result is a fairly rough surface (compared to millwork or lathework) but in the end what matters most here is that it can be done at all.
If you require precision then you're probably going to have to do a little bit of re-work (just like you would with casting), if all you care about is looks then you may have to do some sanding to get a really smooth surface.
3D printing right now does not allow fabrication to the same tolerances that we have gotten used to with even minimal metal working tools unless you are prepared to reach very deep into your pockets.
See above about renting, good quality equipment can be rented for a fraction of the cost of buying bad quality equipment and will produce workpieces that are very good due to the tight control over the head position and the increased control over wire feeds and temperatures.
Consistency (feeds, speeds, temperatures, materials) and precision (accuracy of position, repeatability) are flip sides of the same coin, you can't really have the one without the other, and when both are good then you will have good quality workpieces. If either is not so good then you will end up with rough work, low quality and/or a lack of precision.
Not all work requires precision or good surface finish, it is rare to have work that requires neither.
I don't now a lot about 3D printing, are there any serious competitors?
Easy cleanup, low noise, very little moving parts.
That entire paragraph is subjective. Some people do want WiFi (I certainly wish my printer had it, because my workflow options right now are a pain). Some people do want it to tweet or otherwise give live progress updates. And so on.
Sure, some people want those things (and I can see where they would be nice to have), but I think the point is that if it does not print well, the wifi, tweeting, coffee, etc. do not matter.