Anybody used the 3D printer at the UPS store?

They aren't terribly expensive and I think they could be useful for replacing broken parts of various household items.

I'm thinking of buying this one just for play/experiment...

...Jim Thompson

me too, but, making something on a mill takes a lot more planning and think ing than doing a 3D print, with a mill you have to figure out what tools to use , how to hold the part, what order to mill the part and you have to start b y designing something that is actually possible to mill

t.

if you are looking for something you can spend an endless amount of time fiddling with to get the settings right so you can make a somewhat flimsy sorta usable part on it is probably a great idea

From what I've seen at the local Hackspace, the home printers with PLA/ABS are just about suitable for convincing yourself that you haven't got the dimensions too wrong. The parts themselves have poor finish, are brittle, and just not nice. The home machines are great if you enjoy endless fettling with them.

Commercial parts and materials are much better in just about every respect, and some are astoundingly cheap. Sintered nylon is a "strong and flexible plastic", as Shapeways puts it, and the SLA resins have fine detail and a good finish.

Before you spend too much time/money, go and watch something being made in the same material.

Yes, but...

The whole field is at the same point that microprocessors were in ~1980. They are only suitable for enthusiasts, and evolution is rapid.

The key point for me is that I want to use different materials for each job, and I can't do that with one home printer.

So I have less than zero interest in buying one.

Days??? The Tormach starts at $7000 and can't make many of the things you can make with a 3D printer.

What do you recommend? Budget is adjustable ;-) ...Jim Thompson

n:

ment.

send your design to one of the numerous online 3d printshops?

-Lasse

You missed the "play/experiment" part ?? ...Jim Thompson

People have different toy budgets.

You might check and see what is available in your area to use. Here the local library has a three D printer and will run off things for library patrons.

Dan

I bought the two-extruder version from that company (shipped direct not via Amazon). The parts are pretty crude, but sometimes useful (and the material is inexpensive for smaller parts). Figure maybe an hour to make a small part, many hours to make larger parts. There are a bunch of settings and if they are wrong the part may not come out good enough to use.

I've made a tube with two sizes of ends to adapt a camera to a microscope, a binocular microscope eyepiece cover, a prototype of a product for barmaids (friend's design) a binocular frame that uses Nikon teleconverters (from thingiverse) and some other things which turned out usable. A few other things were too fine or required too much strength in a thin layer to be producable. A vintage BMW replacement part (just a cover) turned out pretty good after painting.

I think QIDI (China) is really quite a decent company, but the technology is about the same in all those machines and it has serious limitations. It's nowhere near as good as commercial SLA 3D printing, for example.

--sp

In what way are the parts not good? Is it strength? Accuracy? What are the settings that can goof it up?

You can buy a Sieg (Shanghai, China) small CNC mill for $4K (on special at the moment) from Little Machine Shop in Pasadena. You need to add a desktop computer and software. And tooling!

Of course the tooling and raw materials and sofware will add to that but you can actually get fine metal parts out of it that won't snap between your fingers. It has proper ball screws by the look of it, and Sieg is a well-known company.

The 3D printers incorporate a lot of intelligence into the way they slice the .stl file- the parts tend to be more-or-less hollow with an internal honeycomb structure. It's also an inherently easier problem than figuring toolpaths and workholding for a CNC mill. Because the material is relatively crummy they can print support structures that can be broken off (or you can use a dual extruder and print the support structure in PolyVinyl Alcohol (PVA) which is water soluble.

I would not really compare the two .. I have not seen much that could be sold with a 3D printer. It might be good enough to make some jigs etc. Even a small manual machine tool can make production parts.

--sp

ther have a Tormach.

There's a front end for various 3D print places (aka 'hubs') some of which are just some dude in his basement I suspect. Scores in any big city.

A friend used them and got 'okay' results.

Many of them will ship cheaply or allow you to pick up for free.

Make sure you know what technology you want- the SLA type (eg. Formlabs) costs signicantly more but is rather better in some regards.

Ponoko is pretty good/cheap -- but slooooow.

--sp

But you can make serious metal or plastic stuff, like production/test fixtures, tough prototypes and short-run production.

One interesting process is to make silicone rubber molds of wax or other prototypes, than cast things from epoxy. The resolution is literally microscopic, and you can make a bunch of parts relatively fast. Tap Plastics has all the stuff cheap. Of course, you can't make the "impossible" internal-feature parts that a 3D printer can.

We recently evaluated a prototype of a benchtop thermal imager whose case was 3D printed. It was unfinished, so felt gritty, like sandpaper.

The easiest material to print is PLA, which is quite stiff (usually a good thing) and doesn't shrink much but it's a bit brittle and the extruded filaments don't bond perfectly together. The parts have a strong pattern on them from the 0.4mm nozzle and unless you have a wall 1.5-2mm thick it will be very fragile. Forget making snaps and such like unless they're very crude and big. ABS is more trouble and not much better if thin sections. Commercial printing is enough better it's worth the extra dollars in many cases, because it makes something you are not going to want to hide.

Accuracy is less of a problem than resolution (which is something you can do nothing about basically). The accuracy is pretty good (stepper-based) and you can fiddle the scaling to account for shrinkage.

The very first part I printed came out fine with default settings, and a number after it have as well, but 'fine' is really not very good.

Not accurately leveling the bed is one problem. There are scores of settings for extrusion temperature, bed temperature, speeds (>10) and so on. The defaults are not too bad, but as I say, the best is not very good. I did have a significant problems as discussed below that were solved by setting a long delay between layers. The problem was that the extruder head reverses or changes from outline to fill at the same point around the circumference and tends to build up a blob which catches the head and if it's still soft it tears the part. There might be a randomize setting somewhere in the software but I didn't find it.

That part is maybe 2" in diameter at the bottom. You can see the break-off honeycomb support structure for the cantilevered hinge in the 2nd photo.

From what I've seen it's not as if you can fiddle with the settings and make a wonderful part- the range is more from cruddy to unusuable.

I admit I have not spent much time fiddling with the settings- but what I've seen is consistent with results I've seen from other machines in this class. The recommended software has many more settings than the basic settings shown above... well over 100 different parameters.

I manually fabricated a few aluminum and steel parts this weekend for a G-job and the concerns are totally different- work holding, tool specifications and toolpaths to miss fixtures, feeds and speeds, conventional vs. climb milling and so on. Interestingly, the 3D printer software can spit out G-code which is the usual machine tool language.

--sp

Yes, epoxy or rigid urethane. You might need a vacuum setup to get the bubbles out. Especially with nice SLA printed parts. I played around a bit with the urethane- it's very exothermic when you mix the components. All in all, I'd rather someone else did it- kind of fiddly.

I see a lot of that at trade shows. They often polish them and paint them so they look almost as good as injection-molded parts. The gritty feel sounds a bit like ZPrinters which can print in color.

You can even buy prototype acrylic lenses (like automotive teardrop headlight lenses) with all the compound curves and beautifully transparent. I imagine they cost $thousands even from China.

--sp

For various values of "can". Running a milling machine is not completely simple. I wouldn't do either. If I need something made, I would farm it out to a machine shop or 3D printer shop. Why invest my time and money in developing a capability I'll use once in a blue moon and will otherwise sit around taking up space?

Carbon3d is making a new machine type that uses light (likely lasers) to cure plastic from a liquid pool. They seem to be going great guns and should be shipping machines at this point. I understand aerospace companies are buying them. I also see there are a number of lower end units available from various companies. Interesting.