3D Printing

These parts often have imbedded metal parts, such as a metal

If you want a nut in the plastic part, you can make a slot that allows you to slide the nut in from the side. That would be the easiest way to do it. If you truly want the nut embedded, I suppose you could make the build in two parts, with a pause in between, where you drop the nut into a hexagonal cavity. Then, the 2nd part of the build would cover over the nut.

Jon

I've heard of plastic parts taking 50 hours to make on a 3D printer. Sort of like building a chair out of toothpicks. Maybe machining makes more sense.

--

John Larkin         Highland Technology, Inc 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

It depends.

There has been way too much hype about 3D printing. The hype seems to travel on lack of understanding, or an imagining that what's available today is what will come in the future.

I did see a column somewhere about them, and they guy basically used the example of a table to explain 3D printing, though he said nothing about the time it would take. It was a ridiculous example, since you have to haul all that plastic home first, then wait while the printing melts that raw plastic and puts it into the form of a table, which in this case isn't likely to be ornate, just a block of plastic. Might as well start from scratch with the plastic.

At this point, most of the hype relates to printers that can only do plastic, which limits their value. You can have a nice plastic box that emulates something from the past, but you still won't have a working device, since the printer can't make the insides. Even if it's just plastic gears inside, the printer can only make the plastic pieces, it can't assemble those pieces.

And for the neophyte, it will rely on having a file somewhere that someone else has made so you can get the printer to print it. That's a big stretch from actually building something yourself.

I'm much more likely to need some odd piece of metal, and there are ways of doing that, even before getting a lathe. For plastic, one can salvage old inkjet printers, that seem to all come in a plastic case that can be cut and formed without cracking. When I bought my used HP 4P laser printer a decade ago, I realized on the way home that the cover over the memory expansion area was missing, perhaps that's why it landed at that Rotary Club sale with such a low page count. So I took the cover from some stray inkjet printer I'd dragged home, and cut a piece of plastic to fit that frame on the HP printer. I have to duct tape it in place, but at least there is a cover. That kind of plastic, I think ABS?, is quite workable with tin snips and files.

I did have a need for a piece of plastic a few months ago to try to replace a broken piece, but I'd need the missing piece and skills I don't have to measure it and create a file for the 3D printer. I think in many cases there, people are hoping that the manufacturer will release the files, so you can make your own pieces. I'm not sure that's going to happen.

Some places are started to do 3D printing on demand. I'm not talking about collective workspaces, aka "makerspaces", but libraries and some big box stores, though I can't remember which of those I was reading about last week. You submit the file, they print it, you get the 3D image of the file back, once you pay.

That sort of thing seems to make more sense, a chance to try things out, to get a grasp of what it's all like. And then you can define whether you really need your own printer, or if you'll just have a tiny bit of need for it, and then print endless junk just because you have the printer.

The trick is to see the 3D printers as somewhere in between off the shelf and making from scratch. If you can't yet figure out how to end up with bits of metal embedded in the results, then you need to be thinking in terms of "how can I make this from scratch", which should help to figure out how it would be done with a printer. If you don't have the skills to begin with (knowing how to make a knob is different from making a good looking knob), it's too early to count on the 3D printer to do all the work.

Michael

You can buy one now for $500.

I expect a replacement cartridge will run $1750.

Before you contemplate investing much money in 3D printing of knobs, check out the surface finish attainable by your intended 3D printer. The ones I've seen all have some sort of pattern left on the part, as an unavoidable consequence of the printing process.

You may be able to print a knob and then sand it and polish it, but I don't think you'll be able to just print a nice shiny knob.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

The most common (cheapest and easiest to use) material is PLA (polylactic acid), and next is ABS. I think both of those respond well to an acetone vapor soak, which melts the surface a tiny bit so it flows and becomes smooth and shiny. Kind of like fire polishing glass. Also, someone talked about places to get stuff printed, besides the local makerspaces and buying your own printer. One company that prints your files is shapeways.com, charging by the volume of material and what the material is. They also have a storefront where you can sell your printed items. You upload the file, set the selling price, they put it on their web site storefront, when they get an order they print it, ship it, and at the end of the month send you your money minus the printing cost and a handling percentage. Pretty clever business model, I think. Not a customer, just saw their site.

----- Regards, Carl Ijames

Before you contemplate investing much money in 3D printing of knobs, check out the surface finish attainable by your intended 3D printer. The ones I've seen all have some sort of pattern left on the part, as an unavoidable consequence of the printing process.

You may be able to print a knob and then sand it and polish it, but I don't think you'll be able to just print a nice shiny knob.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

From what I've seen demonstrated , the plastic used is seriously non structural, no strength. You would have much stronger structural parts using expanded metal buried along with bushes etc in epoxy, taking a mould off the original part. Probably take about the same amount of time as a 3D printer as well.

Hi Dave,

If you want to get your own 3D printer and make custom parts I can think of 3 things you'll need:

First is a 3D drafting package, something that will make a .stl file (stereo lithography file). I think Google sketchup might work and be free, but haven't used it. Alibre design used to make a reasonable SolidWorks clone for about $100 US that is good. SolidWorks is one of the professional standard tools, but costs a lot.

Second is a program to prepare the .stl file for printing. These programs are commonly called slicers. Makerbot has one for free. There are other free ones available, but they may require tweaks to work with a particular printer. One good one is called Replicator G.

Third is a printer. One decision you need to make is material type. The two main choices for the class of printers you are considering is PLA (a hard, biodegradable, brittle plastic) and ABS. Some printers print only PLA, some print either. Each plastic has its merits. For me, the main benefit of PLA is less part warping during printing and cooling, and the main benefit of ABS is part workability for secondary operations after printing.

There are many types of professional 3D printers. They come closest to turn-key what-you-draw-is-what-you-get operation. There are many services that will print your parts for a fee and have available printers of various capabilities. Materials for these printers cover a wide range from fused plastic powders to stainless steel. Most companies use printing supplies as profit centers and so even after purchasing, printing supplies can cost you. Using a professional service is the surest way to get parts that are exactly as you've drawn the first try.

If you use an inexpensive home printer, prepare to spend a lot of time. You'll need to become an expert in it's operation to produce reasonable results. To my knowledge, there are two main types available. The first, which I've seen but not operated, uses a photo-polymer and a light projection system, and is best for small parts and high resolutions (microns), but the finished material is kind of soft. The second melts a plastic filament to print the part with resolutions down to about 100 um.

For the inexpensive route, here are some of the things you'll need to get right for reasonable results:

- the build plate needs to be flat to within about 50 um. This means you need a machined plate. The plates are typically removable and adjustable, and it's useful to make tools to help you level the plate. Too small a gap means the print nozzle will clog up, too large a gap means the parts will not stick to the build plate. Look at

for .stl files for various tooling. A very inexpensive printer may not include a machined build plate.

- the build plate needs to be heated for using ABS.

- one main problem with this class of printer, and especially with ABS, is that the parts will warp during printing due to differential cooling. The best fix is a heated build chamber. Unfortunately, this is patented and the working parts in a home-class printer are not up to residing inside a heated build chamber anyhow. One thing you can do to help is to enclose the printer to at least avoid cool outside air from circulating. You can also help warping with part design, by minimizing long solid runs of plastic for example, or putting relief slots into the part to break up long runs, and later filling with epoxy. These printers work best for parts with under a 4 inch span.

- The home-class additive printers (melted plastic) can't print much of an overhang reliably (around 45 degrees or so). You need to take this into account when designing your parts.

You mentioned embedding metal into the parts. Brass inserts are available in various sizes and can be melted into holes in ABS parts using a soldering iron (one you don't love anyhow). Printing ABS with embedded metal parts, while not impossible, is IMHO, impractical in most situations. JB weld epoxy has been useful for modifying parts after printing.

To sum up, I think these printers can be a useful tool for prototyping, making jigs and even for short run production, especially for internal brackets that don't need the strength of metal and where surface finish is not important. The home-class machines require a large investment of time but not a lot of cash. The professional printing services are the best bet if you need only a very occasional part or if you have a reasonable budget for parts. I might compare using the home-class machines to making pcbs at home vs using a professional service, say 20 years ago.

ChesterW

Though The Motley Fool feels otherwise, I don't think 3D printing will have the success //among consumers// people are predicting. At the moment, it's basically a "toy" product -- and will probably remain so for at least a decade. HP has been wise not to rush into the market.

What I'd like to see is a system of "manufacturing on demand", especially if were possible to improve on the item being replaced. For example...

I have two classic Sony radios, the TFM-825 and TFM-117WB, whose paper-cone speakers were okay-sounding 50 years ago, but would sound a lot better with plastic or plastic-doped cones. Wouldn't it be nice if I could order a "modern" version for $10 or $20? (Ditto for the classic KLH full-range driver.)

This is difficult, though not impossible. The problem isn't //making// the speaker so much as it is //analyzing// existing samples so the printer can duplicate the frame, the cone, etc. The cost for a single copy would be beyond unreasonable.

Thanks for your input, Chester I'm well aware of the software and 3D modeling requirements in order to even begin to make a 3D printer work. I've a number of years experience using and supporting CAD modeling systems, although it's been several years ago. And I know that the parts produced aren't perfect, but they can make parts that would otherwise be very to duplicate by other means. As you infer, learning more about the types of plastics that these things can handle is very necessary.

I'd like to see one in action, and get more info on what they can (and can't) do. Right now, as I said in my original post, it's an interest, not something that I have my heart set on doing. Other types of manufacturing have been proven to be cost-effective for hobbyists, such as vacuum forming, and machining, but each process has its own set of pros and cons, as does 3D printing. I think my next move is going to be to find a company that will make parts for me. I didn't know that these these services exist. Do you have any recommendations?

Many thanks for all the input from the others who have responded.

Dave M

I've watched videos of people doing the vapor soak to alter the surface finish. I've yet to see one that did a good job without a loss of detail. Knobs typically have ridges around the perimeter as a grip surface. Such a vapor polish would likely cause the ridges to erode away along with any markings that might be on the knob. But look at some of the videos and decide for yourself.

Rick

--

Rick

I have a RepRap 3D printer. It's a Prusa Mendel that I assembled as a kit for $800 a few years ago.I can give you some firsthand knowledge:

Check out

for designs.

I'm no mechanical engineer, but I was able to assemble it myself. It did take a lot of patience. There is a lot of instuctions online, but I found that the first few steps of the assembly were VERY well documented, but as I progressed through the assembly, the instructions gradually got a little sloppy and vague. It's almost as if the person writing the documentation had to pee really bad near the end, and was rushing through.

There is also a lot of mechanical "tweaking" that needed to be done in order to get decent build quality. The more precisely calibrated it is, the better quality of your output. I was able to get my layer depth down to 100 microns!

The repraps are meant to be easily sourced for parts, so they can be available to anyone. They are also pretty "improvable", so there is a lot of things you can do to make it better: Increase build volume, automated build platforms, auto bed levelling, etc...

All of the software that I use is available for free. You can buy some if you want, but the free stuff does the job too. This goes for design and production software.

It's good for plastic models: bust of Yoda, Model spaceships, etc. It's also great to replace easily broken plastic things, like a latch for a dishwasher, toilet paper roll holder, etc.

Check out

forfiles of almost anything that you can think of.

Things it CAN'T do well: very small parts that need a lot of structural stability. Very small load-bearing clips are not reasonable. Most items that are expected to be under a load or stress will have an unusual girth to them.

Also, designs have to take into account GRAVITY, so there is often support material that you would need to remove after the printing is done.

If there is a Maker Faire (or similar show) coming up in your area, you might think about visiting. I went to the one in Kansas City a couple of months ago and there were a lot of different of 3D printers running there. Some of them were being displayed by vendors, and some of them just by their owners. Plus there was other fun tech stuff to look at and play with.

Some cities now have a "makerspace" or "hackerspace" - basically a club workshop, you join the club and then you can use their machines. They usually have at least one 3D printer. I think you usually have to join to actually use the printer, but there is sometimes an "open house" night where you can watch it running and ask questions.

Standard disclaimers apply; I don't get money or other consideration from any companies mentioned.

Matt Roberds

No, but you can make two parts that interlock to capture the metal, or design for the metal being a press-fit.

--
umop apisdn 


--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

It can build them in-place, so you end up with a working device.

--
umop apisdn

That's fine for a prototype -- but you wouldn't manufacture products that way.

The 3D printer will never be the equivalent of the Star Trek "matter synthesizer".

Dear Dave,

There are many videos on youtube of these type printers working that might interest you. There are also many service companies. Search for 3D printing service and you'll be inundated. If you want an inexpensive sample of printing from a modest machine, see ebay - some people are using these to make gee-gaws to sell.

Here's a close-up of the print nozzle for a Makerbot Replicator 1 with a mm scale:

and here's a photo of a part with a melted-in brass insert:

A couple of other mods, besides those I previously listed, needed to get my printer somewhat reliable were a foil shield to block the extruder fans from cooling the part and replacing the plastic build plate arms with machined aluminum arms.

Someone mentioned the possibility of removing a part from the build plate and then precisely replacing it to continue the build. I wouldn't say it's impossible, especially since someone here would then do it, but it would be very impractical. The part is basically melted onto the build plate as the first layer is printed, and this adhesion keeps the part registered during the build. Loss of adhesion is a major cause of failed builds.

Some printers offer the option to pause in mid-build and then re-start. The feature is intended to allow you to change filament, say to a different color. It might be possible to get close to what you want using a pause as long as you don't remove the part from the build plate. You'd need to make sure that any metal you placed did not extend above the previously printed surface - else your print head will crash into the metal. Another way to embed metal might be to print the part in two halves, place the metal piece, and then glue the halves together.

I think machines like the Makerbot can produce reasonable results, but you pay with your time. You can even get nice prototype cases if you are willing to do some filling, sanding and painting. Alternatively you can get pretty and reliable prints from a service or by purchasing a $20k Stratasys machine, then you pay with your wallet.

ChesterW

Another approach is to make a plaster mould and cast it. You have to make some jig to hold the metal part where you want it while the material cures, but that isn't a fundamental problem in most cases.

Devcon polyurethane is a good material for this.

Cheers

Phil Hobbs

Hobby stores have cool plastic casting supplies.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation