Thread: 3D Printing - My rapid prototyping

Originally Posted by .RC.

Having a wet day today I did some googling and youtubing and it is astonishing what usable mechanical objects you can build these days with these printers.

It would be great for projects for kids to get them into STEM subjects.

From what I have seen there are quite a few already in schools but like a lot of thing their success depends on comfortable the teachers are with the technology. Their not like computer printers and take a fair but of TLC and finessing to see them working properly.

The Uni where I used to work was going to install a bunch of them in the library but the library staff said they did not want to have anything to do with them unless they were provided with extra staff. Likewise at a couple of pubic libraries here in WA. Consumables are a big issue with these and one thing folks don't appreciate - they are SLOOOOOOOOW. It's not like printing off a couple pages after a google search.

But they have come along way already and will only get more user friendly and hopefully cheaper.

Originally Posted by BobL

Consumables are a big issue with these and one thing folks don't appreciate - they are SLOOOOOOOOW. It's not like printing off a couple pages after a google search.

But they have come along way already and will only get more user friendly and hopefully cheaper.

Not wrong on the slooooooow front. Printing the patterns took ages, I was told.

I expect I'll buy one about another 2 generations of improvement down the track. IOW, probably sometime in 2018.

PDW

They seem to be pretty good now

Better not show you this, otherwise there might be a chipmaster soon for sale in Tasmania https://www.youtube.com/watch?v=ueP7hwicZSs

I have actually been blown away a bit by what I now see happening on youtube. Especially the open source ones where you can actually self replicate all the parts to build one, on the one you build. there is a video showing one person that accidentally broke a part, so they just printed a replacement. Then there is the printer farms, making more printer parts. Although I expect there would be a price point where injection moulding would produce parts at a far cheaper cost and you get far superior parts https://www.youtube.com/watch?v=6zFQi_wtBTA

But from what I understand moulds cost a fortune so you need a run of millions to keep cost per part down.

Originally Posted by .RC.

But from what I understand moulds cost a fortune so you need a run of millions to keep cost per part down.

As a rough approximation, a mould for a simple (quite) small part will start at probably $10,000 and go up in increments of say $5k, depending on size and complexity of part (and hence the required complexity of mould). Some of the volume produced printers probably use injection moulded parts but once you need less than several thousand parts, the amortised cost of the tooling does start to add up.
However, if you think about a printer costing say $2000 and that will be 'old technology' in 2 years, if you print one part per week that is still $20 of amortised cost per part plus materials.

I suspect that making some of these parts on the printer as you go is as much hacking cultural as anything else. I looked at a few of those parts and thought that they would probably be better made in Al, but for people without access to machine tools printing is possibly the only way to go.

Michael

Originally Posted by .RC.

They seem to be pretty good now

Better not show you this, otherwise there might be a chipmaster soon for sale in Tasmania https://www.youtube.com/watch?v=ueP7hwicZSs

I think I can restrain myself.... I'm hanging out for a pristine Schaublin 135 or 150. Or a 30" inch/metric 10EE as a Chipmaster replacement.

Originally Posted by .RC.

I have actually been blown away a bit by what I now see happening on youtube. Especially the open source ones where you can actually self replicate all the parts to build one, on the one you build. there is a video showing one person that accidentally broke a part, so they just printed a replacement. Then there is the printer farms, making more printer parts. Although I expect there would be a price point where injection moulding would produce parts at a far cheaper cost and you get far superior parts https://www.youtube.com/watch?v=6zFQi_wtBTA

But from what I understand moulds cost a fortune so you need a run of millions to keep cost per part down.

Yeah the tech is pretty impressive right now. It's only going to get better.

I agree with both of Michael's observations too - a lot of the printing is done 'just because' and quite a few of the bits would be better done in aluminium. But for prototyping, fantastic. My machining/design guys used to model in Solidworks then put a big chunk of HDPE or similar on the CNC mill to proof the model. Then go to aluminium or stainless once the tool paths etc were known to be good. Generally we never made more than 5 or 6 of anything so optimum tool paths weren't critical but ruining a big chunk of 316L or Acquamet wasn't popular. Nowadays I'd have one of these things in the shop in an instant.

That lathe in the YouTube video is impressive the same way that a dancing koala would be - not because it does it *well* but because it can do it at all. Makes even an ancient Unimat look rigid & capable.

PDW

Well OK, maybe not the lathe. But I did find this https://www.youtube.com/watch?v=EDppqGm-x0I

I cannot say I am an aircraft enthusiast, but I did find the lines and the clear plastic design pleasing on the eye. And you are right, this is just another "just because" with no practical application. But it is things like this that can drive development. I have read it is the computer game industry that really has driven computer development towards ever faster processors. With faster processors it then allows raspberry pi and Arduino to come into being as slow processors are cheap to make being old tech.

If all we did was word processing then we would still be on 286's. I doubt CAD is a big enough industry on it's own to stimulate major processor development.

I have just bit the bullet so to speak and ordered a 3D printer, but it is about one to two months away from arriving.

Originally Posted by .RC.

I have actually been blown away a bit by what I now see happening on youtube. Especially the open source ones where you can actually self replicate all the parts to build one, on the one you build. there is a video showing one person that accidentally broke a part, so they just printed a replacement. Then there is the printer farms, making more printer parts.

You mean like the BigBox Every plastic part (with the sole exception of the nut traps - which were injection molded) on the BigBox was printed on a farm of BigBoxes. I'm pretty sure a chunk of their Kickstarter money went into building the print farm. It's all open source and freely available for anyone to print the bits out and make one: E3D's BigBox 3D Printer by Bigbox3DP - Thingiverse

The independent head design is also open sourced. They are publishing as they go. They encourage people to print the parts off and make one themselves. I get the feeling that there is not much profit to be made on these things.

A model aircraft is probably one of the more sensible things that can be printed - the lathe is a 'cause I can' thing that really has no practical use.
A few years back someone printed some change gears for a CVA/ Monarch 10EE (as well as the box to hold them)
CVA Lathe Metric Gear Change set by JBFromOZ - Thingiverse
One way to do it I guess although I'd hate to be there when a tooth shears off.

Michael

Originally Posted by Pete F

Good tip on the radius, thanks I'll try that.

I've done some nylon parts and tested them to destruction and they literally are unbreakable. Eventually they do get kind of "mushed up" but don't ever really break like other plastics do. When I get more time I want to try some of the newer filaments coming out that are extremely strong plus rigid. The issue with nylon is always it's flexibility. Mostly however I just come back to PLA as it's easy to print, the finish is good, and is generally strong enough. If a part really needs to be strong, the print will be a part of a pathway to something else.

I noticed some facets on your pattern's circumference, is that a result of your CAD software?

I have a bunch of different filaments. I have not had enough time to test them. I currently have PLA, ABS, PETG, HIPS, Nylon (a few types), TPE Flexible (from both Fenner drives and Torwell), Carbon and Wood. I have a few different brands from Taulman, E3D, Torwell, Verbatim, NinjaFlex and ColorFabb. One of these days I will get around to testing it all out. I had the idea that PLA would be OK for stuffing around but nothing else. I was wrong!

One of the really interesting things that you can do with dual heads is "mix" different materials. So you might say print a flexible with one head and ABS with the other - to get a less flexible harder rubber like compound. There are some interesting Youtube videos on that type of stuff.

As for the facets - I'm not 100% sure where they came from. I suspect its the slicer. The files for the water brake are large and the slicer does take a fair amount of time to process them. The preview is pretty slow and clunky for them. I'm not running the newest machine so that may well be a part of the problem. I need to do some more research to find out what the go is. For the water brake it really did not matter as I intend to machine a fair amount back - and the other areas are more cosmetic than anything. That said - it would be best to not have them.

Oh that's a good selection of filament you have there. The PETG is the one I want to try next, I've tried most of the others you mentioned, but that has the most promise to me.

Just to address a few comments here, for those who are holding out getting into this area because they think they will wake up in two years, read the morning paper, and push the button on a 3D printer purchase because it's $9.99, prints in 3 seconds, and operates like a domestic toaster, I'd suggest you keep dreaming! I've owned my printer for about 18 months, 2 years(?) and the one Richard just bought is essentially identical to mine! With this type of 3D printing technology you still need to physically move a print head around and lay down 0.1 mm sausages of plastic. There's no getting around that. You still need a motion control system, you still need hardware, you still need power supplies etc etc etc. even now, if you took each component (like a stepper motor for example) and see how cheap you can buy one, they are stupidly cheap when you add up all the bits. Will they get cheaper? Absolutely, but it will be incremental and I don't see any game changing under this technology

However there are other technologies around that also fall under the same 3D printing umbrella. SLA is just one, and there are many MANY others. One for example uses reams of copy paper and cuts them out and glues them together to form a model. Yes believe it or not. It fills the niche market in which it's designed. Laser sintering is probably one of the most exciting, but that's still a looooong way off the domestic scene for anyone outside the geekiverse. Incidentally SLA is NOT slow, and you can get a model far faster than it could be machined if it has even the slightest complexity.

As as I mentioned to RC off forum, there are incremental advances in cars, just as one example, but just because a new car model comes out, or even a whole new technology like electric, doesn't mean we suddenly throw up our hands in the air and torch our existing vehicles. My Hilux is now 12 years old, but it still serves the purpose I bought it for. Lots of people still use VHS tapes to time shift television programs and watch them later. That's 30 year old technology, nevertheless it still serves the purpose for them. So if you can see a use in your life for a 3D printer now there's really no good reason, assuming you can afford it, not to buy one. It's not going to burst into flames in 18 months when a newer model is released! By then will a cheaper/faster/smaller/etc model be available? Yes probably, but yours may still do the job you bought it for just fine.

That model aircraft was excellent. I've seen a ton of 3D printed quadcopters but that's the best fixed model I've seen. The printed quadcopters are used often in racing and are virtually indestructible. As I mentioned, the uses of the printers is limited only by your imagination, and if it can be manufactured with a thermoplastic, it almost certainly can be printed, often better.

Originally Posted by variant22

You mean like the BigBox Every plastic part (with the sole exception of the nut traps - which were injection molded) on the BigBox was printed on a farm of BigBoxes. I'm pretty sure a chunk of their Kickstarter money went into building the print farm. It's all open source and freely available for anyone to print the bits out and make one: E3D's BigBox 3D Printer by Bigbox3DP - Thingiverse

The independent head design is also open sourced. They are publishing as they go. They encourage people to print the parts off and make one themselves. I get the feeling that there is not much profit to be made on these things.

I looked at a big box but it was just out of reach. So I went with a Prusa i3 Mk2 kit. But it is months away from being delivered at this stage. It was this thread that threw me over the edge. I told Pete about twelve months ago 2016 I was going to buy a 3D printer. Well I missed it by that much.


I would also like to print out some mechanical models of things for my nieces and nephews. Just simple things like a planetary gearbox or a model cut out turbofan engine. To show them how things work as some are getting to that age where they start to show an interest in such things.

When you first get the printer you will probably need to start with very basic models while you get a feel for how to use the printer and slicer. You will probably get sick of the sight of 10 mm cubes, as you'll print dozens of them to tune the printer. That's one of the big differences to something like a Big box, where it is as close as is currently possible to an "appliance" in terms of operation. In other words the software etc is all set up, and once assembled you just send your files to it and it should start printing them quite well.

Like so many other things, you get what you pay for, and in this instance it's a turn-key type solution. Many of the printers of his type are capable of very similar results, eventually. However as a generalisation, the cheaper you go the more work you'll need to put in to them to get them to work as hoped.

I don't expect them to get cheaper, I'd be happy to pay $2K to $3K for a decent 3D printer that met my needs.

I need a bigger work space than the Big Box printer can deliver and I'd like faster speeds. Also some software improvements - I find it *really* hard to believe that you can't pause a print then carry on but I'm sure that's what variant22 told me. As a software developer I'd really like to know the logic behind that limitation.

Anyway I simply don't consider them capable of doing the sort of things I want just yet. They'll get there, I'm sure, or some other additive tech 3D system will. I'm not in any great hurry.

PDW

Actually in this case I don't think it is a software issue. After all a pause in a program is simple.

I haven't got one so this is just a thought but the better filament printers have a heated plattern and the material comes out of the nozzle hot. I suspect that if you paused printing the layer that you are bonding to will cool down and the bond strength may not be as good. How badly that cool down would affect things will depend on a lot of variables but as a worse case imagine having a plane of weakness right through your part. It may not matter if you were printing sculptural parts but if they were bits intended to take a load it could have negative effects.

Michael

Originally Posted by PDW

I need a bigger work space than the Big Box printer can deliver and I'd like faster speeds. Also some software improvements - I find it *really* hard to believe that you can't pause a print then carry on but I'm sure that's what variant22 told me. As a software developer I'd really like to know the logic behind that limitation.

The metal nozzle on the print head (hotend) is running in excess of 180 degrees, so it can't just stop wherever it is, as it will melt the already printed material, and dribble plastic on it.

If you leave the hotend powered for too long without extruding material, the material inside the nozzle can overheat and be ruined.

So to 'pause' you'd have to move the print head off to the side where there is no printed material and shut the hotend down. If the heat bed is running, that would also get shut down, to save wasting power while paused (how does the software know how long you want to pause for?)

Which means, when you restart you'll have to wait for the hotend to come up to temperature (potentially 2-3 minutes, depending on type), then lower it down to wipe the nozzle on the bed, and hope that it doesn't dribble on the way back up to where it left off in the print.

This is where the problem really starts though - if its been paused long enough for the previously printed plastic to cool down, the adhesion of those first few layers to the previously printed material is not going to be very good. And if you're printing something like ABS, it gets even worse - ABS loves to shrink, so your first layer is quite possibly going to air print if the head goes back to the same position it was before, which is going to leave a beautiful bunch of spaghetti on your print. The way around that is to have a fully enclosed print enclosure, and bring the temperature up with the heatbed - but you could be waiting 20 minutes for the enclosure temperature to stabilise before printing can resume. And most printers don't run enclosures.

That's the major issues I see with 'pausing', there are however others born of the light nature of construction of these printers which could also cause issues.