How to design moving parts

Hey guys i am very new here and have had a 3d printer for about 6 months. I have designed some of my own simple projects with out issue and glued them together after printing. What i am fascinated with is how people are designing things that print all as one piece but have moving parts. I have tried looking for some good resources on this subject but have been unsuccessful. Does anyone have a good source that they can share or and sort of guidelines that would help out?


The first thing that i have been tinkering with is a simple box with a gear in it that will turn. i have left a gap between the top and the bottom but it always seems to fuse the gear to the base and the top. Thanks for the help I have attached my STL Attachment 8603

Very carefully.


Check out a lot of the moving items on thingiverse, and see how they are designed and printed.

You may have to use supports and a needle file set to get them out to keep parts free.

look at how intentional3D made their jack. it's a real beauty

Personally I would prefer to print single properly toleranced items and assemble them as opposed to fudging tolerances so that something can turn.

Not answering the question but worth mentioning in this kind of thread.

Quote:

---

Originally Posted by Sebastian Finke

Personally I would prefer to print single properly toleranced items and assemble them as opposed to fudging tolerances so that something can turn.

Not answering the question but worth mentioning in this kind of thread.

---

i would imagine that this method would also have the benefit of leading to much more accurate prints. In some circumstances it should lead to better characteristics like tensile strength as well as you an much more easily optimize which axes to print each part on. As an example, filament on the XY axes are usually much stronger the on the z axis.

Quote:

---

Originally Posted by vexmatech

1. Clearance is very important when designing mechanical parts for 3D printing.
2. Parts have to be cleaned also. You must allocate space for this. 3d printing service delhi-NCR bangalore
3. Powder is trapped in enclosed spaces. If you do not design so that it can be removed your part will not work.
4. Two surfaces(printed in the same orientation) that rub against each other cause more friction than one would expect due to the layerwise production in 3D printing("hooking").
5. If you design some kind of mechanical device such as a coil spring please keep in mind that since it is produced in an integrated fashion(ie part of the housing) there is no(or not much) pre-load in the spring. It will be weaker than a traditional spring made out of two parts and then compressed in the housing.
6. Paint changes the mechanical properties of your model.
7. You could manually correct your model with a file but this is tough going.

3d printing mumbai pune india

---

This... is a really good post.

Nice work.

I would only clarify number 3, in the case of FDM there is no powder but the same still applies, support structure material must be able to work it's way out of the joint or it will bind up in operation.

Number 5 is especially nice. Not something most people consider, spring preload is the difference between a sagging part and a nice high quality function that feels solid.

it is fun trying to get parts within parts. Particularly if like me you don't like using supports.

This is my best part within a part design to date:

Well it would be but uploading pictures is definitely not working.
I'll be back ...

http://slinging.org/forum/yabbfiles/...ed_800x600.jpg

http://slinging.org/forum/yabbfiles/...ed_800x600.jpg

http://slinging.org/forum/yabbfiles/...se_800x600.jpg

basically I wanted a spool to wind 10-20 metre sample lengths of stiff filaments on. Figured was quicker to print one that expanded and locked than just make one big spool.

Works really well :-)
I can print the parts seperately as well, but they print together and just pull to lock with no problem.
I think the smallest gap I've used is 0.5 mm.

This is the mark 2 - mark one has holes in the shaft and locked with a bent paperclip :-)

Print these at 0.3mm layer height and around 65mm/s speed - which is quick for my creator :-)

It looks like you've reinvented the wheel...

That's a freakin genius design.

Must remember, steal, and patent.

-Edison

It is nice.

But not a pretty artistic model, so no money in it :-(
lol

In fact that one in the pic the left arm didn't print that well - but still locked in place without any problems.

Beauty is it gives you a good width to wind stiff filaments round.
Once wound you leave it for a few days to get used to the bend and viola.

The clever bit is getting the top rim to print without supports.

You got to love openscad :-)

Probably the best example and model is the Secret heart box. Just a stunning masterclass in designing and printing encapsulated hinges.
http://www.thingiverse.com/thing:44579

And also done in openscad.

Quote:

---

Originally Posted by curious aardvark

It is nice.

But not a pretty artistic model, so no money in it :-(
lol

In fact that one in the pic the left arm didn't print that well - but still locked in place without any problems.

Beauty is it gives you a good width to wind stiff filaments round.
Once wound you leave it for a few days to get used to the bend and viola.

The clever bit is getting the top rim to print without supports.

You got to love openscad :-)

---

Doesn't have to be artistic. The functional prototype is what's worth money.

not in the 3d printing world it's not.

Ask geoff - the money's in artistic models for film and tv tie ins, cosplay accessories, pretty statues, etc.

I haven't got the foggiest idea how to make money out of practical designs, ideas welcome :-)

You patent the design then sell an injection molded part with better tolerances and small modifications suggested by focus groups.

I can already think of one.

if you break it down into injection moulded parts - then yeah, plenty modifications.
But then you'd have to expect people with 3d printers to buy them. And most would simply just print something else - there are no shortage of clever spools around. Or take one look think 'that's a good idea' and design their own - like i tend to do :-)
I think that's the only one print expanding one. But I don't know for sure.

Like i think my bottle bird feeder is the only one with hassle free bottom loading, but don't actually know. Or my 0.3gm trolley tokens are the lightest in the world and cheapest to print, but it's just a guess.

It's an interesting conundrum. Do you put your designs up on a pay per download site and hope nobody else would rather design from scratch than pay a few sheckles. Or sit umming and aahhhing and just not making any decisions at all - like me.

So I guess the other approach would be along your lines. Design something, test with 3d printing and then try and get someone in the injection moulding industry to buy it from you.
Have to think about that one.

There is always a market for convenience. This part, in this case, would probably not see a return. Injection molding a piece like that would require... at least a 100 thousand dollar investment in the tool alone with multiple side actions and cores required, not to mention the fact that you need at least 4 pieces.

If you only project to sell 10,000 of them, then that's a $10 cost just for the tool. Adding in machine time, material, shipping, assembly... you get the idea... that might be a $40 unit. You could save a little money by going with an aluminum mold base, but not a whole lot, and it wouldn't last more than 100,000 cycles conservatively. It would be about a 20-30% savings in the cost of the tool.

I always find this stuff fun to discuss though.

The thing about this industry is that it is just full of people who would rather design and print it themselves, than buy it at a markup.

definitely interesting to hear about injection moulding numbers.
Easy to see why a lot of compenies are seriosuly looking at 3d printing.

For $100'000 you can buy a full colour sls/polyjet machine that would be fine for small runs.

Pretty much the only way i cna see to make any money out of practical (ie: non artistic) desings is to charge for the design download.
But then you have an unprotected stl file that people are free to share.
So I'm basically holding back until a download service that also incorporates strong drm starts up.
And that'sa controversial topic right there lol.

For smaller parts that most people would be printing, the tools don't cost that much. For a simple part you might be able to get a tool down to as low as $20k, but there again that is still a lot more than most 3d printers. Hell, we have molds that are 4" square and are pretty cheap. We do a lot of very small precision components.

The variables in printing, material properties, and production speed are critical. Typically we are only using smaller format printers for prototyping or one off parts, with exceptions being in exotic geometry that can't be otherwise made. If a typical print time is 2 inches of height per hour for an average cross sectional area, and your part is 4 inches tall, then you can do around 4200 parts a year if you run 24 hours except holidays.

For a $100,000 printer, that's around $24 a part for a 1 year ROI on the machine, not including materials, design, setup, and electricity cost. Production rarely justifies that price and speed for a plastic component. You could see a part being $30 if you only run that all year, up to $50+ per part for high buck materials with long print times or larger volumes.

Companies are looking, that's for sure, but we aren't there yet. The machine costs need to come down significantly. Even if you stretch it out to a 2 year ROI, which many people hate doing, you still have pretty expensive parts. For prototyping however, even a $50 print is WELL worth the cost to test the design out for fit and function, make changes, print it again, repeat that 5 times... then move forward to a production injection mold, or in some cases start building the fixtures for production machining that could cost thousands.

well you've got nylon sls machines coming in at the $5'000 mark.

It's getting there.

At some point a hybrid injection moulder with 3d printing incorporated will appear.
It's all heading to a distant amorphous vanishing point of ultimate replicator technology.

I suppose with the a few select high temperature polymers you could 3dprint a mold that would be capable of running a hundred or so shots before it's garbage, then use a robot to slot it into the IMM side. The mold wouldn't be capable of producing certain features though, as the injection pressures required for thinner parts would blow the parting line of a polymer tool. Certainly a living hinge would be out of the question ;). The cycle times would be insane, considering there would be no cooling and plastic will insulate the cavities rather than cool them, but for a short batch run I can see that possibly being cost effective - just send a file to the machine and it prints a tool then molds the parts automatically.

You could potentially do it with metal SLS too, but there are secondary operations that are required for that also, so being a hybrid unit wouldn't save anything in terms of operations there either.

Okay and now rethink your answer assuming there is new technology you don't know about and a whole host of new materials with amazing properties.
ie: it's in the future :-)

I know a lot about material properties actually. I did say it could be done in theory.

ah yes but you're still thinking about existing materiails and technology.

It's the difference between 'proper' science fiction and the stuff where they use existing technology and try to write modern science based fiction. The only person who's ever consistently pulled that off is michael crichton.

Given the ridiculous leaps in un redicted technology in recent years, us imaginative sf writers probably have a better chance in getting a prediction right.