3D Printing stuff - hostile takeover of Planned Obsolescence thread

[julianh72 (Julian)]

Originally Posted by AndrewJ

Gday Julian


I have been following the evolution of these higher quality devices for a while, so have a good idea of what is possible commercially.
I was looking more at making stuff with astro sized threads built in using "consumer" 3D units. In theory, the machine should be relatively cheap once all the design problems re calibration etc get refined, but i doubt the govt would want these things in the wild.
Imagine being able to "print" a custom designed thin wall OTA for a refractor etc "in the shed" :-)

Andrew

There's a few separate issues here - the physical capability to work in metal etc, the precision / repeatability of the manufactured parts, and what the government wants (or doesn't want).

Working in metal and other "serious" engineering materials is already "here and now", either through a commercial service, or through the still-emerging domestic technologies. When you think about it, the government can't stop this - the genie is already out of the bottle. There's no real philosophical difference between "additive" manufacture (3D printing) and "subtractive" manufacture (milling and machining) - and there's no way that the government could (or should) stop people from machining steel. The big game changer is that additive manufacture opens up all sorts of possibilities and efficiencies that subtractive manufacture stumbles on.

Accuracy and tolerances are a different matter - 3D printing currently tends to have relatively poor tolerances compared to milling / machining (but its improving all the time). If you wanted to make a gun, it would be a LOT easier to machine one than to print it using current technology! The optimal technology for home workshops will probably be a combination of additive and subtractive manufacture - e.g. design parts which can be built quickly and cheaply using additive manufacture, and can then be finished using subtractive techniques where high precision is required.

You will already see that part of the art of design for 3D printing is to recognise where the tolerances are likely to be poor by traditional standards, and design around it. This means recognising that building your part in horizontal layers will tend to make it stronger in the horizontal plane than the vertical axis, and designing and aligning your parts accordingly. Another example is that it is hard to get holes made accurately, whether they are vertical or horizontal. The trick is that you design holes to be somewhat under-sized and ream them out after printing; another trick is the classic RepRap "tear-drop" holes for horizontal holes, which allows them to be made without needing any internal support for the overhangs. If you need a mechanical anchor, you create a pocket (even blind pockets are trivially simple for a 3D printer to make!), and insert a steel nut, and it will grip a steel screw very firmly.

When it comes to "astro-parts" there are some parts which absolutely NEED very high stability and precision (I'm thinking of the whole optical train, motorised axis drive trains, etc), and others where overall precision need not be very high - here I'm thinking of accessory mounts and so on. 3D printing is already well suited to the latter, and I see no fundamental reason why it can't be used even for many high-precision parts, as long as you understand the materials and build technologies you are working with, and design accordingly.

For example, when you realise that a home-built RepRap printer can position the print-head reliably and repeatedly to just a few microns in 3 dimensions, you realise that very high precision can be achieved using plastic gears, rubber drive belts, steel threaded rods, and so on. The dimensional tolerances of the finished parts tend to be restricted by the shrinkage after placement, rather than the build-accuracy. You could certainly 3D-print the gear "blanks" for a drive train, but you might need to allow for the final milling and finishing to be done using conventional CNC machining or similar. (Even that's not necessarily true - the extruder on my RepRap uses a 3D-printed gear train which has had no post-processing or finishing whatsoever.)

(Somehow, we're a long way from a strictly astro-related thread, but 3D printing is another of my passions - and it fills in the cloudy nights very well!)

[AndrewJ]

Gday Julian

There's a few separate issues here

Just a few.
Being a Mech Eng, i understand a lot of the tradeoffs that go into this stuff, but once the process is refined to the point that metal printing becomes as normal as photocopying or printing a letter, it will be a gamechanger.

3D printing currently tends to have relatively poor tolerances compared to milling / machining

Dunno there. The tolerances of some of the nano printed stuff i linked to is certainly getting into the realms of fine turning/milling. The biggest problem is to get the accuracy, it costs time, and time is money, so it may not be cost effective commercially yet, but for a personal unit in the shed, it would be irrelevant.
My only concern is how strong the finished product would be for thin products like optical train adapters etc.
Sintered products may not cut it, but the newer techniques of laser melting of mixed raw element powders to give "designed alloys" looks really promising.

If you wanted to make a gun, it would be a LOT easier to machine one than to print it using current technology!

Dont think so anymore ( and they have a govt license)
https://www.stratasysdirect.com/blog...d-1911-pistol/
http://3dprint.com/21109/3d-print-metal-gun-reason/
The big advantage of getting the printing process right is it effectively makes a jobbing machine shop obsolete for msny things, as anyone can make something in their spare room. Thats the scary bit.

Somehow, we're a long way from a strictly astro-related thread

Thats why its called "General Chat"
However, to try and get it back to something astro
I also see they are now printing glass
http://www.technologyreview.com/news...glass-barrier/
whilst its still crude at present, imagine if they can get to the point of printing mirrors to nano precision using various ceramics. Lots of new optical designs, esp for mirrors may open up.

Andrew
its an amazing world we are living in

[julianh72 (Julian)]

The PiKon - a 3D-printed telescope (from about 12 months ago):
http://pikonic.com
http://top43dprinting.com/university...n-photographs/

[AndrewJ]

Gday Julian

Sorry no wabbit
They bought the mirror

Andrew

[julianh72 (Julian)]

Originally Posted by AndrewJ

My only concern is how strong the finished product would be for thin products like optical train adapters etc.
Sintered products may not cut it, but the newer techniques of laser melting of mixed raw element powders to give "designed alloys" looks really promising.

When you are working with new materials and fabrication technologies, you need to develop a different mindset, and re-design the parts accordingly.

I'm reminded of when the first cast iron bridge https://en.wikipedia.org/wiki/The_Iron_Bridge was constructed in 1777-1781 (at Ironbridge, Shropshire) - while all the structural members were made from the new-fangled cast iron, all of the jointing and detailing followed established practices for working in heavy wood (use of dovetails and dowels, etc). It took many years for new design details to be developed which were optimised for the characteristics of the new material.

In the case of 3D printed parts made on a "low precision" hobby printer from extruded plastic, rather than striving for the precision and strength that can be achieved with machined aluminium, it makes more sense to accept that extruded PLA or ABS is less strong, less stiff and less dimensionally precise (in its unfinished state), so you make your parts thicker (they will probably still weigh about the same as a functionally-equivalent thin-wall aluminium machined part), and provide a "machining allowance" and / or attach machined metal parts for the critical face-to-face connections. You really need to start with a proper "functional definition" of the part (critical vs non-critical dimensions, strength, stiffness, mass, etc), and redesign it to meet the functional requirements and material / fabrication characteristics, rather than just replicating the metal prototype.

To put a few numbers to it - a 50 mm ID / 1 mm wall thickness tube 100 mm long made from aluminium (e.g. a camera extension tube or similar) would weigh about 45 g (although the full part with T-thread adaptors etc will be rather heavier). If you print a PLA tube with the same internal diameter and length but use a 3 mm wall, it will have a mass of about 60 g. The 3 mm PLA tube will have about 60% of the physical strength (bending or tension) of the 1 mm aluminium tube, but a part such as this is probably more governed by stiffness and durability than it is by strength. A better design for the PLA might actually be a 2 mm tube with a few longitudinal stiffeners, for example - which is trivial to design and make on a 3D printer, but quite challenging to make from aluminium stock if your only machining equipment is a lathe.

[AndrewJ]

Gday Julian

You are making me recall long hours in the "Design for Production" lectures :-)
I have no problem in mindset re designing things to suit the production methods. My original wooden wedge looked nothing like the one i designed when i got a mill and could work alloy plate accurately :-)

A better design for the PLA might actually be a 2 mm tube with a few longitudinal stiffeners, for example

Or a 1mm internal and external wall with say 3mm airgap between em and internal honeycomb stiffeners, as that gets the second moment of the cross section up a bit more, with a massive increase in rigidity.

My only query at present is how long before things like strong thinwalled bits with fine threads could be printed cost effectively.
( As an example, an adapter to go into a 2" barrel ( like the Ted Agos Focal reducer carrier ). You want to keep the wall thin so as not to lose any of the lightpath, so designing thicker isnt an option unless you get rid of the std 2" barrel entirely.)

Again, a few years at the current rate of research and it should be trivial.

Andrew

[pmrid (Peter)]

Gentlemen, as much as I have enjoyed these diversions through the virtues and vices of printing this and that, may I (ever-so-gently) draw you back to the beginning of the thread in which I was bemoaning the evils of planned obsolescence and inviting discussion along those lines.

Peter

[AndrewJ]

Gday Peter

Threads do wander a bit :-)

However, if you had a little laser scanner, you could scan the broken part, and immediately print a replacement.
No more obsolescence due to little plastic bits breaking :-)

Andrew

[AussieTrooper (Ben)]

Originally Posted by julianh72

The commercial 3D printers are indeed designed as "appliances" with built-in obsolescence - built down to a price, and designed to lock you into buying the manufacturer's filaments and spare parts. (Just like 2D printers - they practically give them away, but the ink cartridges cost as much as liquid gold!)

Then perhaps we could 3D print spare 3D printer parts in advance.

[julianh72 (Julian)]

Originally Posted by AussieTrooper

Then perhaps we could 3D print spare 3D printer parts in advance.

That's EXACTLY what happens when you build your first 3D printer - print a set of spares!

(Which might instead become the basis your second - improved - 3D printer, so you need to print some spares for that ... )

3D printing can very quickly become a self-fulfilling hobby - you never actually need to look for anything "useful" to print, because there is no shortage of tweaks and upgrades for the printer itself!

[julianh72 (Julian)]

Originally Posted by AndrewJ

However, if you had a little laser scanner, you could scan the broken part, and immediately print a replacement.
No more obsolescence due to little plastic bits breaking :-)

You can do the scanning with a free app on your smartphone, but it is generally better to use a cheap digital caliper to get some accurate measurements of critical dimensions.

Create a 3D CAD model, send it to your 3D printer,and for a few cents' worth of plastic filament - Voila! No more built-in obsolescence.

[Ausrock (Chris)]

I have no doubt that "planned obsolescence" exists but I suspect that in many cases it is just as likely that the particular plastic formulation used is not entirely suitable for the job at hand........ie: it's too brittle, too flexible, insufficient or incorrect UV stabilisers used, etc., etc.

[julianh72 (Julian)]

With respect to the change of title to this thread:

I, for one, welcome our new 3D printing overlords!