What would a replicator due to todays economy?

[Kinnison]

Timo said:
Provided, of course, that energy costs weren't prohibitive. Yet I doubt the replicators really expend anything like E=mcc, or require an antimatter battery to hook up to. The Uxbridges on a rather explicitly infrastructure-free planet received a tabletop model (supposedly "limited in capabilities", but still able to provide food, clothing and fresh water among other things) in TNG "Survivors", after all.

How do you know it didn't have an antimatter battery, though?

 

[Timo]

Even if it did, it couldn't be as energy-hungry as E=mcc, because then every kilogram of food they made would cost at least half a kilogram of antimatter. The device could thus provide a dozen meals at most before it ran out of fuel.

More generally, I don't see how the device could have a "battery" of any sort when it needed to operate for a considerable length of time. It would need a proper "power source" instead, a system that can sustainably extract energy from some inexhaustible fuel source such as sunlight, or seawater, or geothermal tap, or something like that. And the small device we saw probably couldn't have any such system capable of meeting the E=mcc requirement - today, such a powerplant would be several city blocks in size.

But the odds are, replicating (or transporting, an apparently related technique) doesn't require the full E=mcc involved in turning energy to matter. Instead, it probably somehow "cheats"...

Timo Saloniemi

 

[Plecostomus]

I'm sorry, but these are NOT the same as "real" parts. The physical properties of the parts produced are not the same as, say, parts machined from raw steel stock, or cast... or of plastic parts which are injection molded.

It's sufficient for "demonstration" parts, to check fit and form... but you don't get "real" parts.

I use this in my job. Virtually every system I work on is "prototyped" this way first. It's great for checking fit/form/function. And for illustrating stuff to people who can't visualize in 3D from looking at 2D drawings.

Just wanted to clarify.

Depends on the technology. Stereo-Laser Lithography is not suited for production unless you have very specific properties in mind that can be served by one of the several SLL-specific resins. As polymer-science marches forward we'll be moving twords production-scale SLL.

Selective Laser Sistering on the other hand, company I work for has three of them running around the clock doing quick-turn production. This process lays down a thin layer of material (metal, ceramic, plastic) and fuses it, then lays down another layer, and fuses it. Eventually you build up a 3D part. They are buying a forth machine that can do dissimilar materials in the same part. SLS is suited for small-scale production, and of all the technologies this is the one that companies are hot for right now.

Takes about 190 seconds for me to make an item the size of a BIC-pen out of metal. We're limited to items the size of a beer-can or many smaller items. Tray-size and chamber size is what limits us with this technology... and the amount of money my company is willing to spend.

I'm assuming Star Trek ships have advanced forms of these machines in place in addition to the replicator systems. Perhaps the terminal we see is the end-point for a system of waveguides. Depending on what you order it's either assembled out of resequenced protein, atomic-manipulated matter, or other raw materials depending on the process required to make said item. Said item is then beamed from the replicator array below decks to your terminal.

Portable units would be far more limited in scope, perhaps limited to "just" molecular re-sequencing of proteins and polymers instead of atomic/quantum work

 

[Cary L. Brown]

Depends on the technology. Stereo-Laser Lithography is not suited for production unless you have very specific properties in mind that can be served by one of the several SLL-specific resins. As polymer-science marches forward we'll be moving twords production-scale SLL.

Selective Laser Sistering on the other hand, company I work for has three of them running around the clock doing quick-turn production. This process lays down a thin layer of material (metal, ceramic, plastic) and fuses it, then lays down another layer, and fuses it. Eventually you build up a 3D part. They are buying a forth machine that can do dissimilar materials in the same part. SLS is suited for small-scale production, and of all the technologies this is the one that companies are hot for right now.

Takes about 190 seconds for me to make an item the size of a BIC-pen out of metal. We're limited to items the size of a beer-can or many smaller items. Tray-size and chamber size is what limits us with this technology... and the amount of money my company is willing to spend.

I'm assuming Star Trek ships have advanced forms of these machines in place in addition to the replicator systems. Perhaps the terminal we see is the end-point for a system of waveguides. Depending on what you order it's either assembled out of resequenced protein, atomic-manipulated matter, or other raw materials depending on the process required to make said item. Said item is then beamed from the replicator array below decks to your terminal.

Portable units would be far more limited in scope, perhaps limited to "just" molecular re-sequencing of proteins and polymers instead of atomic/quantum work

Okay... let me rephrase my comment slightly.

It is true that SLS (the aforementioned "selective laser sintering") systems could, eventually, be used in lieu of more conventional powdered-metal fabrication... since both are based upon sintering processes.

But SLS cannot... EVER... create a part that has the same strength as a forged part, or a cast part (well, unless you have unbelievably poor casting quality!).

SLS creates a semi-porous matrix of particles which are "melted" (the real term is sintered, but most folks here won't know that) together.

In the case of plastic parts, the material is very weak... because the majority of load-bearing capacity in plastic components exists in the very high-density area at the "skin."

An injection-molded polycarbonate part, for instance, may be so tough as to be literally beyond your ability to bend... but an SLS part made from the same base material will snap in half easily.

Now, SLA (that's "stereolithography"... which is the term used in the United States, exclusively, to describe the process you referred to as SLL...) polymer parts can be more robust... these are, at their best, very similar to what some of you may be familiar with from hobby "garage kit" resin models.

As for polymer science marching forward... well... yes, some materials are available now which are far more robust than earlier SLA resins (which were almost "glasslike" in their brittleness).

But the materials used cannot have any "fill" (no glass fiber fill, no mineral fill, very limited coloration, no elasticizers... nothing like that). The coloration is also limited to a very few options, but since that is not structurally-related we can ignore that.

You do not get the "skin effect" ... you cannot get a smooth surface (there's always a stair-step effect along any surface but the very top-most one... assuming it's perfectly flat. And you have to have a support structure "web" in there as well... which further harms the initial surface finish) and there's no compression-effect at the skin layer.

The resins are always THERMOSET materials... principally we're talking about epoxy resins. The resin must be very thin (though not necessarily "near-water-consistency" it can't be very much thicker... think "cooking oil" as the max allowable viscosity) in its uncured state and must be very fast-curing (since it's essentially being cured a few thousandths of an inch at a time, layer by layer, by a laser).

You cannot use SLA for metals. You can use SLS for metals, but it gives you a porous metal-particle matrix (to envision it, imagine a piece of cake, where the little particles are all sort of stuck together but it's not SOLID). There are situations where the reduced strength of this part is acceptable. You can do a few things to "seal" the outer surface (typically, this involves melting copper pellets which infiltrate into the matrix... sometimes they'll "coin" the outside... hitting it in a stamping press... to compact the outer layer).

Just like with conventional sintered parts, the porosity can be a BENEFICIAL thing. Sometimes this is used in bearings... you heat the part, apply a vacuum, then dip it into oil and reapply the pressure... the oil gets sucked into the matrix giving a "self-lubricating sleeve bearing". Also, this is sometimes used to create filtration systems... you can control the average pore size pretty easily, and thus filter out particles from the atmosphere much more effectively than with most other technologies.

But the parts are ALWAYS much weaker than "real" parts.

One last thing... you said that you guys sinter CERAMICS? This seems very unlikely to me. Sintering involves putting adjacent particles against each other under sufficient pressure and/or temperature to cause some migration of the particles across the boundary... essentially "locally melting" without melting the entire particle.

I am unaware of ANY ceramic material which is capable of being sintered, under any circumstances whatsoever. I'd be VERY interested in hearing how you guys are sintering ceramics. Unless you're essentially using clay and using the laser as a layer-by-layer "kiln"... (it becomes "ceramic" through a chemical conversion process during the baking operation... prior to that it is not ceramic at all) but I've never heard of anyone doing that... and would question the mechanical properties of any part produced by that process as well. So, share... I'd like to hear more.

By the way... here's the biggest SLA part I ever worked with... this was an electronic-power-assist-steering package I worked on years ago. We were trying to confirm that the part, as designed, would actually physically fit (the body-and-assembly folks were notorious for giving away real-estate that they'd already promised to other teams).

 

[Plecostomus]

You know your stuff. The ceramic material is a very fine powder that we fuse with the laser, then we take the item and bake it in a kiln for maximum strength... which still isn't very high. The SLS wiki article mentions ceramics, maybe this is a recent innovation.

Also I just ran some glass-filled nylon parts last week. Granted we're not talking Long Glass Fiber but it did indeed have a light loading of glass fiber in it. Problem is it tends to separate out and jam the feeder and there absolutely no way to orient the fibers. With injection molding the fibers orient in the direction of the flow.

We're always pushing the limits of these machines, I'm not sure what was running in the one that caught fire but it smelled like rotten eggs. Never heard an engineer yell so loud.

All our metal items are compressed further in a press and most of them get an additional machining operation.

You're partly correct about the injection molding... It's not just the skin. We're packing the part with upwards of a ton of pressure sometimes more, so under that skin is a densified polymer structure. I majored in plastics before life forced me to switch to metalworking...

However you'll get no argument from me on the rest, as I said you know your stuff.

And that's an impressive part. Our machines can do an item the size of a beer can.

I suspect we could go on for hours about this, what say we take it to PM and let everyone return to the Trek aspect.

 

[Kinnison]

Aw, no! It was getting interesting!

 

[Plecostomus]

Pfft it's nothing more than a case of "my pseudo-replicator is bigger than yours" at this point.

 

[Bill Morris]

Nice photo, CLB.

Two other types of present-day replicators that I mentioned in a previous thread are (1) what is called a bread machine (Panasonic calls the one I have Home Bakery; my wife bought it here in Japan and uses it to make bread and teacakes) and (2) a dish replicator that stores dishes as discs, which it forms on demand into plates, bowls, etc. Once returned to the machine, each dish is flattened again for compact storage. I posted a YouTube demo before. That’s a working prototype, not yet commercialized.

http://en.wikipedia.org/wiki/Bread_machine

But a concept of mine for fast prototyping, which probably no one has independently conceived and implemented (be my guest), would fairly quickly make parts of quality similar to aluminum engine blocks (so possibly requiring machining after initial forming). So its usefulness would be limited in that way, but it might be cheaper, at least. And it doesn’t build the object quite the way the previously mentioned systems do (so it’s less related to Trek replicators). This involves setting a block of polystyrene foam in your chamber, where it would be whittled into the desired shape by drills (rapidly sweeping side to side and working their way from top to bottom, controlled by a system similar to that of a printer, following your CAD file), then the walls close, the chamber is filled with sand, and you do lost-foam casting with aluminum, Or you can remove the foam and do the casting elsewhere. So you’re left with an aluminum part in the shape of the foam carving. It’s crude and low-tech, perhaps, but at least another idea.

 

[Cary L. Brown]

Nice photo, CLB.

Two other types of present-day replicators that I mentioned in a previous thread are (1) what is called a bread machine (Panasonic calls the one I have Home Bakery; my wife bought it here in Japan and uses it to make bread and teacakes) and (2) a dish replicator that stores dishes as discs, which it forms on demand into plates, bowls, etc. Once returned to the machine, each dish is flattened again for compact storage. I posted a YouTube demo before. That’s a working prototype, not yet commercialized.

http://en.wikipedia.org/wiki/Bread_machine

But a concept of mine for fast prototyping, which probably no one has independently conceived and implemented (be my guest), would fairly quickly make parts of quality similar to aluminum engine blocks (so possibly requiring machining after initial forming). So its usefulness would be limited in that way, but it might be cheaper, at least. And it doesn’t build the object quite the way the previously mentioned systems do (so it’s less related to Trek replicators). This involves setting a block of polystyrene foam in your chamber, where it would be whittled into the desired shape by drills (rapidly sweeping side to side and working their way from top to bottom, controlled by a system similar to that of a printer, following your CAD file), then the walls close, the chamber is filled with sand, and you do lost-foam casting with aluminum, Or you can remove the foam and do the casting elsewhere. So you’re left with an aluminum part in the shape of the foam carving. It’s crude and low-tech, perhaps, but at least another idea.

Your idea is a good one, but it's not "rapid prototyping" at all. That's a real-world PRODUCTION PROCESS, for all purposes.

As for the milling process you mention, that's called CNC machining (computer-numerical control) and it works very well. It's used to make low-volume machined plastic or machined metal parts.

For the casting process, it's very common to do the sort of thing you mention. Typically, though, you use SLA or SLS to make the "pattern" and, depending on the material you use, can either use the original part as your "lost wax pattern" or can make a crude mold off of that pattern and cast multiple copies to use as patterns from real wax.

The company I was at 'til a few weeks back did very low production volumes (in the range of less than a hundred of any particular part per year, with full production runs over several years of less than a thousand). So permanent production casting wasn't always justified, and we used these processes for our PRODUCTION work.

You're thinking right... but it's not really anything new. On the other hand, each casting cost significantly MORE than they would have cost had there been permanent "production tooling."