Replies

"the skills required to run a manual CNC mill"

I wasn't talking about "manual" CNC (computer numeric control) mills. You can build one (with software control similar to 3D printing, minus automatic bit changes) for a couple/few thousand if you know what you're doing. Yes you would need some expertise/skill to source parts for and build the CNC mill.

For prostheses, 3D printing is the way to go for weight savings and relative ease of producing complex infill patterns especially for one-off parts. There are a few niches where 3D really shines, and medical prosthetics are one of them.

I doubt it's anywhere near as "low skill" an endeavor as you believe. Possibly somebody wrote and supplied a software package and equipment which makes them able to take a 3D scan of the patient's body and produce a model to print? Or qualified biomedical engineers and doctors spent numerous hours developing 3D-printable universal prostheses in conjunction with the company that produces the printers? That takes considerable technical know-how to pull off, and possibly a good bit of money that isn't included in your costing. This kind of support would probably not be available free of cost to a typical consumer or business in the 3D printer market.

For 95% of typical machine parts, the printers commonly sold as consumer/prosumer models for under $10,000 are less useful than CNC mills because it is impossible or nearly so (at least from what I've read/experienced with current models) to correctly place and dimension clearance/dowel holes and tap bores with the kind of precision that demands. And as you said yourself, 3D printed parts require touch-up after printing. In the case of machine parts where you must have things like perfectly flat or round surfaces and precise alignment for parts to mate correctly in complex mechanisms, they require considerable touch up. They often require the kind of "touch up" you need a vertical mill or lathe for. They require enough touch up that parts are at least as time-consuming and expensive to produce as they would be with subtractive methods. They take an added layer of skill/expense, because you still need to to do some basic machining on top of having a printer and being able to get relatively good results out of it. Add to this that you will not get the full mechanical strength out of the material. It currently takes considerable skill, effort, and expensive software to design around this in applications where a part might be stressed near failure. In some cases (like gunsmithing) you'd have to have a screw loose, or be in a pretty desperate situation to even bother trying.

That's where the affordable 3D tech (extrusion and low-end SLA) is currently at. I'm not saying they're completely useless, but they're limited unless you have alot of time on your hands and access to other tools/equipment. The other tech (high-end SLA and SLS) has not budged in price much the last 3-5 years. It's currently way out of the "consumer" price range. It may come down to earth at some point, or it may not. We'll see. Some of the patents on those technologies expired a few years ago, but they're not easy to reproduce from a technical standpoint. Alot of the individual components which make them capable of doing things like fusing powdered metal to within +/- a couple microns are expensive in their own right even without the development costs and markup.

Possibly somebody wrote and supplied a software package and equipment which makes them able to take a 3D scan of the patient's body and produce a model to print? Or qualified biomedical engineers and doctors spent numerous hours developing 3D-printable universal prostheses in conjunction with the company that produces the printers? That takes considerable technical know-how to pull off, and possibly a good bit of money that isn't included in your costing.

My daughter is a bio-mechanical engineer and works with prostheses in her paid job. Nights and weekends she works with this charity organization. Spanish doctors head the outfit. There's an Indian software engineer working in the Netherlands that has written most of the code. Basically, a doctor working in a 3rd world hell hole takes a few measurements of the amputee's arm or leg and emails them to the s/w engineer who runs it thru his custom coded preprocessor and the G-code for a properly sized prosthesis is generated and emailed back to Calcutta or wherever. If there is a problem with the printout my daughter runs it on her machine at home and fixes up any glitches. So, yes, lots of time and effort and skill went into developing the system but the end result is a desktop 3D printer on a desk somewhere in the third world printing out prostheses in a low tech environment for just a few dollars a pop.

This kind of support would probably not be available free of cost to a typical consumer or business in the 3D printer market.

Not yet, but I remember the days when an Epson dot matrix printer cost north of a thousand dollars and required a service/maintenance agreement. Now an ink cartridge costs more than the printer. What used to be called a photo-realistic printer is now a non-repairable commodity item. 3D printers will be heading down that same path. Looking at the cost curve and the feature growth of 3D printers over the last several years I see Moore's Law in full effect.

Add to this that you will not get the full mechanical strength out of the material.

I think you are only looking at the downside of a 3D printed part here. If the entire assembly is designed to use 3D printed parts from the get go a designer has a world of options only limited by his skill and imagination. Did you see my response to the gunsmith who thought springs could not be 3D printed? Take a look at that article. Those springs don't look anything like a typical coiled spring and they probably couldn't have been made without a 3D printer but they have features that are hard to duplicate with a traditionally manufactured spring. 3D Printed Spring Exceeds Traditional Manufacturing

I was talking to a Honda engineer a couple of days ago that was very excited about some of the part designs that he was getting approved that were simply not possible to be made on mills. Double and triple undercuts. Parts that are enclosed inside of a shell. Parts too delicate to support themselves until their structure is completed being printed with a supporting media that is evaporated away when the printout is done. He said, that for all of their technical expertise Honda, until recently, had been very reluctant to approve a design that could only be made on a 3D printer. But not any more. He was like a kid in a candy store talking about what he was going to be able to do now. I really don't think they would have to pay him to show up to work if they'd heard the excitement in his voice.

I'm not saying they're completely useless, but they're limited unless you have alot of time on your hands and access to other tools/equipment.

Have you ever been inside a MakerSpace facility? For about $200/month you have access to ALL of their equipment. Laser sintering, top of the line printers, the works. Sometimes I wish I was a kid again so I could go play.