Apple Experimenting With 3D Printing to Create Devices

[Jay Tee]

This is an increasingly common way of manufacturing small, high precision metal parts. Several years ago I designed specialized small 316 screws for an instrumentation product. It was much cheaper to sent the drawings to Germany where a company I had spoken to, made them in their million dollar 3D metal printer. So print and sinter. That’s the main way they’re made these days. But bother metal printing processes are coming out which could prove to be faster and cheaper. Apple is big enough to help that along with money, as they often do.

If Apple increase the amount of components that are 3D printable, this should improve the repairability of the product. Inventory costs should come down too.

The future looks interesting.

 

[jlc1978]

If Apple increase the amount of components that are 3D printable, this should improve the repairability of the product.

I’m not sure how it would impact repairability, which is primarily a design function rather than manufacturing issue.

Inventory costs should come down too.

I doubt it will have much impact on inventory levels. I suspect the printers are not design for JIT manufacturing and are most economic when a multi-unit run is done. Small runs with multiple changeovers to make different parts is likely to cost more than large runs. If it can make parts quicker than today’s methods you could have lower inventory levels, for example enough to meet 2 weeks instead of 4 on hand, since you could more quickly make parts for anticipated demand. It would also mean parts would become unavailable quicker after the 5 year vintage date.

The future looks interesting.

That’s a fact, jack.

 

[melgross]

Yeah, I was under the impression that pricing for titanium powder for the 3d metal printers was even worse (in the hundreds of dollars per KG), so it is strange to see people complaining about Apple not passing on some hypothetical savings on a future process that is still just a rumour.

Because people are often uncritical. They don’t question something if that’s what they want to believe. Or they’re eager to prove someone wrong, and jump when they find something that agrees with their premise.

they also forget that making something from a material can raise the price several times.

slso, there are a lot of somewhat anti Apple posters on this site. It’s known for that. They jump on anything that seems to make Apple look bad.

 

[3xBoom]

what data do you have to support this statement?

First-hand experience with in-house 3D printing of plastic spare parts for scientific equipment. Exceptionally high failure rates require item-by-item attention turning manufacturing into a huge time suck and source of frustration. Granted, we are not using top notch printers but also not walmart crap. I am all for Apple using 3D and hope that this is an indication of them having a robust process.

 

[melgross]

I’m not sure how it would impact repairability, which is primarily a design function rather than manufacturing issue.



I doubt it will have much impact on inventory levels. I suspect the printers are not design for JIT manufacturing and are most economic when a multi-unit run is done. Small runs with multiple changeovers to make different parts is likely to cost more than large runs. If it can make parts quicker than today’s methods you could have lower inventory levels, for example enough to meet 2 weeks instead of 4 on hand, since you could more quickly make parts for anticipated demand. It would also mean parts would become unavailable quicker after the 5 year vintage date.



That’s a fact, jack.

You can make parts with 3D printing that you can’t make any other way. assemblies that now have to be made from two or more parts can be made as one part. Designs that can’t even be made otherwise, can be. 3D printing gives one a vastly greater range of design possibilities. That means lighter weight but greater strength and durability.

it’s also correct that some boards could simply be stored as a file, could be printed on demand. This is something that manufacturers are very interested in for product support for longer terms. The more that can be 3d printed the more useful the concept will be. Right now, a lot of parts can be made this way including some circuit boards.

 

[melgross]

First-hand experience with in-house 3D printing of plastic spare parts for scientific equipment. Exceptionally high failure rates require item-by-item attention turning manufacturing into a huge time suck and source of frustration. Granted, we are not using top notch printers but also not walmart crap. I am all for Apple using 3D and hope that this is an indication of them having a robust process.

What printers are you using. They start at about $200 and go up to almost $2 million.

 

[jlc1978]

You can make parts with 3D printing that you can’t make any other way. assemblies that now have to be made from two or more parts can be made as one part. Designs that can’t even be made otherwise, can be. 3D printing gives one a vastly greater range of design possibilities. That means lighter weight but greater strength and durability.

I agree, and why I said repairability is largely a design issue. Unless you design items to be repairable, how it’s made is somewhat irrelevant. For example, you could make a component one piece that was previously 2 or three, but if it is now 80% of the structure is the design truly repairable? Or if other components are glued in?

it’s also correct that some boards could simply be stored as a file, could be printed on demand. This is something that manufacturers are very interested in for product support for longer terms. The more that can be 3d printed the more useful the concept will be. Right now, a lot of parts can be made this way including some circuit boards.

While I agree, and that was the approach DoD took, how economically feasible is it in terms of the costs of small runs vs large ones where, unlike the DoD, cost is a significant driver? Can the printer print 10 different designs in a row at the same costs, other than material, versus 10 of the same or do setup and other time and cost issues make it uneconomic?

I can see where small runs for very expensive long lived products makes sense since the costs may be small relative to the item’s replacement costs, but does it make sense for an item whose useful life is 3-5 years?

 

[Apple_Robert]

Will those reduced production costs be passed onto the customers?

Absolutely not.

 

[melgross]

I agree, and why I said repairability is largely a design issue. Unless you design items to be repairable, how it’s made is somewhat irrelevant. For example, you could make a component one piece that was previously 2 or three, but if it is now 80% of the structure is the design truly repairable? Or if other components are glued in?



While I agree, and that was the approach DoD took, how economically feasible is it in terms of the costs of small runs vs large ones where, unlike the DoD, cost is a significant driver? Can the printer print 10 different designs in a row at the same costs, other than material, versus 10 of the same or do setup and other time and cost issues make it uneconomic?

I can see where small runs for very expensive long lived products makes sense since the costs may be small relative to the item’s replacement costs, but does it make sense for an item whose useful life is 3-5 years?

Generally that one piece will be more reliable than the two, or more, it replaces. While it does cost more to produce parts that way (now), it’s much cheaper than to have to stock parts on the shelf for older products. A buyer would be more likely to pay more for that part so as to be able to keep the product working than to have to junk it And buy a replacement.

i don’t understand the second sentence in your second paragraph.

 

[jlc1978]

Generally that one piece will be more reliable than the two, or more, it replaces. While it does cost more to produce parts that way (now), it’s much cheaper than to have to stock parts on the shelf for older products. A buyer would be more likely to pay more for that part so as to be able to keep the product working than to have to junk it And buy a replacement.

Thanks, it certainly would make avaiablity of old parts easier assuming manufacturers use 3D printing for them or the engineering drawings are convertable easily to 3D instructions.

i don’t understand the second sentence in your second paragraph.

I was wondering if setup costs and time make short print runs, along JIT printing, more costly than do a larger print run and warehousing product.

 

[melgross]

Thanks, it certainly would make avaiablity of old parts easier assuming manufacturers use 3D printing for them or the engineering drawings are convertable easily to 3D instructions.



I was wondering if setup costs and time make short print runs, along JIT printing, more costly than do a larger print run and warehousing product.

Generally no. It’s being used now in various industries. Shelf cost is very high. We would be talking about hundreds or even, for a big company, thousands of different parts in storage. That really adds up quickly. If all of that could be in solid state storage, one Desktop drive could store hundreds of thousands of parts, along with their individual programs to print them out. Most of this could be done automatically as the part printing could be generated as the order comes in and than it’s little more than a matter of picking and packaging the part for delivery, as it done now when being picked off the shelf in the warehouse. But you won’t need that big warehouse, a much smaller area would suffice.

remember though, that 3D printing is undergoing major changes. Better quality at higher speeds are already here and that will continue getting better with time. I always tell people to not judge by what is currently available, but what will be available a few years hence. After all microcomputers were considered to be toys for a number of years after they came out, then laptops, phones and tablets. But all are now extremely hseful and even necessary.

 

[jlc1978]

Generally no. It’s being used now in various industries. Shelf cost is very high. We would be talking about hundreds or even, for a big company, thousands of different parts in storage. That really adds up quickly. If all of that could be in solid state storage, one Desktop drive could store hundreds of thousands of parts, along with their individual programs to print them out. Most of this could be done automatically as the part printing could be generated as the order comes in and than it’s little more than a matter of picking and packaging the part for delivery, as it done now when being picked off the shelf in the warehouse. But you won’t need that big warehouse, a much smaller area would suffice..

Thanks. I can see a time where 3D printing is done on an assembly line fashion; where say first the case is printed, then the internals put in place, and the the closure printed and installed.

Besides the impact on manufacturing it will have political and social impact as well. Rather than long supply chains and parts crossing oceans, they’ll be produced closer to the ultimate point of sale. Instead of needing tons of cheap labor and designed for purpose tooling, a factory will be run by technicians with some skilled labor; enabling more on-shoring as well as more rapid design cycle. Design chains and updates could be loaded into 3D printing locations instantly around the globe.

remember though, that 3D printing is undergoing major changes. Better quality at higher speeds are already here and that will continue getting better with time. I always tell people to not judge by what is currently available, but what will be available a few years hence. After all microcomputers were considered to be toys for a number of years after they came out, then laptops, phones and tablets. But all are now extremely hseful and even necessary.

I agree. Today’s tech is mereley a shadow of what’s on the horizon.

 

[Jay Tee]

The price for titanium is not 0.35 cents a pound. That’s the price for scrap. The price is around $25 a pound, or around five times that of copper, which is pretty expensive. Your error is what happens when you look something up in Google and accept the first answer as the correct one, which often it is not, as in this case.

i work with titanium. It is difficult to work with and requires special handling. Many workings require an argon atmosphere. It’s difficult to machine and my milling bits are very expensive. One small titanium screw can cost several dollars.

I'm hoping for a silver alloy in the 16 Ultra.

 

[bgillander]

Because people are often uncritical. They don’t question something if that’s what they want to believe. Or they’re eager to prove someone wrong, and jump when they find something that agrees with their premise.

they also forget that making something from a material can raise the price several times.

slso, there are a lot of somewhat anti Apple posters on this site. It’s known for that. They jump on anything that seems to make Apple look bad.

I just happened to come back to this thread and I fully agree with your statement, but the way you worded it as “uncritical” actually struck me with something that has bothered me… that the people criticizing the most (the ones I would refer to as critical) seldom seem to fact check their own data (or as you would say, uncritical). That critical / uncritical dichotomy is as hard for me to understand as Schrödinger’s Cat. 😀

 

[StellarVixen]

All I want to hear about the Apple Watch is one thing:

Blood Glucose, when and at what price.

Aye, now that would be actually disruptive. Classic glucose meters and the strips that you need for them and that expire so quickly would go to history