Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Automotive Towards the *fully* 3D-printed electric car

  1. Jul 3, 2017 #1
    An article from 2015:

    3-D-printed car could hit streets next year. Chris Woodyard, USA TODAY 4:48 p.m. EST November 12, 2015
    http://www.usatoday.com/story/money/cars/2015/11/10/3d-printed-car-local-motors-swim/75530830/

    Several companies have come out with what they call "3D-printed" cars, but none have 3D-printed the most important part, the engine.
    This would be difficult to do with an internal combustion engine, with its high temperatures, multiple moving parts, and high tolerances.
    But it shouldn't be too difficult with an electric engine. In fact considering there are now miniature 3D-printers on the market for the home, an amateur could be the first to produce an entire, scale-size, 3D-printed car. And then it could be scaled up to produce a full-size, working, fully 3D-printed automobile.
    This would revolutionize the industry, obviously.
    The two most difficult parts would be the engine and the transmission.
    This video shows how you can make your own simple electric motor:

    How to Make an Electric Motor at Home - YouTube.


    Looking at the steps in the video, it appears they could all be accomplished by 3D-printing.

    Bob Clark
     
  2. jcsd
  3. Jul 3, 2017 #2

    Mech_Engineer

    User Avatar
    Science Advisor
    Gold Member

    Your arguments do not logically follow each other; the technologies involved in consumer Fused Deposition Modeling (FDM) 3D printers are miles different than Direct Metal Laser Sintering (DMLS) processes. DMLS requires very expensive machinery ($300k+) that sinters parts using a powerful laser in an inert gas environment; home 3D printers squeeze melted plastic out of a nozzle like a hot glue gun.

    I see no obvious path towards a fully 3D printed vehicle as you claim. Do you have any examples of a printing process capable of "printing" a functioning electric motor? Even just the rotor or stator?
     
  4. Jul 3, 2017 #3

    JBA

    User Avatar

    Metal power deposition printing and successive sintering (a modified method of creating essentially powder metal parts) currently performed by a number of 3D printing companies should be an easy method for creating the required electric motor core and body components but would not eliminate the need for the copper wire windings and possibly not the motor shaft. No gear system required for 4 wheel direct motor drive systems but gears have been produced by the same method.
     
  5. Jul 6, 2017 #4
    You are correct that metal 3D-printed parts by amateurs were only designed by them, but had to be actually printed by one of the large 3D-printing companies.

    Still, that leaves open the possibility that a scale-model car could be designed by amateurs to be fully 3D-printed by one of the large companies.

    The largest of the professional, metal 3D-printers common now can 3D-print parts about a foot across and cost about $250,000. So you can imagine a 3D-printer that can 3D-print parts, say, 10 feet across, would be 103 = 1,000 times larger in volume and mass, and perhaps a thousand times more expensive, to $250 million.

    An expensive proposition. But if it can be shown a scale-model car can be fully 3D-printed then it might be worthwhile for a large industrial company to invest in this when it would mean any car of any model could be 3D-printed on demand.

    Bob Clark
     
  6. Jul 6, 2017 #5
    Yes, this is the method that could most easily be copied by amateurs for 3D metal printing since it similar to the method used now by amateurs for doing plastic 3D-printing:

    Desktop Metal Production System.


    The only complication is that it is a 3 step process: the part is 3D printed, then put in a solvent bath to dissolve the binder, then finally put in a high temperature oven to sinter the metal together.

    However, the method itself probably is not patented since another company Metal X is using the same process:

    Solid Metal 3D Printing Under $100k - Markforged Metal X | CES 2017.


    Many amateurs working independently could come up with improvements to the process.

    For instance, is it possible to combine the metal deposition, dissolving the binder, and sintering all into a single step?

    Bob Clark
     
  7. Jul 6, 2017 #6
    Ideally, you would want the entire car to be 3D printed, but you could have most of the electric motor be 3D printed and the core wire windings placed separately.
    The method of metal powder deposition printing, most similar to the method amateurs use for plastic 3D-printing, would be difficult to create also the wire windings at the same time as the electric motor. The reason is it requires an additional step of heating at high temperature and the wire insulation would melt. Perhaps there could be used high temperature ceramic coating for the wires.

    Other methods for 3D-metal printing such as electron beam and laser deposition do not require this extra step of furnace heating so should be able to do the copper wire winding, assuming the 3D-printer has the capability of rapidly switching out the material being deposited between metal and plastic as needed. For instance if you imagine a horizontal slice through the electric motor. You would have wire insulation, then the copper wire, then insulation, then copper wire, and this pattern would be repeated, then you would have the core, then the insulation, copper pattern repeated again. A 3D-printing method that could rapidly switch between metal and plastic insulation material should be able to do this.

    Bob Clark
     
  8. Jul 6, 2017 #7
    Direct drive systems do not require gears? I was not aware of that. I would think gears or some method of "mechanical advantage" would always need to be used to match the motor speed to the speed of the vehicle.

    Bob Clark
     
  9. Jul 6, 2017 #8

    russ_watters

    User Avatar

    Staff: Mentor

    I don't think you heard his objection. It doesn't have anything to do with scale and doesn't even primarily have to do with the different methods used, its that there are no methods in use or even in development that can create the key parts (the moving parts, wiring and electronics. There is no path currently in development that comes anywhere close to a fully 3D printable car on any scale.
    Why? Why is this a desirable goal?
    Your description of the complexity level is overly simplistic and even then there is no such technology on the drawing board.

    To repeat: there is no currently active technology development path that ends in a fully 3D printable car.

    ...and not for nothing, but the OP reads like a failed attempt at bait and switch: we can't 3D print a gas engine, so let's switch to an electric motor which....er.... we also can't 3D print!

    To be frank, there is considerable usefulness with 3D printing, but right now it seems to me like its scifi/fad value is swamping its actual value.
     
  10. Jul 6, 2017 #9

    russ_watters

    User Avatar

    Staff: Mentor

    The term "direct drive" means no gears (or pulleys). The reason gas cars have gears is:
    1. They have one engine and have to power at least two wheels. Electric motors can be placed at each wheel.
    2. They have a limited range of RPM where they function well. Electric motors produce about the same torque at any RPM.
     
  11. Jul 6, 2017 #10
    Exactly. This is what I call "Falling in love with a technology". That approach doesn't help us, we should be technology agnostic, and look to results. Why would we want a car to be 3-D printed? Why?

    3-D printing can be a great approach in certain cases, generally small production, or custom/semi-custom parts. But anyone who has seen the production speed and low per part cost of metal being poured into molds, or stamped, or plastic blow molding (generally with multiple items produced in one step) should be impressed with how very slow and expensive 3-D printing is in comparison.

    Use each technology where it fits. Don't force square pegs in round holes.
     
  12. Jul 6, 2017 #11
    Impracticalities abound - financial, economic and technical.
    A positive return on investment is what most entities consider very important for the go/no go decision.

    It the technicalities were to be overcome, then,
    If your cost example of the just the machine of 250M$ is anything to go by, which it may or may not be, if that one machine could perhaps produce one car a week, that's a retail price of 500,000$/car just to re-coup the $ for that one investment with no profit. Add in labour, material cost, plant, and maintenance, operating costs, and profit, then one could be looking at $!M to $2M per car as a wild guess.

    Feasibility analysis
     
  13. Jul 6, 2017 #12
    The reason you don't see this happening is that it's just not economically feasible.

    Why wear out your million dollar SLM machine making an engine block when it can be cheaply cast.

    I also don't see SLM making a crankshaft stronger than forged steel.
     
  14. Jul 7, 2017 #13
    As to why, because of the sci-fi dream of just dialing in your vehicle and see it created before you. Got bored with your Porsche? Download the specs on a Ferrari and watch as your home 3D-printer builds up your new Ferrari layer by layer before your eyes and has it ready in a couple of days.

    DeskTop Metal says their production level machine will be able to produce multiple metal parts at 8,200 cm3 per hour. With say the steel content of a car at, say, 1,000 kg, 1,000,000 grams, and the density of steel at 7.8 gms/cm3, 1,000,000/7.8 = 128,000 cm3. The DeskTop Metal production level machine could 3D print this in 128,000/8,200 = 15.6 hours.

    The DeskTop Metal production machine though only has a manufacturing volume of about a foot across. So this would have to be scaled up for producing large parts. The DM chief engineer says the process could be scaled up to produce something like a car chassis:

    VIDEO: How Additive Manufacturing Can Produce Metal Parts en Masse. James
    Anderton posted on June 06, 2017.


    Bob Clark
     
    Last edited: Jul 7, 2017
  15. Jul 7, 2017 #14
    Another 3D-printing technique that doesn't use laser sintering has better promise for producing metal parts at high speed:



    Bob Clark
     
  16. Jul 7, 2017 #15

    jack action

    User Avatar
    Science Advisor
    Gold Member

    When I'm bored with my Porsche, I go to the Ferrari dealership and get my new Ferrari, at worst, in a couple of days.

    So why do I need - at home - a machine larger than a car, that will certainly cost more than the Ferrari itself? And I still have to go out to buy the material and do maintenance on the machine!

    Yah, I really hate when I'm printing a car and the process stops with the error message «The copper cartridge is empty.» ?:):mad:
     
  17. Jul 7, 2017 #16
    Yes. The cost of these machines is high now. But as I said the "binder jetting" method used by Desktop Metal is something that could be copied by amateurs since it is so similar to what they are doing now with plastics. With many amateurs doing the same thing as the commercial machines the price for these machines would come down.

    Bob Clark
     
  18. Jul 7, 2017 #17
    You are just talking rich guy, hobby level "cool wants", nothing with any practicality at all, even for high end cars.

    I doubt you have ever worked in high volume manufacturing. The speed and capability of the equipment is incredible. You simply cannot serially lay down material as fast as you can stamp it, die cut it, mold it etc. And you can do things like mold a part, and then do a little CNC machining for the surfaces that require higher precision. It's why we have so many functions on silicon chips now (both electrical and micro-mechanical), those are batch processed in parallel, they don't build them up transistor by transistor. OK, I'm not sure, but they may actually use a light beam to serially move along and outline/define the transistors - in the old days that was a batch light/photo-mask process. But they will batch everything they can.

    Will cars have specific parts made with these 3-D printers? Maybe, and most likely for expensive, low-volume or custom cars. But printing the entire car? There is no reason, no motivation for it. No. Not in our lifetimes. And while these technologies advance, so do the competing technologies. Maybe not as fast, but they are so far ahead in so many ways for some of these parts, it's not even comparable.

    Again, the right tool for the job. You are trying to shoe-horn a technology in where it doesn't fit. Fine if you want to do that for personal reasons, but no way will I invest in your company if that is your goal!

    It's like asking "Instead of reaching into the cabinet and pulling out a glass to get a drink of water, could I have a 3-D printer print a custom design glass for me every time I'm thirsty? And it could remelt the material and recycle it to make the next one!" Well, I suppose you could, but it is a solution looking for a problem.
     
  19. Jul 7, 2017 #18

    jack action

    User Avatar
    Science Advisor
    Gold Member

    I thought you were going to say it's like asking «Instead of opening a faucet to get water from the municipal aqueduct, you have a "water printer" that builds ##H_2O## from your hydrogen tank.»
     
  20. Jul 7, 2017 #19

    russ_watters

    User Avatar

    Staff: Mentor

    That's what I figured. Please don't confuse sci-fi dreams with technical and economic realities. This is a technical forum, not a sci-fi forum and posting a sci-fi dream on a technical forum does not make it technically or economically (or legally) possible.
     
  21. Jul 7, 2017 #20

    JBA

    User Avatar

    Production cars (US) and trucks (Daimler) already have 3D printed parts. As described in an earlier post, GE's gas turbine division acquired two 3D companies that it is now using to print parts of it gas turbine engines and they have been able to replace 35 suppliers with just those two 3D sources, as reported by a GE Division executive in a presentation speech.
    Daimler is now expanding their 3D printing by placing machines at their spare parts supply centers for providing 24 hr delivery of replacements for their 3D printed parts.
    One US auto manufacturer (I believe it is GM but I do not have the sources material to check) has also installed a large volume 3D machine in their design division to replace the full clay models of their prototype body designs.
    The idea of using 3D printing for a complete car is not necessarily a practical idea but many corporations are already deep into this technology.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Towards the *fully* 3D-printed electric car
Loading...