Hacker News new | past | comments | ask | show | jobs | submit login
3D Printing Vending Machine on the UC Berkeley Campus (3dreambox.com)
90 points by kaptain on June 4, 2013 | hide | past | favorite | 16 comments



I pass by this machine all the time, so hopefully I can offer some insight:

* It's pretty much just a Makerbot in a enclosure. There is an Android tablet that you enter your information into which allows you to retrieve your object.

* The machine is surprisingly busy, even though a lot of people I talk to on campus don't know about it. I suppose that's one of the perks of having a lot of people in one area.

* It's great for people to make 3D models without the hassle of setting up and maintaining a printer of their own.

I'm curious to hear the metrics to see if it's in line with my casual observations.


What a great minimum viable product!

Ignoring the fact that there's obviously going to need to be more breakthroughs in 3D printing before this can scale I think that this setup could shed some light on the demand for consumer 3D printing. A large spike in demand followed by a bunch of complaints about the user experience/wait time/build quality/etc could tell you which part of the printer process needs to be optimized first before this can gain widespread acceptance. Excited to follow this in the months to come.


Great concept, but it's a bit ahead of it's time. We'll get there, but consumer-grade 3d printing is nowhere near unattended operation like this.

Stereolithography maybe, but additive 3d printing tech like the one shown in this video suffers from all sorts of issues requiring operator assistance - like the object unsticking from the build plate, and the head needing frequent cleaning. Unless this "vending machine" is manned, I doubt it could be useful/successful until technology improves.


Exactly true. Today's tools* require far too much maintenance & operator involvement to be fully automated like this.

Shapeways.com & Ponoko & others have "upload & print" (by mail) services, & they work well, since they have operators in their facility.

*But, things change fast. It might just be a few months or years before the machines can reliably produce without maintenance & operator steps. Is this it already? I'm curious & signed up & will give it a try.


I like this model. It makes a lot more sense to get people used to the technology without a huge upfront investment.

The biggest reason that I don't have a 3d printer is after the first week of making things, I have no idea what I would possibly use it for. Having a setup like this would be perfect as parts would still be rapid (compared to having them made & shipped) and it would save the large upfront expense.

Sure there may be some quality & operational issues to address, but that's fine.

A webcam might be useful for both the end user to watch his part being created and for the operator to remotely check for jams, clogs, messes, and other disasters.


I believe there is a webcam inside the machine that allows you to watch parts being made.


Though I like the idea, I feel like this would work much better as a business if you had professional printers instead of (what appears to be) Makerbots in the vending machine. The fact that this is limited to small, one colour designs greatly lessens the appeal. Also, I am concerned that objects would break coming down that shoot mechanism.


So some back of the envelope calculations here...correct me if I misstep:

They average 5.70 per product. Assumption - 1 product per hour. Assumption - 20hrs/day (leaving time for adding more plastic and down time)

This totals about $114/day and at 7 days a week this would take approx. 16 days to pay for the box itself. Now include expense of filament (again, approx $20 per day for 20 pieces or so, this is a guess and highly depending on the product and I am assuming they took that into account) for a total of about 20 days of constant 20hr/day operation to pay for just the materials and printer. Now include the costs of creating a queuing system for procrastinators who don't pick up their stuff right away. And technician time to manage errors, refills, maintenance, etc. If you can get 250 people to buy a product this sounds like a decent idea.

Now just make sure neat things can be printed and you are a winner.

I may have to try this.


On this scale, it seems like the retail frontage space would be a non-trivial expense.


It does take a considerable larger investment though to get that started compared to a makerbot.


This could be really cool, although as other people have said, achieving unattended operation seems like it will be challenging.

Does anybody know how the metal printing is supposed to work? Selective laser sintering machines that I've seen all seem way too big to fit in a vending machine.


We had one where I used to work. I wasn't responsible for it or anything, but I'd guess it's base was around 8ft by 4ft, and it stood about 6ft tall.

This one worked by depositing a thin layer of aluminum/magnesium powder over the working base. A laser would then fire, igniting the magnesium which would fuse together the aluminium. The laser draws the shape of that layer of the feature. When the layer is complete, the process repeats for the next layer.

The biggest thing I'd worry about is liability and cost of operation. Because their raw material mix is basically thermite, they have a not insignificant potential to explode during use. While removing parts, for instance, you need to worry about sparks igniting the airborne mixture creating a firebomb. The material is also quite hazardous to your lungs and eyes even when it's not aflame.


This seems cool, but I was wondering: Why is it called "3D printing" machine and not "sculpting" machine?


The object is from the bottom up by layering extruded plastic. The nozzle that dispenses this plastic moves back and forth, just like a printer that has the ink nozzle slide back and forth on the carriage.


"Scuplting" implies something being carved away from a block of larger source material. 3D printing builds an object by adding layers of material in the exact pattern of the object.


In trying to figure out what I was trying to say, I found the following on wikipedia: "Durable sculptural processes ... used carving (the removal of material) and modelling (the addition of material, as clay) ... A wide variety of materials may be worked by removal such as carving, assembled by welding or modelling, or molded, or cast."

It seems that building up the material could fall under sculpture, which is called "modelling."

Maybe we should call them 3D modellers instead of 3D printers. Except that "3D modellers" is kind of ambiguous. It sounds like a person who makes 3D models.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: