Text Size

Printing Large Stuff - tables, chairs, etc.

Printing Large Stuff - tables, chairs, etc.

Postby SoftwareAssassin » Fri Dec 06, 2013 2:06 am

I'm interested in 3D printing large objects, for example a table, or a chair, all in a single piece of plastic. What's the largest 3D printer I can buy????

I'm having a hard time finding one with more than 2ft cubed of printing volume and have already started to design one myself.

I'm expecting there to be concerns with printing on a scale like this that don't necessarily present themselves when printing on a smaller scale which seems to be more prominent - such as the cooling of the plastic for example. Anyone know of any other problems I might encounter?

Thanks in advance for all your help!

Chris
SoftwareAssassin
 
Posts: 1
Joined: Fri Dec 06, 2013 1:42 am
CNC Machine: None

Re: Printing Large Stuff - tables, chairs, etc.

Postby PHoodDaniel » Fri Dec 06, 2013 7:30 am

The first thing that comes to mind is the time it would take to print an object of that size. If you are really set on using additive fabrication techniques rather than subtractive (CNC), then you should consider other materials, and/or larger deposition apertures. The resolution will be lower, but the fabrication time will be more tolerable.

The machine can be of a standard CNC with a deposition tool rather than a spindle. The frame of the machine could be open with a long z-axis.
PHoodDaniel
 
Posts: 395
Joined: Sun Feb 21, 2010 7:56 am

Re: Printing Large Stuff - tables, chairs, etc.

Postby Awesomeness » Fri Dec 06, 2013 11:32 pm

To expand on what Patrick is saying, printing something the size of a baseball takes like 8 hours. So something the size of a chair, at the speed of conventional 3D printers, would be months or possibly years. Then the factor of cost, where material ranges $0.50-15.00 per cubic inch, something the size of a chair could be a couple hundred bucks... but at least you'd have several months to pay for it! :-P
If you don’t have time to do it right, when will you have time to do it over?
Image
User avatar
Awesomeness
 
Posts: 2886
Joined: Wed Feb 24, 2010 7:46 am
Location: Firestone, CO
CNC Machine: BlackFoot v4 4x8

Re: Printing Large Stuff - tables, chairs, etc.

Postby keith » Sat Dec 07, 2013 6:24 am

ONe answer to the question you have asked is yes you can print almost ANYTHING of ANY size.

BUT only if you have LOTS of money.

Do a Google search for "Enrico Dini 3d printing" he has a machine which prints using concrete and he is printing houses and house sized objects.

But for home 3D printers this sort of printing is going to be too costly unless you have a VERY understanding wife......
The Titanic was built by professionals; the Ark was built by Amateurs..........
keith
 
Posts: 596
Joined: Mon Nov 23, 2009 11:18 pm
Location: North Wales, United Kingdom
CNC Machine: BlackToe v2 4x8

Re: Printing Large Stuff - tables, chairs, etc.

Postby keith » Sat Dec 07, 2013 6:37 am

Sorry but I just got interested in it this topic and again the answer is YES if you have a lot of money :(

Have a look at what they are doing at Loughborough University here in the UK:

http://julianh72.blogspot.co.uk/ :D

http://www.constructiondigital.com/gree ... f-concrete :D

Of course if you look around the Universities in Canada and the USA I am certain that a few of them are doing research into this.

I think that in the future it will be an extremely valuable addition to the construction industry. I know that the Oil & Gas industries are making very good use of Laser sintering stainless steel to make valves without joints or seems....

Maybe I should look around the local university to see if they need anyone to help with a concrete printer project, hmmmmm..... :)
The Titanic was built by professionals; the Ark was built by Amateurs..........
keith
 
Posts: 596
Joined: Mon Nov 23, 2009 11:18 pm
Location: North Wales, United Kingdom
CNC Machine: BlackToe v2 4x8

Re: Printing Large Stuff - tables, chairs, etc.

Postby davecove » Mon Feb 17, 2014 1:35 am

"printing something the size of a baseball takes like 8 hours. So something the size of a chair, at the speed of conventional 3D printers, would be months or possibly years."

That sort of assumes that you are trying to print that large object with the same 0.3mm nozzles as you use on small stuff right? If you were printing the same object, scaled up by a factor of 3 and used a 1mm nozzle the print time should be similar, no?

Dave
User avatar
davecove
 
Posts: 11
Joined: Thu Apr 18, 2013 4:32 am
Location: Galveston County, TX
CNC Machine: Custom

Re: Printing Large Stuff - tables, chairs, etc.

Postby Awesomeness » Mon Feb 17, 2014 6:15 am

davecove wrote:"printing something the size of a baseball takes like 8 hours. So something the size of a chair, at the speed of conventional 3D printers, would be months or possibly years."

That sort of assumes that you are trying to print that large object with the same 0.3mm nozzles as you use on small stuff right? If you were printing the same object, scaled up by a factor of 3 and used a 1mm nozzle the print time should be similar, no?

Dave


Yes, that's true. However, printing with a large volume nozzle mostly defeats the purpose of 3D printers (and CNCs), which is precision. Generally speaking, there are faster ways to do all these things, but CNCs do them very accurately. If you don't need the accuracy, you're giving up the chief advantage.

For example, another topic that has been discussed on these forums is "CNC vs. Table Saw". If I need to cut a bunch of 3" wide boards, I can do it with either of these tools. On the table saw, I set the rip fence to 3", and push the sheet of plywood through a dozen times, for a total time of 5 minutes. With a CNC, I probably can't even draw the CAD file in 5 minutes. Even with a super fast machine (e.g. ShopBot) that cuts at "table saw speed", has auto-zeroing mechanisms, etc., I'm never going to beat that 5 minute time. Where I start to gain ground with the CNC is cutting large quantities of something, or cutting really obscure shapes that would be tedious and time consuming to lay out and cut individually.

So yes, you could stick a "hot glue gun" on the 3D printer, and print large objects fast, but inaccurately. I can't think of many circumstances where this is particularly helpful, especially considering the cost too. Maybe if you're some kind of artist, trying to design an automated piece of art, that prints objects to "demonstrate to humanity the psychological effects of how robots have taken the pride in workmanship away from the common man", or something bizarre like that.

As the volume of what you're building increases, your resource consumption (e.g. time, cost, etc.) goes up cubed (length x width x height).

As an aside, thinking about the problem more, going to a bigger diameter nozzle only increases the output speed in two dimensions, while your build volume increased in three. So in order to really use a 3x bigger nozzle to build a 3x bigger object in the same time, your nozzle's extrusion rate (movement speed) also needs to multiply by 3x. In turn that means you're consuming significantly more energy to melt the material, in a fused deposition modeler. I'm not sure if it's realistic to scale the extrusion speed (or related melting energy) that far.
If you don’t have time to do it right, when will you have time to do it over?
Image
User avatar
Awesomeness
 
Posts: 2886
Joined: Wed Feb 24, 2010 7:46 am
Location: Firestone, CO
CNC Machine: BlackFoot v4 4x8

Re: Printing Large Stuff - tables, chairs, etc.

Postby davecove » Mon Feb 17, 2014 3:41 pm

Why would it be inaccurate? It seems to me that the using of a 0.3mm nozzle with a 0.1mm step is no better than using a 1mm nozzle with a .3mm step if the object being printed is 3x larger in its cartesian directions. That is, the relative size of a flaw seems to be the same.

Let's take something like an R/C airplane wing. If you are designing something yourself, you are going to need to CAD the design regardless of whether you plan to print it or hand build. Once that is done I would rather click 'print' and go to bed than start cutting out balsa or coroplast. It seems to me that a 36" wing with a 1mm nozzles would be just as functional as a 12" with made with a 0.3mm nozzle.

I'm interested in this because I am in the middle of building a 'steel Blackfoot' and I would like to be able to switch the router out for an extruder at some point.

Dave
User avatar
davecove
 
Posts: 11
Joined: Thu Apr 18, 2013 4:32 am
Location: Galveston County, TX
CNC Machine: Custom

Re: Printing Large Stuff - tables, chairs, etc.

Postby Awesomeness » Mon Feb 17, 2014 5:11 pm

davecove wrote:Why would it be inaccurate?


It's not the positioning of the extrusion that is inaccurate, it's the ability to replicate the fine details of a model that inaccurate. It's like trying to build large, detailed objects out of Legos. Because a Lego block is a certain "large" fixed size, you can't have any details that are smaller than the block. Also, the items you build must be in sizes that are multiples of the blocks. If you try to make something that is 1/2 way between two block sizes, your only choices are to either go up or down one block, leading to a 1/2 block inaccuracy.

The issue I mentioned at the end of my last post really has me thinking this will be even harder than it sounds, too. The extruder on a 3D printer doesn't have a hard time thoroughly heating through the thin filament quickly. There is some fixed amount of energy required to melt a fixed volume of plastic. A thick extruder (e.g. 3x thicker) requires 3^3=9x (3x wider, 3x taller) more energy to melt the filament. The way you get more energy in is by raising the temperature, but plastics are poor conductors of heat. So this single 3x larger extruder, that you must heat 9x hotter, at some point just starts burning the outer layers of material without melting the core of the filament. This would lead you to need to develop some other system, such as several smaller extruders that melt thin filament, and get combined into the one larger melted stream.

I don't know at what point this would happen, but since your heat/temperature/energy requirements are going up squared (exponential) in relation to the diameter, it's going to happen rather quickly. And then you have to tune the machine to print slowly enough to allow the previous extrusion to cool/solidify sufficiently, before laying down the next layer, which is happening exponentially slower too.

It's also worth observing that large extruders, like you describe, aren't in use on the open market. That brings up the question "Why?". Is it just that nobody has thought of it yet? Maybe. It could also be that it's not useful, or has pragmatic challenges that are difficult to overcome (like I described above).

A major issue is still cost. FDM 3D printing costs $0.33-0.82 per cubic inch. So printing a solid "chair leg", 1" diameter, 20" long, costs $20.72-51.50. That probably makes the whole chair in the "couple hundred bucks" range, of material itself. You also have to print support material structures to hold up any unsupported free spans, which worst case means printing a 30% solid lattice under the entire object/volume... time and cost cubed (exponential) again. (And, before someone asks, could you print the chair leg hollow, instead of solid? Maybe. But remember that your weight/mass bearing requirements are going up cubed as the build volume increases, but your material's physical properties are staying the same. So you need to thicken everything up, the bigger it gets. The larger your extrusion gets, the less contact surface you also have between extruded layers, making it weaker.)

To me, this just seems like a wall of challenges to try to overcome. It is possible in some specific instances, and impossible in others. You would need to have a very clear vision of what you want to do exactly, and how much resources you are willing to invest in accomplishing it.
If you don’t have time to do it right, when will you have time to do it over?
Image
User avatar
Awesomeness
 
Posts: 2886
Joined: Wed Feb 24, 2010 7:46 am
Location: Firestone, CO
CNC Machine: BlackFoot v4 4x8

Re: Printing Large Stuff - tables, chairs, etc.

Postby davecove » Mon Feb 17, 2014 6:53 pm

I think your costing math is off... a 20" chair leg with a diameter of 1" has a volume of 15.7in3. At $0.33 to $0.88 per cubic inch, that's $5.18 to $12.87. It looks to me like if one were to use an extruder and buy ABS pellets (~$5/lb) you could get the cost down to about $0.19 per cubic inch and a total cost of $2.98 for the leg. That seems OK for a chair leg.

Raising the temp isn't the only way. To get a volume of plastic up to temp you could also heat it at the same temp for a longer time. Given the same feed rate, this just means a longer 'oven' (or as you mentioned, using a 'parallel' oven). A combination of a moderately higher temp and moderately longer oven (serial or parallel) could be just the ticket.

What the heck anyway... we're here to experiment right?

Dave
User avatar
davecove
 
Posts: 11
Joined: Thu Apr 18, 2013 4:32 am
Location: Galveston County, TX
CNC Machine: Custom

Next

Return to DIY 3D Printing

Who is online

Users browsing this forum: No registered users and 1 guest

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 158 on Thu Mar 08, 2012 8:54 pm

Users browsing this forum: No registered users and 1 guest