I'm asking people with experience - Is my big progect even possible?

[TotalNEWB 0]

To start, I have no experience of my own, only what I have learned from YouTube videos.

I'm asking for your opinions before I bite into something I cannot chew.

I want to make a big statue, that is also very detailed using 3d printing while maintaining a reasonable cost.

 

The way I intend to make it, is by utilizing an onion like layer structure. In the center, there will be a steel column with a bunch of welded steel rings to distribute the weight of the filler.

 

I will surround the metal "skeleton" with a rough 3d printed inner shell and fill the space between the skeleton and the inner shell with "SOMETHING"(concrete/plaster)

 

The last 2 layers will be an outer shell made of photopolymer resin 3d prints and Polyester resin to bind the inner and outer shells.

To summarize: Steel Skeleton > "SOMETHING (concrete/plaster)" filler > rough plastic 3d shell > Polyester resin > 3d printed outer shell.

 Amazing visual representation here: 

Why this layout?

I want the statue to be both big - 15 feet(4.5 meters) and detailed - Something that only resin 3d can do, however the prints are very small. And before you suggest, at this scale it would be cheaper in man-hours to make it in resin, than to make it in plastic and have every nook and cranny filled with plaster and smoothed.

 

The way it would be assembled is like layering a cake.

1st, the base would be poured and the metal skeleton welded. After that, I will work on 40 centimeter increments. Put the plastic shell >pour concrete and let it dry >Assemble the outer shell >Pour resin and let it dry >repeat until finished.

Plaster >polish the joints of the outer shell and primer>paint.

Alternatively, I was thinking about removing the rough 3d plastic shell after the concrete sets and have the Polyester resin on top of the concrete.

 

Ok now let's dig in into the calculations.

I have skipped the Base and the metal skeleton parts because they will not be that expensive compared to everything else.

 

The average human body is 180 by 50 by 30 centimeters if put into a rectangle box. In order for the statue to remain proportionate, I am multiplying the dimensions of the rectangle by 2.5. Giving new dimensions of 450 cm by 125 cm by 75 cm.

Note this is an overkill. Realistically, both the surface area and the volume are about 40-50% less, since the space around the head and legs is mostly empty. But It's better to plan for the worse and hope for the best.

 

I'm using ChatGPT to help me with the calculations because I am bad at math. I try to double-check them myself, but since I have no experience, I just try to spot the chat's mistakes.

 

The volume of a 450x125x75cm rectangle is 4,218,750 cubic centimeter

How many tons of cement do you need in order to fill 4,218,750 cubic centimeters?

ChatGPT:

The density of cement can vary depending on the type and manufacturer, but a common density is around 1440 kilograms per cubic meter. Converting 4,218,750 cubic centimeters to cubic meters: 4,218,750 cm³ = 4.21875 m³ To find the mass of cement required, we can multiply the volume by the density: 4.21875 m³ x 1440 kg/m³ = 6075 kg Converting the mass to tons: 6075 kg ÷ 1000 = 6.075 tons

For a 6-ton concrete mixture using a 1:2:3 mix ratio, you would need 1 ton of cement, 2 tons of sand, and 3 tons of gravel.

Based On the prices where I live, i'm looking at about 2-300 USD for everything.

 

For the inner plastic shell, I''m getting a bit lazy with the calculations. It will not have to be pretty, and it would be mostly hollow, with the exceptions of some supporting cables to keep it from expanding/deforming when the concrete is poured. However, I'm thinking that those cables should be made from wire, so the concrete doesn't just melt them. The shell will be made out of modules that fit into each other and are about half a centimeter to one centimeter thick.

 

How much would it cost to 3d print a sheet that is 125 centimeters wide, 1400 centimeters high and 0.5 centimeters thick?

ChatGPT:

As a rough estimate, we can assume that the sheet will be printed using FDM (Fused Deposition Modeling) technology, which is one of the most widely used 3D printing technologies. The most commonly used material for FDM 3D printing is PLA (Polylactic Acid), which is relatively inexpensive compared to other materials.

Assuming a printing speed of 60 mm/s, and a layer height of 0.2 mm, the estimated printing time for the sheet would be around 58 hours. The amount of PLA filament required would be approximately 8.75 kg, assuming a filament density of 1.25 g/cm³.

The cost of PLA filament varies depending on the quality and brand, but a rough estimate would be around $20-$30 per kilogram. Therefore, the total cost of the filament would be around $175-$262.

 

I don't know how accurate that is. Probably not that much. This guy did a 6-foot master chief for about $850. https://www.reddit.com/r/3Dprinting/comments/wmj42o/my_3d_printed_life_sized_master_chief_is_finally/ However, his statue is solid, and I want just a shell. So I am putting a round $1500 for the shell as an estimate.

 

The next 2 calculations are about the resins. For the 3d printed shell, I'm using a thickness of about 4 millimeters. For the polyester resin, I will double that.

So here is what I calculated for volume:

450x175x4 = 22500 centimeters3 = 22.5 liters x 2 sides = 45 liters

450x75x4 = 13500 centimeters3 = 13.5l x2= 27l

125x75x4 = 3750 centimeters3 = 3.75l x2 = 7.5l

=77.5 liters total

x2 = 144 liters polyester resin

 

This is what ChatGPT had to say

Unsaturated Polyester Resin (UPR) typically has a density of about 1.1 to 1.3 grams per cubic centimeter (g/cm³) when it is in liquid form. However, once it cures or dries, its density increases and can vary depending on the type of UPR and the curing conditions used.

Assuming a density of 1.3 g/cm³ for the cured UPR, a ton (1000 kg) of UPR would occupy a volume of approximately 769.23 liters (1000 kg / 1.3 g/cm³ = 769230.77 cm³ = 769.23 liters).

 

So based on this calculation I will need about 200 kg in order to fill 150 liters of volume. At a price of 1.5 USD per kilo, I am looking at 360 USD plus some unknown shipping.

 

That is for the 3d photopolymer  resin. I have read that it has a conversion rate of about 1 to 10. For every 10 liters of liquid, you get 1 liter of solid. Again, I'm not sure if I'm right here, but if the calculation is correct, I will need 780 liters of resin. They sell it in kilograms for about 20$ at Alibaba so I will need 850 kilograms since I have read that a kg is about 0.9 liters. That is 17 000 usd + about 20% vat. So 20 000 USD plus some shipping cost.

 

Now for the printing I made calculations using Phrozen Sonic Mega 8K Large High Resolution Resin 3D Printer

https://phrozen3d.com/collections/resin-3d-printer-phrozen/products/phrozen-sonic-mega-8k-3d-printer?aff=93#specs

It has a Print Volume: 33 x 18.5 x 40 cm

 

I made my calculations with the following assumptions: Since the sculpture is so big most of the parts will be largely flat with a little curve. I gave them a bit of space on both sides, so I imputed 16 centimeters width 32 centimeters height and about 4 centimeters depth accounting for curvature and supports. (the parts themselves are 4 mm thick as stated)

So I made the assumption that I can fit 8 on the plate at the same time.

 

What is the surface area of a rectangle that is 16 by 32 centimeters, 32 cm * 16 cm = 512 square centimeters.

How many of those rectangles can you fit on 199,500 square centimeters,

Number of rectangles that can fit = Total area / Area of one rectangle = 199,500 cm² / 512 cm² ≈ 389.06

 

If a 3d printer prints at a speed of 70 mm/hour, how many hours would it take it to print 32 centimeters?

ChatGPT

To convert 32 centimeters to millimeters, we can multiply by 10: 32 cm * 10 = 320 mm

To find the time it takes to print 320 mm at a speed of 70 mm/hour, we can use the formula: time = distance / speed

time = 320 mm / 70 mm/hour = 4.57 hours

Therefore, it would take approximately 4.57 hours to print 32 centimeters at a speed of 70 mm/hour.

 

/ 8 pieces on a plate = 48 prints 1 print = 5 hours = 10 days

This is without any rest or leeway. And obviously I need to calculate the cost of the 3d printers 2700$ + 400$ and Base plus steel another 1500-2000$ and paint probably few hundred bucks.

 

One of my worries is that the concrete will deform the 3d shell while curing. But I am assuming if I spray the outside of the shell with water, it will prevent deformation.

So I'm asking you. Are there any big flaws in my logic or calculations?

 

[gr5 2,170]

I am told that ChatGPT is particularly bad with formulas and math.  Just a warning.

 

@rooiejoris made a life sized elephant using Ultimaker 3D printers (about 200X200mm print area but he extended it so that it's unlimited height).  He's also an artist.  He might have ideas.

 

This project sounds quite doable.  It's hard to get concrete (which usually has gravel as you mentioned to cut costs) to fill into small spaces.  You could skip the gravel (then it's called mortar?).  Either way you want some kind of "gentle" vibrator to put against your scupture and vibrate it to get the air bubbles up to the top.

 

It seems to me it would be best to do this in many layers of maybe about 1 foot each.

 

I recommend you do a 1/4 scale version first.  At 1/4 scale it will have 1/16th the surface area and 1/64th the volume.  You will learn a lot and you won't be upset when you make mistakes and you won't lose much money either.

 

Also think about 3d printing nozzle size.  I think Joris used a 1mm nozzle (that's the opening).  Typical printers use a 0.4mm nozzle.  If you double the nozzle diameter (and layer height) you can print 4X faster which will save you weeks of time.  You can do test prints with different nozzle sizes to see how you feel about the resolution.

[TotalNEWB 0]

12 minutes ago, gr5 said:

I am told that ChatGPT is particularly bad with formulas and math.  Just a warning.

 

@rooiejoris made a life sized elephant using Ultimaker 3D printers (about 200X200mm print area but he extended it so that it's unlimited height).  He's also an artist.  He might have ideas.

 

This project sounds quite doable.  It's hard to get concrete (which usually has gravel as you mentioned to cut costs) to fill into small spaces.  You could skip the gravel (then it's called mortar?).  Either way you want some kind of "gentle" vibrator to put against your scupture and vibrate it to get the air bubbles up to the top.

 

It seems to me it would be best to do this in many layers of maybe about 1 foot each.

 

I recommend you do a 1/4 scale version first.  At 1/4 scale it will have 1/16th the surface area and 1/64th the volume.  You will learn a lot and you won't be upset when you make mistakes and you won't lose much money either.

 

Also think about 3d printing nozzle size.  I think Joris used a 1mm nozzle (that's the opening).  Typical printers use a 0.4mm nozzle.  If you double the nozzle diameter (and layer height) you can print 4X faster which will save you weeks of time.  You can do test prints with different nozzle sizes to see how you feel about the resolution.

Very good suggestions, I want as many opinions as possible because the initial setup is 3-4K for printers alone, not to mention resins and other materials and equipment, so knowing whether it's doable from the start is very important. 

I'm worried that the resin calculation speed might be off. I saw a video of a guy who printed a 30ish centimeter tall helmet, and it took him 30 hours, which would bump my  printing time to 60 days

[gr5 2,170]

We are mostly all FDM (aka FFF) people here as UltiMaker doesn't sell any resin printers.  The speed of a resin printer is controlled mostly by the thickness of each layer.  Whereas the speed of an FFF printer is mostly controlled by the size of the nozzle and the volume/weight of the plastic printed.

 

Other things to keep in mind.

 

Assuming you go FFF, ff it all possible, use PLA.  You will get better quality results, you don't have to keep it insanely dry before printing, it can withstand sun and weather, however it does have a drawback: it starts getting soft (like hard clay) around 130F which is usually not a problem outdoors but could be in theory.  A bigger worry is that sometimes people use various forms of 2 part epoxy or resin that gets above that temp while hardening - say if you use the PLA for a mold.

 

That's another reason to do a practice part.

 

The full size elephant by Joris was printed in about a week (maybe?) with 5 printers.

 

I think he has done full sized humans as well with a different technique.

 

 

 

 

[rooiejoris 5]

Hi,

 

Thanks @gr5 for tagging me... : )

The size is really doable. I work a lot for/with artists and advertising agencies. There are some things I would see different I think.

First of all is the costs. The focus here is on materials. The real investment is in 3d modeling and print strategy (knowledge). No matter what printing technique you use. In most of my projects, material costs are mostly a small fraction of the total project.

About the finishing... If you have no experience, you might get very disappointed since your expectations might differ than what could be feasible. And what if the expectations are higher, then it could very well be the lack of print strategy knowledge...

 

The layered structure looks very complex and why is the concrete a must? sometimes it easier to make it cheaper/faster and maybe print a new one if the older one failes than design and make a very robust and 'hooligan proof' object. Also since the print costs are mostly a fraction of the project cost/time investment. (Not always, but yes if you design in a way where you need as less as possible post processing)

 

 

8 hours ago, TotalNEWB said:

However, his statue is solid, and I want just a shell.

 

His statue is most likely not solid...! FDM is almost never solid and uses infill structures.

 

Printing large things is not difficult. Printing large things within a reasonable amount of time is the challenge. Therefore the print/slicing strategy is very important. And when reducing the time, a reprint (of some parts) is also not a big problem.

 

Just some thoughts......

 

cheers / joris

 

 

 

 

[TotalNEWB 0]

7 hours ago, rooiejoris said:

Hi,

 

Thanks @gr5 for tagging me... : )

The size is really doable. I work a lot for/with artists and advertising agencies. There are some things I would see different I think.

First of all is the costs. The focus here is on materials. The real investment is in 3d modeling and print strategy (knowledge). No matter what printing technique you use. In most of my projects, material costs are mostly a small fraction of the total project.

About the finishing... If you have no experience, you might get very disappointed since your expectations might differ than what could be feasible. And what if the expectations are higher, then it could very well be the lack of print strategy knowledge...

 

The layered structure looks very complex and why is the concrete a must? sometimes it easier to make it cheaper/faster and maybe print a new one if the older one failes than design and make a very robust and 'hooligan proof' object. Also since the print costs are mostly a fraction of the project cost/time investment. (Not always, but yes if you design in a way where you need as less as possible post processing)

 

 

 

His statue is most likely not solid...! FDM is almost never solid and uses infill structures.

 

Printing large things is not difficult. Printing large things within a reasonable amount of time is the challenge. Therefore the print/slicing strategy is very important. And when reducing the time, a reprint (of some parts) is also not a big problem.

 

Just some thoughts......

 

cheers / joris

 

 

 

 

As I said, I lack practical knowledge, that's why i'm asking people who have. The statue will be outdoors, It would be covered from UV exposure, but there will still be some wind. That's why I figured that concrete would be appropriate, since it's cheaper. If the structure is not rock solid, the outside shell would bend and crack eventually. That's why I was thinking of removing the PLA shell after the concrete sets and having the 2 resin layers on top of the cement. 

Since you have been doing this, how much would the design of the outside 3d resin sculpture cost. Ballpark numbers 10-20K? more?

[rooiejoris 5]

 

57 minutes ago, TotalNEWB said:

That's why I was thinking of removing the PLA shell after the concrete sets and having the 2 resin layers on top of the cement. 

 

ah, I thought you wanted to print with a resin, but I am told photopolymers get quite brittle outdoors...

 

 

58 minutes ago, TotalNEWB said:

Ballpark numbers 10-20K? more?

 

For me the challenge is to get it as affordable as possible. 20K is a challenge, it is not difficult to spend 50K on the same object. But you will get different details. On the other hand, if you can do more things yourself, also the (metal) construction and have enough time, 5-10K should be feasible.

Usually the more time you spent the better it gets. But many times good enough is perfect.

 

 

I have a PLA sculpture hanging outside for more than 8 years now and it is still hanging there. This is only a single wall print and mostly completely hollow.

 

You can consider to cover the pla print with a coating and keep the PLA.

If using concrete, maybe interesting to do a sort of roto moulding on the separate parts before assembling. So the objects have for sure the concrete more to the outside.

 

Printing moulds is more difficult for the details. Also becaause of the postprocessing of the prints (grinding/sanding/welding), that is more difficult on a negative object.

 

I printed a lot of objects, then made a mould (eg silicon) and then casting the final object in the mould.

In this case chocolate: https://www.rooiejoris.nl/chocolat-sculptures/

 

 

 

 

 

 

 

 

[TotalNEWB 0]

23 hours ago, rooiejoris said:

ah, I thought you wanted to print with a resin, but I am told photopolymers get quite brittle outdoors...

I do want to, as I said, I don't know enough about the material properties. I was hoping that if I give them a few layers of protective paint, they would be fine.

 

After reading a bit more, I'm starting to think that PLA prints that are smoothed and then cast with iron would be more suitable and cheaper at the end.
Do you have any experience with that process, does the PLA shape need to be removed or does the metal just melt it off?

[gr5 2,170]

PLA

 

Plain PLA will last for many years in the sun, wind and rain.  Probably a few decades, out in the sun untreated.  A simple coat of clear or opaque paint can protect from UV rays but if there is a cover then you don't even need that.  Resin is different and I think would need some type of paint coating (clear or opaque).  My biggest concern is heat but if there is a covering from the sun in the middle of the day then you would have to be someplace crazy hot like the UAE before heat is a concern.  Or located in the middle of a parking lot.  PLA is fine up to 130F/52C.

 

casting

 

Casting metal with PLA is done with sand.  You can do "investment casting" where you melt the PLA out of the cast.  This requires an oven to put the entire cast in.  Or you can do multi part casting and preserve the original PLA positive where the portion holding the sand is disassembled into two or more parts, the pla is removed and then the cast is reassembled before pouring in the molten metal.  If you don't remove the PLA it will be a mess.  The PLA would burn creating gasses and smoke bubbles.

 

You can't make a mold with PLA and then pour in something hotter than say 60C (150F).  You would even have potential trouble with two part epoxies which may get above 52C as part of the hardening.  You certainly couldn't pour liquid metal directly into a PLA mold.

 

Designing mold shapes is complex (vents, internal air pockets) and you should watch some youtube videos to learn some of the rules and you should also practice making some smaller ones so you learn from some mistakes.

 

resin printers

 

These have a much higher resolution than FFF printers and people are drawn to that and this makes them great for jewelry (for example) but for something larger than 2 feet high it seems a bit silly since FFF can easily have millimeter and much smaller details.  Whereas resin printers you would need a magnifying glass to appreciate the details.  If people are observing from more than 4 feet away then FFF resolution is fantastic.

 

resin itself, before hardened, is a sticky, gooey, toxic mess.  You need gallons of (I think IPA?) to clean it up along with disposable gloves, many rolls of paper towels, etc.

 

I'm biased to FFF so I'm not the best advisor but I'd avoid resin printers for a project this large.  But they are fantastic for smaller art.

 

[robinmdh 100]

I would not use concrete the weight and heat during curing will break your PLA mold/outer shell.

Small layers (if you only pour small layers at a time the temperature of the exothermic reaction will not rise very high) of polyurethane foam seem like a better plan.
Maybe instead of PLA use PETG or ASA since the higher temperature resistance is worth a bit even if it only has to stand in the sun for a short bit. I'd try and print most of the outer shell, with the FDM techniques with some sanding, filling etc you will be faster and cheaper/faster then covering everything with resin prints, keep the resin prints for details where precision matters, like the face and other small details.

Edited April 4, 2023 by robinmdh

[gr5 2,170]

2 hours ago, robinmdh said:

I would not use concrete the weight and heat during curing will break your PLA mold/outer shell.

I think the heat is fine.  I've never heard someone say - don't touch that concrete - you'll burn your fingers!  But the weight might be a problem if your pours are more than a foot or so thick.

 

I think you really need to do a test piece. 

 

Polyurethane foam sounds like a fantastic idea.

 

[ahoeben 1,948]

1 hour ago, gr5 said:

I've never heard someone say - don't touch that concrete - you'll burn your fingers!

You can burn your fingers on concrete while it is setting. Here is some concrete steaming, 15 hours after pouring:

 

Polyurethane also has an exothermic reaction while curing. The pressure from it expanding in combination with the heat may cause deformation of the "mold" if it is not properly supported.

 

Pour in small batches, not in one go. That way you keep the heat under control.

Edited April 4, 2023 by ahoeben

[Dadkitess 19]

I quite like the approach gr5 shown with the elephant video : by slicing the whole statue in parts, you would be able to print it in large 3d printer (depending on how much slice you're willing to do, it might even fit in a UMS5 even if I would not recommend that printer at all for that). Then you would be able to mount theses slices onto a rigid metal (or anything sturdy enough for your showcase conditions) with anchor parts designed from scratch and allowing for some adjustments.

 

This way you would have a light and cheap metal internal skeleton and easy-to-print and cheap PLA parts to assemble with as much details as you want. A final touch would be to coat everything alltogether to fill in the small gaps between parts, sand it to make it really nice if you need to and then add a protective finish to whistand the wether, UV, any conditions that the statue might meet. 

 

I'm not sure to see the advantage of concrete and multi-layer things.

[TotalNEWB 0]

You guys are a great source of knowledge. So, let me throw you a curveball.
 

I was thinking of making the surface of the statue into a mirror. 

Maybe I could print the 3d parts with PLA, then create negatives using clay and finally get some glass sheets and soften them until they conform to the negatives. Then some silver nitrate and some lacquer to make it a first pass mirror.

[ahoeben 1,948]

1 hour ago, TotalNEWB said:

Then some silver nitrate and some lacquer to make it a first pass mirror.

That would give you an incredibly fragile surface Even removing dust from the surface - let alone finger prints - will damage the mirror.

Edited April 5, 2023 by ahoeben

[robinmdh 100]

you, you can add mirror finishes with sprays,

 shows this process, but I'd add a resin clear-coat to protect it from scratches and fingerprints like @fieldOfView said, it's a fragile and tiny mirror layer.
There are tools used to chrome automotive plastic parts as well that work similarly but may be cheaper.

you can also do a slightly worse mirror like finish with graphite powder on a glossy paint base.

Edited May 3, 2023 by robinmdh