3D cement printer

> The first stage- the metal parts, motor, hardware,
> software, firmware- is done. Everything looks and
> moves nicely. I was able to print, though the
> extrusion is unstable. After several tries, I have
> come to the conclusion that it's better to use a
> progressive cavity pump for the extruder.
> Here is where I run into a problem, as the
> cheapest price I could find for this pump is
> $5,500 for a new pump (and I can't find any used
> ones).

Maybe try something else. Why not print a very thin layer with a dry mix and wet the printed part with a mist sprayer(fogger) as the layer progresses. Then you can use a gravity feed from a small tank fed with a small conveyor belt system and the water is easy to manipulate. I do not know if this will work but it may be worth a try.
By going the dry print route you can manipulate the water very accurate in order to get a strong concrete - less water means stronger concrete and if you for some reason have a pause in printing you can easily proceed after a while without worrying about a lot of wet concrete in pipes and pumps. There can be a mist sprayer ahead of the printer head to wet the part to be printed for better adhesion of the layer following. The mist will also act as cure for the concrete - not let it dry out to quickly.
Just a thought.

Edited 1 time(s). Last edit at 07/07/2013 05:26PM by Xabbax.

Isn't there some way of having a wheel with scoops push the cement down a tube? Or like a gearwheel with many small teeth so it goes very progressively? Ofcourse, the wheel will get VERY dirty, but if you make sure it doesn't touch the machine on any other part, it's manageable.

How about using a auger drill bit, rotating in reverse, with a close tolerance inside a piece of DOM (drawn over mandrel)(it is seamless) pipe. Drive the auger with a stepper to control flow. The hopper would need a vibrator on it to keep the concrete moving to the auger.
It would be trial and error to see how big the auger you will need to get enough flow (pressure).

Rodney

Yes, my first design was exactly like you said: Auger bit, vibration, etc. Unfortunately, it doesn’t work. The cement mortar releases water at any vibration; thus, the extruder gets clogged, especially towards the end. It works and pumps mortar only if there’s no stopping. If it pauses, even for a few seconds, it starts to clog again, and eventually clogs up completely. Mortar is a very strange, unpredictable material.
Also, the scoops, gears , etc might work for only thin, low-velocity mortar, but for high solid mortar, they do not.

Xabbax Wrote:
Why not print a very
> thin layer with a dry mix and wet the printed part
> with a mist sprayer

About the dry mix- I like the idea, But here is disadvantage- From real-life experience, during construction, the area can be very windy; thus, it is very likely the dry mix will be blown away. But either way, I’m going to experiment with dry mix and water anyway, and see how well it works.

when do you think v2 of your printer will be done by?

Yesterday I saw a technique of spraying on concrete with a spray nozzle that had the cement being extruded from one side slowly and an air compressor that was connected to the side of the spray-nozzle that pressed out the concrete in a spray.

This worked incredibly well, I know it's still the problem of getting the concrete there, but it looked like something useful.

AndreyR Wrote:
-------------------------------------------------------
> 3D Cement Printer
>
>
> I am prototyping a cement printer. So far, I have
> the first version. It’s 1/3 of full scale and
> the major


In case no one pointed it out (sorry I didn't real the whole thread), there is a professor who has a ted talk on 3d printers and has already extruded about fist-sized tracks of cerment for building cheap houses in poor areas of the world. You should find out how they did it !

6:45
[www.youtube.com]

P.S. just to vent. Say you've just invented the first 3D house builder that changes the game in terms of cost and labor and structural design. And what do you do - build a rectangular wall. I'm just saying. When the technology (and cost) for modeling a Pinto is nearly the same as building a Ferrari, you know the problem is going to become artistic design very soon. There just aren't enough good models out there.

Edited 2 time(s). Last edit at 07/10/2013 12:46PM by Simba.

> About the dry mix- I like the idea, But here is
> disadvantage- From real-life experience, during
> construction, the area can be very windy; thus, it
> is very likely the dry mix will be blown away. But
> either way, I’m going to experiment with dry mix
> and water anyway, and see how well it works.

The movement of the head along certain paths is not really a problem. I think the design of the printer head will be the crucial element of the printer.
Some ideas: different feeds for the water, cement, sand and the aggregate. computer control of each of the 4 elements. mixing compartment for the cement and aggregates and then the controlled printing of the mix and finally the addition of the water by spraying. If you can control all the elements at all time you can change the design of the mix to have a stronger concrete ( more expensive) on certain sections of the print eg. above the windows and the doors while the infill can be printed with a more economical mix - very effective use of materials.
The head must be able to rotate to the direction of the print - can be complicated and maybe a bit bulky at first.

Also just wanted to point out that even in the construction of skyscrappers, the cement engineer has to manually validate each batch, and will frequently through out entire batches. Where the sand came from, on what day, how much water was in it before...it all affects structural integrity. So its one thing to build it and say it worked, vs. building something that will stand up to code, and for this field, I don't know of any cement making process that has been perfected and can work without the human element.

Most of our furniture is square, I see no problem here.

My bed is square, my door is square, my table is square..

If the man is trying to create a way of cheap housing, it's very logical to keep it square.

I understand your comment, but why should you 'not' want to have it square? I think just making something organic so it's not square, a kind of crappy idea it has to fit concept.



Simba Wrote:
-------------------------------------------------------
> AndreyR Wrote:
> --------------------------------------------------
> -----
> > 3D Cement Printer
> >
> >
> > I am prototyping a cement printer. So far, I
> have
> > the first version. It’s 1/3 of full scale and
> > the major
>
>
> In case no one pointed it out (sorry I didn't real
> the whole thread), there is a professor who has a
> ted talk on 3d printers and has already extruded
> about fist-sized tracks of cerment for building
> cheap houses in poor areas of the world. You
> should find out how they did it !
>
> 6:45
> [www.youtube.com]
>
> P.S. just to vent. Say you've just invented the
> first 3D house builder that changes the game in
> terms of cost and labor and structural design.
> And what do you do - build a rectangular wall.
> I'm just saying. When the technology (and cost)
> for modeling a Pinto is nearly the same as
> building a Ferrari, you know the problem is going
> to become artistic design very soon. There just
> aren't enough good models out there.

What I'm trying to say is, you have the ability to make fantastic organic work, but because it is hard to draw in 3D most prototype work is made out of squares and circles in cad or sketchup. Whether you like angular/square designs or organic is totally individual decorating preferences and a la mode design - but now that we have 3D printers I see more squares and less curvy designs, and what I'm saying is that's a shame. I would have expected to see more organic design, so that means we are still in a very early/calibration phases.

Interesting enough, BTW, it is hard to make a shell of a rectangle than it is an organic shape. Organic shapes 3D avoid warp better, because they do not pull in the same direction, allowing the plastic to relax. Of course in the case of cement, I do not think it is thixotropic enough to stand better than a 5 or 10 degree slope, most 3D printers are now doing 40-50 degrees with ease, and even 60 degrees (30 from the horizontal) sometimes cleanly.

Today, I performed an experiment with dry mix and water. In the beginning, it was very promising- the dry mix was extruded flawlessly- but when I added water supply to the nozzle, the nozzle started to clog, and as you can see in the video/pictures, this technique completely failed. At least for now. As you can see its failed for few reasons..
[www.youtube.com]

Here are two different videos I found that got success with dry mix/sand and liquid. But from what I can see, they are able to build art sculptures but not houses.
My goal is to make a real machine that is able to build real houses in real life. So it looks like I still need a progressive cavity pump; it appears to be the best choice for pumping high viscosity mortar.
[www.stonespray.com]
[www.youtube.com]

What if the water and dry mix weren't mixed inside the nozzle? Could you have a spray of water on the side of the nozzle to avoid clogging?

Fl0yd

Simba Wrote:

> > there is a professor who has a
> ted talk on 3d printers and has already extruded
> about fist-sized tracks of cerment

they are from the University of Southern California, and they have quite a bit of sponsors. Their team has several professors and a couple dozen students. However, the only video of their printer functioning I've come across has been circulating the internet for many years with no info about any progress. It looks like this video is from 2007. I'd really like to know what they're working on now, and in the several years since the video, have they actually built anything? Based on their videos that I've watched, this project reached its peak in 2007 and hasn't made much progress since. There have been videos of new ideas, but no videos of actual results that I've found.

Ohmarinus Wrote:
-------------------------------------------------------
> Yesterday I saw a technique of spraying on
> concrete with a spray nozzle that had the cement
> being extruded from one side slowly and an air
> compressor that was connected to the side of the
> spray-nozzle that pressed out the concrete in a
> spray.

good for existed walls but how to build a wall without any support?

AndreyR Wrote:
-------------------------------------------------------
> Today, I performed an experiment with dry mix and
> water. In the beginning, it was very promising-
> the dry mix was extruded flawlessly- but when I
> added water supply to the nozzle, the nozzle
> started to clog, and as you can see in the
> video/pictures, this technique completely failed.
> At least for now. As you can see its failed for
> few reasons..
> [www.youtube.com]

I do not think this was a failure. The concept worked perfectly well.
Some suggestions:
1. some vibrational tool added to the head to ensure the flow of the mix.
2. nozzle size big enough to let the mix flow ( think of hourglass where the hole must be big enough for the sand not to clog )
see comment here[forums.reprap.org]
3. rectangular hole with the width a bit smaller than the width of the wall
4. Water - a fan type sprayer that will cover the width of the wall ( garden micro irrigation nozzle 180 degree or smaller)
5. some way of rotating the head to follow the movement of the head to keep the watering at the back (and possibly in front) of the mix
6. a much thinner layer at a time to avoid piramide forming

KEEP UP THE GOOD WORK

Edited 1 time(s). Last edit at 07/11/2013 02:43AM by Xabbax.

Imagine a chain pulled up through a bucket of mortar, and compressed air blowing the mortar of the chain,
the chain could be replaced with a strip having holes or a rotating disc with holes, this would give the advantage of DOD,
although the dots may be large.

---

Random Precision

AndreyR Wrote:
-------------------------------------------------------
> ... In the beginning, it was very promising-
> the dry mix was extruded flawlessly- but when I
> added water supply to the nozzle, the nozzle
> started to clog,...

Is it not possible to avoid adding the water inside the nozzle? Why not just spray it on the dry cement /after/ it comes out of the nozzle?

johnrpm's idea with a chain or belt being pulled through the mix and then pushing the mix out of that conveyence also sounds promising. Perhaps there is a better method of dislodging the mix than air... A belt with holes going around a sharp bend at the nozzel exit should drop the mix.

---

[www.linistepper.com] Open source stepper motor drivers.

Something of this size you do not want to use Chain, belts or even threaded rod.

The distance it too large.

Belts/Rods will sag under their own weight.

You will have to move the driving mechanisms to the carriage itself, like an overhead crane.

OverHead Crane

I have lots of ideas for the cement printer, and I thought I’d share one of them here. It’s very likely a crazy idea, but might be a boost for the next generation of cement printers. You can see my idea in the attached picture.

The most complicated part of the cement printer- how to make layers dry/cure quickly in any weather condition. If we make an additional electric arc, it might help with this process, and I’m not just talking about the heat, but about an electrical/chemical reaction.

Electricity can help activate or start some chemical reactions in mortar. The question is, what reactant/additive/chemical/powder to mix with the cement mortar? If we can find such an additive, the layers will cure quickly and we could print year-round; as currently, the printer is limited to few seasons during which it can print, this will open up a huge number of opportunities.

What do you think about limestone?

Wow, this is really neat... nice work.

I wonder if it's possible to incorporate reinforcing wire in future versions ? concrete is not very strong without some form of reinforcing... yes ?

I am no expert on concrete, but various accelerators are available, chemicals, but they can adversly effect the final strength.

---

Random Precision

I've been thinking about this the last few weeks since my last comment on using the auger.

Back 20 or so years ago, I did drive a mixer truck for a few years so I do a know a little about concrete.
I was on a jobsite where they used a pump truck to pour a smoke stack basically the same way a printer would do it. They did use what is called a slip form that as the concrete was pump in, it would move up. With this type of pour there is no cold joints (set concrete joining new concrete) and vibrators (used to mix old concrete(a few minutes old) and fresh concrete) were used as the hose of the pump truck was moved around the stack.

Start with a dry mix using the auger idea and then have a mixing area at the nozzle area where you could mix water and some calcium chloride (will make it set real fast) at the nozzle . Using a real hot mix (lot of Portland cement) with calcium chloride and you could have almost a instant set. The more Portland you add the stronger the mix will be also so the need for steel is gone. The one thing is that you'll have a cold joint on every layer. I don't know how this will affect it. Maybe have another nozzle in front of the concrete nozzle with it spreading a thin layer of adhesive.







Rodney

I wonder if a A Moineau pump could work for your concrete... "A Moineau pump is a type of positive displacement pump with steady flow (no pulsing). This type of pump is common in the food processing industry because it can efficiently pump slurries (like soup) without crushing the contents. "

[www.thingiverse.com]

Looks like it can be used to extrude clay and ceramic mixes
[www.youtube.com]

---

[www.linistepper.com] Open source stepper motor drivers.

update?

I really want someone to 3d print a cement structure for me one day so here is my .02c

You can try to invent your own cement nozzle but why reinvent the wheel when you could modify something that is similar to do the job for you. Take a look at these videos. It seems like you haven't researched what has been done enough.

Spraying concrete

Spraying lime cement

Mortar Sprayer

Cement extruder

clay extruder


For added strength chances are the cement might need some filler/binder. Rather than spraying lime cement on bales of hay you could design a dual nozzle for chopping fiberglass matt or hay and cement in one line. Think of it as a fiberglass gun but with cement and tuned into a nozzle.

Fiberglass gun

This just got me thinking that fiberglass could probebly be pretty easily 3d printed with a bit of UV curing resin, uv lights and a fan to keep things cool.

Years ago I did an unrelated project (electronic rugby goal verification, before high speed, hires digital cameras and DSPs) for a chap that had developed a way of high speed building fabrication.

They built small 2/3 room houses out of a set number of standard panels. They used full wall size moulds.

The process was something like the following.
* 1: Apply mould release
* 2; Spray concrete skin with chopped fibre (polyester roving was fed into a cutting machine) with fast cure additive
* 3: Fit electrical and plumbing pipes
* 4: Spray core concrete mixed with expanded polystyrene beads and an additive to prevent beads floating up
* 5: Spray other concrete skin like the first
* 6: Hand float the surface to be smooth
* 7: Move mould down the line and do the next wall
* 8: After 12 hours lift mould past vertical and tip out wall and cure humid in vertical position
* 9: After 2 days transport panels to site and erect on 7 day old waffle foundation in 4 hours

They had a set of moulds for two houses and a truck crew that would assemble two houses a day and space in the yard to cure 6 sets wall panels. Last I heard he relocated to the other end of the country to build some classrooms.

While this was a bit of a rough operation, and the owner was an inventor investing his own money, it many not have been quite as successful as he hoped or he got tired of the bureaucracy that was already plaguing him when I met him (I have not heard of someone building two houses a day in the last 10 years) it goes to show that processes and additives for fast cure and fibre reinforced concrete do exist.

With enough light weight fillers concrete may be almost thixotropic and set soon enough to do a layer on top while it keeps curing. I would be happy with a 3D house printer that does the wall core to replace the brick work and then still need trowel or spray on plaster (stucco) or siding outside and plaster/drywalling/panelling on the inside. If the heavy lifting and accuracy is done by machine then the variable finishes could be done by hand giving the best of both worlds.
[en.wikipedia.org]

There is a technique of mixing chopped metal (often stainless steel) wire 20 to 50mm long (with a few kinks) into concrete to prevent stress cracking during drying. It allows for casting (non-stop) of infinite slabs for factories, warehouses, reservoirs that would do better without expansion joints. Concrete shrinks on setting and unless one cuts expansion joints in a slab during the first few weeks the concrete will crack where it likes before it reaches full strength in 28 days. With the chopped wire the slab is strong enough to get to full strength without cracking even though it is under tension.

Introducing interesting additives, reinforcing fibre or insulating fillers will be the long term goal of 3D printed concrete structures. Perhaps the large aggregate could be 'shot' or dropped in after the cement/sand/water mix is extruded.

The goal here is to put crazy ideas out into the open so they cannot be patented and to give the big money ideas to develop the first commercial units. They are the ones who will have to certify the code compliance of the concrete mixes and additives and in the beginning it will be proprietary for a while but in the end a village will lease a printer for 3 months and build all the houses they want and then the machine moves on. We need to remember that the DIY aspect of tinkering with the process will remain but the need to own a house printer will seldom be a private need and few will have the money to even DIY build one while the thought is entertaining. In South Africa the building code is pretty good. It has specifications and then it has 'deemed to comply' means of achieving them. Such as a double brick wall is adequate for a wall of such and such height and width, if one stays within the accepted limits one does not need a civil engineer. If trying something new one just has to find a tame civil engineer to calculate the strengths (he will want to see test results) and sign off on the safety of the design.

With regard to the university projects you can be sure someone somewhere is continuing on with the development. It is quite possible that one day we will wake up and see a machine on site printing a house. More likely it will be used out of public scrutiny for as long as possible to prevent what happened to all the Jacquard looms when the weavers guild got upset. Builders should remain happy, even in countries with a negative population growth there is still a need for builders and repairs and renovations will still require builders so there is little chance that any disruptive technology will be able to eliminate all builders overnight. Alternately it will be publicly announced at some disaster relief site.

---

Kalle
--
Lahti, Finland
The only stable form of government is Open Source Government. - Kalle Pihlajasaari 2013

hackaday has an interesting article about Paste extruders including concrete:
http://hackaday.com/2013/10/16/3d-printering-pastestruders/

---

[www.linistepper.com] Open source stepper motor drivers.

After several months of experimentation, I finally made a custom extruder which works in the way it needs to for the cement printer (stable on any speed). This was not an easy task, and I made 5 different versions before the current working one. In the photo, you can see the very first print. Unfortunately, my garage workplace is close to freezing right now, and it would be hard to print several layers in one trial. But it will be even more interesting to experiment with cold conditions. I hope to be able to make these prints within the week, and when I do, I will post pictures or videos of my final results.
Thus, the first version of the prototype is done, with the prototype working just fine. My idea to make a cement printer based on the Rep Rap project and using attainable parts has finally become a reality.
Now on to version two.