Wax printing

Hi All,

Just an idea to throw out there, perhaps plant a seed with someone.

Many years ago (1988) there came out a range of Tektronix Phaser printers (Xerox took over the line and they also sell laser printers under the same name). These came out at a time when high-speed true colour lasers and inkjets were rare and are still used for proof printing and other niche markets. They were CMYK solid wax printers and had an industrial grade print head that can last up to a million pages (the black in was free but power cycling used a bit of the colours as well).

[en.wikipedia.org]

Basic function was a 4 colour page (210mm) wide print head with nozzles every 5mm across or so (about 42 for each colour) This print head would move across only 5mm during a page print. The printed image was sprayed (piezo pumps) onto a heated smooth (silicone oil wiped) drum that had a circumference of page height (about 300mm). It would print one line from each nozzle (all colours) around the drum and then step over a bit and do the next line, repeat 60 (guess) times to cover the 5mm nozzle spacing and cover the whole drum. The still soft image was then offset against a sheet of paper in one turn and it would set instantly and the drum was wiped clean with a (silicon) oily felt. All this was pretty quick and the quality was pretty good with vibrant wax.

Now the issue, this nozzle has to print horizontally (usually onto the drum because it has galleries with molten ink that is automatically topped up as it is consumed by melting drop by drop from solid blocks (they look and work like jumbo wax crayons, the black ones were free). The head is likely very expensive to make but in quantities could be affordable. or availbale as a spare part. Moving a build platform up and down in front of the standard heads might be possible to test the technology.

What if one had a regular wax and a water soluble poly vinyl alcohol in just two sets of jets. Arrange it so the openings point down. Print one layer at a time in wax and support (with voids where savings would make sense as only the wax and PValcohol shells are needed) on a bed that moves in one axis and the head moves across slowly from side to side to cover the jet spacing.

Dissolve the support and you have an investment casting pattern that is limited in size by your head width.

Working with a puddle of molten wax and a piezo head may initially scare people from experimenting but it has all be done so it is possible. Piezo heads can be arranged in arrays and gaps between the jets is not a problem as one can scan across to cover the spaces

As an aside I recall there was a range of IBM mainframe colour inkjet printers [Canon Colour Ink Jet printers. Model PJ-1080A. OEM by IBM for mainframe terminals] that has a single jet per colour and were low resolution (60/72 dpi) like early dot matrix printers, I remember a biology lab dude once purchased a few I had picked up to use the piezo pumps for spraying samples onto electrophoresis or chromatography substrates or something like that, this was 15 to 20 years ago.

Kalle
--
Johannesburg, South Africa

Yes, I am fimiliar with them, I occasionaly search on auction sites to try and get one to play with, a friend did
buy one some time ago, but I am not so lucky, I think they have a lot of potential for hacking, if you get your hands on one please post any info.

Edit, Canon Colour Ink Jet printers I had one from the skip, the printheads were tiny, must be very rare now.

Edited 1 time(s). Last edit at 08/19/2013 01:31PM by johnrpm.

---

Random Precision

I poked around inside one of the Phasers once for a friend, I cleaned out a wax spill and changed the main drum timing belt, got it going again. Hacking would be a bit beyond me but the resolution should be around 300 to 600 dpi on the early models. The latest might be better and faster. The layers will still be very thin and there may differences in jet characteristics that would cause uneven deposition bands. Might be possible to scan and compensate though to keep the thing on a even keel. Would be nice to see if a poly vinyl alcohol os some water soluble meltable stuff could be run through the jets though.

They have always been over priced in my opinion and never had a chance to strip one.

---

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

If I understand this correctly, the proposition is to print using melted wax that cools down and solidifies.

When I was a kid, we used to drip candles over a glass of water. The wax floats and cools and you get these decorative wax plates. Anyway...

What if a regular plastic printer was placed in a refrigerator and made to print wax. This could potentially be very high resolution.

What if this printer printed a mold of wax and filled the mold layer by layer with resin that then cured.

Also it could print more like a ink-jet as stated earlier.

I am still a bit caught up in my earlier idea about Cold 3D Printing, so I'm seeing it everywhere. Also there are some good ideas under the topic of "3d printing using paper".

For example, if it was difficult to get the dots to line up precisely maybe one could layer the model with a fine mesh of some material similar to how one makes fiberglass. This could also add strength.

A number of 3D printers on the market use wax inkjet technology. For example Solidscape. On the wiki, I wrote up some experiments on subtractive wax machining using a heated nozzle. It might be possible to "reverse" the process and do wax deposition with similar hardware, although getting an acceptable viscosity would be challenging.

If I remember right, the forum search doesn't match anything with three characters or less, so you won't ever get any hits if you search the forums for "wax". Instead, google something like "reprap wax". Then you find interesting things like this Heated Piezo for Jetting Wax, hidden away on the wiki. I wonder if it works...

Hi,

Bright idea alert.

gmerz: Your idea of dripping of wax on water and working in a fridge gave me a revolutionary synthesized idea. The general problem with wax is that it is runny while hot and will not stay in place long enough to set. Using cold air will not help as it does not cool fast enough. But what if one had a runny wax extruder that was working in chilled water or silicone oil or whatever will not mix, dissolve or add a barrier to wax. Have the support liquid with the same density as the wax and there are no forces to move the wax from where it was placed and the cold liquid will conduct the specific heat of fusion away much faster. This would work just like a FFF RepRap and have no problem with layer heights because the nozzle will deposit at the right height and push any stuff that is too height out of the way. One would just need an ultra smooth flow peristaltic pump in a heated chamber and insulated pipes to extrude the wax.

The liquid in Lava lamps comes to mind, can handle high temperatures, almost the same density as molten wax and does not prevent the molten wax from sticking together again. It smells a bit like a paraffin/white spirit but may be a slightly less common petroleum fraction that will not mix with the wax. The wax in the lamps may also be special and one would want to use the standard investment wax so the casting house does not get upset with contamination/

So many experiments just waiting to be done. Remember you saw it here first and it is now public domain unless patented previously.

MattMoses: Solidscape use machining after deposition to keep the layers even as the wax can flow unevenly.

I suggest that it might be easier when using a single head to adjust the amount of wax depending on how high the point is where one wants to add more. Use a small laser or camera 3D scanner to check the point before ejecting the drop and correct for it at each spot. This might mean that layers have some small amount of unevenness but are kept level on each layer. If adjusting the drop size is difficult one could position the dot away from the outside edge and position the drop towards the hollow middle.

The Piezo jet you linked is just the thing for experiments but a single jet with tiny drops might be too slow which is why I was thinking of the page wide Phaser multi jet. I saw the DIY piezo jet on the Wiki but did not see a link to a heated wax one, must see if they can be liked together.

The Waxuum idea I have looked at before and it is neat and except for the difficulties of having the hot needle melt nearby sidewalls it should be pretty accurate. Using a very thin needle with a Teflon sleeve to insulate it a bit might allow for less unintended heating. It would certainly be less messy than a router and will work fine on the normal RepRap mechanics. Using the idea with a 5 axis robot arm would make it into a powerful tool as the standard milling XYZ setup possible with the RepRap does not allow for overhangs at all so limits the geometries a lot though parts should be built in parts with all voids exposed on the top surface and then just joined together with heat or solvent.

---

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

A few months ago, Mike Alden uploaded some pictures and video of his interesting heated piezo wax inkjet. The wiki links to the videos are functioning now, so if you're interested in printing with wax take a look at his results on the wiki page! Heated Piezo for Jetting Wax

MattMoses:

Thanks for updating the page. Those process pics and video clips are just amazing.

In the one showing the process from below through a sheet of plastic I notice something interesting. There is a second smaller spurious drop that ejects with each drop. I expect this is due to the burr inside the tiny orifice. I have seen similar splatter on ink-jet printer heads when they have been cleaned badly, printing a microscopic halo around solid lines from jets that have dirt.

It is amazing just how consistent in size the drops are. Even if this cannot work for building in 3D because the top surface level cannot be guaranteed for the next layer without scanning or milling it should be a very good option for printing of etch marks for PCB fabrication. It is a non contact head so should be able to be carried by most RepRap mechanisms without too much trouble.

-----
[www.reprap.org] Lava-Bot

I will use this name to describe any submerged nozzle zero buoyancy melted wax/metal/? systems I ponder on.
The warm wax in the supporting fluid idea might allow for 'infinite' bridging if the fluid is cold enough and the wax sets as it comes out of the nozzle. It will be working under zero gravity like in the ISS.

There are issues with the generic peristaltic pumps in that they have some pulsation on the output when the roller lifts from the tube. However at typical pumping rates this will be characterisable and with a index pulse on the pump shaft each time a roller passes one could compensate for the fluctuation in flow rate with modern software and micro-stepping motors. I believe it should be possible to get near perfect volume measurement from such a pump. Granted there may be patents on such flow compensation in peristaltic pumps in dosing applications but this would be a new and novel use.

---

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

Found a few links for light reading regarding ink and wax jet print heads.

[www.slideshare.net]
Some interesting formulas for drop formation in an continuous inkjet printer
Page 72 has a diagram showing the Phaser arrangement. Turns out there are over 300 jets per colour and not 48 as I estimated. This means that the head has to move less than 1mm from side to side so the nozzles cover the whole drum width. Lots of other really interesting pictures about ink jet technology.

[www.qucosa.de]
A involved document relating to direct offset printing with hot melt inks, kind of like a Phaser but high volume. Good for bed time reading if you want to fall asleep, I looked at the pictures and they were pretty (interesting).

[www.scribd.com]
For those that want to decide if they can use one for experiments there is a Xerox Phaser 8560 Service Manual on scribd (I think you have to be a member to download the PDF) that you can read. No details on the print head.

---

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

What is the benefit of poly vinyl alcohol ? What is it and what would you use it for?

I'm working on a single nozzle version of the machine you describe above. The advantage is that there is an Open Sources platform that I can use with it. How would you deal with a system that has more than one nozzle? Do you have someone that can create code for such a system?

Most piezo systems should be able to deal with almost any fluid providing it has the right viscosity. The wax is solid at room temperature and flows well above the melting point.

Thanks,

Malden

Malden Wrote:
-------------------------------------------------------
> What is the benefit of poly vinyl alcohol ? What
> is it and what would you use it for?

It is an extrudable water soluble support material. A bit expensive and hygroscopic from what I hear.

> I'm working on a single nozzle version of the
> machine you describe above. The advantage is that
> there is an Open Sources platform that I can use
> with it. How would you deal with a system that has
> more than one nozzle? Do you have someone that can
> create code for such a system?

Glad to hear there is activity in this area. Let everyone know how you get along.

I do not know of anyone working on a multi nozzle software but if the hardware was available then I expect a competent coder would be able to make it work. The powder bed printers use lots of nozzles and scan across the work surface.

> Most piezo systems should be able to deal with
> almost any fluid providing it has the right
> viscosity. The wax is solid at room temperature
> and flows well above the melting point.

The piezo head is doable, the uneven surface and uneven drop size from multiple nozzles may be the limiting factor.

---

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

What if we took a wax piezo heads and mixed it up with a powder printer. Use the hot wax piezo head to spray a wax binder on wax powder to make homogeneous wax objects. Instead of heating with a laser that causes some shrinkage or building with liquid wax alone that has an uncertain layer top profile, what if one mixed the two technologies and used the fresh wax powder to level the surface each layer and used the wax binder to add enough material to counter the consolidation shrinkage os a SLS wax printer.

Just a silly thought.

---

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

I think that adding technologies together will be the next big thing. The others who inspired me to continue this work were using sugar as a dry substance and actually got to see it work well enough to make raised letters ....... "Hi". So once this is refined the sky is the limit.

It only take a couple of people to get it started and keep it going. But I'm sticking to my original focus.

Cheers,

M

You want a new idea, here is another one. I call this one submerged hot point fusion SHPF :-)

Take a soldering gun type thing and modify it so all the heat occurs at the tip (solid copper conductors and a tiny tungsten coiled filament). Now submerge this in a tray of loose wax powder and move the hot tip around leaving a trail of melted wax behind. Held in place by the wax powder surrounding it and binding together al the wax that is close enough to the melt pool. Move it around at a steady speed to build and move it fast to jog to the next spot (add a small cold air jet to hold the wax away from filament when jogging if required) and then start melting again.

Have a hopper with a pipe or two around and above the hot point that keeps supplying loose build material at a level higher than the hot point and all around like a sand pyramid. So using the powder as a support and as raw material and adding it where it is needed while keeping the XYZ head mostly above the material as you start building in the bottom of a box.

Again this should be able to be done with standard hardware and firmware (almost) use the retract to turn on the air blower and use the extruder steps to adjust the hot point current.

The surface features will be coarse like sand but it could be as accurate as a sand casting if the powder and other parameters were carefully tuned. A quick blow with a hot air gun or a dip in lower melting temperature wax could smooth out the surface suitable for casting.

---

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

The Submerged Hot Point Extrusion (SHPF) can be married with hot piezo wax onto wax powder and with the Lava Bot principle.

One could try use the wax powder as the buoyancy support and extrude liquid wax submerged in powder.

We can go one better and extrude cyanoacrylate glue submerged a sodium bicarbonate powder (a common super glue fast activating filler powder use by the modelling fraternity) or acrylic dust (as used in the nail parlour industry, expensive in small quantities but cannot be too bad in bulk, available in a range of fashionable nail polish colours) from a PTFE tipped nozzle that should not clog.

Basically any liquid that will set when injected into a powder could be tried. Perhaps sawdust could be used with super (or other) glue to make wood objects.

This should work with the standard electronics and firmware. You need a peristaltic pump on the extruder axis (can be a remote unit as there is no heat involved) and a powder dispenser that will build a powder pyramid around the nozzle, could completely surround it or just have a few exit pipes a bit higher up.

The submerged hot point could be controlled at higher speed if it was heated with a fibre laser that shines onto the back of a small metal or ceramic shpere. It would be possible to control the temperature of the hot point very accurately if the sphere just happened to be the welded bead of a thermocouple.

The loose powder bed systems are very tempting to try out as there is no powder surface levelling required to keep the layers flat, the print point/nozzle does this automatically as in the regular FFF systems. They do need a XY head and a robust Z platform or better yet an XYZ (delta bot) that has a fixed bed. Keeping the bed and feed powder warm (on the wax/melt systems) might allow faster build speeds but perhaps not, a cold powder might freeze the wax better and allow more movement..

One could have some stackable rings the size of the build platform and as the powder starts to spill out the sides one adds another ring and the Delta Bot can keep building the part higher. When finished, lift the rings and all the powder falls on the floor.

---

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

Definitely food for thought. I'd kind of like to see some sort of sketch that shows these basic concepts.

If we have a way to apply a flat layer of dry material like wax, we could simply use white light and a magnifying glass focused to a super small spot. The spot of light would easily be hot enough to melt the wax in place. I think of the days when I took a hand help magnifying glass out in the sun on a warm day. Once I focused the wax I would inscribe with a burn mark in fresh wood. And this could have a method of controlling how hot the spot of light would get.

Keep the creative juices flowing.

Mike

I second that, I am building a dlp 3dprinter, so playing with projection lamps, a car headlamp bulb, 100watts focused down to a small point burns the fingers.

---

Random Precision

Super cool. Is the dlp printer with some sort of resin liquid? I know a friend who left my company and was well on the way of getting his system running before he left.

Do you have any links for how your system works? (I know it's off topic for the wax printing subject but I'm interested in following that technology too.)

Thanks,
~M~

Google dlp 3dprinter and you should find plenty of stuff, even a complete build on instuctables, creationworkshop, software, as well as others, I found a projector forum on how to defeat bulb detection, so you can use your own.

sorry but to not have the links to hand.

---

Random Precision

Quote
johnrpm
a car headlamp bulb, 100watts focused down to a small point burns the fingers

Could something like this be used instead of a laser in a SLS system? (We probably all remember the Solar Sinter. I wonder if something similar could work with a strong light source instead of the sun. Maybe use glass fibers to bring multiple light sources to one spot? Maybe use mirrors instead of lenses, so as to collect the infrared in addition to visible? Lots of options...)

... there are lamp-heads focussing on a spot of 5-3 mm diameter with powers up to 300 Watts used for soldering or heat-curing ... this heads can be used for SLS too ... but the prices are around or above 3000 Euros ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Does someone know the thicknes of plate in wax head?