Selective Thermal Sintering using DLP and instense, non-laser light source

A problem with inhibition sintering is that it often wastes most of the sintered material. The problem with mask based sintering is that although it is very fast, you can go through many, potentially expensive masks for a single part. The problem with laser spot focus sintering is that although it may be more precise that thermoplastic FDM, it is also very slow, doing one minute spot at a time. Heating the complex pattern of each layer all at once using an LCD instead of a large set of masks risks melting the LCD, or at the very least bleaching the liquid crystals of their polarization shifting properties when heated beyond their room temperature operating range.


A solution to the temperature sensitivity of an LCD panel is to switch to a different current display technology, DLP. The micro mirrors of a digital light pipe, or more accurately a multi-mirror device, can withstand much more intense light and heat, and still spread the heat over the surface in a controlled way. With the very short time constants for changing the mirrors from on to off or back again, they could even be sliding down the work material shining a shifting pattern constantly to only illuminate the material to be sintered. (Hey, is this a patentable idea? If the but companies make them, I want my share!) With the ability to illuminate most or all of the work surface at once, this would be a very fast buildup technique. Current DLP panels are less than an inch across, in the 1 to 1.5cm range, so optics would still be needed to spread the light (heat) around to make a larger work area. And as for whether or not they can stand the heat, just consider how hot the light must be for a 10,000 lumen projector when the light intended to fill a 12 square meter screen is all focused onto one (or three) 2 square cm panels.

This idea sounds so good I am going to start a separate thread, which we should have done long ago for the methods beyond thermoplastic FDM 3D printing

Mike

... you have to check what's cheaper, easier to handle and more accurate: - a DLP-projector with enough power for sintering the material, or a laser with a scanning head.

Actually 2D-galvo-scanners can be bought at ebay for some ten to some hundred Euros (depending of speed and accuracy) and lasers with powers above 30Watts can be obtained for some hundred Euros too.

To achieve the same accuracy with a DLP-projector you need a really good optical setup, so a DLP-projector suited for sintering with the same accuracy and speed as a scanning laser could be even expensiver than the laser-setup ...

---

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

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

True, which is why I was guess professional production first. But it would be MUCH faster. Whether with a stationary DLP head or one that slides back and forth between layers being added, you could sinter almost as fast as you can layout fresh, even layers. But yes, the optics would have to be custom to move the focal point from meters away to centimeters. You would probably also want a bouncing mirror to interpose between the work area and the optics to keep stray heat from fusing the whole surface.

Still, I would love to see this in action. And it might be possible to hack the optics and do a home brewed version. Wish I had the money to blow on a 10,000 lumen projector!

Mike

10,000 lumen projector? Seems to me it might be worth investigating how to get something like a solar death ray or Solar Light Sharpener to focus at the .1mm or lower .. Then again, maybe not... the tolerances may be beyond what could be done with something at this scale..

My Archimedes Solar Death Ray for Mythbsuters was bigger. But after the shipping company ruined it, the focus was no where near as good.

Mike

Perhaps a good start to the thread would be to provide a few links to some good sites so that we can all become familiar with the basics of the process.

If the principle can be proved I'm sure there would be many here that would then carry on the work.

Edited 1 time(s). Last edit at 04/30/2010 06:57PM by martinprice2004.

Quoting myself from the Generational Degradation thread:

"In addition to removing the horizontal movement axes by illuminating the entire layer at once, your idea also might remove the vertical movement axis. One could simply refocus the projector for each layer. The lens is moving up and down, of course, but the entire projector wouldn't need to be moved. Three caveats: refocusing instead of moving the projector would limit the build area to a sort of pyramid shape (with the top cut off at the minimum focal length), and, because the total projection surface gets progressively smaller with each layer, the layer images would need to get progressively larger to compensate. The light intensity would also increase with each layer."

I hope this helps.

Edited 4 time(s). Last edit at 05/01/2010 12:41AM by fgrams.

I think that the folded optics light paths needed to separate light/heat from dark with only a 10 degree deflection, the high intensity lamp and cooling system, and probably a second cooling system for the underside of the projector to prevent partial sintering just from radiant waste heat will make the projector system, even if well designed and purpose built, large enough to make it easier to move the surface (or actually drop the floor out from under the flat work surface and layer adding raking system) and keep the projector fixed at optimal focus.

Martin,
I am proposing whole new areas to explore here, so there is no prior work to show if it is feasible. However, I just found out is can cost less than $35,000 to buy the parts and get started testing.

$3.49 MEMS micro mirror array
$5 TV projection lens
$6 Plano Convex Lens
$6 Plano Convex Lens
$3 Condensor Lens

Not sure if these all add up to one layer-at-a-time sintering burner, but when I get time I will try. I have already ordered the parts (I wanted to get some before everyone cleaned them out!)

Mike

---

Team Open Air
Blog Team Open Air
rocket scientists think LIGHTYEARS outside the box!

here's an odd train of thought I had, don't whether it is feasable, but it seems lo-tech enough:

1) The simplest DIY multi-kilowatt lamp I know is an open-air arc lamp (typically built from two carbon rods + a simple welding transformer). The output of such a lamp is hard UV (cover your eyes and your skin! protect your neighbours too!).

2) you cannot use lenses to focus the light since glass absorbs UV ==> use mirrors (perhaps some contraption of bent metal sheets can do the job?)

3) Since glass absorbs UV you could use that as building material :-)

The glas used to provide UV transparancy is called Woods Glass.

So lenses and mirrors are not totaly out of the question.

Though finding some where to get Woods Glass mirrors and lenses might be more dificult. Posibly expensive to buy as they will be reserch lab quality and price.

UV cure dental fillings are becoming more common place so that might be another source to start looking at for getting components that handel UV in an industrial manner.

---

Bodge It [reprap.org]
=======================================

---

BIQ Sanguinololu SD LCD board	BIQ Stepcon	BIQ Opto Endstop
BIQ Heater Block PCB	BIQ Extruder Peek clamp replacement	BIQ Huxley Seedling
BIQ Sanguinololu mounting	BIQ standalone Sanguinololu or Ramps mounting	Print It Stick It Cut it

My rep strap: [repstrapbertha.blogspot.com]

Buy the bits from B&Q pipestrap [diyrepstrap.blogspot.com]
How to Build a Darwin without any Rep Rap Parts [repstrapdarwin.blogspot.com]
Web Site [www.takeaway3dtech.com]

I had this very same idea the other day.

The only way for the deposited the powder to remain flat and even is to have the standard base lowering in to a tube that fits it method. The advantage here is that it is east to heat the whole thing. If you preheat it to a little below the melting point of the plastic then it requires much less power to melt and cheaper projectors could be considered for hacking. I could be easy for it to melt it with a infrared bulb of moderate wattage. Would the optical systems of these projectors work the same projecting infrared light?

The bulb could be dimmed or switched off between each blast of heat so the projector components do not have to survive continuous running. This would allow for the projector to cope with more heat than it was designed for.

For someone with the inclination and a bit of money to burn (Hopefully not literally after modding.) it would not be so expensive to buy a cheep or second hand DLP projector and mod it to see how much heat and power was needed and if it could be made to survive.

I think it may be worth doing the opposite to inhibition sintering, that is to say print with black ink onto the powder
and fuse with an infrared lamp, the powder is held just below fuse temperature and the lamp takes it over the edge,
the black absorbs more heat than the white powder surrounding it.

I have not tried, but it may be possible to focus the infrared into a stripe so its a bit like a laser and make it sweep across????

About 3 years ago I worked on making a powder bed printer, mainly the software side, although I collected the bits for hardware also, recently someone asked me to help them build one, I dug out the stuff I did, it used the parallel port though,
so I have tinkerd with it to print with usb instead, the software slices to images, then sends each image to the printer after the recoat stuff, I have used the velleman usb board, " because I have a couple" which will then go to an arduino or straight
to stepper boards, its not finished and it could be done better not doubt, but if anyone is interested I am happy to share.

It is written in Euphoria, which may put people off, its open source but if I can learn it, anyone can.

---

Random Precision

I quite like the DLP idea! Even if you cannot fuse a whole layer at a time it might be viable to scan the DLP head as you might a laser - but much faster - and scroll the image across the DLP as it moves.

Even better might be a line of mirrors with one of the Infra Red heat lamp tubes which moves over the surface. On it's return trip, it could spread out the next layer of powder.

Why don't you have a go on a larger scale to test the idea as with better optics it would probably scale down. You might be able to use a big version to attract venture capital to make a small one!

Si

So why not use a bath of UV sensitive resin and shine the DLP projected layer onto the top of the bath with an intense UV-lamp source projecting through a normal DLP projector optics? then drop the resin bath down 0.2mm at a time and shine the next DLP (slice) image... repeat and you may end up with an object out of Resin similar to the way Laser-Resin 3D printers do it.

One problem is that the model is being built under the resin so it may not end up looking as good as raising it up out of the resin the way laser-resin sls works.

It would be interesting to try it as you are making a whole layer every exposure of the UV lamp through the DLP projector, you may even be able to use gray-scales on the DLP image to get more or less cooking of the UV cure resin.

It would even get around the problem of support material as you are making a whole layer at a time and it has some support in the resin bath.

I would love to try this.

Not sure how long this would all take to make an object? any ideas?

Rich.

---

[richrap.blogspot.com]

[blog.makerbot.com]

The UV resin curing has been done and it worked perfectly

Thanks, I should have googled it first... quite a few people are doing it already

You could use a bath of PCB solder resist?, this is UV curable and lower cost than the resin at $90/ lb but a little thick so may need to be thinned down.

---

[richrap.blogspot.com]

It would be a good idea but solder resist is transparent is it not? That resin used in 3d printing is opaque

No, solder resist is opaque and thick like honey

I have small pots of Green and White but I'm not sure what you would thin them down with

I used an old EPROM eraser to cure it on some home-made PCB's some years back

---

[richrap.blogspot.com]

Some are water based while some are solvent based. Why don't you try a thimble of solder resist and some paint thinner?

[www.coates.de]

Here's one that's UV curable: [www.toyobo.co.jp]

Maybe you can ask the company directly what they use to thin it down.

Just a footnote: Yag lasers can be fiber optically delivered. And crystal lasers are a bit easier to work/deal with then a gas laser (CO2)

... this is essentially the 'non laser'-thread

For lasers: - 8Watt@808nm-diodelasers with atached fiber of 0.1mm core-diameter will cost around 350 Euros (or 200 Euros when bought in qantities above 5 pcs)

---

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

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