Electrostatic Sintering?

Hi all. I'm just digging into rapid prototyping and I've caught the bug. I can't wait to build my first reprap.

Here's a thought for a (possibly?) novel method for 3d printing. The idea is to use the guts of a laser printer to deposit 2d layers of toner. Here are the details. I'd love to know if you think there's anything to the idea:

* Take apart two laser printers (or one colour laser printer)
* Mount two photosensitive drum assemblies and one fuser on an x/z stage
* Float the x/z stage over a build table that contains a strong static charger
* Fill one one drum assembly with low melting temperature "build" toner, and the other with high melting temperature "support" toner. Each type of toner should have the same charge and particle diameter.
* Take the first layer of your model's STL file and send it to the build drum. Take the exact inverse of the first layer and send it to the support drum
* Run the x stage over the build table. The two drums should transfer the build and support toner to the table in exactly the desired pattern.
* The fuser is kept just above the sintering temperature of the build toner. As it passes over the completed layer, it fuses the build toner, but keeps the support toner undisturbed.
* Repeat for each layer.

The advantage I can see for this method is that the layers can be controlled to a very consistent, small thickness. And the printing process should be blindingly fast since we're printing 2d layers per pass, instead of 1d strips. The major challenge I see is getting the drums to transfer their toner images to the build table as well as they do to paper, keeping the two drums aligned so the build and support toners don't end up with "registration errors", and sourcing the toner. How do you think we might surmount these?

OK, now that that's off my chest - I need to get some sleep!

... this could be an interesting approach, but is hard to realize and adjust.

With an automated powder-bed and SLS you can make nearly the same with a single powdered material and the layer thickness can be varied much more ...

---

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

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

NuclearNerd,
Welcome to the fascinating world of reprap, I think some university did some work on the sort of thing you mention, but have not seen any results, I did some experiments with electrostatics some time ago, using an ioniser, or so called air purifier, I drew lines on acrylic sheet and dusted it with powder, the powder, was attracted to the drawn lines.

---

Random Precision

Quote

With an automated powder-bed and SLS you can make nearly the same with a single powdered material and the layer thickness can be varied much more ...

True, but the surface quality of SLS printing suffers because of the porosity of the sintered part. Here I'm suggesting a method that allows us to densify each layer more compactly using heat and pressure from the fuser. When you combine that with very thin layers, you should theoretically end up with a very high quality part.

Plus we don't have to engineer powder handling mechanisms - that's all taken care of by the laser printer drum assemblies.

... true, but even easier is LOM-fabbing with thin sheets of material glued/fused on the surface and lasercut contour-traces and separating lines ... you have to search an optimum between accuracy, density and fabbing speed.

I was tinkering with some more fabbing methods over the past decades and found that every one has its own advantages and drawbacks ...

I think for DIY the easiest to realize methods are CNC-milling, FFF, paste-dispensing and lasercutting/-engraving/-etching.

Powderbed-printing, SLS, ILS, tonertransfer and UV-curing are much more sophisticated, so should be the next step after you have one or more of the previous methods running ...

---

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

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

Actually, I've built such a machine a few years ago. With one drum only, but it was capable of doing a print.

- Using the printer's electronics to image the layer works fine. All that laser and powder feed stuff can be taken as-is.

- Ordinary toner is fine for printing 3D, so you can stay away from material experiments on the first machine.

- For melting the most recent layer, we used a tubular infrared-lamp moving right behind the printer drum. They're a bit hard to find, but are reasonably priced.

- The toner drum was left inside the printer, so the whole upper half of the printer moved over the build bed. That worked fine.

- Layers are _very_ thin. Like 0.01 mm or so.

- Layers aren't even, laser printers tend to make larger areas thinner than narrow ones. You'd need a way to compensate for that.

- As the "controller" we used the lower half of the printer, sitting at the side of the apparatus for feeding it with dummy paper. That was subject to change, of course.

That experimental machine exposed a problem with this electrostatic approach: To get the particles off the drum onto the paper, laser printers have a high voltage wire opposite to the drum, on the back side of the paper. As Paper is always relative thin, this always works.

As you increase the distance between this wire and the drum, not only the required voltage on the wire quadruples with double distance, the image transfer process drum->part also becomes very unsharp. You'd have to focus that electrostatic field similar to a light beam, somehow. No idea how this could happen.

These last two points lead to the decision to do a major redesign of the apparatus (which has to happen, yet). The new idea was to print somehow onto an intermediate film, transport that film outside of the printer and to iron it onto the build bed there. That's actually closer to printing on paper than the first approach, so the printer has to be modified even less.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

... toner-transfer printing was one of the first methods which was tested for 3D-printing in the 1980-ies!

I had then an Okimate-20 colour printer with thermal transfer system - a plastic ribbon coated with thermoplastic paint was pressed on the paper and a head with separate heated pins was moved over the back side of the ribbon so the paint melted in the hot spot and fused to the paper.

This can be realized with actual laser printers printing on an endless plastic roll and 'ironing' the actual print on the fabbing area.

But this is nearly the same principle as LOM-fabbing - the only difference is that with transfer-pronting you have to print all points embedded in the solid, while with LOM you have only to remove the contours and add some cutting lines for separating the completed object out from the block.

So LOM will be faster (lesser points/lines to process) and is not limited to the toner materials, but can be made with any material you have as sheet and the laser can cut through ... even green ceramic sheets or metal foils!

---

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

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

The difference between LOM or SLS and toner transfer is, the selection of build material or not build material doesn't require to melt or even burn away anything with toner transfer. So you need only a very small amount of laser energy to do this selection. For comparison, a powder sintering machine like a SLS has a laser in the $20'000 range and consumes electricity in the kW range, while laser printers get along with something like $200 for the entire machine and consume about 400W, mostly for the heating drum. Still, laser printers tend to be faster.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

... i'm melting powders and cutting thin foils with diodelasers in powerranges from 1W to 9W and prices from 30€ to 300€

---

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

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

Actually, I've built such a machine a few years ago. With one drum only, but it was capable of doing a print.

Great post Traumflug. It's really cool to hear about your experience with this. I figured that getting the toner off the roll would be tough. We would need to keep the part at a static potential greater than the roll. Maybe we could connect the build table to a Van De Graaf generator. The fabber could look like a 50s sci-fi movie set piece I didn't want to go to a paper intermediate precisely because that's just LOM, but more difficult.

About layer thickness - I'm surprised you found it so inconsistent. I would have thought that laser printers would have controlled the amount of toner on the drum pretty tightly, otherwise they wouldn't be able to print a consistent grey level. What am I missing?

If you only used one roll, how do you get the toner to overhang?

Edited 1 time(s). Last edit at 02/22/2011 08:40AM by NuclearNerd.

Quote

I didn't want to go to a paper intermediate precisely because that's just LOM, but more difficult.

LOM uses the paper as the build material while with toner transfer, it's just a (reusable) transport medium. None of the paper is intended to end up in the final part. Additionally, a plastics film like kapton will probably work better. You don't want the toner to stick to the paper, after all.

Quote

About layer thickness - I'm surprised you found it so inconsistent. I would have thought that laser printers would have controlled the amount of toner on the drum pretty tightly, otherwise they wouldn't be able to print a consistent grey level. What am I missing?

Grey levels are the diffcult part of laser printing. You see it when printing with an old cartridge or old printer: the inner part of an area is light grey, while it's darker at the borders of that area. Maybe modern printers are better in this regard, but I still doubt it's sufficient to rely on it when 5000 layers (50 mm) stack up.

Quote

If you only used one roll, how do you get the toner to overhang?

Once you have a single roll printer working, you can think about making one allowing overhangs. One step at a time ;-)

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Traumflug Wrote:
-------------------------------------------------------
> LOM uses the paper as the build material while
> with toner transfer, it's just a (reusable)
> transport medium. None of the paper is intended to
> end up in the final part.

OK, I understand now. Kind of like those iron-on t-shirt transfer sheets you can buy for your printer. If we can't solve the charged-model transfer issue, that would work as a backup (but more complicated?) process.

> Grey levels are the diffcult part of laser
> printing. You see it when printing with an old
> cartridge or old printer: the inner part of an
> area is light grey, while it's darker at the
> borders of that area. Maybe modern printers are
> better in this regard, but I still doubt it's
> sufficient to rely on it when 5000 layers (50 mm)
> stack up.

That's tough. It sounds like laser printers suffer from the same issue as electroplating - edges tend to collect more charge. I would like to experiment with this though. It sounds like something that could be overcome in software (if the mathematics are understood).

> Once you have a single roll printer working, you
> can think about making one allowing overhangs. One
> step at a time ;-)

Gotcha.

Do you have any pictures of your previous experiments Markus?

Quote

It sounds like laser printers suffer from the same issue as electroplating - edges tend to collect more charge.

The good news is, pronter toner (or any other powder) isn't 100% dense, so you can "iron" it flat to some extents. Having a lower density in the middle of a part wouldn't hurt that much.

Quote

Do you have any pictures of your previous experiments Markus?

Sorry, no. This was at a time when digicams weren't in widespread useage, yet. Think of a carriage about 300 mm above the table with the upper half of a LaserJet 3 (the part you open to swap the cartridge) moving back and forth. All cables to the right and at the bottom, the lower half of the printer. Elongating the calbes between both halfs wasn't exactly difficult.

On the left, below that carriage, a liftable table with a precision screw. That table has to be made of some non-conductor, like epoxy, of course. Such a setup should get you started.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Thanks again Markus. I've almost got enough enthusiasm to start hacking something up.

I searched the archive and saw that this has been discussed a little before, although they envisioned using the toner transfer method you suggested upthread. I'll PM jdoll and see if he got anywhere with it.

So step 1 is to get toner transferring from a laser printer drum to a build space, either directly or though a transfer drum / film. Time to search kijiji for a laser printer....

(I hope Step 1 is easy, because step 2 - getting uniform layers, and step 3 - getting soluble / non-fusible "support" toner will be more challenging )

Here's another obstacle - toner isn't the only consumable in a laser printer. The photostatic drum wears out too. I did some research and found that typically the drum and the toner are both replaced after about 3000 pages at "5%" density - At about 0.01mm thick, that's only a few mm thick of build material per $50 cartridge. That's not exactly economical!