Sintering thermoplastic - what wavelength?

I've been thinking about DIY SLS machine and doing some research, but couldn't find an answer to a seemingly simple question. Are wavelengths of ~800 nm of an infrared diode laser enough to heat clear/white thermoplastic powder such as PLA or ABS and, hence, can be used for sintering, or it needs to be considerably higher, like 10um of a CO2 laser?

More generically, is there any easily accessible table of transparencies of common rapid prototyping materials for various wavelengths?

... you can look at the tables on side 61+62 in the RepRap-magazine #2 - [reprapmagazine.com]

Here I've posted some specs to common diode lasers.

With clear/transparent and bright plastics you'll need really high powers with 808nm to affect them.

For melting/evaporating plastics you have to add IR-absorbing pigments or use a CO2-laser, that will be absorbed with nearly 100 percents ...

---

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

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

Wow, great article, many thanks!

Adding some black coal powder looks more attractive then messing with water cooling and 20 kV electronics for a CO2 laser. At least to start with

... when mixing clear or bright plastic powder with coal dust you will get some funny results

I've tested some of the IR-diodes with icing shugar mixed with coal - depending on power and energy density in the spot the coal evaporated locally different and the molten track was between dark grey and totally clear again.

With black mineral microspheres I've received molten tracks from pure black over green to clear (again depending on energy density) ... with recondesing green or clear 'nano-spheres' all around the spot ...

---

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

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

This is all a bit confusing. So infrared lies between 700nm to 1000nm. Above 1000nm is microwave. Yet on the wikipedia page it says "people can see infrared up to at least 1050 nm in experiments".

So is a 975nm laser diode the ideal then for heating even white / clear plastic? Isn't that "deep infrared"?
EDIT: Oh wait lol I just realized that infrared goes up to 1mm, or 1000um, or 1000000nm. So there is a much larger range of "deep infrared" and a CO2 lasers uses 10um or 10.000 nm. Sorry! So 975nm isn't in the "ideal" range yet at all.

You can definitely find those on ebay between 1 and 10 watt ($50 to $200). Here the value of 140mW was thrown around as being "enough" for sintering plastic. It seems the laser is easy and rather cheap to get (although I'm not sure if the german customs office wouldn't confiscate them).

Edited 3 time(s). Last edit at 04/03/2015 09:16PM by Dejay.

... the common IR-diodes with several Watts of power are in the NIR (Near-Infra-Red) range, which is highly colour sensitive for energy absorption -- white or transparent materials will simply let the beam pass without melting.

You'll get better absorption rates for plastic materials with wavelengths below 500nm and above 1500nm ...

---

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

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

@VDX thank you! I've read a bit more and came to the same conclusion (mostly thanks to your posts).

I guess the best way for a DIY SLS printer project is to just go with something like a 2.3W laser 445nm and dark Nylon PA12 like Sintratec does. You could always upgrade later.

Honestly I see no real reason why for this limited use case SLS should be more expensive than a FDM printer. You need a laser, a laser controller, a galvo, a surface heater and a IR surface temp sensor. It's more complex than a hotend, an extruder and a heated build plate and a linear carriage but the parts shouldn't be that much difference. Of course I have no clue how to build a galvo / servo and a IR heater / sensor

... for galvos with big focus distances you'll need a pefect beam shape, what's not so easy with the 445nm-diodes.

When focussing a 445nm-diode you'll get a good round focus-spot when the focus is in max. 50mm distance ... with bigger distances you'll get the 'real' beam shape, what's more a square with maybe 4:1 ratio.

This will be different with fibercoupled 445nm-diodes, but this modules are much higher in price and not so common.

For this I've focussed on the fibercouples NIR-diodes - here you can find different types with wavelengths of 808 to 975nm and optical powers of 1 Watt to 10 Watts (available with up to 100Watts!) out of a fibre with 105micron core-diameter, what's pretty easy to refocus to a small spot again ... but again, better/smaller spots with lower distances, so not so good when using galvos with mostly >200mm focal lengths.

So my recomendation is - try with a diode (445nm or NIR) on a cartesian style frame with a moving head (carrying the diode with focus-head) and focal distances of max. 50mm ...

---

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

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

Thanks again for your valuable insight!

But what is the problem with focusing a diode laser to a spot? Is it just the changing distance of a galvo? I've seen those beam shapes (e.g. here) but can't you just focus the beam with a lens to a spot on e.g. 500mm? I'm just trying to understand where the problem lies.

Are those diodes with fiber optics just generally better and "prefocused" or does the fiber cable somehow help focus the beam?

I would think theoretically you could shape the beam to 0.1mm and "parallel light" but I guess this is expensive with lenses and might be hard on the galvo mirrors?

I guess a gantry is definitely easier to do. Sintratec specifies a laser speed of 70+ mm/s for heated nylon sintering so this shouldn't be a bottleneck for sintering.

... with focussing you simply 'transmit' the 'image' (or shape) of the emitter onto the target - the effective size of this 'image' in the focus distance is depending on the geometrical parameters ... e.g. the distance relations between emitter<>lens<>target and the max. beam diameter passing the lens.

With an emitter-shape of a typical 445nm diode (what is representing a square with 100microns width and 1 micron height) you will get this shape again as image in the 'focus point' - when the focal distance is short enough, it will burn a nealy round hole in the surface, as the high energy in the center will 'outshine' the image shape.

But with bigger distances the best possible 'focus diameter' will grow too, so you'll start to see the real shape of the emitter, what will result in a line instead of a round spot ...

---

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

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

Thanks for explaining, I think I get it now. The focused image has a certain size. Makes total sense I guess to shrink this focus image further down at a certain distance you'd need additional lenses. I've seen that you can reduce the "ellipse" in one direction with two prisms like this one.

I've looked around and read a bit from the laserpointerforums etc. Some crazy youtube video's out there!

Apparently the most powerful cheap blue laser diode is the Nichia NDB7875 445nm-450nm 1.6W-2W Blue Laser Diode/9MM for $54 with free shipping. It can be boosted up to 3.5W but probably you don't need that much and it has a rather large bar. The M462 goes up to 2.3W and has a better focus spot but it's also more expensive and more rare on ebay. Or maybe the OSRAM PLTB450B 450nm 1.6W ($39)?
An IR diode might be more powerful for a low price but I guess visible light is safer for experimentation (near IR can blind you just as well, right?). And fibre coupled lasers are more expensive. Can you fiber couple a laser yourself? It would be great to have only have a very light lens and a fiber cable mounted on the "print head".

What do you think of the Nichia NDB7875? From the videos it seems you should even be able to sinter / melt plastics without a surface heater. If you move slow enough that is.

Anyways, some laser diodes:

Type	       Focus	Wavelength	Diameter		Power		
        M140	+	445nm	5.6mm	$84	2.2W	$45	[sites.google.com]
        M462	++	462nm	5.6mm	$144	2.3W	$105	[sites.google.com]
Osram PLTB450B	++	450nm	?	$80	2.0W	$43	[sites.google.com]
Nichia NDB7875	-	445nm	9mm	$104	3.1W	$65	[sites.google.com]
Nichia NDB7A75	--	450nm	9mm	$225	5W	$185	[sites.google.com]

Now I'd need a laser mount with a glas lens, a heat sink ($6), and a ttl driver with max 3A ($12). Or this heat sink with a fan ($15). All free shipping thanks to china post! For $71 this seems incredibly cheap to me, and the shipments under 22€ to 27€ should be free of import taxes. I understand that these aren't quality components but it would be a worthwhile test to see if it works with "cheap".

But do these things fit together with the Nichia NDB7875 445nm-450nm 1.6W-2W Blue Laser Diode/9MM? I don't really have a clue about lasers.

For example do I need to adjust the max ampere the diode gets? Can I simply measure the current amp setting with a multimeter while turning the tiny poti? Does the TTL input also adjust the ampere up and down to the max?

I guess I should also buy a better lens like the 405-G-2 Glass Lens... or the cheaper G-2 lens from china? I guess china it is

Oh and 2x405nm-473nm Laser Protection Goggles ($14) of course. Or should I get better ones?

Any advice would be welcome Maybe my "shopping list" will help someone else too. I don't know if I'll ever actually build a SLS printer but in the very least I'd have a cool laser, and that is something I always wanted And even better I could use it as a laser engraver / balsa cutter on my delta.