DIY Project: SLS Sintering? (Guidance Much Appreciated)

[shintashi]

OK, so I got pretty heavy into electronics, 3D fabrication, and more recently, the little kits like Arduino/Genuino and Raspberry/Banana/Orange Pi.

I like DIY projects, and I especially like to "stick it to the man" whenever possible in the $$$$$ department. As most of you know, Laser Sintering machines are great for making metal stuff that otherwise looks like 3D printed plastic stuff, except unlike the pasta company's bronze and steel PLA spools, laser sintered metal powders, as opposed to polymer bonded metal powders, are really strong, and unlikely to warp in the sun, etc.

OK, so some of you guys have lab and bench laser experience. The highest powered 445 Laser I've seen was something like 12-20+ watts, and I can't even recall the company's name.

I was wondering if using multiple diodes, or one of those big gas lasers is better. I know nothing about YAG lasers (and forgot most of what I knew about 405 and 445 nm lasers). All I know for certain is I want something around 200 watts red, give or take. I seem to recall different frequencies cook faster than others.

This isn't an overnight project, more like a Great White Whale, but Sintering machines currently retail for the same as sportscars. I think we can get a bucket of metal powder on rods and worm gears with a laser for less than $100,000. Don't you?

 

[Isaac Clarke]

hmmm look for Styropyro....he have a laser that can ionize air...

 

[diachi]

OK, so I got pretty heavy into electronics, 3D fabrication, and more recently, the little kits like Arduino/Genuino and Raspberry/Banana/Orange Pi.

I like DIY projects, and I especially like to "stick it to the man" whenever possible in the $$$$$ department. As most of you know, Laser Sintering machines are great for making metal stuff that otherwise looks like 3D printed plastic stuff, except unlike the pasta company's bronze and steel PLA spools, laser sintered metal powders, as opposed to polymer bonded metal powders, are really strong, and unlikely to warp in the sun, etc.

OK, so some of you guys have lab and bench laser experience. The highest powered 445 Laser I've seen was something like 12-20+ watts, and I can't even recall the company's name.

I was wondering if using multiple diodes, or one of those big gas lasers is better. I know nothing about YAG lasers (and forgot most of what I knew about 405 and 445 nm lasers). All I know for certain is I want something around 200 watts red, give or take. I seem to recall different frequencies cook faster than others.

This isn't an overnight project, more like a Great White Whale, but Sintering machines currently retail for the same as sportscars. I think we can get a bucket of metal powder on rods and worm gears with a laser for less than $100,000. Don't you?

There's a couple of reasons SLS machines are so expensive - one is due to patents, then there's the complexity. The powder is usually held in what is essentially an oven to keep it near melting point, making the lasers job easier. Then there's the laser - which is typically a rather beefy Yb:Fiber laser (40-50W in experimental setups, I imagine it's more in commercial machines) - not cheap. Some systems appear to use CO2 as well. Then you need to take into account the control systems and a way to manipulate your laser beam.

I'm not aware of any 200W Red systems on the market, or if any red systems at those powers have ever even been constructed. You *might* get close with 445 if you have a lot of cash. You'd need ~40 diodes, combined with knife edging or some other means, all running at 6-7W (Need that extra power to account for losses). Not an easy build by any means. If anything you'd go for a BIG fiber coupled IR diode array. You're either going to want to go fiber, gas (CO2) or DPSS - beam quality out of a diode array that large isn't anywhere near good enough.

There are some open source projects working on doing SLS with plastics and waxes using off the shelf laser cutters and other parts that we have available to us (And for relatively cheap, thanks China!). Plastic and wax is much easier due to the lower temperatures required, and by extension lower laser powers. I don't think anyone has made any progress with metal SLS though. http://reprap.org/wiki/OpenSLS#Laser_Power. I'd say metal SLS is going to be a while for the DIY market.

Don't mean to rain on your parade.

hmmm look for Styropyro....he have a laser that can ionize air...

Not suitable for SLS if we're thinking of the same laser - repetition rate is too low, peak power is likely too high. It'll ablate the powder before it has a chance to fuse ...

 

[Hemlock_Mike]

If Styro has an SSY laser, it has high output but low
pulse rate due to heating of the flash tube cavity.

We need the new fiber lasers !!!
HMike

 

[diachi]

If Styro has an SSY laser, it has high output but low
pulse rate due to heating of the flash tube cavity.

We need the new fiber lasers !!!
HMike

It's not an SSY - It's one of those big Er:YAG lasers that come up on eBay all the time. Rep rate is a bit higher than the SSY I believe and I imagine the peak power is a good chunk higher.

 

[shintashi]

There's a couple of reasons SLS machines are so expensive - one is due to patents,

Expiry of Patents in 3D Printing Market to Decrease Product Costs and Increase Consumer Orientation - 3D Printing

This is one of the things I was looking forward to happening. Complexity obviously a problem, heating up the powder also a problem. Power consumption not so much.

You're either going to want to go fiber, gas (CO2) or DPSS - beam quality out of a diode array that large isn't anywhere near good enough.

Fiber lasers? like fiber optic bundles or something? Now that you mention it, the Blue Diode beams are pretty large diameters, which would yield poor resolution prints. Green lasers are pretty tight beams, but are red lasers or infra red lasers cheaper?

I think one of the important tricks is melting, rather than vaporizing the powder layer, then having a fine enough grain powder to lay down the next layer. There would have to be a delay for the melted layer to cool, but it seems to me the thinner the layer, and depending on the material, it might bleed the heat really fast. How might the lower sintered layers and surrounding powder act like a heat sink?

 

[diachi]

Expiry of Patents in 3D Printing Market to Decrease Product Costs and Increase Consumer Orientation - 3D Printing

This is one of the things I was looking forward to happening. Complexity obviously a problem, heating up the powder also a problem. Power consumption not so much.



Fiber lasers? like fiber optic bundles or something? Now that you mention it, the Blue Diode beams are pretty large diameters, which would yield poor resolution prints. Green lasers are pretty tight beams, but are red lasers or infra red lasers cheaper?

I think one of the important tricks is melting, rather than vaporizing the powder layer, then having a fine enough grain powder to lay down the next layer. There would have to be a delay for the melted layer to cool, but it seems to me the thinner the layer, and depending on the material, it might bleed the heat really fast. How might the lower sintered layers and surrounding powder act like a heat sink?

They don't use fiber lasers the way you are thinking, at least not in commercial systems. They use doped fiber (Usually Yb doped) so the fiber is actually the gain medium, not just a delivery method.

https://www.rp-photonics.com/fiber_lasers.html

Green diode lasers have similar beam characteristics to blue diode lasers - any high power diode is going to suffer the same issues - thus the need for fiber coupling. DPSS is going to be expensive in the powers required - at least if you want to go for metal printing. Plastic or wax wouldn't be nearly as bad. Fiber coupled IR diodes are available commercially in very high powers. The Coherent FAP 800 series goes up to 70W @ 810nm out of a 800um fiber bundle. Not sure how well that wavelength would work for metal.

Yep, need to melt the powder, vaporizing it is no good. Keeping the whole print bed at just below the melting point of the powder would be a big challenge but is required for effective printing. For example, you'd need to keep the bed at over 600*C for aluminum. Compare that to your oven at home - that likely doesn't go above 300*C.

 

[shintashi]

They don't use fiber lasers the way you are thinking, at least not in commercial systems. They use doped fiber (Usually Yb doped) so the fiber is actually the gain medium, not just a delivery method.

https://www.rp-photonics.com/fiber_lasers.html

Green diode lasers have similar beam characteristics to blue diode lasers - any high power diode is going to suffer the same issues - thus the need for fiber coupling. DPSS is going to be expensive in the powers required - at least if you want to go for metal printing. Plastic or wax wouldn't be nearly as bad. Fiber coupled IR diodes are available commercially in very high powers. The Coherent FAP 800 series goes up to 70W @ 810nm out of a 800um fiber bundle. Not sure how well that wavelength would work for metal.

Yep, need to melt the powder, vaporizing it is no good. Keeping the whole print bed at just below the melting point of the powder would be a big challenge but is required for effective printing. For example, you'd need to keep the bed at over 600*C for aluminum. Compare that to your oven at home - that likely doesn't go above 300*C.

for some reason that only seems reasonable up to certain thresholds. I get the feeling something like titanium and steel powder printers don't have beds heated to 2500F and 3125F respectively. I'm guessing they do something different.

Here's a 100 Watt in action:
https://www.youtube.com/watch?v=VImKhUD-8hk

I can see getting the table to like 800 degrees, but getting something into 1000F+ without some kind of serious changes in design, I dunno. They say that's where the real power suck goes. One of the things I thought up though, is the area of overlapping circles.

We initially think of lasers with a poor focus as a bad thing, but if you have 3 nearly focused lasers slightly overlapping, there will be a spot that gets more intense heat than the other regions. But perhaps the pass of the three lasers, across the surface of the powder, can heat them up along the periphery. As the laser moves, the areas where the beams don't overlap would pre-heat the powder around them, but it wouldn't be hot enough to melt them until the second or third beam converged? And if the beams were anything like bright flashlights, they would have a very powerful core (throw) and a diffuse flood corona type effect. That could, in our wishful thinking model... preheat much of the surrounding powder, which might (again, wishful thinking) save a lot of electricity requirements needed otherwise for heating the whole bed.

Again, I don't know if there's some laws of physics I'm missing here, I'm just thinking of it like how my lasers work - if i purposely put them out of focus, they don't burn things as quickly. I got a surface up to 700 degrees with about 1 watt. If we overlapped 3 x 40 watt lasers, or something, would that save anything?