RepRap Central Announces 2015 3D Printer Challenge

Imagine a 3D printer which features rapid build times, super smooth vertical operation, incredible accuracy, and super quiet operation. The team at RepRap Central, a UK merchant dedicated to the RepRap community, are asking you to take on the challenge of engineering and possibly building just such a machine. They call the idea an FLM printer, or Fused Layer Manufacturing, and the winning entrant will need to complete one full layer at a time using only the vertical Z Axis. You can read the whole story here: http://3dprint.com/39021/reprap-central-flm-challenge/

Let us know if you have entered this contest and feel free to post details of your entry below!


http://3dprint.com/wp-content/upload.../01/RepRap.jpg

Don't DLP based resin printers already do a complete layer at a time? Granted DLP may not be as fast but they do handle one layer at a time. Resin is also expensive so I guess it would depend on what the additional details were.

Very interesting indeed. If there are any interested parties (businesses or individuals) willing to start a conversation (could lead to a project?) then please get in touch with me at scott.knowles @ objectform . co . uk

Thanks

It seems pretty sketchy when their page doesn't explain what it really is or how it works. It reads like a common marketing ploy of an advance announcement to pump up some hype.

I know Boyer is in it, I would have hoped that he would be better than that.

I admit, i had to read the full post at couple of times to understand what Dr. Bowyer is looking for... and it would be a game changer. DLP resin based printers and FFF printers work in essentially the same fashion. In fact they all do. The material is hardened or extruded at a single point (x,y,z) at one moment in time. What the good doctor is hoping to find, is a method by which an entire layer (z) is laid down at a single point in time. As ObjectForm said, "very interesting indeed".

I'd be looking to simplify this by first handling two axes at a time, either X || Y && Z. That in itself would be an order of magnitude faster.

And that's all i'm going to say at this point :)

(Edit: My bad, DLP printers do work differently)

Quote:

---

Originally Posted by truly_bent

DLP resin based printers and FFF printers work in essentially the same fashion. In fact they all do. The material is hardened or extruded at a single point (x,y,z) at one moment in time. What the good doctor is hoping to find, is a method by which an entire layer (z) is laid down at a single point in time.

---

I think you're confusing DLP for SLA. DLP cures a whole layer at a time. Laser type SLA cures one point at a time.

I'm not going to get excited until they start giving specifics.

You're right JRDM. Thanks for straightening me out on that.

It does sound very much like a DLP printer, except if their plan is to extrude a layer at a time having millions of little nozzles pushing out hot plastic. Why would you go through that complexity to have so many more parts that can fail?

Yes a DLP projector itself does contain more parts than a complete FDM printer, but to a DLP printer a DLP projector is just one component that can be replaced easily. It is a component that is proven to be stable and usable for many hours at a time without needing replacement.

Resin isn't as expensive as it used to be. You can get it at $65 per litre. Your failure rate with DLP is much lower than with FDM so your mileage will be alot better with 1L of resin than it is with 1kg of filament. I have kilograms of failed ABS prints in bags sitting in the garage.

Quote:

---

Originally Posted by edu3d

Don't DLP based resin printers already do a complete layer at a time? Granted DLP may not be as fast but they do handle one layer at a time. Resin is also expensive so I guess it would depend on what the additional details were.

---

Layer print needs to be rapid, one of the criteria 50 to a 100 times faster

Quote:

---

Originally Posted by taskman

It does sound very much like a DLP printer, except if their plan is to extrude a layer at a time having millions of little nozzles pushing out hot plastic. Why would you go through that complexity to have so many more parts that can fail?

Yes a DLP projector itself does contain more parts than a complete FDM printer, but to a DLP printer a DLP projector is just one component that can be replaced easily. It is a component that is proven to be stable and usable for many hours at a time without needing replacement.

Resin isn't as expensive as it used to be. You can get it at $65 per litre. Your failure rate with DLP is much lower than with FDM so your mileage will be alot better with 1L of resin than it is with 1kg of filament. I have kilograms of failed ABS prints in bags sitting in the garage.

---

Complex nozzles may not be the way to go, DLP tech too slow

Quote:

---

Originally Posted by JRDM

It seems pretty sketchy when their page doesn't explain what it really is or how it works. It reads like a common marketing ploy of an advance announcement to pump up some hype.

I know Boyer is in it, I would have hoped that he would be better than that.

---

A surprisingly negative approach to something so positive

Well, this problem has been percolating in the back of me wee brain for a few days now. In my first posting above, i suggested a single sweep of either the X or Y axis (whichever is the narrower) to lay down a layer in the Z plane:

Quote:

---

I'd be looking to simplify this by first handling two axes at a time, either X || Y && Z. That in itself would be an order of magnitude faster.

---

Like taskman, i initially thought of a multitude of tiny nozzles pushing out hot plastic. That lead me to the old dot matrix technology and the line printer matrix:

http://3dprintboard.com/attachment.p...tid=4543&stc=1

I was thinking something like this could be adapted, with much hard work, to pushing out hot plastic along an entire row as the horizontal plane was swept. Sounds like you'd end up with a real mess, doesn't it? I thought so, too.

Okay, so what about something like inkjet printing, where tiny blobs of ink are shot onto a planar surface? Well, it turns out that Xerox is still developing something similar, called solid ink printing. I say 'still developing' because the technology has been around since 1987 or so. You'd have thought the whole world would have dropped it by now, but not Xerox.

It's about shooting tiny blobs of primary coloured (yellow, magenta, cyan, and black) wax onto a drum, which is transferred onto your paper. What a great idea. Take out the drum, substitute PLA for wax, and you have the makings of a 3D PLAJet printer, or maybe PlastiJet.

I'm not the first to have thought along those lines though. If you Google "xerox solid ink 3d printer", the first story to come up is: 3D Systems to Acquire a Portion of Xerox's Oregon Based Solid Ink Engineering and Development Teams. They paid 32.6 million dollars for access to Xerox's technology. Imagine that. A 3D PlastiJet printer, which prints a complete Z axis layer in a single pass. The resolution would be ridiculously high and it's in full colour no less!

Is this old news? I keep thinking i must have missed the announcement already. :)


Edit: About a week after posting this, i ran across something called Polyjet printing, which is similar to SLA. This Polyjet process lays down dots of liquid photopolymer which is cured by UV light, almost immediately after being laid down. It looks great, but running at 16 microns (or 0.0006") per layer i expect it takes forever to print a part. I can only assume that these printers are so expensive as to be beyond the means of most consumers. And so, the search goes on.

So how is FLM different than DLP printing? Wouldn't a high quality projector achieve the same result? A full layer in one image. Tune the power and the resin properties correctly and I bet it could be very fast. Or stitch multiple images together with multiple projectors?

http://www.prodways.com/en/technolog...ht-technology/

I have a feeling they're mainly interested in a "melted plastic" type printer as opposed to resin, correct?

I like that Prodways technology. Makes sense that they could get great resolution, but at what cost?

Perhaps they feel that there's some buyer resistance to resin machines. Although i've never worked with one, i tend to think of resin based machines as expensive, having a small work envelope, and messy. Although i could be wrong on all counts, as an average consumer, i would still tend toward the "melted plastics".

Just speculatin'.

Quote:

---

Originally Posted by truly_bent

I like that Prodways technology. Makes sense that they could get great resolution, but at what cost?

Perhaps they feel that there's some buyer resistance to resin machines. Although i've never worked with one, i tend to think of resin based machines as expensive, having a small work envelope, and messy. Although i could be wrong on all counts, as an average consumer, i would still tend toward the "melted plastics".

---

All that is still relatively true. I think you can build decent machine for $1000 USD. Some resins are $60/liter, which is less than a 1kg spool's worth of filament equivalent. The mess is still unaddressed in the inexpensive machines, though there's always going to be some clean-up.

I have no idea what Adrian, et. al. are really trying to promote. I get the feeling even they don't know, that this is some kind of X-Prize-like thing where they want to offer rewards for figuring out how to make it happen. I question the feasability to make a huge grid of nozzles though, going strictly by their description. Even a long linear row of nozzles for a single sweep per layer seems like asking a lot, and one stuck nozzle means a bad day.

There are other whole-layer technologies, such as those that layer sheets of cut paper. I don't know if that's desirable. Mcor has one that prints where the edges of the part goes, so you have a multi-color model when it's done. It's hard to say it's full-color because it has a faded look.

Quote:

---

Originally Posted by JRDM

All that is still relatively true. I think you can build decent machine for $1000 USD. Some resins are $60/liter, which is less than a 1kg spool's worth of filament equivalent. The mess is still unaddressed in the inexpensive machines, though there's always going to be some clean-up.

I have no idea what Adrian, et. al. are really trying to promote. I get the feeling even they don't know, that this is some kind of X-Prize-like thing where they want to offer rewards for figuring out how to make it happen. I question the feasability to make a huge grid of nozzles though, going strictly by their description. Even a long linear row of nozzles for a single sweep per layer seems like asking a lot, and one stuck nozzle means a bad day.

There are other whole-layer technologies, such as those that layer sheets of cut paper. I don't know if that's desirable. Mcor has one that prints where the edges of the part goes, so you have a multi-color model when it's done. It's hard to say it's full-color because it has a faded look.

---

Just to clarify, this Challenge originates from RepRap Central. Adrian Bowyer has not been involved but of course, any input from Adrian and the RepRap Community would be most welcome. We are still formulating the Challenge details, it's going to be open to anyone, Individuals, Teams, Companies around the World and we are looking to challenge the current status quo. If we can revisualise current 3D Printing methods and move to instantaneous layer based 3D Printing it really could result in a very significant step forward. It won't be easy but then real challenges are rarely easy.