Hi, what do I need to know?

[wilsonj 0]

Hi all, I'm deciding on my first 3D printer, and I think I want to go the ultimaker route.

Probably a really silly question to ask on this forum, but is this a good value for money first choice ? I realize there are cheaper solutions, but speed and precision are important to me. For half the money I can buy a mendel unit. But I was under the impression they take much more time and effort to get working well, and possibly not capable of the print quality and definitely not the speed. Or have I got this wrong?

Secondly, how strong are the parts these machines make ? I'd really love to see a youtube video of someone breaking a piece, just to show its strength. Know of any? I did read somewhere that parts are about 30% of an injection molded part. Does that sound about right?

As for software. Is netfabb worth the extra over say replicatorG ? I have downloaded replicatorG and it seemed fine. At least I understood it.

Thanks for reading.

Cheers

Jamie

[ddurant 0]

Probably a really silly question to ask on this forum, but is this a good value for money first choice ?

For printers of this size, I think it's probably the best value for the money if you want to get up and running quickly. Some of the RepRap machines are quite nice but they're a bit nerdier than an Ultimaker. For the smaller machines, the MakerGear Mosaic (~5x5x5 build area) is probably the best printer out there.

Secondly, how strong are the parts these machines make ? I'd really love to see a youtube video of someone breaking a piece, just to show its strength. Know of any? I did read somewhere that parts are about 30% of an injection molded part. Does that sound about right?

Depends a lot on the settings and probably the shape but I'd be surprised if it was only 30%.. This number sounds familiar - I think I've heard it in reference to quality. If so, it's out of date and we can do a lot better now.

As for software. Is netfabb worth the extra over say replicatorG ? I have downloaded replicatorG and it seemed fine. At least I understood it.

There are 2 pieces of software: the slicer and the host. The slicer takes the 3D model and generates the tool path file (gcode). The host sends the gcode to the printer.

RepG is really just host software that also sorta wraps in skeinforge, a very powerful, free and open source slicer that's also pretty slow.

Netfabb is mostly a slicer but they're adding host support to it, too. The host part is a bit flakey right now but should be getting better soon.

I think skeinforge is probably the most powerful/capable slicer around today with netfabb a not-too-distant 2nd place. The feature set isn't one-to-one so it's a little hard to compare but if you're going to be doing lots of complicated shapes in high-resolution, netfabb is probably worth getting.

Thanks for reading.

You're welcome!

[wilsonj 0]

Thanks ddurant, that was very helpful.

One other thing I did forget to ask. What shapes can't you build with an ultimaker or kin ?

[ddurant 0]

One other thing I did forget to ask. What shapes can't you build with an ultimaker or kin ?

Of course, things bigger than the Ultimaker build platform can't be done in one piece.

X/Y resolution is limited to the nozzle size: 0.4mm diameter. Z resolution is much finer.

Parts of an object that are unsupported (like if you were to print the letter T standing up, the top left/right parts are unsupported) require you to either use support structures (which can be tricky and eat up plastic) or reorient the model.

Right now, we haven't quite sorted out 'reversal' which is used to prevent little strings between unconnected sections of a print. I'm hopeful that that'll work before much longer and not require hardware changes..

[wilsonj 0]

Thanks again ddurant! Makes sense.

Incidentally, now the UP! 3D printer is available for about $1250, would that outperform the mosaic ? or ultimaker for that matter. Of course with a much smaller print area.

Cheers

Jamie

[ddurant 0]

Dunno much about the UP! except that it's going to be a lot more proprietary than a Mosaic. UP! uses its own software and hardware - Mosaic uses similar (or even the same) software and electronics that we use and is all open source.. It should be a lot easier to find support for Mosaic. UP! support may also be good - I have no idea.

I just don't know how they compare performance-wise but suspect both are pretty good. You could always hit the #makergear IRC on freenode and ask Rick - he'd know (Rick owns MakerGear).

[Daid 306]

For half the money I can buy a mendel unit.

I heard in the #reprap channel a few days back "While you could build one for less, realistically expect to spend 700$ on a Prusa Mendel."

A RepRap is also harder to get going, as there is no "single kit" every machine is a bit different, so there is not a single build guide. There are tons of options. Which can be good or bad, depending on if you want to tinker with your machine or just print stuff with it.

As for the UP!, on Thingiverse you sometimes see things made with UP! printers, and I think quality wise they match good ultimaker results. (judging from the photos)

[wilsonj 0]

Thanks ddurant and Daid for your replies.

I'm assuming that even though the UM is incredibly quick, to get high quality parts the speed would need to be significantly reduced? It seems it is difficult to do a direct speed comparison between machines, when it comes to quality.

The UP for instance says it can print from 10-100cm^3/h. I have no idea how that converts to mm/s.

I guess what I am asking, is what sort of quality can one expect at high print rates?

[joergen 2]

Thanks ddurant and Daid for your replies.

I'm assuming that even though the UM is incredibly quick, to get high quality parts the speed would need to be significantly reduced? It seems it is difficult to do a direct speed comparison between machines, when it comes to quality.

The UP for instance says it can print from 10-100cm^3/h. I have no idea how that converts to mm/s.

I guess what I am asking, is what sort of quality can one expect at high print rates?

if you consider some simplifications (extrusions are rectangles), it's not that hard:

if you have a 0.5mm nozzle, and print 0.1mm layer height as a moderate pace of 100mm/s

extrusion: 0.5mm * 0.1mm * 100mm/s = 5mm^3/s or 0.3mL/min or 18mL/h (18cm^3/h)

this is a very moderate speed for a UM I think... the upper extreme of what the Joris UM is capable of (with a 0.8mm nozzle?) is about 1kg/h

1kg PLA is 0.8L (or 800cm^3 or 800000mm^3)

printing it in 1h translates to 222mm^3/sec, making it about 8x faster than the UP printer (in theory and special occasions and bigger nozzle).

in reality, I would either say that the UP printer is taking much thicker layers into consideration to print at 100cm^3/h (125gr PLA/h) than the UM usually does.

it mostly comes down to how much PLA/ABS can you melt and squeeze through the nozzle per sec/min/hour and since the UP and the UM use similar nozzles, they both run in the same range...

[wilsonj 0]

THanks Joergen, I did do some maths, but wasn't sure if it really translated to real world results.

Quality is very subjective also, which makes it almost impossible to compare performance/quality without having both printers side by side.

Or can the UM clearly print at the same quality as the UP at 3, 4 or 5+ times the speed ?