Hello & another before I buy :-)

[printbot 1]

Hi Ultimakers,

I am just about decided that Ultimaker is the machine for me but before I buy one I wanted to ask if you think this model is printable - I have been informed that it should be OK but wondering what the community thought - I have no experience of 3D printing, as yet :-)

I have attached the model in STL in case you want a closer look but the model is a timber barn structure so lots of voids within....

Thanks in advance for any thoughts.

[Daid 306]

With a FDM machine like the Ultimaker, this is a very difficult print. Because of the thin sections in each layer, and the long thin pillars which break easy during printing.

[printbot 1]

Hi Daid,

Thanks for getting back to me - do you think that this rules it out as a possibility or that it is difficult but possible?

[Daid 306]

I think it would be possible, but very hard to print at once. But printing it as multiple parts and glueing it together would be a lot easier. As you could put the parts flat on the printer bed. Bit the same as a sprue building kit.

[printbot 1]

I would love to be able to print this & similar in 1 go - if there are any skilled and brave Ultimaker users out there willing to give it a shot I would love to know how it goes..... Daid? :-)

[snowygrouch 1]

That print looks like a nightmare.

I will try it tonight overnight, but I suspect that it will look very "salvador dali" in the morning....

But I dont have anything else to print right now so I will try it for fun.

It would help if your STL wasnt scaled to be the ACTUAL barn.....

EDIT:

You have alot of issues with this STL, and I am unable to work with it. Even though I have

Netfabb, CURA, and KISSLICER - so if you want anything to happen please resend with some proper

native CAD data that I can convert to STL.

[printbot 1]

Hi snowygrouch,

Thanks for looking at this, much appreciated.

The barn size STL was in case anyone wanted to print a full barn :-)

Sorry about the STL quality - I'm new to this - I have attached a 3DS version (still barn sized), is that any good - if not then I could do dxf?

Look forward to your thoughts & thanks again.

[snowygrouch 1]

Ok will try again tonight.

I dont mind leaving the printer running overnight, that doesnt cost me anything, but if I need to

spend 2hrs tinkering with the CAD its probably not going to happen.

Anyway lets see how the different CAD file works out.

C.

[printbot 1]

Thanks - completely understand and agree, messing with dirty CAD files is no fun.

I have attached another file which has been scaled to 200mm long in dxf format - should also be cleaner geometry but a slightly different design along a similar theme.

Cheers.

[snowygrouch 1]

Nope no good, I spent 15mins in Netfabb doing various mesh repairs but its still a mess in the slicer.

I cannot tell if the problem is due to overlapping geometry, or some mirroring or rotation operations you

may have performed before, but the slicer seems to think the model contains a very different boundary to the one you really have.

Sorry but I dont have time to explore any further. You will need to either make a much cleaner CAD model

or get someone prepared to spend an evening on repairing the mesh.

Better luck next time

[printbot 1]

OK, sorry about that and thanks for trying. As a last hope I have now modelled an example from scratch which is hopefully very clean, the right size and in several formats including STL which I hope is good right off the bat?... understand if you can't face looking at more broken models though :-) Cheers.

[snowygrouch 1]

ok your new model is basically fine.

I dont know what you did with the other one, but whatever it was...never do it again if you want to

have a chance of printing it !

Anyway, it says 3.5hours - which will work out about 5. So will post result tomorrow.

C.

[printbot 1]

Hey, your're a star - thank you - can't wait to see the results :-)

[joergen 2]

OK, sorry about that and thanks for trying. As a last hope I have now modelled an example from scratch which is hopefully very clean, the right size and in several formats including STL which I hope is good right off the bat?... understand if you can't face looking at more broken models though :-) Cheers.

At first impression, the SAT file looks clean, but I would still suggest to make a boolean out of all your parts, and then export it to STL.

in regards to printing, you could try laying one of the roof sides flat on the bed, that way everything is at a printable angle, and you could get away without support. it'll be a stringing, small island nightmare, which I would print in ABS, since it has less strings.

[printbot 1]

Thanks for the tips Joergen - the idea of printing on one of the roof planes makes good sense. Cheers.

[snowygrouch 1]

So the result is....nearly...but its not very pretty.

Actually the fact it didnt quite work isnt the printers fault, its mine for not quite tuning the support options

to get it right. I think the areas under the horizontal top beams were just too small to register with the

support setting I had and so it didnt print any support there. Hence it screwed up.

Also having very long thin vertical rods is very hard because the nozzle touches the tip with the blob

of plastic thats usually attached to the end and vibrates the rods. So as they get longer the surfaces line up less and less.

So overall I would say, that I could probably print it "OK" given another go with a different support structure config - but I would say that if this is the sort of thing you want to do all the time....probably its going to be hard going.

One other problem with tiny spindly stuff is getting it off the bed in one piece.

So I would say that if you split the building up into about 8 major sub-components and bonded it together after you could get quite a nice result. Although I do not have one, I would guess a hotbed would allow easier removal

of tiny parts without damage after they cool down.

Sorry but I dont have time to try again. But I think you can see enough to get the general idea.

Possible ? Yes

Advisabe...not really (at least in this method of printing all in one go)

(I added a picture of the sort of thing I usually print, and the sort of result I can get - just for reference)

C.

PS> This is photographed "as is" when the print stopped. So its realistic to get no strings (even for a shape like that)

[printbot 1]

Hi snowygrouch,

Thanks for exploring this and feeding back, very much appreciated. The results, whilst lacking the quality of the example you showed, are not a complete disaster and I can see the potential for getting a usable result :-)

Whilst this would not be the main purpose of the printer it sure would help me justify spending out and placing my order. I think I'm pretty close now to coughing up, just wondering if I should hold on until a heated bed gets added as standard.

Might I also ask what software you can recommend to take geometry from the STL stage to the printer? Not sure if I should be considering the NetFabb Basic offering or something else.

Thanks again

[snowygrouch 1]

Hmm well thats a whole different question and you can find plenty of discussion on this forum in the software section.

But if you are just about to buy a machine, I recommend CURA because:

1) Its easy to use

2) Its made by Daid at UMaker so its all set up for the machine

3) It has nice tools to check the printer is working after you build it

4) It is an easy way of uploading new firmware to the electronics

So just use that, then once its all working well you can start worrying about all the other choices !

C.

[printbot 1]

Sound advice - thank you.

Hopefully the fun can begin soon.

[msurunner 2]

I agree with Snowy on Cura. Until you get familiarized with what your printer is saying to you, don't bother with paying for/experimenting with the other slicers.