Reprap open source 3D printer - Page 7

SLATYE · 24th June 2012, 9:49 PM

Quote:

Originally Posted by Gazob

Do you have an expected delivery on this yet, looking at the makibot website it looks a little light in the forums on people actually having completed builds and probucing items. The price is impressive, the build area is workable, not so tiny as to be a real issue but always nicer if it was bigger.

No, not much info yet. Still not printing.

Quote:

Originally Posted by UncleRes

I'm interested to hear everyone's thoughts on the new wave of DLP UV Activated Resin 3D Printers (B9Creator, MiiCraft, etc).

Personally I see these machines as a revolution in resolution for the Open Source 3DP crowd, and I'm excited to see where it takes the industry as a whole.

I may be wrong, but isn't the resin for those (a) expensive, and (b) toxic?

I suspect that the competition between these resin printers and the cheap extrusion models will be similar to the competition between laser and inkjet printers. One option will be widely regarded as better, but also more expensive. Laser printers are (finally) getting into the price range where they can seriously compete with even the cheapest inkjets ($100 for a colour laser, $50 for a monochrome one).

With 3D printers, I think that it'll be a few more years before the extrusion ones are on the shelves at OfficeWorks and included in Dell's system builder (ie common, every-day hardware). A few years after that the resin-based ones will become practical for consumers, and a few years after that they'll be cheap enough to largely wipe out the extrusion printers.

paulvk · 24th June 2012, 11:06 PM

Quote:

Originally Posted by SLATYE

($100 for a colour laser, $50 for a monochrome one).

Its $37 for a monochrome !

mtma · 24th June 2012, 11:11 PM

Mind you much like the laser and inkjet printer analogy there's also pros and con's to both, for example whilst laser is widely regarded as better for black and white and where high quantity of prints are necessary over quality, inkjet is greatly superior when it comes to full colour printing.

btw, was the granule extruder always going to be a part of the the makibox? I don't recall seeing anything about that until recently. Personally I do like the idea though, as it would definitely open up a wealth of materials that you could potentially use in an extruding machine.

SLATYE · 24th June 2012, 11:23 PM

Quote:

Originally Posted by mtma

Mind you much like the laser and inkjet printer analogy there's also pros and con's to both, for example whilst laser is widely regarded as better for black and white and where high quantity of prints are necessary over quality, inkjet is greatly superior when it comes to full colour printing.

Yes. I suspect that the advantage of the extrusion printers will be their ability to handle several different types of plastics. We've already got ABS and PLA; I've seen several others being tested, and there's even some work being done on mixing plastics.

Colours are also going to be much, much easier with extrusion.

Quote:

Originally Posted by mtma

btw, was the granule extruder always going to be a part of the the makibox? I don't recall seeing anything about that until recently. Personally I do like the idea though, as it would definitely open up a wealth of materials that you could potentially use in an extruding machine.

The pellet extruder was Jon's weekend project for a long time (Jon is the guy designing the Makibox). When that got to a stage where it looked practical, he decided to make it an official part of the Makibox and drop the filament feed.

It's certainly a very nice addition, although it's probably pushed the project back by another month.

UncleRes · 25th June 2012, 9:19 PM

Quote:

Originally Posted by SLATYE

I may be wrong, but isn't the resin for those (a) expensive, and (b) toxic?

Most of these initial DIY attempts at DLP printers are selling their own propietary blends at the $70-100/kg range, but places like BuckTown Polymers are selling them at around $40/kg. Expensive compared to Chinese filament? Yes. But expense is relative given the jump in resolution.

Toxicity is certainly a concern - however they're far less dangerous in their cured form (ie, the final part). Infact some commercial UV Cureables in the SLA/Inkjet 3DP markets have USP Class VI certification (3DS' Accura ClearVue and Objet's MED510).

Quote:

Originally Posted by SLATYE

I suspect that the competition between these resin printers and the cheap extrusion models will be similar to the competition between laser and inkjet printers. One option will be widely regarded as better, but also more expensive. Laser printers are (finally) getting into the price range where they can seriously compete with even the cheapest inkjets ($100 for a colour laser, $50 for a monochrome one).

There's substantial room for the price of these machines to come down, especially in the DIY market compared to FDM methods. For one, there's only a single stepper and a single axis to move, the Z. There's no extrusion head and there's no heating elements. The biggest expense is in the procurement of a suitable DLP Projector.

Quote:

Originally Posted by SLATYE

With 3D printers, I think that it'll be a few more years before the extrusion ones are on the shelves at OfficeWorks and included in Dell's system builder (ie common, every-day hardware). A few years after that the resin-based ones will become practical for consumers, and a few years after that they'll be cheap enough to largely wipe out the extrusion printers.

The day is coming fairly rapidly - 3DS is now shipping the Cube. While most RepRappers scoff at the device, it does represent the first real consumer levelled FDM machine from an industry leader.

IMO the consumer market will probably bypass DLP altogether, as Inkjet based systems will probably mature to a suitable price point by that time. Given the Razor and Blades model the industry already uses, Inkjet will be pursued with vigour in the future. DLP is simply too "messy" (difficult to contain the resin etc), but will provide a good stopgap for the DIY crowd in the meantime.

Quote:

Originally Posted by SLATYE

Yes. I suspect that the advantage of the extrusion printers will be their ability to handle several different types of plastics. We've already got ABS and PLA; I've seen several others being tested, and there's even some work being done on mixing plastics.

Colours are also going to be much, much easier with extrusion.

Well that would depend on the technology, but certainly DLP/SLA will remain single colour (at a time) technologies, though they can be effectively any colour. That is unless someone cracks the PhotoChromic UV Cureable Photopolymer chemistry

Inkjet however has already proven its capabilities in the Full Colour 3D Printing space with the Z-Corp - not a technology that will catch on in the consumer space (again, too messy) - but there's certainly no reason why the tech couldn't be integrated into the Objet/Projet technologies and thus filter down into the consumer space when those technologies become affordable. Objet infact recently demonstrated 2 Colour mixing/printing through their Connex (multi-material) machines.

Sorry to ramble on. 3DP is my hobby (and it also pays the bills), I tend to get a bit carried away sometimes :P

SLATYE · 25th June 2012, 10:30 PM

Quote:

Originally Posted by UncleRes

Most of these initial DIY attempts at DLP printers are selling their own propietary blends at the $70-100/kg range, but places like BuckTown Polymers are selling them at around $40/kg. Expensive compared to Chinese filament? Yes. But expense is relative given the jump in resolution.

That makes sense. It'll be interesting to see how much prices can drop.

Quote:

Originally Posted by UncleRes

Toxicity is certainly a concern - however they're far less dangerous in their cured form (ie, the final part). Infact some commercial UV Cureables in the SLA/Inkjet 3DP markets have USP Class VI certification (3DS' Accura ClearVue and Objet's MED510).

Yes, much like laser printers - toner is awful stuff, but the final product is largely harmless. If the input material is contained (as it is for laser printers, in a toner cartridge) then it's not a big deal.

Quote:

Originally Posted by UncleRes

There's substantial room for the price of these machines to come down, especially in the DIY market compared to FDM methods. For one, there's only a single stepper and a single axis to move, the Z. There's no extrusion head and there's no heating elements. The biggest expense is in the procurement of a suitable DLP Projector.

Even so, the DLP projector is an extremely complicated (and therefore expensive) bit of hardware. I suspect it'll be a while before those become really cheap.

Quote:

Originally Posted by UncleRes

IMO the consumer market will probably bypass DLP altogether, as Inkjet based systems will probably mature to a suitable price point by that time. Given the Razor and Blades model the industry already uses, Inkjet will be pursued with vigour in the future. DLP is simply too "messy" (difficult to contain the resin etc), but will provide a good stopgap for the DIY crowd in the meantime.

Okay. It'll be good to see future developments here.

It's rather exciting to be keeping an eye on the birth something that is very likely to become a world-changing technology.

mtma · 26th June 2012, 12:26 AM

Don't UV systems require quartz glass optics? I can't really see them becoming much cheaper than current mid to high end projectors today (these of course being projectors that use full glass optics). Even those projectors show visible barrel or pincushion distortions which would make the end result not as enticing as the theoretical resolution suggests. But I guess the overall speed:quality benefit may be worthwhile in some instances.

UncleRes · 26th June 2012, 8:56 PM

Quote:

Originally Posted by SLATYE

Even so, the DLP projector is an extremely complicated (and therefore expensive) bit of hardware. I suspect it'll be a while before those become really cheap.

It certainly is - most systems are using $500 ebay projectors and getting great results, although because of the cost the resolutions being used are only around the 1024*768 range. Of course with these systems projector resolution is directly proportional to build size/part resolution - a higher resolution projector (such as those used in the commercial Envisiontec machines) can produce better results in terms of size and part resolution.

There has been some experiments done as far as using LCD panels and replacing the backlighting with xenon bulbs, offering both very good economy and high resolutions. IIRC the biggest challenge posed by this method so far is the tendency for the polarizing filters present in panel blocking and absorbing the overwhelming majority of the UV light. It's an interesting approach to the problem anyhow.

If anyone's interested in keeping up with the developments of DIY Photo Curables, the Yahoo group is where it's all happening. A popular topic at the moment seems to be trying to repurpose the cheap and readily available polygonal scanning mirrors inside laser printers for DIY SLA tech.

mtma · 5th July 2012, 11:17 PM

The Ultimaker is here! Well at least in kit form anyway. 6 to 20 hours of assembly time incoming according to the guide

Gazob · 9th July 2012, 8:57 AM

Quote:

Originally Posted by mtma

The Ultimaker is here! Well at least in kit form anyway. 6 to 20 hours of assembly time incoming according to the guide

Keep us updated, would love to see a buildlog. Ive gone for a self built, but small number of sources style kit, Ordbot Hadron build platform, qu-bd hotend extruder, and still trying to finalise electronics. not expecting anything to start arriving for a while yet.

GaZ

mtma · 9th July 2012, 8:08 PM

The problem is that I'm not really good at making logs, and there's not much to look at, though I'll give it a shot:

Day 1

Click to view full size!

I had about 1 and a half hours to take a crack at building the Ultimaker on Friday, I managed to assemble the frame and install the stepper motors and limit switches.

The lasered wood had a few burrs and splinters here and there but nothing major, it was a bit like a light fir and wasn't substantial enough to actually stab into your skin and give you grief. A minor scrape with a blunt metal edge (like the back of the box cutter) and it's all cleaned up. They do recommend you give the parts a light sand though.

The quality of the wood is very good, as is the overall laser cutting of the wood, they've obviously tuned the process fairly meticulously here.

In hindsight perhaps painting the wood might have been a good idea, because for this kit anyway ( I note that on their site they now have examples made of coloured wood ) the wood is basically as is. Although it is marine ply which is meant to be moisture resistant, but certainly any accidental spills on this machine are probably not going to go away if you don't paint it.

Day 2

Click to view full size!

Click to view full size!

The keen eyed of you will see the little brown bits stuck to the bearings. I at first thought that it might have been just some excess grease muck that was inside the bearing tubes (see right) - there's 10 of these 604's in the machine) but as I found out rather disturbingly, it's actually dirt that's stuck on the bearings. Bearings and sand mixed in grease is a most displeasing sight.

There was more than is shown in the photos, I removed some originally before discovering it was actually dirt. It appears that whoever handled the packaging of the little tube has been a bit careless in the handling of the bearings. I wiped and checked all the bearings for smoothness but luckily no damage has been done - the 2RS saves the day - might not be so lucky next time though... ... and for what it's used for in the ulimaker you could definitely use ZZ's - that's if you don't push dirt into them

Click to view full size!

Click to view full size!

2 hours later, the bearings axes, pulleys, belts and bearing blocks are now all fitted.

The linear bearings on the X and Y axes along the size are of a bronze type. I believe they're sintered and oil impregnated by there's a bit of talk about lubricating the shafts to get better results, which with correctly sintered bushings isn't something you need to do. But nevertheless they slide fine on the axes.

That night

Click to view full size!
I finished up about 11:30pm that night, though I didn't get around to taking this pic later at 1:30am the next day,

At this stage, the head is fitted to the printer and the extruder is mostly assembled.

I managed to assemble the extruder head wrong. And realised only after mounting it to the X-Y axes, when I looked at the fan and though 'ooh that's going to interfere with the motor'. After remounting it and checking it looked reasonably good.

Luckily the whole thing can be disassembled without removing the linear bearing blocks. Whilst it was a disaster, it wasn't a huge disaster. If I had gotten the X and Y orientation of the body wrong though, that would have been a bigger disaster. But luckily the front faced the front - I wasn't quite that daft.

The head, along with the Z table sits on real recirculating ball linear bearings. very smooth and neat stuff.

I hate spiral wrap.

At this point I'd like the point out how a good idea turns into a mediocre one: The cable guide on the right rear of the machine is getting full, and it's only going to get more full the next day when I finalise the electrics. It's still OK but it is getting to the point where guidance is a bit finicky for the last cable that I need to get through (the thermocouple and fan power cable, with three plugs) when I install the electrics next.

That aside, the axes too me much longer to align than I had expected. As some of you may know this sort of mechanism needs to be synchronised between each sliding block otherwise there would be a tendency to jam. That's well and good, but unfortunately Ultimaker don't provide you with a real alignment tool to actually achieve this. Their suggestion is to use a little part that has rounded corners, that has a tendency to fall as you adjust the alignment and in reality doesn't really align well to any datum point on the machine.

About here the online guide which has actually been quite good starts to fall to bits. There's about five independent things you have to do here to get the axes aligned and setup correctly, and it's not all that clear from initial reading which ones you're supposed to do, and what they really might mean by 'not too tight' and 'don't overdo it'.

The next thing after mounting the head was the Z table. This was a chore to be honest - an unfortunate side-effect of the laser cut construction. Regrettably the Z table contains about 5 layers of the plywood stacked together, which then grips the linear bearings in the back fairly tightly both vertically and horizontally. This means interference. Interference with the only method of guidance being screws means difficult to assemble - very difficult to assemble. 2 parts, yeah ok, but 5 parts :S. That was for each arm, then the arms themselves had to be joined to one another.

If the Ultimaker is to be offered direct from these guys as a pre-built kit I have a feeling that this part of the machine will be a moulded or machined part - this might be true for the bearing blocks as well as the motor mounts, but in those cases it wasn't so bad.

Overall at this point I would have preferred it greatly if they had made the guide using exploded drawings and whatnot. In the pictures at times it can be unclear the orientations and whatnot due to colour or camera angles, along with the ill mix of small and large details in the photos which means you have to click to flickr in some cases and expand something to figure it out.

One point to note here by the way is that characteristic of the laser cutting process the wood is all tapered from one side to the other. It's so slight you might not notice it but it's definitely there but you'll really feel it when you start installing the push fit things. The vertical axes rods are these things, and their design suggests to push them from the top down. Don't do it this way, it's bloody difficult. The machine has actually been lasered from the bottom side in both instances of the bottom panel and the top panel; uncapping the holes at the bottom and pushing the shafts up to the top (though the machine had to be put on it's side) was a substantially easier way of installing these shafts.

The same could have been said for the axis bearings, in some instances I installed them from the inside out because it was substantially easier doing it that way. But beware, you shouldn't put all that much pressure on the centre ring alone. I found the larger side of the lasered circles that fell out of the holes to be a fairly good implement to press the bearings a little deeper into the wood.

The extruder was mostly straightforward to assemble here.

Day 3 - Completion!

Click to view full size!

On the home straight!

Unfortunately I didn't get pictures of the electronics itself. The guide really falls apart here because they haven't updated this section of the guide for the ultimaker version that I have.

First thing they have you do here is install the electronic board. Now a comment on the electronics board, the assembly feels really amateur. I don't know if they really hand assembled it or not but the board edge connectors are all at funny angles, there's TO-220 packages standing up ass high and bent down in the air (OK as I understand it this is for cooling purposes) but still... Come on you have access to a laser cutter, make some assembly jigs at least. Small things can make all the difference.

The foldy hybrid fan and wood cooling system was fiddly. The mechanism is keyholed to slide out so you don't need to remove all the screws, but the resulting thing after you squish the folded air-guide under the wooden panel is that the whole thing bows a little bit - seizing the thing on the screws that are meant to slide on the keyholes. At this point I'm also missing a screw. I don't know if this was my error or something else happened but it happened. It wasn't a major drama though, it's not like all the screws were needed to secure the cooling panels to each other.

Guiding the cable with the three plugs (thermocouple and fan) was also a bit of an annoyance because as I mentioned the cable guide was quite full. I ended up putting blue tape around the plugs to give it some controllable shape and pushing it all up through the guides.

I hate spiral wrap.

Printing

I rather unfortunately didn't get around to printing much that day. But luckily that's not because my machine didn't work

The software I used was Cura, which is apparently RepG and Skein packaged into a neat thing with a UI. That installed in a breeze and I was up and running in a matter of minutes software wise.

One problem that arose was that the Marlin firmware that Cura wants you to download and upgrade to has inverted limit switches on some axes to the ones indicated on the laser etching on the Ultimaker panels and what the Ultimaker guide tells you to connect the switches as. I don't know why, but I found out during the machine function check in the wizard. Unfortunately it seemed to have some problem with function checking the Z axis switches (the test terminates before it does those checks oddly enough) so I ran the machine then found out that the Z axis was wrong AFTER it made a hideous noise from homing but finding itself locked out of home and forcing itself into the top Z switch ad infinitum in an attempt get there. No damage done though, turned it all off and inverted it and whatnot and everything was fine after that.

Perhaps it would be an idea if the firmware/software makers would code in sanity checks there. If you're moving machine in one direction and suddenly the opposing limit switch turns on for more than ~2mm of estimated travel, somethings gone wrong... duh!

As a test I downloaded a gear for a reprap something or other and scaled it down by half for my first print. At first it didn't work ( no adhesion ) but I adjusted the (very finicky) Z-home switch to make it close enough to stick.

Later I had something that looked like a gear. I have a video actually, but the following screenshot will have to do for the time being:

Click to view full size!

Shortly afterwards I stopped having something that looked like a gear. It appeared that insufficient material was coming out of the extruder, and there wasn't enough outer layer overlap at this scale so the resulting outer ring wasn't actually attached to the inner ring. The second problem was that it was much too well adhered to the blue tape. After mauling the outer ring of teeth (being not attached except for at one point) and a lot of tapping I actually tried to remove the centre two pieces of tape to take the gear off - this wasn't the solution, the tape broke at the edge of the thing I had printed. It was stuck tight and too low profile like a barnacle to a rock to take off.

In the end I heated it and pulled it off with a pair of pliers first print ruined unfortunately, but at least I knew the printer seemed to be working fine mechanically

Later I tried to print a 1:1 of the same gear, but with raft and stuff. I declared it would take too long and too much material after about 3 layers, and found out that raft sucks for print quality too.

This was it for the day (was only about 2-3pm) so all up it took me perhaps 12-14 build hours.

Read up a bit since then about print quality and tuning, so much to do in the coming weeks!

P.S: Having someone help is really good for some parts. Assembly isn't really an activity fit for a group (more then 2 then it would really be too many hands), but I suppose if everyone is competent you could have all the subassemblies (e.g extruder) prebuilt for a really fast assembly.

dirtyd · 9th July 2012, 11:20 PM

Thanks for the write-up mtma

Been putting off a Prusa build for 6 months but it's getting harder and harder to resist, looks like great fun.

Gazob · 10th July 2012, 9:13 AM

yeah nice work, thanks for taking the effort to write it up, my stuff is going to take forever to turn up, and reading bits like this even if not directly relateable to what ill be building may end up speeding my build.

Do you have an actual use for the printer?

GaZ

mtma · 11th July 2012, 12:27 AM

I have a use for it, but mainly it's just a hobby machine - it supports all the things I'd want to make for my other hobbies. I have some things in the pipeline I'd like to do for work as well, but that's a ways off yet - need to get familiarised with the process and the machine first.

At the moment I'm making a mount to replace another thing I made from laser cut processes. Laser cutting is for all intents and purposes restricted to 2D and you have to tile up 2D pieces to make 3D objects - and certainly what I did here to address the problem was towards the complicated side of it - that's probably why you find my comments about stacked laser cut pieces to be quite negative above - it's more of a been there done that perspective - there is a point where you should consider alternative manufacturing processes, though I pushed the limits more on the size capability of the equipment rather than the difficulty of assembly.

Not having support material is kind of restrictive, you certainly need to think about how you make things. For example the following could have been monolithic, but because of how it is I need to divide it into pieces. But luckily because of what it's doing that's also a double edged sword.

Click to view full size!

Click to view full size!

There's an error in the print compared to the model. The model is how it should be... do you see it?

Spoiler below, highlight to read:
I used the software Z-reverse button because it wasn't modeled the right way up - meaning that the printer tried to print it upside down and that just doesn't work. The trouble is that it mirrors - not flip - so the left and right are the wrong way around.

The print quality obviously isn't crash hot at the moment I was playing around with the speeds and had some problems with the filament seizing. The dreaded bowden cable slip problem had arose, however that's in part my error and not the root cause of the seizure at the time. I cleaned out the hot end easy enough and put that all back together.

As it turns out you shouldn't put all that much pressure on the knurled bolt because it only takes a minor change in size of the filament (in this case I tightened it thinking it was too loose because I had only feather engagement before) to result in the filament not going through the bowden. In this case I measured it to have been crushed so it was 3.2mm on one dimension - and that seems to be quite enough variance to result in problems.

Additionally it appears that whilst the temperature of the thermocouple is well regulated, the actual extruder nozzle temperatures are varying quite substantially from a combination of cooling fan flow and operating speed. Unfortunately in PrintRun the fan speed is not something you can dynamically control (I'm guessing I could give it an M command or something though) so later I printed something at 50% speed which turned out to be much less than 50% actual cooling.

Still quite some parameter tuning to do yet.

SCwerTA · 11th July 2012, 1:33 AM

Good to see you up and running so quickly!

Not having support material isn't so much of a problem. You can incorporate support structures into your design that break away or let the slicing program create the structures for you.

Slic3r does this ok, but I like to design them myself so that they break away easily and cleanly. When you have a 3D printer, traditional design thinking goes out the door. I attended a Additive Manufacturing seminar recently where they had a simulation of a model in all possible rotations to optimize the orientation for minimal or even no use of support structures. I thought it was quite interesting...

Looking at your print, you need to up your flow rate a small amount as I can see gaps in between the shells and the infills. I'm not sure what program you're using to slice with but, generally you can either;

increase filament dimension size
calibrate your extruder stepper steps/unit in firmware over a set measure distance (usually 100mm is enough)

If you find after increasing the flow rate that your layers are getting too much plastic but the gaps are only just ok, then it'd be a clip issue. "Clip" for those of you who don't know is the distance between the tool paths, particularly important between shells and infill.

Also slowing down the printer in small layers does wonders, especially in the region where you have your holes. You can sort of see the print is not as clean there. In slic3r and skienforge that setting is cool. You can adjust fan speeds in that setting in slic3r as well.

Just out of interest how is the bed leveled on the Ultimaker? Any noticeable deflection on the far end? Its a cantilevered plate if i recall correctly.

Edit: for those interested cheap chinese filaments here: http://www.repraper.com/

9th July 2012, 8:08 PM	#101
mtma Member Join Date: Aug 2009 Posts: 2,593	The problem is that I'm not really good at making logs, and there's not much to look at, though I'll give it a shot: Day 1 Click to view full size! I had about 1 and a half hours to take a crack at building the Ultimaker on Friday, I managed to assemble the frame and install the stepper motors and limit switches. The lasered wood had a few burrs and splinters here and there but nothing major, it was a bit like a light fir and wasn't substantial enough to actually stab into your skin and give you grief. A minor scrape with a blunt metal edge (like the back of the box cutter) and it's all cleaned up. They do recommend you give the parts a light sand though. The quality of the wood is very good, as is the overall laser cutting of the wood, they've obviously tuned the process fairly meticulously here. In hindsight perhaps painting the wood might have been a good idea, because for this kit anyway ( I note that on their site they now have examples made of coloured wood ) the wood is basically as is. Although it is marine ply which is meant to be moisture resistant, but certainly any accidental spills on this machine are probably not going to go away if you don't paint it. Day 2 Click to view full size! Click to view full size! The keen eyed of you will see the little brown bits stuck to the bearings. I at first thought that it might have been just some excess grease muck that was inside the bearing tubes (see right) - there's 10 of these 604's in the machine) but as I found out rather disturbingly, it's actually dirt that's stuck on the bearings. Bearings and sand mixed in grease is a most displeasing sight. There was more than is shown in the photos, I removed some originally before discovering it was actually dirt. It appears that whoever handled the packaging of the little tube has been a bit careless in the handling of the bearings. I wiped and checked all the bearings for smoothness but luckily no damage has been done - the 2RS saves the day - might not be so lucky next time though... ... and for what it's used for in the ulimaker you could definitely use ZZ's - that's if you don't push dirt into them Click to view full size! Click to view full size! 2 hours later, the bearings axes, pulleys, belts and bearing blocks are now all fitted. The linear bearings on the X and Y axes along the size are of a bronze type. I believe they're sintered and oil impregnated by there's a bit of talk about lubricating the shafts to get better results, which with correctly sintered bushings isn't something you need to do. But nevertheless they slide fine on the axes. That night Click to view full size! I finished up about 11:30pm that night, though I didn't get around to taking this pic later at 1:30am the next day, At this stage, the head is fitted to the printer and the extruder is mostly assembled. I managed to assemble the extruder head wrong. And realised only after mounting it to the X-Y axes, when I looked at the fan and though 'ooh that's going to interfere with the motor'. After remounting it and checking it looked reasonably good. Luckily the whole thing can be disassembled without removing the linear bearing blocks. Whilst it was a disaster, it wasn't a huge disaster. If I had gotten the X and Y orientation of the body wrong though, that would have been a bigger disaster. But luckily the front faced the front - I wasn't quite that daft. The head, along with the Z table sits on real recirculating ball linear bearings. very smooth and neat stuff. I hate spiral wrap. At this point I'd like the point out how a good idea turns into a mediocre one: The cable guide on the right rear of the machine is getting full, and it's only going to get more full the next day when I finalise the electrics. It's still OK but it is getting to the point where guidance is a bit finicky for the last cable that I need to get through (the thermocouple and fan power cable, with three plugs) when I install the electrics next. That aside, the axes too me much longer to align than I had expected. As some of you may know this sort of mechanism needs to be synchronised between each sliding block otherwise there would be a tendency to jam. That's well and good, but unfortunately Ultimaker don't provide you with a real alignment tool to actually achieve this. Their suggestion is to use a little part that has rounded corners, that has a tendency to fall as you adjust the alignment and in reality doesn't really align well to any datum point on the machine. About here the online guide which has actually been quite good starts to fall to bits. There's about five independent things you have to do here to get the axes aligned and setup correctly, and it's not all that clear from initial reading which ones you're supposed to do, and what they really might mean by 'not too tight' and 'don't overdo it'. The next thing after mounting the head was the Z table. This was a chore to be honest - an unfortunate side-effect of the laser cut construction. Regrettably the Z table contains about 5 layers of the plywood stacked together, which then grips the linear bearings in the back fairly tightly both vertically and horizontally. This means interference. Interference with the only method of guidance being screws means difficult to assemble - very difficult to assemble. 2 parts, yeah ok, but 5 parts :S. That was for each arm, then the arms themselves had to be joined to one another. If the Ultimaker is to be offered direct from these guys as a pre-built kit I have a feeling that this part of the machine will be a moulded or machined part - this might be true for the bearing blocks as well as the motor mounts, but in those cases it wasn't so bad. Overall at this point I would have preferred it greatly if they had made the guide using exploded drawings and whatnot. In the pictures at times it can be unclear the orientations and whatnot due to colour or camera angles, along with the ill mix of small and large details in the photos which means you have to click to flickr in some cases and expand something to figure it out. One point to note here by the way is that characteristic of the laser cutting process the wood is all tapered from one side to the other. It's so slight you might not notice it but it's definitely there but you'll really feel it when you start installing the push fit things. The vertical axes rods are these things, and their design suggests to push them from the top down. Don't do it this way, it's bloody difficult. The machine has actually been lasered from the bottom side in both instances of the bottom panel and the top panel; uncapping the holes at the bottom and pushing the shafts up to the top (though the machine had to be put on it's side) was a substantially easier way of installing these shafts. The same could have been said for the axis bearings, in some instances I installed them from the inside out because it was substantially easier doing it that way. But beware, you shouldn't put all that much pressure on the centre ring alone. I found the larger side of the lasered circles that fell out of the holes to be a fairly good implement to press the bearings a little deeper into the wood. The extruder was mostly straightforward to assemble here. Day 3 - Completion! Click to view full size! On the home straight! Unfortunately I didn't get pictures of the electronics itself. The guide really falls apart here because they haven't updated this section of the guide for the ultimaker version that I have. First thing they have you do here is install the electronic board. Now a comment on the electronics board, the assembly feels really amateur. I don't know if they really hand assembled it or not but the board edge connectors are all at funny angles, there's TO-220 packages standing up ass high and bent down in the air (OK as I understand it this is for cooling purposes) but still... Come on you have access to a laser cutter, make some assembly jigs at least. Small things can make all the difference. The foldy hybrid fan and wood cooling system was fiddly. The mechanism is keyholed to slide out so you don't need to remove all the screws, but the resulting thing after you squish the folded air-guide under the wooden panel is that the whole thing bows a little bit - seizing the thing on the screws that are meant to slide on the keyholes. At this point I'm also missing a screw. I don't know if this was my error or something else happened but it happened. It wasn't a major drama though, it's not like all the screws were needed to secure the cooling panels to each other. Guiding the cable with the three plugs (thermocouple and fan) was also a bit of an annoyance because as I mentioned the cable guide was quite full. I ended up putting blue tape around the plugs to give it some controllable shape and pushing it all up through the guides. I hate spiral wrap. Printing I rather unfortunately didn't get around to printing much that day. But luckily that's not because my machine didn't work The software I used was Cura, which is apparently RepG and Skein packaged into a neat thing with a UI. That installed in a breeze and I was up and running in a matter of minutes software wise. One problem that arose was that the Marlin firmware that Cura wants you to download and upgrade to has inverted limit switches on some axes to the ones indicated on the laser etching on the Ultimaker panels and what the Ultimaker guide tells you to connect the switches as. I don't know why, but I found out during the machine function check in the wizard. Unfortunately it seemed to have some problem with function checking the Z axis switches (the test terminates before it does those checks oddly enough) so I ran the machine then found out that the Z axis was wrong AFTER it made a hideous noise from homing but finding itself locked out of home and forcing itself into the top Z switch ad infinitum in an attempt get there. No damage done though, turned it all off and inverted it and whatnot and everything was fine after that. Perhaps it would be an idea if the firmware/software makers would code in sanity checks there. If you're moving machine in one direction and suddenly the opposing limit switch turns on for more than ~2mm of estimated travel, somethings gone wrong... duh! As a test I downloaded a gear for a reprap something or other and scaled it down by half for my first print. At first it didn't work ( no adhesion ) but I adjusted the (very finicky) Z-home switch to make it close enough to stick. Later I had something that looked like a gear. I have a video actually, but the following screenshot will have to do for the time being: Click to view full size! Shortly afterwards I stopped having something that looked like a gear. It appeared that insufficient material was coming out of the extruder, and there wasn't enough outer layer overlap at this scale so the resulting outer ring wasn't actually attached to the inner ring. The second problem was that it was much too well adhered to the blue tape. After mauling the outer ring of teeth (being not attached except for at one point) and a lot of tapping I actually tried to remove the centre two pieces of tape to take the gear off - this wasn't the solution, the tape broke at the edge of the thing I had printed. It was stuck tight and too low profile like a barnacle to a rock to take off. In the end I heated it and pulled it off with a pair of pliers first print ruined unfortunately, but at least I knew the printer seemed to be working fine mechanically Later I tried to print a 1:1 of the same gear, but with raft and stuff. I declared it would take too long and too much material after about 3 layers, and found out that raft sucks for print quality too. This was it for the day (was only about 2-3pm) so all up it took me perhaps 12-14 build hours. Read up a bit since then about print quality and tuning, so much to do in the coming weeks! P.S: Having someone help is really good for some parts. Assembly isn't really an activity fit for a group (more then 2 then it would really be too many hands), but I suppose if everyone is competent you could have all the subassemblies (e.g extruder) prebuilt for a really fast assembly. Last edited by mtma; 9th July 2012 at 8:15 PM.

9th July 2012, 11:20 PM	#102
dirtyd Member Join Date: Jan 2006 Location: 127.0.0.1 Posts: 3,680	Thanks for the write-up mtma Been putting off a Prusa build for 6 months but it's getting harder and harder to resist, looks like great fun. __________________ "It is worthy of remark that a belief constantly inculcated during the early years of life, whilst the brain is impressible, appears to acquire almost the nature of an instinct; and the very essence of an instinct is that it is followed independently of reason." - Charles Darwin \| trustworthy traders

10th July 2012, 9:13 AM	#103
Gazob Member Join Date: Apr 2002 Location: Sunbury Posts: 1,164	yeah nice work, thanks for taking the effort to write it up, my stuff is going to take forever to turn up, and reading bits like this even if not directly relateable to what ill be building may end up speeding my build. Do you have an actual use for the printer? GaZ __________________ Currently an Active Virtual Sailor with http://www.vsk-aus.com/

11th July 2012, 12:27 AM	#104
mtma Member Join Date: Aug 2009 Posts: 2,593	I have a use for it, but mainly it's just a hobby machine - it supports all the things I'd want to make for my other hobbies. I have some things in the pipeline I'd like to do for work as well, but that's a ways off yet - need to get familiarised with the process and the machine first. At the moment I'm making a mount to replace another thing I made from laser cut processes. Laser cutting is for all intents and purposes restricted to 2D and you have to tile up 2D pieces to make 3D objects - and certainly what I did here to address the problem was towards the complicated side of it - that's probably why you find my comments about stacked laser cut pieces to be quite negative above - it's more of a been there done that perspective - there is a point where you should consider alternative manufacturing processes, though I pushed the limits more on the size capability of the equipment rather than the difficulty of assembly. Not having support material is kind of restrictive, you certainly need to think about how you make things. For example the following could have been monolithic, but because of how it is I need to divide it into pieces. But luckily because of what it's doing that's also a double edged sword. Click to view full size! Click to view full size! There's an error in the print compared to the model. The model is how it should be... do you see it? Spoiler below, highlight to read: I used the software Z-reverse button because it wasn't modeled the right way up - meaning that the printer tried to print it upside down and that just doesn't work. The trouble is that it mirrors - not flip - so the left and right are the wrong way around. The print quality obviously isn't crash hot at the moment I was playing around with the speeds and had some problems with the filament seizing. The dreaded bowden cable slip problem had arose, however that's in part my error and not the root cause of the seizure at the time. I cleaned out the hot end easy enough and put that all back together. As it turns out you shouldn't put all that much pressure on the knurled bolt because it only takes a minor change in size of the filament (in this case I tightened it thinking it was too loose because I had only feather engagement before) to result in the filament not going through the bowden. In this case I measured it to have been crushed so it was 3.2mm on one dimension - and that seems to be quite enough variance to result in problems. Additionally it appears that whilst the temperature of the thermocouple is well regulated, the actual extruder nozzle temperatures are varying quite substantially from a combination of cooling fan flow and operating speed. Unfortunately in PrintRun the fan speed is not something you can dynamically control (I'm guessing I could give it an M command or something though) so later I printed something at 50% speed which turned out to be much less than 50% actual cooling. Still quite some parameter tuning to do yet. Last edited by mtma; 11th July 2012 at 12:31 AM.

11th July 2012, 1:33 AM	#105
SCwerTA Member Join Date: Jan 2005 Location: SE Melbourne Posts: 227	Good to see you up and running so quickly! Not having support material isn't so much of a problem. You can incorporate support structures into your design that break away or let the slicing program create the structures for you. Slic3r does this ok, but I like to design them myself so that they break away easily and cleanly. When you have a 3D printer, traditional design thinking goes out the door. I attended a Additive Manufacturing seminar recently where they had a simulation of a model in all possible rotations to optimize the orientation for minimal or even no use of support structures. I thought it was quite interesting... Looking at your print, you need to up your flow rate a small amount as I can see gaps in between the shells and the infills. I'm not sure what program you're using to slice with but, generally you can either; increase filament dimension size calibrate your extruder stepper steps/unit in firmware over a set measure distance (usually 100mm is enough) If you find after increasing the flow rate that your layers are getting too much plastic but the gaps are only just ok, then it'd be a clip issue. "Clip" for those of you who don't know is the distance between the tool paths, particularly important between shells and infill. Also slowing down the printer in small layers does wonders, especially in the region where you have your holes. You can sort of see the print is not as clean there. In slic3r and skienforge that setting is cool. You can adjust fan speeds in that setting in slic3r as well. Just out of interest how is the bed leveled on the Ultimaker? Any noticeable deflection on the far end? Its a cantilevered plate if i recall correctly. Edit: for those interested cheap chinese filaments here: http://www.repraper.com/

24th June 2012, 11:11 PM	#93
mtma Member Join Date: Aug 2009 Posts: 2,593	Mind you much like the laser and inkjet printer analogy there's also pros and con's to both, for example whilst laser is widely regarded as better for black and white and where high quantity of prints are necessary over quality, inkjet is greatly superior when it comes to full colour printing. btw, was the granule extruder always going to be a part of the the makibox? I don't recall seeing anything about that until recently. Personally I do like the idea though, as it would definitely open up a wealth of materials that you could potentially use in an extruding machine.

26th June 2012, 12:26 AM	#97
mtma Member Join Date: Aug 2009 Posts: 2,593	Don't UV systems require quartz glass optics? I can't really see them becoming much cheaper than current mid to high end projectors today (these of course being projectors that use full glass optics). Even those projectors show visible barrel or pincushion distortions which would make the end result not as enticing as the theoretical resolution suggests. But I guess the overall speed:quality benefit may be worthwhile in some instances.

5th July 2012, 11:17 PM	#99
mtma Member Join Date: Aug 2009 Posts: 2,593	The Ultimaker is here! Well at least in kit form anyway. 6 to 20 hours of assembly time incoming according to the guide