gr5

Well.... experts (aka other posters) claim that soldering is the entire problem. Those connectors are supposed to be crimp only. But if you solder then the solder wicks up the wire making it stiffer and giving it a smaller area where it can flex. If you *do* end up soldering, then you should use the strain relief well. I put a blob of hot glue gun to connect my temperature wires to the strain relief. I don't know if that was a good idea or not.

I see they have it also! Just be careful to get "adhesive" and not "paste".

I guess there are 3 things going on. If you are printing high volume of PLA you need to increase temperature to get that much to flow through the nozzle. The max you can do with a .4mm nozzle is around 10 mm^3 which works out to around 120mm/sec at .2mm layers. But you need it hot to get those speeds. But when you are at these limits of printing speed (whether hot and fast or colder and medium speed) results in high pressures of PLA in the nozzle. This can result in underextrusion. Also issues where you can get blobs or leaking when the head slows down for a corner as there is still high pressure in the nozzle (so over and under extrusion at the same time). The 3rd issue is stringing which is also worse at higher pressures, but even at low pressures stringing is better the lower the temp (I like 190C to get rid of all stringing). So there will always be a tradeoff between printing speed and quality. The UM has fantastic design and can move the head at 300mm/sec very accurately but for excellent quality you need to print more like 50mm/sec or 20mm/sec. I usually prefer to print fast and don't care about quality as much. Note however! That if you are printing .1mm layers you can print twice as fast and get the same quality regarding blobs and stringing and still avoid underextrusion as well.

Very nice! Those hours and hours of sanding and filling paid off. Please post what filler, primer and paints you used.

Yeah - it looks like a Cura bug to me. When I ran the "first run wizard" and told it that I was not using an ultimaker and said that the center was 0,0 it did the same thing - centered the part it at 0,0 and put it in the corner instead of the center of the bed. I think it will slice fine - it just won't show it visually correctly. Cura 13.03 works fine but it is a different slicer engine and runs much slower.

By the way that upper print - the white one - is very impressive. I like that one a lot.

I consider 225C to be medium. I often print at 190C which I consider cool. And I often print at 240C which I consider hot. 225C is just normal for me. Oh! I laughed at that one! Well if you want to avoid "spiders" print cooler. And if you want to avoid blobs print slower. If you aren't picky about quality you can print very fast but if you want very very good quality there is no substitute for printing nice and slow. I would have printed both of these objects at .2mm layer and either 100mm/sec at 240C (or 50mm/sec at 190C if I don't want the spiders). But if I want it to look even better I would print 70mm/sec at 240C or 40mm/sec at 190C. And as you go slower and slower it will look better (less blobs) and as you go cooler and slower you get less stringing.

Excellent photos by the way. I should also mention that Z screw issues often repeat every 3mm vertically (the vertical movement for one rotation). Often the Z screw has a huge wobble which causes the Z nut to slide around which can cause these kinds of issues. Also did I mention to try adding more grease? The green tube that came with the UM.

You have 4 problems caused by 2 thing. First the minor thing: You levelled slightly lower on the newer print so the first layer is underextruded. This also means the first layer didn't make as good contact so the part was able to warp a little. These are not a big deal and can be fixed by levelling more carefully probably with the nozzle hot (the nozzle expands when it is hot). The major problem: Well you have over extrusion on one of the lower layers -0 maybe layer 5 or so, and then you have it again on the layer at the first "top" section about 1/3 of the way up the part. ALL of the other areas look pretty good. This overextrusion causes blobs when the filament explodes out of the over pressurized nozzle and also causes extra stringing due to too much pressure in the nozzle. So what causes the overextrusion? And why only in two vertical areas? It's almost certainly related to the Z screw. Home the Z axis, then disable the servos and turn the z screw by hand. Turn it until you have moved at least the height of this part. Notice if there is a sticky spot. Turn the Z screw back the other way until you're head touches the limit switch and repeat a few times. Try to notice: 1) Is there a sticky spot or two? 2) Watch the Z Nut carefully - that big huge nut that is embedded into the Z stage. Does it slip around suddenly at all? 3) Consider putting more grease on the Z stage 4) Does the Z stage hit anything? Is there something that the stage hits on it's way down? Maybe some wires? Look all around the edge and under the Z stage for any obstacles that might slow it down briefly. If none of this shows you anything you could try experimenting with pronterface or cura to move the Z stage with the stepper. Listen carefully for any noise changes. Consider 1-4 above during this. A good test print for you would be a rectangular box that is 2cm by 2 cm in X and Y and 20cm tall. Slice and print that at .1mm layers (to show off the problem more) and also make darn sure your layer cooling time is enabled and set to at least 5 seconds. Basically you need to fix the Z but if you are in a rush you could simply change your layer height to .2mm and you might not get such severe overextrusion.

Earlier you said everything was too loose. I would have just used a hot glue gun and smothered the z-nut (and anything else that was moving) in hot glue. It would have been even uglier.

Just? Wow! Cool! You get the logo for free as a bonus! I am highlighting this post in my bookmarks for when I need to do this some day. I've used openScad which I assume is the hardest step anyway so the rest should be easy I suppose. Don't know inkscape at all but maybe I can use photoshop.

Well, it sounds like you know some electronics. If it were me I would try to figure out this puzzle myself even though it sounds like you could just get UM to send you a new board. Attached is the relevant part of the schematic (but the 1.5.6 version but should be the same I would think). I would examine the circled pin and make sure the logic changes when the X moves left versus right on a long move (long move so there is time to measure the voltage without needing a scope). The pololu like stepper driver uses the same 2 wires to move left versus right and does so by reversing the voltage across the 2 signals that move the stepper. So I don't think it's likely that the pololu could break in this manner. So I think it must be this "dir" signal (direction). It could be another bad arduino or a bent arduino pin (most likely) or a bad solder on either end of that signal. Or a bent pin on the driver. The driver itself could be bad - this seems very unlikely but easy to test by swapping X driver with any other driver. later edit: just to be clear - this direction signal should be around 5V (logical high) for one direction of moving and 0V (logical low) for the other direction.

You can power up the board just fine without that heatsink but *do not* move any stepper motors until you get that heatsink properly reattached with thermal adhesive. Do not use normal glues, don't get thermal paste. It must be "thermal adhesive". You can't get this at bestbuy but you can get it at some walmarts I think: http://www.amazon.com/Arctic-Alumina-Thermal-Adhesive-Tube/dp/B0009IQ1BU/ref=sr_1_1?ie=UTF8&qid=1379219795&sr=8-1&keywords=thermal+adhesive

I'll bring a few. Nick - if you are going to MakerFaire can you bring maybe 5 meters of printbl green that you like? I'll bring the printbl purple that I love and we can trade some samples to try out.

Google it and you will find the answers about PLA - there are great summaries out there on all the wonderful and crappy PLA out there. I use printbl.com and they have very good quality.

The nice thing about running HB through UM is that when the print finishes it automatically shuts off the HB. That's probably the only reason but it's a pretty good reason. If you start a print and then go out to lunch or go to sleep or get a phone call or whatever, it kills HB power when done so when you come back your print is ready to pop off the bed.

Uggh! This happens all the time with the new Cura. It drives me crazy. The quick solution is to play with the 4 "Fix Horrible" checkboxes. Note that there are only 11 possible ways to check as checking A and B at the same time doesn't do anything differen than B alone. I'm going to guess you need the last checkbox only to fix this. But every time is different. The problem is that the new steamEngine is much dumber about inside versus outside (solid versus air) than the older slicer. The new slicer ignores layers above and below the current layer. So. Any tiny error (extra interior walls) freaks it out and causes issues just like this. The other solution is to use the older cura: 13.03 is my newest cura of the older slicer. The final solution is to clean up your CAD model - if you look at your model in Cura "xray" view - anything that is red is a problem spot that you would have to fix. Instead I recommend you try all 11 combinations of the fix horrible checkboxes one at a time. Maybe write down what you tried as you go. Also try to get the problem to occur with thicker slices (e.g. .4mm) to save time until you find the correct combination of checkboxes to fix this.

Well - not what you asked - but you could reduce the polygons by a factor of 2X or 100X potentially without any reduction in resolution (resolution after printed). I love using meshlab: Click on menu item: "filters" "remeshing..." "quadratic edge collapse decimation". I set quality to 1 as recommended in a shapeways article about this subject. I had a model of a human with over 10 million polygons but I wanted to print it only .5 inches high. I reduced the polygon count to 10K polygons and it was still more than I needed. After reduction Cura handled it nicely (before reduction Cura had issues).

Ah! I see - there are rev 4 instructions now but you have a rev 3 wooden piece. Yes, I would widen the hole. I hope you have drill. Or keep looking - you might have both. My rev 3 kit came with some rev2 parts that I just threw away. The Ultimaker is a kit for hobbyists. In theory you shouldn't buy it unless you feel you have the skills (but maybe not all the tools) to build the whole thing yourself. It's kind of an attitude: Hey all you guys building these 3d printers using our parts list and stuff - why not just get all ( most? ) of the parts from us as we put this kit together for you. Having said that, if you write a ticket explaining that you have the wrong top, you should be able to get UM to send you a new top for free. But you might have to wait a few weeks. The sooner you get the support ticket started... HOWEVER, there is a chance you have it but it came in a different package. I was amazed to find some of the very important parts I needed for my UM were in with the ulticontroller pieces. These were pieces that weren't in the original design so they were packaged up separately. So this part ("top") may be in a totally different section of parts than most of your plywood pieces. Plus when you break up all that plywood into so many pieces it is easy to lose one or two.

It sounds like a cura bug for non-ultimakers. Please run the "first run wizard" under expert menu, and let me know what you fill in for all the values. I will then do the same thing here and then place two simple cubes and slice and see if I get the same problems as you.

I think after a few posts the approval process stops occurring.

NO - don't tighten the belts. Tighten the little tiny set screws that keep the pulleys from spinning freely.

1) Good about the tape. I assume this means you used isopropyl alcohol. Either that or by the time you do the 3rd print the tape is getting sticky. If you rip the tape obviously you just put down new tape on the areas that you rip. To get the tape off the PLA soak it in a puddle of isopropyl alcohol for 5 minutes and then it just slips off. 2) The pyramid is just to get stringing down to minimal. It's a good test peice before printing eifel tower. The tower will have thousands of strings if you don't print it right. More details on stringing experiments here: http://umforum.ultimaker.com/index.php?/topic/1872-some-calibration-photographs/ 3) To change plastic, heat up the head to 180C (or hotter but 180C is fine). Then release the extruder and pull out the filament. Sometimes I forget and the head is cooling and is down to 150C and I pull nice and hard and it pulls out more plastic than if I pull it out normally so there is less in the head when you put the new filament in. You will need to prime the head with the new filament. For that try 240C. Or at least more than 180C. This is a good time to feel the pressure - turn the gear slowly and watch filament come out and at the same time set the temp to 240 after being at 180 and watch the relationship between temperature and viscosity (180C is like toothpaste, 240C is like honey). 4) Don't do ABS yet. lol. If you print anything larger than 2 inches in X or Y you will probably need a heated bed. Plus I hear that ABS clogs your nozzle real easy if you print too hot. Clogged nozzles aren't fun.

Cell phone case came out great! I published the cad files (stl and sketchup) and instructions on how I did it at youmagine: https://www.youmagine.com/designs/galaxy-s4-ballistic-case This technique of using 2 colors with a single head works really well. I've learned some things. The first layer needs to have no stringing so you need to keep the temp cold (190-200C) and print slow. Leveling is super super critical. If you are off by .05mm on one side then that side will be thinner and the color behind will show through. Whatever you print on top (totally different gcode file from cura) can just be started from the begining and just put a layer height of .4mm in Cura and it will just ooze over and around the first layer color. I printed the second model (and second color) at 240C for the first layer so it would flow into the cracks among the first layer colors nicely. And bond well. I used a heated bed for this at 70C and covered with kapton tape as this gives a better finish than blue tape. I think the technique should work okay for cold bed and blue tape but didn't try it. I rarely use the heated bed because I rarely care about the bottom surface finish. But in this case the bottom surface is the most important layer.

If you calibrate with paper and a cold nozzle you don't have to turn the screws and extra 1/8 turn to compensate for the thickness of the paper. If you have a calibration method that works just stick with it. The biggest problem I have with cold nozzle is that often there is a tiny ball of plastic on the tip of the nozzle but it's too small to see. So when I finish calibrating and then heat up the nozzle, calibration is off.