gr5

Well Nick, not having to level is a major improvement to me. I am very good at it and I still hate it. Regarding pictures: here are pictures of some early prototypes where things were printed and not yet injection molded: https://www.ultimaker.com/spree/uploads/19/original/presskit.zip Here's some pics and video I took of UM2 from last Friday night: https://www.ultimaker.com/spree/uploads/19/original/presskit.zip More here: https://www.igo3d.com/blog/ultimaker-2-pictures-comparison/ Some discussion here: https://groups.google.com/forum/?hl=en#!topic/ultimaker/clvd7U3ltMs Discussion and photos here: https://groups.google.com/forum/?hl=en#!topic/ultimaker/HDR8FyLlMw0 There's a third fan on the print head you can't see that cools the plastic parts just above the heater. Even smaller than the 2 side fans.

It's too easy to leave the nozzle hot with the printer not printing and end up burning and sometimes getting a clog. Personally every 30 seconds I turn the "big extruder wheel" by hand to keep this from happening if the nozzle is hot. It seems like there should be something similar. Perhaps the extruder should *always* be running when not printing - maybe at a very very slow speed. I don't know. I just know that going down 2 to 4 submenus is going to be a pain versus reaching over and spinning the "big wheel". 5 seconds versus .5 seconds. It adds up. Also when I start a print, if I have recently primed the nozzle, all is fine, but if it has been 5 minutes and noone turned the "big wheel" then 3mm of "priming" before printing is not enough. So I usually turn the wheel just before I start a print. Although I also increased the 3mm to 6mm in my start gcode (and slowed down the extrusion speed). So maybe it doesn't matter anymore. - George

Not everything UM2 has been published yet. They are busy - give them a little time.

Daid means 20-25mm/sec is a more *reasonable* speed as 45mm/sec is too fast for some (most?) Ultimakers and they will skip steps. I use 25.

The nozzle size also affects the positioning of the head. For example if you are printing a 10mm cube with a .4mm nozzle, in the gcode, the head will move 9.6mm because the idea is that the plastic will stick out an additional .2mm on each end. At least this was true with the older slicer. Haven't tested steamEngine. - George

If you take the end of the filament, and then tuck it under another loop - you will get a disaster. An always moving knot. The easiest way to get rid of it is to remove the filament from the UM and fix it and then re-insert it. As you use up filament (like someone already said), things get easier.

>Do you print directly on the glass? No more hassle with the blue tape would be a REAL improvement. Glue stick? The UM people were printing directly on glass with glue stick and with the bed at a "low" temp. I think around 50C. The heat helps it stick. >How much power does the machine draw? if you run it a lot, a heated bed can make for quite an electricity bill. Don't know, but if you print PLA - you mostly can keep the temp low. >Do the two fans cool the top part of the hotend as well as the object? No. There is a 3rd fan not visible in any photos. In the back. It's even smaller than the 2 side fans. It cools the upper region of the hot end to keep heat from rising up and causing a partial melt followed by a clog. >I think having injection molded pieces holding the linear bearings is an improvement on the gantry if there is a way to >tension the belts. Belt tensioners hidden inside. Built in. >What is thew processor on board? Arduino. Same thing. With Marlin. >Have some doubts as to the impossibility of advancing the filament by hand.... Me too, but I watched it happen many times and it worked. I think it should be the MAIN control on the display when it isn't printing instead of the feedrate. So if you do nothing, turning the dial controlls extruder. It should not be burried in menus. But the concept of skirt and never needing to level again helps a lot. One thing to note - the current is controllable now in Marlin to all the steppers. So the current to the extruder is set to a value that will skip steps before grinding filament. So you can have it "auto load" the filament where it sucks the filament in and sends it right into the head and lets it skip steps for a few seconds and you are primed with no filament grinding. - George

>The Bowden tube popped out from the print head during the print. Bummer. I'm thinking maybe your head is cooler than the UM is reading so you might want to raise temps by 10 or 20C next time. And/or slow down the print some more. >I may have to remove the print head assembly from the axes If you unscrew the brass nozzle, make sure to heat the head to 180C or so first and be very gentle. Brass is not very strong and many people have twisted it into 2 pieces. Not fun.

Wow. Sounds like a firmware bug to me. Maybe related to your colorful LED's?

That's the fun part for me. I like puzzles. I like figuring things out.

Oh wait! I take it back. I just noticed that you printed several different pieces all at the same time. *that* is what caused those blobs. That you could definitely improve with lower temp (maybe 195C) and by printing slower. Or you could print the parts individually. In Cura you can click on a part and then explode it into it's pieces. Then Cura may or may not print one at a time in a long series. Depending on gantry height setting and other settings ("file" "settings").

Something is definitely wrong. But I don't know what. By the way, when you post a topic, do you click "follow this topic" at the top or will my post never get read? Watching your earlier videos of the squeaking bowden I noticed your extruder appeared to be spinning AMAZINGLY fast. I'm surprised it works at all at that speed. There was a bug in Marlin that kept you from going over 20mm/sec in the extruder drive and that bug is in the version that comes with Cura. Or it did up until recently. The bug was fixed about 6 months ago and if you built with ginge's builder or if the latest cura has the bug fixed then maybe that explains your video. Anyway, people are recommending a max extruder speed of 25mm/sec. I put that in my firmware so I don't have to worry about it although in Cura it defaults to 40mm/sec. I would go into Cura and change it to 20 or 25mm/sec. This one thing may fix all those blobs. Don't know. But I can tell you for sure that you shouldn't have all of those on such a simple print. And it's not the temperature. It could certainly be the print speed. Try 50mm/sec as Ian suggested.

Works for me. I actually saw this behavior recently - maybe a week ago. I just loaded it in a different browser and I was fine. But checking it in all 3 of my browsers it's fine now. If it happens again I'll pay more attention to the browser and will check the browser for javascript errors or errors loading js files.

It would help to have a specific question and/or show us an example of what your mold will look like. The first obvious thing to me is that your first layer Z height is going to be very crucial and will probably take 4 or 5 attempts until you have the print bed at the perfect height. Another thought is you might need a smaller nozzle hole size. Right now the UM comes with a .4mm or 400um nozzle diameter. If you print for example a square, the inner corners will be quite sharp, but the outer corners will have a .2mm radius (.4mm diameter = .2mm radius). See this thread for very visual, very clear explanations of the problem of printing small things: http://umforum.ultimaker.com/index.php?/topic/2717-cura-xy-resolution/ Also be aware that although using a .25mm diameter nozzle hole will greatly increase the resolution of your print, there is a much higher likelyhood of clogging your nozzle. So print quite slow! I would stay under 10mm/sec and I would also modify the start/end codes to not extrude so fast before the print starts.

I printed several of those clips and they all broke within 2 weeks until I printed the zip tie version which works great for me.

There's a direction pin that goes from the arduino to the stepper driver. There could be something wrong with that signal or with the stepper driver. I think it's pretty much impossible to be anything wrong with the stepper itself in this case. But another possibility - is it at all possible that one of the pulleys is slipping - make sure the 6 (yes 6) set screws on the 6 pulleys are all super tight and mark each pulley and rod with a permanent marker to make sure they didn't slip. This is very very common on a newly assembled UM. Back to the electronics. Could it be that this happens when something warms up? Consider swapping the X and Y stepper drivers. There are 4 of them for the 4 steppers. They each have a black heatsink. Pull them off very carefully and slowly without bending any pins - best to pry one side a little, then the other and back and forth. And reinsert very carefully. With power off of course. If the problem moves to your X axis after that then you know for sure it is the stepper driver. You could try lowering the current. Try googleing how to do that - there is a tiny little potentiometer that you can turn to lower the current but make sure you find which model you have as some have higher current when clockwise and other stepper drivers are the other way.

More like 90%. I'm always amazed how an open tube can create such back pressure. You need a special kind of fan (more like an impeller) to fight back pressure. Like those blowers they use to inflate a bouncy castle - that kind of fan wouldn't be affected much by a hose. But the fans used for cooling electronics can't handle hoses or restrictions well. But 10% is better than 0%. Much better. And is probably enough in many situations.

I recommend you use the builder to make a Configuration.h that is close, then take that file and add it to the latest ErikZalm build and edit configuration.h to add the dual extruder options. Or maybe you should take Daid's version of Marlin - not sure. Anyway building Marlin is surprisingly easy. It comes with an instruction file. You download and install the Arduino IDE which is a gorgeous GUI. You install the other stuff in the instrucitons, click the checkbox and you have a hex file (it's hard to find but google helps you through all these steps). Then use cura expert settings to upload the hex file. Or let the IDE upload marlin for you. Or you can do it the hard way (but every step explained in detail) here: http://www.extrudable.me/2013/05/03/building-marlin-from-scratch/ Once you've built Marlin once, it takes seconds to change a setting in Configuration.h and rebuild it.

First of all you should "follow" this topic (button on the top right corner). In fact follow any topic you post. Looking much better. I think you have some "backlash" aka "play". You should tighten up your belts just a little. Probably the short belts (the ones connected to the x,y motors) which fortunately are the easiest to tighten (slide the motor down with more pressure). I can tell by the pattern in the top of your cube.

There are so many possibilities unfortunately: 1) Nozzle plugs (probably not likely but this could be the problem which requires removing nozzle and burning everything out of it). Make sure your filament is clean and doesn't carry any dust into the nozzle. 2) Temperature variations. For many people something goes wrong with the temperature reading (after a few months) and the nozzle ends up being much colder than expected. Do you have a way of testing the temperature? What you show is classic of printing too cold and/or too fast. Try the settings illuminarti suggests but at 240C. I print at 240C often so it's a safe temp to print PLA. Also try setting it to 105C and make sure watter sizzles/boils when it touches the print head. Also have you had any mintemp or maxtemp errors lately? If so then probably a bad wire. Many people deal with the temp variations by simply recalibrating every 10C and replacing the temp table in Marlin. 3) Feeder - check the hobbed bolt. Is there plastic in there jamming it up? Please send photos of the feeder both open and closed and inside where the hobbed bolt is (with good lighting and focus - maybe use reading glasses on your camera if it can't focus that close). I think #2 is most likely.

The thicker layers is just nicely masking the problem. I think. The problem is with z movement. You tell it to move .1 mm and sometimes it moves less than that and you get overextrusion and fatter walls, and sometimes it moves more than that and you get underextrusion and thinner walls. And this pattern alternates in a semi random pattern. I still think it is Z movement related - I just don't know how to prove it. I guess you need to put a piece of paper taped to the side of your UM and put a mechanical pencil on the bed. Move 1mm at a time and make a mark each time. You should see the same/similar pattern with some marks closer together and some marks farther apart. Watch the Z nut very carefully as you move .1mm at a time. Did you grease the Z screw?

1) Your initial problem was you simply had support turned on. Now you should turn on .8mm thick shell thickness (minimum 1.2 possibly but .8 should be fine). Also .4mm top/bottom thickness. Also - and this is the key here! - go to expert settings and check "solid infill top" as well as bottom. I think maybe you left off the top. Or maybe your print didn't finish. 2) You are now also underextruding a bit. Tighten the screw on your extruder a few turns. The spring should be compressed to around 11 to 11.5mm. You are printing at a moderate speed that should not experience underextrusion. You can fix this by slowing down another 30% but you should be able to print easily at 50mm/sec .2mm layers and 200C or hotter. Which seems like that is what you are doing. So your extruder isn't pushing hard enough. Most likely you need to tighten that spring. If that doesn't help send us a photo of your extruder in closed position. It should be able to pull 22 pounds on the filament before slipping.

To reduce underextrusion, print slower, hotter, and with thinner layers. Hotter because the filament is less viscous. Typical speeds for me (I like to get close to the limit): 100mm/sec 240C .2mm layer 150mm/sec 210C .1mm layer Or for a part that needs to look beautiful: 50mm/sec 200C .2mm layer 100mm/sec 200C .1mm layer (these last two print at the exact same volume per second)

Flow is very simple. If you set flow to 200, you get double the normal amount of plastic. If you set it to 110 you get 10% extra. For example if you are printing .2mm layers and your first layer is also .2mm and you did a bad levelling of your bed and are off by .1mm, you can increase flow to 150 and you will get enough plastic for .3mm (instead of fixing the levelling).

Well is it possible there is a single move on that level that goes beyond 205mm? I would look at the gcode. Maybe you can search for "X 0020" or something? Maybe there is a simple search you can do to find all the moves beyond 205 easily. I would also look at the slice-view in cura carefully and see if around that layer the print head moves beyond 205 or beyond 0 (goes negative) in X or Y. Find the farthest out move, note the layer, then look at the gcode in a text editor. To be off by 10mm it needs a move that is 10mm beyond the end (-10 or 215).