illuminarti

Steve - The '0.1mm' number that is quoted comes from a reply that Daid gave on here shortly after the Ultimaker² was launched. Looking at the code that you pointed out, it seems that in fact it assumes a 0.2mm thickness - and indeed, it only applies that correction to the currently 'live' version of the homing offset, when bed leveling is completed (when the second pass of the front-right leveling has been done). By then, the homing offset has already been stored in EEPROM - that happens after the second pass at the back-center leveling. Therefore, for the rest of the session after running a bed leveling, the bed homes to 0.2mm closer to the nozzle than the height set in the bed-leveling - i.e., about 0.1mm too high, if the paper used really was 0.1mm thick. Conversely, after power-cycling, the 0.2mm offset is gone, and now bed homes too far from the nozzle, by the thickness of the paper used during leveling. I'll put in a patch for this, along with the changes I'm working on to finish fixing the start-up sequence.

What Operating System is this? And what type of printer are you connecting to?

It's good to hear that you are getting good results from your new printer. You do have to push pretty hard to get the filament in sometimes. It seems that sometimes there can be some sticking of the extruder tension arm/roller that makes it harder to move than it should be. Loosening the four bolts that hold on the extruder, and then retightening until you just feel resistance can sometimes help with that, as can putting a lever (like the allen key) into the hole in the middle of the grey tension wheel, and levering it against the adjacent plastic ring to push the tension wheel a little to the left. Before your printer shipped it would have been fully tested, and the plastic went in ok for that, so maybe something settled slightly in shipping. Your trick will be to find the right level of tension, so that you can get the plastic in and out without a problem, but still get sufficient pressure that it doesn't grind the filament during printing. If you have retraction-related under-extrusion that is caused by the filament tending to grind away, rather than the motor clicking because it can't push the filament any harder, then you might get better results by slightly increasing the tension. In theory, with the current extruder design, and the right amount of tension, you should get very little grinding of the filament. I guess the rumors that I've been hearing and mentioning for a while about Ultimaker expanding their presence in the USA are true after all... :wink:

See here for my advice on how to reinsert the Bowden tube in the hot end... http://umforum.ultimaker.com/index.php?/topic/4586-can-your-um2-printer-achieve-10mm3s-test-it-here/?p=40399 As Robert says, once it's all assembled, the Bowden tube shouldn't move either up or down. It may be that the Bowden tube has been scraped by the retaining clip during the removal and reattachment process. If so, cut a few mm off the bottom of the tube, so that the retaining collet has fresh plastic to grip.

Is the power plug fully inserted into the back? It should click into place, and not come back out until you pull the sleeve back on the connector.

The poor sides on that rook piece are due to retraction not working right (probably because you lowered the distance), and also a lack of cooling when printing all of those tiny pieces. You need to set a longer minimum layer time, and maybe print multiple pieces at once so that there's more time for cooling between layers. Also, you might find you get better results with thicker layers. I rarely find a need to go below 0.1mm on the Ultimaker².

My guess would be that probably some part of the print curled up, and caught on the head?

Haha, yeah. I mean 30mm³ volume. Not speed. Thanks for catching that.

I think a picture would help. Here are the instructions for picture posting: http://umforum.ultimaker.com/index.php?/topic/4525-how-to-upload-an-image-to-the-forum/ However, it's an inherent limitation of this type of technology; the print is built up in layers off a base platform. You can't have an infinitely small contact area with the platform; indeed you need quite a large contact area to get the print to stay stuck down during printing. Printing off a raft or support might help. Finding ways to orient the part differently and/or assembled it from multiple prints is probably a more viable option. In any case, some amount of post-print finishing/clean-up work is going to be necessary.

The stringing is due to the nozzle oozing slightly during the moves between each island. A tiny drip forms, and then wipes off when it gets back to the edge of the print. That's why the fronds gradually move away from the shell - the wipe happens on the edge of the previous layers wipe blob. Your settings look basically ok, and its a nice looking print apart from the ooze. I'd make the retractions a bit less enthusiastic - set a minimum travel distance of maybe 1mm, so you aren't retracting even for tiny infill moves. That's not relevant to the problem, but it will help speed the print up and avoid grinding. The other thing I'd do is turn off the z-hop setting. Z-moves are much slower than x-y and are probably slowing down the movement, allowing more time for the head to drip. Finally, check your retraction settings in the printer. I generally use 5.5mm at 35mm/s.

Lol. I wholeheartedly agree about the value of comments. But, in my defense, the constant in question is called PRIMING_MM3_PER_SEC, and in configuration.h it's defined thusly: #define PRIMING_MM3 30 // number of mm^3 of plastic to extrude when priming // (Ultimaker 2 hot end capacity is approx 80 mm^3) #define PRIMING_MM3_PER_SEC 5 // Rate at which to prime head (in mm^3/s) // (Ultimaker 2 upper limit is 8-10) I made the conscious decision to define these things in mm³/s, precisely because these are factors that are solely related to pushing plastic through the head, and the limiting factor is related to, and understood in terms of, the total volume of plastic per second. A speed of a few mm/s doesn't sound like much when we're used to moving all the other axes at speeds that are 10's of mm/s. But we need to remember that in normal use, the filament typically moves at 1mm/s or less. Having tested it, I don't think that 30mm³/s is quite enough. By all means increase that to 50 or so.

One thing to check is that the fan shroud isn't pressed up against the nozzle; make sure that the nozzle is centered in the hole in the fan shroud.

Yes, the front control panel has a 'tune' menu that is available during printing and allows you to change the speed multiplier, flow rate multiplier, head and bed temperature, retraction speed settings, fan speed, and LED brightness while the print continues.

Could it be a firmware related partial-grind at start up that finally makes its way to the apex of the bowden, or the hotend, or something, and gets stuck? Are you getting the motor skipping back, or just renewed grinding? If grinding, try increase the tension on the spring a bit - the motor should be able to grip the filament well enough that it skips, before it gouges the filament.

The white indicator in the corner of the extruder sets the tension that is applied to the filament to press it against the knurled wheel. The lower the indicator mark, the higher the tension. It is adjusted by inserting a hex screwdriver in the hole above it, and turning the screw. You turn the screw anticlockwise (as seen from above - 'unscrewing it') in order to increase the tension.

The difference after a print finishes normally compared to when you abort a print is because when you abort, the filament gets retracted 20mm. When you end a print normally it doesn't. It should, and in my version of the firmware, it does. In the latest latest release firmware (which you don't have, Rick, if you're using firmware from 14.06RC-something), the priming at the start of the print happens in two phases. First, a fast movement, to undo the 20mm retraction, and then what was supposed to be a slower true priming. However, two things got unfixed compared to the code I submitted. Firstly, the 20mm retraction at the end of the normal print doesn't happen - so there's an attempt to extrude 20mm of filament pretty fast - which can cause stripping on newer-model printers that have weaker springs. On older printers, it just leads to skipping usually. And then Daid changed the speed of the slower true priming, so it actually now happens super fast, kinda defeating the point, and leading to yet more stripping.

So, it sounds like you are saying that the correct bed level is only achieved *after* you have power-cycled the printer?

The offset on the bed that you are seeing is probably just caused by the fact that the print nozzle is offset from the center of the head. The single nozzle actually covers an area of the bed that is shifted slightly to the left of the entire physical plate. When a second nozzle is added alongside, it will cover an area slightly offset to the right. The print area that is defined and shown in Cura is actually the area that the left hand nozzle can reach, not the entire build plate (which extends a bit further to the right). So everything prints slightly to the left of where it looks like it will print in Cura.

Is your firmware version now 13.06.2? Thats the one that comes with the latest Cura (13.06.1). Most likely you just need to rerun the bed leveling routine on the printer. There's definitely no need to re-save the gcode in the newer Cura. There's nothing in the slicing process that would change from version to version about z-height.

Because the speed and the amount were both defined in cubic mm. The volume factor converts between linear mm of filament and volume of filament. The function being called expects a speed in linear mm of filament/sec. The constant is defined to be 7, iirc (I'm not in front of the code). I want to prime at 7 cubic mm per second. Because that's in the range the head can handle. By taking out the volume adjustment (which multiplies the constant by about 1/6.5), you're feeding the filament into an already primed head at 7mm/s. That's about 45mm^3 per second. The head can't handle that. And with the new weaker springs, it's not guaranteed that the extruder motor will slip back. In a lot of cases, it just grinds.

The red and black wire is for the fan on the back of the head. That needs to plug into the single receptacle that sticks up from the middle on the far side of the board. (Which will require you to unscrew the board from the frame to get at it). The blue and white cable pair is unused. The yellow green pair powers the side fans, so check that connection in your photo, and then go look at the other end of those wires above the print head. Pull the sheathing back at the printhead end, and look at the connectors there. the yellow and green pair supplies the power to the two side fans. The yellow and green each go into a connector at the print head end, and a short green cable jumpers them together to wire the fans in series. The side fans connect to these two via their own red/black pairs. The red/black pair coming from the electronics board connects to a third red/black pair that goes to the rear fan.

Its a bit hard to see whats happening before the drop... is it just that it's moving/falling too far when that big clunk happens, or has it not been moving properly in the previous layers, and now it's catching up? It looks like it's happening on the intended layer change, as it immediately goes to printing perimeters afterwards.

Which firmware upgrade are you using? When does the 'horrible noise' happen? it sounds like maybe you aren't triggering the end stops properly? Have you checked whether the cross rods that go through the head are fully seated into the sliding blocks. If you power off the printer and move the head by hand, does it move smoothly? Do you feel and hear the end stops clicking in the back right corner before the head/fans crash into the frame? You can download a previous version of Cura, and install the firmware from that... but it may be best to try performing a 'factory reset' on the printer. Remove the filament first, and then go into the maintenance advanced menu and perform a factory reset. Then go back through the leveling and filament load process.

If the print is coming off the bed, it might be due to temperature or first layer height - you certainly want the first layer to look slightly squashed into the bed, and not like the plastic is being laid on top of it. Small parts, or parts with sharp corners may benefit from using brim to add contact area, and spread out the forces on at the corners. Glue can also help; apply gluestick thinly before the print, and then spread it with a wet paper towel, so that it dries into an almost-invisible film.

The two follow up commits broke the 20mm retraction at the end of a print, and the slower priming extrusion.