gr5

print speed. Infill is this cross pattern with mostly open space.

Daid while we are talking about speeding up infill without hurting quality... It would be nice if for those times when someone sets the infill speed at a different rate than the print speed, it would be nice if for the last 1 second worth of travel, or the last 50mm or something like that - it would be nice if it would gradually (or suddenly) switch back to the print speed. So that there isn't this bad quality the moment it changes speeds (typically over extrusion when it starts the skin again).

You mean .4mm nozzles? Which outer edge? For a cube, fine, all three layers stack on top of each other but for most prints they don't. So which of the 3 outer edges do you run the infill to?

In my opinion this is a big deal because those 2 outer layers do not "stack up" on top of each other so do you take the more "inner" layer? Or the higher layer as your contact for infill? I think kisslicer might already do this and it has a nice gui just like Cura and it is free.

That sucks. Maybe you have a steel bristle brush where you can cut off one bristle? Another solution is to heat the head to 180C, remove filament, loosen the 4 screws about 4 turns each, then remove the blue clip, push down on the outer ring that the blue clip held up and lift the bowden tube out while pushing down on the ring. Now with head at 180C cut a short piece of PLA and insert it. Then change head to 90C. Walk away and come back after the head has been at 90C for a full 3 minutes and pull firmly the PLA out. This should hopefully pull out some or all of your plug. Reheat to 180C and see if you have good flow. If not, repeat several times. Or, take the head completely apart (not really that hard after you got this far!) Including removing the thermocouple and heater (very carefully!) and do this while at 180C so the pla doesn't lock any threads. Then take the nozzle and put it in a hot gas flame.

The third fan (not to be confused by third eye?) comes on the instant you switch on the UM2. Even before the lights. So that is useful debugging info. Usually you simply lift up the woven cover above the print head and that exposes the connectors. The side two fans are in serial on the same circuit (you can see a short wire connecting the connectors. The rear fan has it's own separate power.

This does nothing for PLA but works for ABS. Are you printing with ABS? If PLA, then THF can be used as a solvent. If you buy THF make sure you get it "with inhibitor". Be careful with Acetone and THF as I believe these are quite toxic and quite flammable.

So which versions of Marlin now calculate the D factor (slope of the error) before it hits the +/-10C window? I assume the versions of Cura that come with UM1 don't have this fix, correct? Also UM2 has the fix, correct? Also the latest ErikZalm version has this fix, correct?

Then you need the next advice - how to get the damn parts off. First make sure it cools down to at least 50C as hotter than that and you risk deforming the part. I usually use a putty knife or knife with a razor blade. Once I get a corner up I always switch to a putty knife if I didn't start with that. If you use the glue stick it dissolves nicely in water so sometimes I take the whole sheet of glass out and put it in the sink. You could just put the glue stick under the corners that tend to lift. Also freshly cleaned glass sticks better than glass that has oil from your fingers on it.

Robert gives great advice but also know that you will not cause damage. The UM is designed to withstand this kind of force and the steppers definitely don't care at all if they skip steps (which is partly what is happening).

Regarding those rings: It's tough to go from zero extrusion (retraction even) to 70mm/sec printing instantly. It takes a 100millseconds or more to get up to speed. So for larger things like the tree itself you don't notice as much the bad spot before quality improves. For the tiny ring it's obvious. The simple solution would have been to slow down the printer to 40% feedrate when you got near the top of the tree. The quality of the trees would have been improved greatly at 30mm/sec but it also would have taken more than twice as long so it's a tradeoff.

Actually that creates little vertical spikes. "non-printing" is harder than one would think.

Again, can't comment on readers. They are probably all fine. You will get a free SD card with your printer so you don't need to order one.

If you are printing a circle with 10 segments, jerk allows the printer to not slow down much at each vertex so it can print the circle fast. If you are printing a square it slows down more at the corners (vertexes). This is controlled by "jerk" which is a Marlin specific term and not the same as what a physicist would call "jerk".

Not sure what part you are talking about. UM Original? or UM2? Can you take a picture of the part? Does this help? https://github.com/Ultimaker/UltimakerOriginal

I believe the problem with the ribbon cables is usually "cross talk" where signals from one cable pass to the other as radio waves. So a meter won't show the problem. Try keeping the cables at least 10mm apart. Or just wait for Sander's new cable.

Ian has great advice - all those things help the part *stick* better. But there is another technique that instead reduces pulling forces. Another technique is to do all that but instead RAISE the bed temp a little and keep the fan off (although 70C is usually plenty). This second technique keeps the bottom 10mm of PLA *above* the glass temperature so that it is soft and mushy and the pulling/shrinking stesses deform it a little but not enough to pull it off the bed. This second technique usually requires the fan to be *off*. However only use this second technique for large parts - parts that take more than 15 seconds to print a layer. If the top layers of the part print less than 15 second you need to turn the fan back on at that point. A third technique creates a heated chamber. This isn't usually necessary.

I like Daid's theory - slipping on the extruder. You could print at 1/5 the speed and see if that helps. My theory is Z axis movement. It sticks a little or something and when you command it to go down .1mm it actually goes down .07mm and so you get 30% over extrusion and the layer sticks out. Then you command another .1mm and it corrects and goes down .13mm and you get 30% under extrusion. Something like that.

In the photos it looks like you *did* use gluestick. I much prefer brim over raft. Also I recommend raising your print bed a little warmer to 110C to keep the ABS above the glass temp. You can test this: put a thin piece (preferably flat) of filament on the glass when it is at 110C. Let it sit there for 5 minutes and then try to bend it. It should bend into a new shape. If the heated bed is at 90C it should *not* bend. Being above the glass temp means the shrinkage stresses are spread out more throughout the softer material and so it stays on the bed better. The fan can wreck this though - some people use zero fan for large ABS parts.

Nice! Yah! You finally got your own printer and tried it yourself! Did you: 1) use gluestick or hairspray? 2) try to use brim? Even just 2 passes maybe? I assume the "back" two corners are the worst - those should have room for some brim - you can add it in the cad if you can't do it with cura. If you can't do that you could try a "raft". I've never done it. I'm not a fan. But a lot of people swear by it - it allows a little flexing and warping without actually seperating from the print bed. Usually. Raft is very popular among cheaper printers and among people who do ABS (which has much worse warping than PLA). 3) Is the bottom layer squished a bit into the glass? This is important - if the bottom layer just is laid down without being pressed in it won't stick as well. I recommend .2mm layer height for this part since it is so large and there isn't a lot of critical detail I would think. Either that or increase the speed to maybe 80mm/sec (but don't do both). This recommendation is only to save time - it won't help the quality (it might hurt but it should be okay).

Cut your speed in half and it will click less. Keep cutting the speed in half and eventually it will stop clicking. Here are tests I did. I found that my max speed at 210C .2mm layers (for ultimaker blue filament) was around 40mm/sec. Of course with .1mm layers you should be able to go twice as fast. http://umforum.ultimaker.com/index.php?/topic/3418-um2-extrusion-rates/

If it were me I would take the 2 screws off the bottom cover and remove it. Then trace the power cord and power switch and make sure when the power switch is on the power gets past the switch. I would test this using a meter. Then I would trace this signal to the main voltage regulator - a large part with 3 large pins. I would check to make sure the part generates 12V. But truly the problem is probably before this part because I think the 19V goes directly to the rear fan and so if the rear fan doesn't spin then something VERY SIMPLE is broken. I think. Possibly the switch. More likely a cable or connector right there where the power comes in.

Ah. Good picture. Okay. This implies the darlington transistor is blown. Q4. It has enough power to drive the multimeter to 17V but not enough power to drive the fan. You attached 12V to the wire under the printer right? This transistor is easily damaged by normal use. I recommend you simply replace it.

Sometimes there is electrical noise/interference between the fan signal and the temperature signal. Sometimes the problem is worse when the fan is *not* at 100% or 0%. The solution is usually to move the fan wires in the print head away from the thermocouple wires. The problem area is not above the head. The problem area is between the thermocouple and the tiny circuit board on the print head. As a test, set the fan to 0% and temp to 200C. Watch the temperature curve. After it settles, turn the fan on to 50%. See if it starts to oscillate or swing randomly. Then set fan to 100% and watch again. Another common problem is the cable and connector. AFter the above test, while head is at 200C push it around to the 4 corners several times. Pause in each corner to see if the temperature changes suddenly. If it does you have bad wiring above the head.

You didn't answer my question but I can infer it from the fact that you turned on support *and* the top surface looks good. You need to do #1-6 in my post above. Also: 7) Turn off support 8) Rotate the spoon 180 in Cura so that the flat part faces down.