gr5

My UM1 belts are much looser than I would have expected belts should be. I get good quality. I have heard that tight belts cause high friction and may cause other problems.

1) After slicing did you look at the stool in layer view? 2) What? does it just sit there? Is the printer quiet? No steppers moving? Are fans still running? It sounds like you lost power to everything but the arduino maybe? Does this "pause" take one or two seconds? Or does it last for many minutes?

The belts go through a "block". It sounds like the 2 X blocks on the front and back side are not lined up. Either that or the 2 Y blocks aren't lined up. You can open up these blocks - I haven't done it yet - I'm not sure how you do it but it should be pretty simple. Inside there is a belt tightening mechanism and some way to let the belt slip. You can either let the belt slip through so that the axis is now lined up... Or alternatively you can loosen the set screws on the two pulleys on one of the sides and you can then move the belt on one side without the other. Get everything nice and square and lined up and then tighten those set screws tight. Tighten them very very very tight. Else they will slip.

MB2 acceleration settings must be very very low with its heavy heavy print head. This explains a lot, thanks.

@etienne - UM1 or UM2?

I often print only 3 or 4 layers and abort the print. Then measure the actual size and go back to the cad and adjust.

Much discussion here: http://umforum.ultimaker.com/index.php?/topic/3418-um2-extrusion-rates/ But no, M907 as in: M907 Z255 is full current THIS IS WRONG - 255 WILL DO 255MA M907 Z166 is default (1.3 amps) ALSO WRONG - USE Z1300 TO GET 1.3A Actually I think M908 will work also: M908 P2 S127 (or maybe P3???) should set Z to 1.0 amps Now I tested these commands a while back as you can see in the link above but they didn't work. But Daid said he would enable this feature in a future version of Marlin so maybe these work now.

My quick summary. If "jerk" is zero then it's simple - every line segment would start from a stop, accelerate to max velocity, maintain that velocity as long as possible and then decelerate to 0 at the last possible moment so that it ends the segment with a deceleration ramp. But in practice it doesn't come to a complete stop at each segment so it's similar but starting at a particular velocity, ramping up, maintaining max speed (if it gets there) and then decelerating to be at the proper next vertex speed when it gets to that vertex. The "jerk" setting is not zero. The jerk setting is not the official definition of jerk but similar. It sets the velocity at the vertexes where any two line segments come together - the idea is that for a circle of 20 line segments the change in angle is minor and you shouldn't have to come to a complete stop 20 times. So it calculates the "delta" velocity vector from one instant before the vertex to one instant afterwards and keeps that delta below the jerk value. So if jerk is 20mm/sec (default i think) and two vertexes are in a straight line you can go unlimited speed from one vertex to the next. If the vertexes are a 90 degree corner then the minimum speed at the vertex will be sqrt(2)*20 or 28mm/sec. The more gradual the change in direction the faster you are allowed to go. The worst case - 180 degree change in direction implies max speed at vertex at 10mm/sec. Now you have to calculate all these max speeds 16 segments ahead and work backwards to "now" and figure out what your max speed is along the whole path.

I said it already - but consider assembling the Z stage now with the parts you have.

Probably not. More likely the later - nothing comes out and pressure gets high enough that the extruder motor simply "clicks" and jumps back and misses steps. No big deal as long as there are a few more layers with similar foot print. The default for cura I think puts the bottom layer at .3mm.

You can't? You don't need to heat the bed to print - but you *do* need to heat the nozzle. I would think there is enough power for the nozzle and the steppers. No? You can print cold on the blue tape (you have to relevel of course) and there are tricks for printing on cold glass here: http://umforum.ultimaker.com/index.php?/topic/3404-printing-on-glass/

The UM2 has software controlled stepper current. There is a way to control it but you would have to add a special gcode command to your gcode file before copying it to the SD card.

Dual nozzles just doesn't work very well yet. The "off" nozzle drips and makes a terrible mess. Also increasing the speed on infill reduces quality as you get over extrusion when it goes back to the skin. But if you are making parts that no one will notice (like the feet on ultimaker or gears burried in a mechanism) this could be useful.

I used to see this often when I used to do USB printing (I don't any more so it might be that) and also I upgraded my firmware last January and haven't seen it since (also did more upgrades this past spring). The version of marlin you get from here is much much newer than that which comes with cura. And it has some very nice and minor feature improvements - try it and if you don't like it you can just reinstall the version that comes with cura: http://marlinbuilder.robotfuzz.com/ So I'm thinking it is a bug related to USB printing or maybe a bug that was fixed in Marlin.

Because of the flexibility of the clear bowden tube there is a springiness to the pressure on the filament in the nozzle. So when you speed up you are getting higher pressures getting that PLA through that tiny nozzle, and when you slow down it reduces pressure. But there is a delay to the pressure. So when you have a path that is a long rectangle, on the long edge you are extruding faster at higher speeds. When you get to the corner it slows down and over extrudes a bit as it approaches the corner, it then goes around two tight corners slowly and then as it accelerates out of the second corner for the long edge you get underextrusion because of this delay. That's why the angel wings have that underextrusion just on the corners. You can even tell that it was a counter clockwise movement on the right wing in the photograph at head height (underextrusion). The upper portion of that wing is probably showing a z-scar. All of this is fixed with a slower speed as you don't have such extreme pressure changes in the nozzle because you don't have as extreme speed changes (the short distance is at the same speed but the long distance is slower than before). This "half way through" has nothing to do with the height of the part that is being printed and has everything to do with the geometry or path that the print head goes in.

I don't remember that at all. That's a pretty amazing print speed for 200C. I had NO IDEA it was a makerbot. I was assuming all this time it was a reprap. Do you know how he got the antennas so good? With such a small area the nozzle never lets the plastic cool long enough - maybe he has a better slicer for antennas - cura prints 2 layers on one antenna, then 2 on the other. Instead of alternating. Saves time but you get more melted antennas. So Weiin - how did my robot photos compare to your sample from the rep2? You never answered that question. Please compare the pictures in post #6 above to the rep2 made robot in your hand. And realize that if my pictures don't fill the screen you aren't zooming enough - first click on a picture, then right click and select "view image" then click again to get full size. They look about the same to me - I don't see either being better than the other.

Wow! Nice paint job! I thought this was an 8 inch part until I looked at the other image that includes your fingers!

Ha! I don't buy it. I'm not convinced the UM1 can't do just as good if you print just as slow and have your retraction "tuned".

I personally think it's too early to say. I suspect the UM2 extruder drive isn't as powerful as the UM Original. Quality and speed of prints is going to be similar - identical maybe even. However... The UM2 is much more beautiful. Probably much more reliable - especially - well everything from electronics to clogs (um original more likely to clog). I suspect it can print hotter. It has slightly larger build volume It is quieter. It is easier to use (the leveling part - almost everything else is about the same but levelling is a pain so this is sooo much better on the UM2). Also extruder is easier. And you can print more materials than just PLA. UM Original is pretty much pla. But if you wait 6 months their (hopefuly/likely) upcoming heated bed for UM Original will fix that. UM 1 is much more hackable. And if it breaks it's less scary for most people to just fix it (not that it's easier but who cares if you have to add a wire tie or drill a hole in the UM Original).

I love printbl.

pc or mac? Is this for a class? Is this for a project? Will you only be doing console (text in and text out) stuff or gui (like buttons and graphics and text boxes and such) also?

Wow. Wow. Yeah it looks compatible. Usually these types of boards need minor modification (swap a cable around or cut a trace) to work with Ultimaker but in this case it appears to be compatible (don't know for sure). I hope they published their design and parts list and so on. If they didn't they may be in violation of the license.

I'm guessing the panel slipped at some point while it was being cut. I know one other person who got a bad panel where it doesn't align but I think this only happens to maybe 1%.

You can send gcodes by hand also if you learn them. There are descriptions here: great reference: http://reprap.org/wiki/G-code scroll way down to the gcodes on this page: https://github.com/ErikZalm/Marlin/

:-P Yes, it's very nice. Notice that when it connects it determines all kinds of things about your printer. For example the PID values of the nozzle and the max acceleration and so on.