gr5

Your post gives me all kinds of ideas. It's common to have under-extrusion during acceleration (start of a string) and over extrusion at the end. This is caused by the delay between when you ask for more filament and when it comes out (delay mostly caused by bowden tube and other stretchy/springy physical elements). This "error" is reduced if you print at slower speeds but should also be reduced if you greatly lower acceleration parameters. So I'm thinking that simply lowering the speed when you get near a bridge would help. Also lowering the XY acceleration for the section of print that involves a bridge might help. You want to lower it a LOT. I would lower it to about 1/10 of the current value for a test. I have many experiments on my list right now so I might never get to this one, but this is very intriguing!

Wow. First of all, your photos don't show #2. Did you do many more tests? Can you upload more pictures to the same album - put notes with each picture that describes what it is showing (if you have these pictures already). Oh! Daid please! Please add comments to show each individual bridge line! That way this guy can write a plugin to do some experiments to improve bridging string strength!

Great post - I rarely edit a post but I did yours - hope you don't mind. Let me know if you want me to revert to original. I did some bridging tests also recently but your zoom-in-photo is fascinating! It makes me think Cura should output a little extra flow for the first 2 millimeters maybe? Currently post #17 (but that will likely change within a month): http://umforum.ultimaker.com/index.php?/topic/1872-some-calibration-photographs/?p=25304

That is very good photography! I zoomed way in and could see quite good detail! 1) I'm pretty sure those lighter layers is white PLA. It's strange but sometimes when you change colors, the other color can stay in the print head for an hour before coming out. If you use higher temperature (240C-250C ) for a few minutes while changing filaments and pushing as much filament through as possible you might be able to get the other color out sooner. 2) How did you get the white ring inside the pink part? Was that printed separately and then assembled later? 3) How many line segments do you have for each circle? Wow. Very nice. I had a similar issue to you where I printed an orange part (pumpkin) and then later printed these 3 white people and at their waist (half way up) some of the orange PLA came loose and changed the color: Too zoom in, you need to do 3 things: First click the image, then right click it and choose "view image" then click it again. This is what I had to do to see your image full size.

a quick test? lol! There are known problems with Marlin reading the SD card on UM2. What version of firmware do you have (go to the advanced menus and it's in there under "version"). I think I have 13.10-5 but I"m not sure if that is the version with the fix or not. I'm relying on memory of Daid's posts without looking for it.

Whatever works!

The shipping to the US part was pretty fast. 3 days. Actual days. DHL doesn't seem to stop on Saturday or Sunday. It kept moving along as I could see on the tracker. It shipped at 3pm Netherlands time on Friday and arrived 3pm Boston time on Monday.

Yes I want to try PVA/wood glue and I want to try hair spray but most important was to try the glue stick that comes with the UM2 to help future UM2 users (and me). Glue stick cleans up very easy with water: wet sponge or wet rag. You also don't have to wait for glue stick to dry as far as I can tell. And glue stick is very fast to apply (don't need to unscrew a jar lid).

I keep notes on EVERY print. I have a notebook that sits in the bottom of the printer. In cura speed was 50mm/sec and I ran at 50% so: 25mm/sec and temp 195C. I was printing 3 people at once so I don't think I hit the "minimum layer time" but I'm not sure on that as it isn't calculated perfectly. When the printing got to the hips I noted excessive stringing which was surprising. So I lowered the temp for the upper half to 190C and slowed down to 40% (20mm/sec). It still only took probably 10 minutes to print. Not sure (don't usually write that down). Normally at 20mm/sec and 190C I get zero stringing with this particular PLA but I got lots of very thin strings. I think it has to do with the fact that it doesn't get much time to cool down. Fan was at 100%.

So... Now that I've seen Ian's photos and looked my UM2 over carefully - there is a system of locked pulleys, black spacers and pulleys against the walls of the inside of the UM2. 8 of them. NONE of the rods should be sliding around. If they are then either you are missing a black spacer, or more likely one of your set screws in one of your pulleys is loose. The one with the problem will actually always be the one closest to where the rod is poking *out*. So push the rod back in and slide the pulley against the spacer (if there is one) and then against the outer wall, and then tighten the pulley set screw.

I did a big experiment today using the UM2 and printing on glass. I learned a lot. I tried at room temp (21C) through 75C. I only researched how well parts stick to the bed. I did *not* research how temp affects warping/shrinking issues. I only printed PLA on glass with and without gluestick. I would print the bottom of the UM Robot (first 4 layers) and then remove it with a screwdriver pushing horizontally. I pushed the screwdriver with a scale that goes up to 30 pounds which was barely enough for one of the prints. Here are my conclusions: You can print on freshly cleaned glass no problem - you need the filament to stay hot long enough to fill in the gaps and make a good surface if you want it to stick. So for example 220C nozzle, 30C bed was not enough - 1/2 pound force was enough to remove. Wimpy. Yet 30C/240C nozzle took 14 pounds force. At 220C printing, the bed needs to be at least 45C to stick well (14 pounds). If you lower to 40C it only took 5 pounds force to remove. If you add gluestick you don't need any heat (but it helps a little). I printed at bed temp 21C (room temp) nozzle at 210C with glue stick and it took 19 pounds force to push the robot bottom off. With 45C bed, 220C nozzle, and gluestick it took 30 pounds force! Thats enough to put holes in the base of the robot print. The other thing I discovered is bed temps at 60C and higher were bad. The part was still soft and it didn't take much force to remove it and when you did it wrecked the part. 220C/60C no gluestick took 11 pounds but it destroyed the part. 220C/75C took only 1 pound of force and it folded the robot base almost in half as I pushed. Also if you use gluestick the amazing thing is you can soak in warm water for just a few seconds and pry it off slowly and the water disolves the gluestick nicely and the part comes off with only 1 pound of force even if otherwise would take 20-30 pounds. edit: I also printed on plain fresh blue tape and it took over 30 pounds force. I couldn't push hard enough and had to turn the screw driver sideways to pry it off (then it came off with 30 pounds force). With alcohol it sticks much more - probably over 100 pounds force but I would probably damage my bed. So here are my recommendations. For medium/reasonable stickiness print at 45C-50C bed temp and use no glue stick. If you want it to stick like crazy use 45C with gluestick but beware it is hard to get off without soaking in water. All of this ignores warping issues! I want to do warping/overhang tests at some point but I'm not yet sure how I will test this in an easy, repeatable, objective manner. Here are some of the results in photo form, lol: I cleaned the glass with windex and a rag before every print! Here's all the raw data: bedtemp/nozzle temp pounds to remove, notes 220C/50C 11 220C/65C 5 220C/75C 1 (part just kind of folded up - much too soft - see photo) 220C/50C .1 220C/50C 1 Above tests were with putty knife that sometimes slipped under part too easily so I kind of threw that data out. Remaining were with screwdriver and blade vertical 220C/50C 10 220C/40C 5 220C/60C 11 (but part was wrecked - see photo - part with notch) 220C/50C 14 220C/30C 0.5 (part had warping/lifting on corner before I started to remove) 240C/30C 14 (hotter filament helped a lot!) remaining are with gluestick 220C/45C 30 (bottom right in photo - part has 2 notches - first notch was only 20 pounds force) 240C/21C 16 210C/21C 19 (even at room temp - wow) 220C/45C 20 (mostly repeatable, yay!) 220C/45C 1 (soaked glass plate in water in sink for a few seconds while pushing/lifting lightly with fingernail)

You need to level to within about 25% of the first layer height. In cura you can set the first layer height thicker but if the first layer is .06mm you need the bed accurate to about .02mm. This is one fifth the width of a piece of paper. Also when the print head warms up it expands so you have to level with a hot print head, then after you get it perfect with paper, you have to twist the 3 knobs a little more to get it a little closer. How much to twist? I don't know yet. I guess one can figure it out with 2 pieces of paper to see how much to rotate those thumb screws to achieve one width of paper in height change. Your photo looks EXACTLY like what I would expect. In some places you are too close to the bed and the pressure builds and builds until suddenly it can't take it and you get lots of extra filament squirting out the side of the nozzle which causes blobs, or thicker areas. What I do with the UM Original is print an extra large skirt and twist the Z axis by hand (difficult to do on the UM2) fighting the stepper. This does not dammage the stepper motor. By the time it does the final skirt pass either I have levelling perfect or it's time to abort and start over.

Just to explain further - after it finishes one layer of the main part, it goes to the second part to print a layer there and gives the first part time to cool down.

You linked to a "dial indicator" and your link simultaneously links to the spring portion - some spiral part. I don't see any cylinders. But the biggest problems are usually cooling related. You might need to combine two tricks. The first trick is to lower the temp to 190C and make sure fan is on at 100% min and max and set layer time to at least 5 seconds. The second trick is to print 2 at the same time - set the gantry height in your machine configuration and print two. It sounds silly and wasteful but it makes a huge improvement for very small objects such as these people I printed recently:

That sounds crazy. You cancelled a print because it *might* fail? I removed one of the endcaps of my old UM Original and the rod sticks out but if I just ignore it, it never gets worse. It's in a happy place now. Some day I will look into it and see if I can adjust the pulleys (slide them along the rod and re-tighten) but for now it can print for months and months like this.

One more post and you can do anything! So just reply to this and say "replied" so I can approve it. On gallery page there is a blue button on far right that should say "upload". You can alternatively post a pic on the web - for example imgur. Then click the "image" icon where you are posting and then insert the URL when prompted. If you use the gallery, after you upload, start a new post and click on "my media" next to the smiley face to insert your picture into your post.

Oh - now that I think about it I'm not sure if you can use kisslicer with UM2. I really don't know much about UM2 yet. I'm more a UM Original expert. The "joris" setting is now called "spiralize" in Cura 13.11.

By the way these "outer dimension" errors are caused 90% due to how much PLA shrinks as it cools. Very little of the error comes from errors in the printer. Different PLA shrinks different amounts. PLA45 (from germany) shrinks much less and is much easier to print because of this and your cad model needs less adjusting. But it costs much more and only comes in one or two colors (black and white I think).

When one firsts prints something simple like a cube, cylinder or cube with a round hole, the tolerance is annoyingly off. Converting to mm, a 20mm by 5mm part will probably print out to about 19.5mm by 4.6mm. However this is easily fixed by increasing the part dimensions by .5 and .4mm respectively. After doing that you can easily get tolerances (repeatable tolerances) to about .1mm error. The z distances are usually spot on because the lower portion of the part is cool while it is printing the last few layers. Inner holes in the part (like a shaft or a cylindrical hole) are much worse. For many reasons they come out quite a bit smaller than expected and a 3mm hole will come out probably 2.4mm. If you change filament, change printing speed, tighten your belts, change printing temperature then you might have to readjust your settings by another .1mm if you need that kind of accuracy. If you will be printing lots of small parts (under an inch long) and you don't need to print quickly you can switch your nozzle to a .35 or .25mm nozzle and get finer detail. Your corners of your cubes will have tighter radius. You can't switch nozzles on the UM2. But on the UM1 you can get lots of other nozzle sizes for well under $5 per nozzle including shipping.

He has a fan aimed at the print head instead of at the print bed. For some people, especially with V1 print head design, heat slowly goes up the print head. Especially if you have lots of retraction that pulls the heat up allowing warmer filament to touch the inside walls. Then eventually it gets hot enough to soften the PLA and you get upper-print-head clogs. Usually the clogs happen only after a long time. For example after 5 hours of printing.

I love that thingiverse part. While you are printing it, or on your next print, remove the extruder from the back of the UM. Don't panic - it's much safer than riding a bicycle with no hands! Notice how much quieter the UM is. If it is still noisy then this part won't help you. For me removing the extruder from back panel makes a big difference.

I suspect all the confusion is over which relay pin is which. I think either you have it wrong, or the engineer had it wrong. I would triple check which relay pin is where and mark up your little proto board with pin numbers. Then check it again. Then check it again. It's confusing if you think the wiring diagram is top view but it is actually bottom view. Check that when no power is applied there is zero ohms resistance (or at least less than 1 ohm) between pin 30 and 87a. If this is not true, then you are confused about the relay pin numbering. You could additionally check the coil resistance - it should be something on the order of 10 ohms to 1000 ohms. And with power off ping 30 to 87 should be an open (very high ohms) although your other electronics may make it look like as low as 3 ohms. But certainly it should be more than 1 ohm resistance with everything powered off. I can certainly say with confidence that I have no idea where for example pin 30 is on that thing.

Oh! And please post your handle! We'd love to see it. Maybe post a photo of the cad drawing in Cura.

You can either set shell thickness to 200mm or you can set fill density to 100. Both give very different results but in both cases you get 100% solid part. Solid parts take MUCH longer to print, are not much stronger, and tend to warp much more (shrinking plastic pulls). Your part may lift off the bed if it is solid (while you print) or you may experience other troubles. Definitely wipe your blue tape with isopropyl alcohol to clean off the was that comes with the blue tape. This will increase your sticking power by around 5X. Consider setting shell thickness to say 5mm instead of completely solid. If this handle needs to lift 200 pounds, maybe 10mm shell? The infill isn't for strength as much as general support to hold it together until the print is done (people will disagree with me on this!). I've printed many parts with 3 passes on the shell (1.2mm shell) and zero infill.

Very beautiful. First I would seriously consider a different slicer. Vases are special cases. Try kisslicer maybe. If you want to stick with Cura, look at the part in slice view and try to find these strings. They will show up as blue lines (blue means non-extruding). Check to see if there is a vertical movement shown in slice view - you need the latest cura 13.11 to see this for a UM2 print as there is a bug in Cura 13.10 where it doesn't show the vertical blue line on retraction for UM2 in slice view (even though retraction is there). If there is no vertical line it is likely (possible) that Cura is not retracting. But if it is then the usual ways to fix stringing are available to you. Basically you can reduce stringing by lowering the temperature to as low as you can go. I would do a stringing test first. But if I want to eliminate stringing typically I just go to 190C and print slower - like 20mm/sec. This may not be an option for you as your Vase might take days to print - not sure. You might want to buy a UPS (uninterruptible power supply) before you start a multi-day print. Here's how I did my tests on stringing (first photo): http://umforum.ultimaker.com/index.php?/topic/1872-some-calibration-photographs/