lars86

Absolutely. As far as I can see, you would just need to change the steps value for Z to match the new lead screw and install/adjust the new limit switch.

So, you stuck with it? The modification and adapter were just means to mount it? Did you notice any difference in print quality?

Some questions: Has anyone used an anti-backlash nut in general or specifically on this new Z screw? http://www.robotdigg.com/product/18/Anti-backlash-Nut-for-Tr8*8-Leadscrew I know that the latest Cura has a compatible firmware build, but was wondering if using this builder would be equivalent: http://marlinbuilder.robotfuzz.com/ I have a few other tweaks; custom extruder drive, direct drive XY, and a bunch of custom parameters for accel, jerk, etc. Is there a way to make a dump of my current firmware as a backup?

Hey guys, Beyond the announcement thread,I didn't find a thread that had some working information on upgrading an original Ultimaker with the official heated bed upgrade kit. I just began assembling mine tonight. First thoughts are that this is a well sorted kit! When I first saw the kit's price, I was surprised, but after being hands on with it, feel that they put a lot of value in. It is thoughtful in design, and in instructions. It improves all the shortcomings I found with the original bed: poor stability of the bed platform and the acrylic, 4 point leveling adjustment, lack of heat, ugly. The 3 point leveling system on this new unit is very clean. The support springs are much beefier. Flange mounted linear bearings gives you the ability to let the bearings be right where they need to be for your specific guide rod spacing (assuming they are parallel). If the platform assembly fully defines the bearing location, tolerance stacking will likely lead to some degree of over-constraint of the system, which induces bind and therefore extra friction/wear/power consumption/heat. It looks great. I gave the aluminum platform plate a quick wet sand on my granite to check flatness. All in all it was pretty good, not amazing. Give it a close inspection, paying particular attention to the linear bearing flange area. Since that is defining both bearing axes, any ding or burr could hold a bearing off angle. I'm not sure why they picked flanged bearings with big counterbores on the backside of the flange. This really reduces the surface area immediately under the four machine screws. I gave the bearing flanges a quick sanding too, and they were definitely dinged and a bit out of flat. Definitely inspect yours before installing. After a little light lubrication, I gave the new bearings a test fitting individually on the 12mm shafts. I found that while moving the bearings upwards, then reversing to move down, both hang up on the shaft. Enough so, so they can support their own weight without falling. I tried flushing one out with alcohol to make sure it was free of contaminants, then hit it with compressed air and re-lubed, but the binding is the same. Did I get bad bearings or are they cheap? Having a very smooth, reliable downward Z movement is critical for print quality. I'll add more tomorrow... Lars

Cool, thanks. How did the belt length work out for you? Are you using tensioners? What kind?

Bump!

Very nice! The motion definitely looks cleaner. I have upgraded pulleys on the original belts, and am considering a GT2 upgrade. My heated bed upgrade just arrived, so I'm excited to get that rolling as well. Where did you get the parts? What sized belts and how many teeth on the pulleys?

I run 4 and 5 axis Okuma mill-turn machines, and they will wait for a line of code to finish before executing the next. So: G1 X4 Z20 Y3 C60 ...is 4 distinct moves. While: G1 X4 Z20 Y3 C60 ...is a simultaneous move of all 4 axes. I'm curious if Marlin behaves this way. It's really the extruder that throws a wrench in it for me. It would make sense to move the extruder's retract/advance move concurrently with the lift/return move, but the feed rates need to be different. I was thinking along these lines, if Marlin could parse independant feed rates on the same line: G1 X63 Y54 E464G0 F12000 X64 Y54G1 F2400 E462 F1800 Z22.700G0 F12000 X61 Y50;TYPE:WALL-INNERG1 F1800 Z22.600 F2400 E464G1 X60 Y50 E464 Or even like this, if it would accept rapid and feed moves on the same line: G1 X63 Y54 E464G0 F12000 X64 Y54G1 F2400 E462 G0 Z22.700G0 F12000 X61 Y50;TYPE:WALL-INNERG0 Z22.600 G1 F2400 E464G1 X60 Y50 E464 @Daid, can you shed any light on this?

Does anybody have experience with these guys? http://www.citizenscientificworkshop.com/ They claim +/- 0.02mm on the diameter using Natureworks Ingeo 4043D PLA. Always love to hear about good quality US based filament distributers!

Here is a segment of code during a layer change: ;TYPE:WALL-OUTERG1 F1800 X59.382 Y50.881 E364.96438G1 X58.957 Y50.811 E364.96977G1 X58.802 Y50.780 E364.97175G1 X58.802 Y49.623 E364.98622...G1 X61.015 Y51.501 E365.10788G1 X60.622 Y51.295 E365.11343G1 X60.163 Y51.106 E365.11964G0 F12000 X60.441 Y50.148;LAYER:79G1 F2400 E363.61964G1 Z15.900G0 F12000 X52.583 Y52.829 Z16.000;TYPE:FILLG1 Z16.000G1 F2400 E365.11964G1 X51.619 Y53.793 E365.13669G0 F12000 X52.257 Y54.192;TYPE:WALL-INNERG1 F2400 X52.109 Y54.699 E365.14330G1 X52.105 Y54.722 E365.14359G1 X51.135 Y54.722 E365.155737 I'm curious why the initial head lift is a feed move executed at the last deposition feed rate. Maybe to make sure a dimple isn't left? Here is a same layer, lifted rapid move: G1 X63.637 Y55.520 E464.05874G1 X63.616 Y55.515 E464.05901G1 X63.456 Y54.755 E464.06872G0 F12000 X64.206 Y54.478G1 F2400 E462.56872G1 Z22.700G0 F12000 X61.311 Y50.839;TYPE:WALL-INNERG1 Z22.600G1 F2400 E464.06872G1 X60.858 Y50.609 E464.07508G1 X60.262 Y50.378 E464.08307G1 X59.785 Y50.245 E464.08927 Can concurrent feed moves for different axes be given on one line, with different feed rates specified? Do feed rates stay modal to each individual axis?

Hi guys, I'm noticing that enabling Z hop (0.1mm) seems to also introduce a bit of a dwell before and after. Combine with the fact that exterior perimeters are done last, this leaves blemishes on the surface. My Z movement generally seems well tuned and snappy: Jerk: 0.1 Vmax: 30 A: 200 Here is slow motion of what I'm seeing: Is this a motion tuning issue, or is it possible that different Gcode strategy could improve this?

I'm currently running these off a separate inverter I had laying around. I'd like to pop a switch into the side of the case and pull power from the UM electronics. If anyone has a suggestion of where would be best to draw power, let me know!

Hey guys, Hope these are of some help in keeping your axes nicely squared. I'm uploading a Youtube video to help show their use. https://www.youmagine.com/designs/ultimaker-xy-axis-square

Hi guys, I have been running these little LED light strips on my Ultimaker for a while. I had them lazily taped onto my frame, but finally whipped up a nice little mounting channel yesterday and thought I would share. LED lights themselves: http://www.amazon.com/gp/product/B0056Z036A/ref=oh_aui_search_detailpage?ie=UTF8&psc=1 Basically, you just take two of them stacked side by side and slide them into this channel, then bolt it onto the UM frame. I have one on the front and one on the right. This really helps with dialing a print in, as the bright, low-angle light helps show over/under-extrusion. These LEDs are pretty cheap, and I have one dead bulb on 2/4 strands, but they have held there for quite a while now. Plus, you never really look straight at them, so it isn't obvious. I had to do a little trimming on the LED strands to have them fit better. Near the wire end of the strands, the shape of the rubber flairs out a bit. You can just take a utility blade and carefully trim that area to keep roughly the same cross section as the rest of the strand. It won't be visible once they are in the channel. https://www.youmagine.com/designs/led-light-bar-channel

Hey guys, I ordered this upgrade for my UMO and can't wait to put it in. There was a bit of sticker shock at first, but especially after reviewing the Github parts list and manual, it looks like a well polished kit. I was already planning to rebuild the Z stage for stability, so two birds with one stone! I just wanted to see how the upgrade was working for everybody, since it's been out for a bit and ask a few questions. I know that the latest Cura has a compatible firmware build, but was wondering if using this builder would be equivalent: http://marlinbuilder.robotfuzz.com/ I have a few other tweaks; custom extruder drive, direct drive XY, and a bunch of custom parameters for accel, jerk, etc. Is there a way to make a dump of my current firmware as a backup?

You bring up some good points. It would definitely be an advanced method, where the user would be responsible for sizing the pattern appropriately. Maybe there could be a checkbox that allowed any cavity equal to or smaller than the grid spacing, to be filled solid. I like the triangular infill pattern idea, and can't wait to try it out.

Swordriff, you are close, but a little off. For the same diameter bar, solid is both stiffer and stronger than a hollow bar. For the same weight bar, a hollow will be stiffer than solid because the mass is distributed into areas where the material properties can stiffen the bar. For a given angular deflection, a greater material displacement is required, the further you get from the centerline. So, material very near center doesn't resist bending efficiently.

I do. It's not something I've tested, but call it an engineer's hunch. I think that the cumulative strength of a thicker rib, can be higher than individual strands added up.

Hi guys, I'm wondering if a different style of sparse infill exists in a slicer. I believe that KISSLicer simply does every other line for 50% infill, every third line for 33%, every 4th for 25% etc; and just over extrudes to "interpolate" for values that fall in between. I'm not sure how Cura handles it. What I think would be very useful is an option like this: You could specify the thickness of the members 't' like a shell thickness (increments of nozzle width), and the spacing between members 'S', independently. The single pass infill normally used is only so strong, but I feel that this would allow for very strong internal ribbing when tuned well for a given part. Does this exist?

Bump!

Hey Jonny, how did this turn out?

I appreciate the help, but I wan't able to get it working. I only wanted a single pass ( preferably not spiralized). Even after enabling "only follow mesh surface" (which also kills your brim) and/or "spiralize", it wants to make 2 perimeters, which badly violates the model. I got what I was after with Slic3r, but wish I could rely on Cura.

I'm having trouble with wall thickness management as well. I tried to print a calibration part which is a 0.4mm thick shell, that is 1" x 1" on the exterior. The STL seems clean, but Cura refuses to generate a path. Even lying to it and entering a 0.3mm nozzle width. I get the same result. Thoughts?

I would start by seeing what the minimum temp is for good deposition with the specific filament you are using. Heat build up is your enemy with such a small print. Use a thin layer height. Maybe .1mm In expert settings, enable cool head lift. Run the cooling fan at full speed. You may even want to put a table fan blowing through the printer for additional cooling.