neotko

You are forgetting an important factor when pointing out the slicer. Cura has control over acceleration and jerk, check a print done with one slicer and other doesn’t prove a thing. Speed vs temperature is the reason of that visual error. I was giving examples of layer heights as a guide, not a recipe. Cura is more focus-based on click-print recipes, but every object has his needs. Just print 3 benchys at the same time and see how at least 1 is perfect, and possibly 1 is meh and 1 is bad

This an issue of any fdm. At 0.15-0.2 layers the time at that point shortens and it’s a small loop followed by different size infill. It’s easy to understand if you check the gcode preview what happens there. Since there’s no slicer out there that can compensate time va heat va flow, the temperature for that 2 layers changes making the print show a visual defect. Its the same effect that happens when a small island is detached from the model, as soon time and head stays more time on an area the difference from flow/temperature makes a visual defect. The solution is to print everything slower/cooler, that’s why this doesn’t show as much on 0.06-0.1 Also some materials show more this than others just because at different mm3 the filament stays ‘hot’ more time, shifting the color. So as soon the filament prints with delay (at that height suddenly it has 4 jerk slowdowns (max speed to change direction) the filament stays more time on the hotend, so is actually hotter than the filament of the lower layers that does suffer less jerks because the nominal extrusion speed is easier to achieve). Chechk your gcode with a good analyzer like https://www.gcodeanalyser.com/ and you will see a big slowdown on the affected area

I did at that time of test, also got a x/y shift months after. In the end like I say on the other post I just moved out to duet3d and been enjoying the quality of life of a proper 32bit cpu with trinamic steppers. Happy printer, happy life

Looks nice and clean, my only 'but' is that the heat craw will go up affecting the small boards and the printed part. The smothers easily get to 60-70C so you better print that on something else than pla or that enclosure will deform just like the plastic covers they sell with them. The other issue of the smoothers is the torque is reduced, so if you get a x/y slipping change the travel speed for non printing moves. I also observed a lot of torque lost on the extruder, so check that if you print at the limit or raise the hotend temperature to compensate the torque. For reference

Also @hugoboc check duet3d forum, lots of users there, very experienced on mods and hacks

I did ‘cut’ the psu to the board cutting the plug, I remember someone making a convert cable from um psu to plug, but can’t find it. They have been working absolutely fantastic on my 3 printers, would never to back to atmega or no wifi boards. I have them plug into a switch with buttons to turn of each printer cutting the power, but frankly I only turn them off in vacation and since you can upload the gcodes by wifi and do everything like reset, etc, I never actually turn them off. The config.g needs some work, and I would recommend to move the Z endstop to the top (like umo+) to save time, but duet3d you can plug a bed z sensor and do a mesh calibration if you want (I don’t do that). Overall the biggest difference is that you can rebuild everything (temp, sensors, fans, etc) by editing a gcode file, no need to reflash or prehistoric stuff.

Impressive extra options. I wish someone fix your mac, great work

That's not 'um3' magic, that's just another program But, you could do that on a Duet3D, more rustic, but totally doable since um3 doesn't pop the print out and like in 99% of the printers you need to go there, remove the print and go to your station. As 'magic' I mean the heat/cold sequence to print, the amps control to manage the weak PSU. Ofc if you mean as 'magic' the start-end sequence that the Core need to avoid clogging hard, ok, that's the linux firmware pythons.

Sorry I love hour idea, but somehow isn’t easier to put a duet3d board, a better PSU (the only real work the um3 linux board does is to control the amps use of each head). 99% of the magic of um3 is all cura. You could also use mark2 firmware (that does that 1% magic cura can’t) and have a um2.5 working with a bit of effort but less money. As for um3 firmware they never released the source, (github style) but you can download the restore boot image and current firmware if you google a bit inside the forum and their website. Anyhow, good luck and keep modding!

Fun fact. I found out about ironing (neosanding) year before writing about it, because forcing the second pass does indeed strength the parts. The way I use it is to print 2 parts that actually hit and reprint on the same area, then neosand it. I never published it because everyone would ask for ‘scientific proof’ and I don’t have the time to do the necessary tests. This being said, my proof is that 2 delicate items stopped snapping on poat office after I realized this 🙂 Slighly better explained on my twitter

Ahh I get it now! Ty!

Why not combing on infill? From precision point of view, it should be desired to avoid dripping on areas where the object should not have material. It's a recommendation for the printer model to avoid grinding or just is from the theory that 'dripping where you don't see doesn't matter' ? Because that's a very subjective matter.

https://www.thingiverse.com/thing:3456261

Oh forgot. Sent me a PM on Twitter so I don’t forget. I have gudo step files of the old bed stabilizer and my simple bed skis, but mind that I can use it without losing Z because I use the gudocorexy and it gives 9cm on Z that I use to place the bed flanged bearings below (making a sweet triangle to distribute the weight).

Afaik that can be done pressing Shift while rotating?

Mi experiencia personal es que o es el calentador que está casi muerto (prueba calentar a 220 sin imprimir nada y observa si pasa) o que la PSU de alimentación está fallando (más difícil de debuggear, pero puedes desenchufarla completamente y esperar 12h mínimo hasta que se descargue completamente la reserva de los capacitadores).

@gr5 might have the good ones (TFM) on usa

Just in case you ever notice weird underextrusion that you can’t debug (specially if you don’t use geared or bondtech motor feeder) I would suggest everyone to set the extruder on duet3d to disable the interpolation on the extruder To do so change the config.g extruder config to: M350 E16 I0 ; Disable interpolation on extruderM350 X16 Y16 Z16 I1 ; Configure microstepping with interpolation for XYZ This way by splitting in two lines you can set I0 for no interpolation for the selected feeders and enable it for the rest. I had some issues with bad filament that comes very tight on the end of the spool and removing the interpolation I could regain enough torque to remove the issue.

This is the stuff I'm finishing for 2019, they are hairpins in two colors that I do with an advanced Pause script that allows me to Atomic plug the filament totally unmanned and waits for me until I go to change the color and click resume. Ofc this is all thanks to Duet3D advanced pause that has Gcode to resume previous position and even Fan, hotend, etc temperatures. What's needed to make a plugging out of this? - Manual setting to define layer number (I know this already exists) - Script to insert the Previous temperature/FanSpeed & Position set by the slicer (so any changes while paused can be resumed if the user needs to move Z, move X/Y or prehead the nozzle to clean any previous color residue. This is important to resume the print without temporal changes affecting the real print. - Area to define custom gcode - for the Atomic - so the users can adjust the atomic pull sequence for their extruders. Mind that I use DirectDrive so I can change the color pretty fast without issues because also my extruder isn't behind the machine or hard to access. Mind that the M106 R1 or G4 S5 are specific Duet3D gcodes, so this can't be just copy-paste into a normal gcode print without manual editing. That's why ideally Cura could have a plugging to mimic all the automatic steps in order to make it user friendly. Check my twitter posts to see videos of how the process works: This is the post I made on Duet3D forums: For this, first I had to find the perfect momento to pause on the gcode. I use Skirt with 12 layers (the same amount of layers for the Pause layer) with 5mm distance and 2 Skirt Outlines (but I suppose for really tall prints it might be better to use 3). Then I do a search on the gcode (mind I use S3D) of the last "; skirt" Then I find something like this: ; skirt G1 X98.056 Y59.801 F10800 G1 Z2.480 F1500 G1 E0.0000 F1500 G92 E0 G1 X99.094 Y59.554 E0.0355 F4200 G1 X99.106 Y59.551 E0.0359 Being that last G1 the beginning of the Skirt Print So I ADD the M25 to do a Pause ; skirt G1 X98.056 Y59.801 F10800 G1 Z2.480 F1500 G1 E0.0000 F1500 G92 E0 G1 X99.094 Y59.554 E0.0355 F4200 M25 ; Pause REPEAT Speed of previous line on next so it knows what speed to resume G1 X99.106 Y59.551 E0.0359 F4200 ; ADDED F4200 HERE Doing this I can be sure that the extruder was on top of the Skirt before doing the pause, so if the filament Strings it won't dirt my print. Then the rest it's on the Duet3D macros. I edited my pause.g to work like this: ; pause.g ; called when a print from SD card is paused M83 ; relative extruder moves G1 E-3 F3600 ; retract 10mm of filament G91 ; relative positioning G1 Z5 F360 ; lift Z by 5mm G90 ; absolute positioning G1 X15 Y15 F10000 ; go to X=15 Y=15 M104 S120 T0 ; Sets to a generic LOW TEMP before calling PLAColdpull Macro G1 E5 F1600 ; Keeps pressure FAST Initial recovers from retract also M98 Preversecowmilking.g ; Calls the COLDPULL macro This is what I do on Resume.g ; resume.g ; called before a print from SD card is resumed M104 S225 T0 ; Sets general Print temp for PLA M109 S225 T0 ; Waits until it reaches 225 for PLA M83 ; relative extruder moves G1 E5 F100 ; Purgues 5mm where it was paused slow to realign gears G1 E5 F300 ; Purgues 5mm where it was paused normal G1 E-1 F2000 ; retracts 1mm before moving to resumed position G1 R1 Z1 F6000 ; go to 5mm above position of the last print move G1 R1 ; go back to the last print move M83 ; relative extruder moves And This is my ReverseCowMilking Macro (automated Coldpull) I add this macro on the SYSTEM area, otherwise I wasn't able to call it but I'm using an old 1.20RC3 version of the firmware so probably just an old bug ;reversecowmilking.g ; Semi automated coldpull for PLA M106 S200 ; SET FANS UP G10 P0 R70 S70 G4 S35 ; Wait 35 secs so the temperature reaches 160-170 if was a 225 M302 P1 ; ALLOW COLD EXTRUSIONS just in case G1 E2 F15 ; Push 2mm slow to keep pressure at semi able temperature M116 ; WAIT FOR TEMPERATURES SET TO ARRIVE before continuing M106 S0 ; SET FANS OFF G4 S5 ; wait for internal temperature to stabilize along filament ; starts milking the cow movement G1 E-0.3 F100 ; retracts 0.3mm G1 E0.1 F2000 ; reajusts 0.1mm - total 0.2 M18 E0 ; Unlocks feeder G4 P500 ; half a sec to allow filament tension to recover & avoid grinding G1 E-0.3 F100 ; retracts 0.3mm G1 E0.1 F2000 ; reajusts 0.1mm - total 0.4 M18 E0 ; Unlocks feeder G4 P500 ; half a sec to allow filament tension to recover & avoid grinding G1 E-0.3 F100 ; retracts 0.3mm G1 E0.1 F2000 ; reajusts 0.1mm - total 0.6 M18 E0 ; Unlocks feeder G4 P500 ; half a sec to allow filament tension to recover & avoid grinding G1 E-0.3 F100 ; retracts 0.3mm G1 E0.1 F2000 ; reajusts 0.1mm - total 0.8 - Enough to dislock filament from hotend M18 E0 ; Unlocks feeder G4 P500 ; half a sec to allow filament tension to recover & avoid grinding G1 E-10 F50 ; retracts 10mm SLOW - To avoid breaking if bond if weak M18 E0 ; Unlocks feeder G4 P500 ; half a sec to allow filament tension to recover G1 E-20 F800 ; retracts 20mm medium speed - Should keep integrity M18 E0 ; Unlocks feeder G4 P500 ; half a sec to allow filament tension to recover G1 E-30 F2000 ; retracts 30mm fast - Should be able M18 E0 ; Unlocks feeder G4 P500 ; half a sec to allow filament tension to recover ;G1 E-30 F2000 ; retracts 30mm ;M18 E0 ; Unlocks feeder ;G4 P500 ; half a sec to allow filament tension to recover M302 P0 ; DONT ALLOW COLD EXTRUSIONS M106 R1 ; RESTORE FAN SPEED BEFORE THE PAUSE NOTES I use directdrive with zge extruder, so the force, speed of the coldpull and timing should be totally different for bowden. I release the feeder constantly to allow the filament natural elasticity to go back so I can avoid grinding the filament I use Bondtech drivegears, so the filament grip it's perfect at all times. Macro comments might not be accurate because I been adjusting this many times until I was able to achieve enough repeatability on my 3 printers. Happy 2019 & enjoy printing

So cool to see what an architect can achieve knowing how to print the parts. Great job!

@korneel I write what I think. And if you want to tak twitter, go twitter. Hugs & kisses

Then send emails to every customer that has one inviting them to join this post and give you feedback. Also, stop packaging new UM S5 with a second glass and wait for a decision and make it public not just on a forum post. Not gonna happen right? Prove me wrong please, show the face and stop saying a second glass is enough compared to a precision aluminum plate, that for example one from misumi of 7mm thickness and top flatness goes for 100-120€ minimum distributor price. So?

Does that means that ONLY who complains gets something better than a glass? I really think UM should do better and release a proper release note instead of hiding this on the forum.

Just buy a neoceram glass plate, cut to size and cheaper than UM parts. Also, from the point of view of giving 1 extra glass, it’s a bit of shameful. UM is aiming at pro market and can’t make their promises. I would understand that from e3d and prusa, but UM thinks of itself as ‘pro’ market. Ofc the buyers could make a legal use of the written promises and ask for more money than just a glass price. Just like what happens to any mayor market seller, if they fail they get sued, clear and simple. Ofc that didn’t happen when was the um2 and the promised second extruder upgrade because that wasn’t Pro market, but now? From the point of view of finding a good replacement I would buy a neoceram glass or vitroceramic glass, no expansion and very sturdy (many shops around the world sell cut to size hightemp glasses). Ofc the issue would be the bed sensor probably? Dunno.

@Rastamonk77 and anyone else interested, I did upload a full step file assembled where you can pick anypart needed (I hope) 😄 https://www.youmagine.com/designs/neotko-fatiroberti-bondtech-feeder-for-um2-um3 File is: Um2plusbondtech Full175 & Um2plusbondtech Full285 - In Step format Happy editing! Cheers!