randyinla

I didn't see anything in the Cura github issues list that matches this issue. I've noticed this about 4 or 5 times in the last few weeks since upgrading to Cura 4.0. (Side note: I had to turn off auto-slicing due to Cura crashing 5 times a day and it hasn't crashed since) When I'm finished slicing and want to save onto the USB drive, I: Insert the USB drive for my U3Ext into my 2015 MBP running the latest Mojave Get a popup from VMWare Fusion (I run Windows-based Solidworks in a VM) if I want to mount the USB drive in Mac or Windows, I choose Mac Wait for the drive to mount and for the, "Save to external drive" message on the Cura Save button and click it I'll go to Finder and open the external drive to move the file into a folder and notice 5 or 6 Cura log files and some other file I delete those files and eject the USB drive and put it back into the U3Ext These log files should never ever be saved to the external USB drive, correct? They should always get saved to my computer's designated Cura folder, right? Is Cura sometimes using the last used folder instead of using the Cura folder on the computer? The next time this happens, I'll add the file names here and maybe copy/paste the contents if they are small enough.

So many people take this approach to print out small things or tall narrow things (the robot, for example). Seems it would be nice to have an option to calculate a layer's area and have the hot end move off of the print for a few seconds to allow the layer to sufficiently cool. I realize there could be drippage, so not sure if it's a viable solution. The goal would be to gain the advantage of printing multiple things at the same time to force layer cooling without the wasting of filament when printing two of something. A wipe tower where the hotend spends much less time on than a prime tower. Just a simple wipe before moving back over the print. I was attempting a two-color print yesterday on my U3Ext and tried it without a prime tower. The first color dripped a few times but the second color dripped every single time it was pulled up and out of the way. Both were PLA and both had a standby temp of 140c. I had an AA core in both sides, so it wasn't that a BB core was the culprit of the drippage. As for line width... it's been my experience that setting it slightly thinner than my actual nozzle draws out lines a little closer together. The extruding seems to be the same and I get lines that are squished more into one another horizontally. I understand the documentation says that the extrusion is adjusted when the "line width" parameter is adjusted and I've seen people quote gcode showing slightly less or more extrusion. However, in looking at what my printer spits out (I typically print .1mm layers), a line width setting of say 0.385mm always prints lines that touch each other and leaves no gaps. Setting line width to .4mm or above (many 3d slicing articles & threads on the Internet suggest always setting "line width" wider than the nozzle, some suggest as high as 120% of the physical nozzle width) will usually leave gaps between my printed lines. If my part needs to be dimensionally accurate, I'll measure the printed piece and possibly adjust the horizontal expansion and print again until it's exact (though material shrinkage plays a huge part in this as every filament can shrink more or less). But mostly, I just reduce the line width a bit and end up with perfect prints.

My cura 4.0 kept crashing on me. I would load a model and auto-slicing would fire off, sometimes crashing. I would other times load a 2nd or 3rd model onto the same build plate and maybe 60% of the time, it would crash. Since disabling auto-slice just over a week ago, I haven't had a crash since. I, too, noticed that the Slice button remained active after clicking. I am conscious of when I click it now and wait a few seconds before doing anything else.

Am currently at work but I left my U3Ext printing something. Do I have to set anything up from Cura at home or from my U3? I can see a device listed in mycloud, but it's listed as "ultimakersystem-ccbdd3001d5a". The device is "IP Null" and doesn't show anything printing in the current group and nothing queued. I'm still running Cura 4.0 at home and it is currently running on my home laptop.

Thoughts/experience of the A4988 solder hack: I recently performed the A4988 hack of shorting out a single resistor on each A4988 chip. I had one cold solder joint on the 2nd extruder that had to be fixed, but aside from that, my zebra stripes have diminished about 90% of what they were. Am very happy with the hack. However, and I don't know if these things were happening before the hack or not, but I noticed my X stepper is about 2 to 3 times louder than the Y stepper. If it were the Y, I'd suspect it was active leveling, but it's the X where active leveling shouldn't be moving the Z at all. ie.when creating a cube and it's printing the walls, both steppers move the same speed and have the same pitch but X is noticeably louder and can be felt by touching the outside of the U3 or the gantry. My wall finishes when printing a cube are different. Front/Back flat surfaces are glossy smooth with some vertical lines (can see them, not feel them) and the Left/Right sides have a sort of pleasant noise pattern, like a tv station tuned to a non-existant channel, looking like sandpaper but still smooth to the touch (love the pattern as it masks any lines and just looks good). Again, can see that, not feel it. When I print the cube at 45 degrees, all sides have a similar noisy pattern that I like a lot. If you don't want to do the solder hack, the TLSmoother boards seems to result in a similar outcome without having to take an iron to the motherboard. You connect them between the motherboard and the steppers and can remove them at any time. For now, you can try rotating your model on the print bed to see if the lines go away. I forget the rotation %. It's mentioned a few times in the main zebra line thread. Between 15% & 45%, I believe. If you do attempt the solder hack, I would suggest desoldering the SMD resistors & cleaning the two pads of the old solder. This will make it much easier to solder a small wire across the two pads. FYI: If the resistor is gone and the two pads are not shorted, the stepper driver still works but it doesn't send enough voltage to actually turn the stepper motors. You can barely hear the stepper receiving the steps, but it won't turn.

Thank you, Phillip for the wonderful nozzles (AA.25, AA.4, AA.8, BB.4) & heating blocks! I was surprised to see Christmas come so early in the year 😃

Fixed it. After taking off the cover and putting on my jeweler's magnifier goggles, I could see from the side that the wire I soldered onto the pads wasn't touching one of the pads. Must have desoldered the SMD resistor as well, I guess. Removed the resistor completely, cleaned the pads and used a different tiny piece of solid wire and made a proper connection. 2nd extruder working fine 🙂 WHEW! 😅

Found the 1548-I Ultimainboard schematic for the U3Ext in the Ultimaker github repo and confirmed that U13 with R103 is the stepper driver for the 2nd extruder.

I performed the stepper driver hack (linked below) to help reduce zebra stripes in my prints the other day and all seemed fine at the time. This morning, I was trying to load filament for the first time since the hack into the 2nd printhead of my U3Ext and the extruder stepper wouldn't turn. If I lean in close to the extruder, I can faintly, barely hear the steps being sent to the stepper. It climbs in pitch and back down again, just as if the extruder were loading the filament but it doesn't spin. Seems the stepper commands are working but the voltage is so low, nothing happens? Is the stepper driver under R103 the one for extruder #2? That was the last short I made over the resistors. Didn't take any longer than the other 4. Looking at the photos I took after, it doesn't appear to be shorting anything else. The SMD resistor didn't move etc. I'll have to revisit it to check if maybe I shorted one of the pads to another. R102 is the one where the SMD resistor fell off. So if any of the steppers was misbehaving, it would kind of make sense it would be that one, though shorting a resistor is the same as shorting two pads with no resistor. Sigh 🤨

When opening up two separate .stl files that are meant to be a single 2-color print, they come in at different locations on the print bed. - Select one and set which printhead will print it - Select the other and set the other printhead to print it - Select both at the same time, right click on one of them and choose to merge them. Cura will slide them perfectly on top of one another and remember which printhead should print which part Edit: I've also noticed that the merged model will sometimes have the Z location changed. Either -Z or put slightly above the build plate. Always check Z position before slicing/printing. You shouldn't ever have to manually move the two prints around, nor would you ever really be able to get them to line up correctly. If you don't have a multi-color capable printer, then this does you no good. It won't pause the print to let you change filament. It's specifically for printers that already support 2-color printing, like the Ultimaker 3 or S5. Hope this clears things up a bit.

Thanks, Smithy.

That’s very nice of you, Philip! If you have paypal or Venmo, I will gladly pay for the nozzle & shipping! Do you happen to also have a BB .04mm? I’d love to have one as a backup.

Are the nozzles on the 3DSolex possibly the same thread size as the Olsen Block nozzles? I'm probably going to buy a 3DSolex hardcore but hope to reuse the OB nozzles I already have.

Yeah. Makes sense.

I switched material at the same time I switched the nozzles and went from too much stringing to almost none and thought it was the material swap & temp change that did it. I was thinking it would ooze more. That makes sense that it could. Thanks for the info. I still have the 3DSolex page open in a tab from Gr5's store. Sure looks sexy! Guess I'll have to dish out the cash. Part of me wants to get another original AA core just so I have a 100% matching pair without possible drippage, as well as a 3DSolex... or two.

I just did the solder solution for my a4988 stepper drivers on my U3Ext. I performed the hack on all 5 stepper drivers (x,y,z, extruder 1 & 2) because the issue is with the 2 missing/softer steps on all steppers, not just the ones that move the gantry/bed. How much those missing steps can affect the actual pressure through the nozzle is up for debate (may not affect it at all), but I figured might as well fix all of the similar devices instead of just x,y,z. Seeing the heat sinks on the smoothers makes me wonder... Do the original stepper drivers also get hot with either the addition of smoothers or the "solder over the resistors" hack I did? I don't want to possibly butn them out because I didn't know.

I experienced a bad print with old, bad PLA. The print had lots of retractions and the filament broke just before going into the extruder. By the time I got home, the melted filament in the nozzle had burned and crystallized under 220c for over 8 hours, clogging the nozzle quite well, indeed. Then I did something stupid. I couldn't clean the nozzle no matter what I tried. Warm pull, cold pull, removed the nozzle and tried scraping. I then thought it would be a good idea to use a drill bit to drill partially down into the hardened plastic and pull it out, like a stripped screw. I should have used my lathe with a vice to control the depth better, because as soon as the slowly spinning bit met the plastic, it got pulled too deep into the nozzle. When I pulled the bit out, I had some plastic ( yay!) but also saw brass (no yay!). Going to my stash of Olsen Block nozzles used reliably on my U2, I realized for the first time that the nozzle used in the new U3 AA/BB print head assemblies has longer & different threads 😞 Guess I could steal the .4mm nozzle from my BB core sitting on the table. (I had already swapped the BB core for my 2nd AA core to do a 2-color print). Having followed the excellent @gr5 youtube video on how to disassemble the print core, I was soon back in business. (I have since also seen the cutaway drawings for the .4mm nozzles from both the AA & BB cores. They are different inside and I don't know if using it for printing PLA/ABS et.al., is worse than using the original .4mm from the AA core?) I went looking for a place that sold just the U3 print core nozzles and found zero. I understand Ultimaker came up with an easily swappable print core system, but the nozzle is still relatively easy to remove/replace that it seems that there sould be a market for it. Having to pay $114 for a replacement AA core makes me sad. Might as well go for a 3DSolex core with multiple nozzles, including the ruby one. Does anyone know of a place that sells just the nozzles for the U3 print cores?

Success! The position of the 5 SMD resistors were easily accessible without having to take the board out of the UM3Ext. Just had to unplug a few cables and move them out of the way. Was able to do the soldering with the machine on its side. Sadly, RIP to resistor #102, who fell to his untimely death. He gave his life for better surfaces. He will not be forgotten 😢 The blurry looking purple text looks that way because the ink seeped into the layer lines of the print. The actual surfaces are in focus. The Before right side has some funky layering towards the top because I was adjusting the speed/heat on the fly after it passed that indented layer line (which happened on its own) to see if that made any difference. It didn't, really. As others have said, the After zebra effect is still there, but it is MUCH less pronounced. Whereas I can still see a faint version of the striping at certain light angles, I cannot feel them. The before stripes can be felt by the fingertips. I am very happy with the results! With the lines still happening, albeit barely, I wonder if there could be a firmware update to adjust voltages to the steps that cause the lines? Sort of try to have them fire with the same strength/duration as the other steps. If the board wasn't designed to have such things controlled, then probably not... maybe a different driver would produce even smoother results?

Thank you, Cohen!

I picked up my U3Ext April 20, 2017, long before this thread had a solution. I assume that means Ultimaker wasn't able to update the board before I got mine. If someone wanted to do the solder fix on an U3Ext, would the same resistors be shorted out or are they different ones? Would it still be R84, R85 & R86 for X/Y/Z on the U3Ext? @Torgeir / @gr5 ? Thanks to everyone who contributed to troubleshooting the zebra issue. Was great to read it all in one sitting.

The 3Doodler "Create" & "Pro" work only with 3mm filament

I just realized my solution doesn't update the pagination links for First page | Previous page | Next page | Last page, just the numbered page links. Here is the 1st line again with an update to the 2nd line to include those: var urlFix = "/en/community" + $('ul.breadcrumbs > li > a').slice(2).map(function(){ return "/"+$(this).text().trim().replace(/\s/g, '-').replace(/[?.]/g, ''); }).toArray().join('').toLowerCase(), newHref;$('a.pagination-page, a.pagination-link').each(function(){ newHref=urlFix+$(this).prop('href').split('/en/community')[1]; $(this).attr('data-replace-href', newHref).prop('href', newHref) }); Again, this would just be a temporary quick fix to put into place for forum pagination links until the backend can be updated to send the right pagination URLs. It seems like all of the other pagination links, like for pages pertaining to an individual post, work fine.

For example, I can type the following two lines of jQuery into the console log of my browser (it works because you are using jQuery in your page) and update all displayed pagination links with the correct "href" and "data-replace-href". *I noticed that changing the href only would update the URL but still display the wrong content in the page and updating the data-replace-href only would display the correct content but the URL would now be wrong. So they both have to be updated to fix the link. var urlFix = "/en/community" + $('ul.breadcrumbs > li > a').slice(2).map(function(){ return "/"+$(this).text().trim().replace(/\s/g, '-').replace(/[?.]/g, ''); }).toArray().join('').toLowerCase(), newHref;$('a.pagination-page').each(function(){ newHref=urlFix+$(this).prop('href').split('/en/community')[1]; $(this).attr('data-replace-href', newHref).prop('href', newHref) }); The first line grabs all of the breadcrumb texts after the first two, (so it excludes: "Community -> Forum") then replaces spaces with '-' and removes '?' and '.'. You'd have to update the 2nd regex to exclude any other character displayed in the breadcrumb that shouldn't be part of the resulting URL. The 2nd line grabs every pagination link and replaces the "/en/community" part of href & data-replace-href with the correct URL. These two lines could be put into a function and called whenever the page initially finishes loading and then after every AJAX call to load a new page via the pagination links.

Thank you for the response, SandervG. Making the forum faster doesn't help much if the navigation doesn't let you get around in the first place Makes me think of tuning the engine of a car to make it faster when the transmission has fallen out. As a web app developer, when I see issues like this, it probably makes me more frustrated than the average forum visitor because I can envision a fix in my head. If saying it's a back-end issue and not easy to solve means you might be able to fix it faster if it were fixable with just the front end, then that might be a possibility. I understand that the backend is sending the pagination info to the front end, but after a quick glance, I see that the breadcrumb just above the first post is displaying the correct current path to the forum. The front end should be able to grab that text after the page loads, parse through it to get the names it needs (eg.General -> Ultimaker.com Feedback), reformat it for the URL ("/general/ultimakercom-feedback) and insert the missing parts into each pagination link... until the backend developer can fix what it is sending the front end. Granted, I don't know what you are using for front/backend (ie.could be some WYSIWYG forum package/framework where it's not feasible to do custom Javascript functions after the page loads), I'm just throwing out suggestions of how I'd approach it if the backend dev was unavailable and my forum was broken

Thank you for the reply, DaHai8, but I already had that information in my original post