GregValiant

I printed up a hole chart with holes from 2 to 30mm. I chamfered all holes top and bottom so my measurements would be of the walls. I found that from 2 to 10mm the holes were .4mm undersize with very little deviation. From 11 to 30mm the holes were consistently .2mm undersize with very little deviation. I was surprised by the results. I had anticipated that there would be a fairly linear error (call it a ratio) with the largest error being on the smallest hole. As a result I was unable to develop a consistent "Snot Factor" for use with horizontal hole expansion. If I have a part with large holes I set it to .1 and for holes 10mm and under I set it to .2. This has worked out well enough for me since nothing I do is mission critical. If a hole requires precise sizing I ream it with a drill bit. Holes that pass through a model anything other than vertically are what they are. Again, a drill bit fixes any size (or out-of-round) problem if necessary. Regarding flat surfaces - is it a ratio? If a machine can be calibrated by adjusting the steps/mm, and it's calibrated correctly, it might not be a ratio but a fixed number. Gotta think on that a bit. I thought the Snot Factor would be my claim to fame but alas, it's not to be.

You will have to see if you can get support from the store you bought it from. Off the top of my head I'd say the red is power to the heater and black is the negative to the heater. The double blue wires are the thermistor and typically they don't have polarity so you can't get them wrong so long as you connect them to the correct two wires on the machine. But check support at the store first and double check my assertion regarding the blue wires that there is no polarity.

In the interest of continuing this "I'll show you mine if you show me yours" thread... I'm attaching two project files. They contain the same model (25x25x25 cube with no features). Settings wise it's pretty vanilla with no skirt/brim and with infill at 100%. One has all flows at 100% and the other has all flows at 50%. When sliced in Cura 4.6.1 on my laptop the Gcode files turn out as I expected. The 50% file uses half as much plastic. If you would be so kind as to open each of the files and slice them on your computer and show the results - maybe we can figure this out. My expectations are that since we are running the same version we should come up with the same results. For reference - from my gcode file of the slice of the 25x25x100.3mf file: ;Filament used: 6.69192m and from my gcode file of the 25x25x50.smf slice: ;Filament used: 3.3331m 25x25Cube50.3mf 25x25Cube100.3mf

It should fit. If you have a 10mm skirt 10mm away from the 200 image it wouldn't slice but the image itself should fit. A screen shot of the Cura build plate with the image on it would help.

No matter how much you were to pay for Cura, the sheer number of 3d printer manufacturers and models out there makes any sort of support an impossible task. And this triangle situation (did anybody figure out why there was an extrusion?) was a purely Creality generated problem that was fixable using Cura. Consider that Creality ran out of 8 bit boards and without telling their customers or anyone else started putting 32 bit boards in. How do you support something like that from afar? Creality won't even acknowledge a problem that is fully their responsibility. In effect, my Ender 3 Pro had no warranty and no support from day 1. It was what I expected when I decided to buy a really cheap printer to get in to this as a hobby and see if I cared for it. I suppose I was having trouble with the word "ignore". I guess it's the right word although I don't think the Cura team ignores the others because they want to, but rather because they have to or they'd all go nuts. Have another Twinkie Longtoke and pass that sucker over before it gets too short.

Exactly. Yes - you've got it!🍺 The printer does not change unless it sees a Gcode M221 (or a change in the LCD flow - an internal M221). That is a Global change to every E number that comes through the printer-planner. It just doesn't care what those E numbers are. In my example above, if I print the 100% file the printer will say "Flow =100" and over the course of the print it will go through 21.7m of filament. If I print the 50% file the printer will say Flow=100 and over the course of that print it will push 10.7m of filament which is pretty near 1/2 as much (it's not exact because of other settings). The E numbers in the first file were CALCULATED at 100% by Cura and the E numbers in the second file were CALCULATED at 50% by Cura and so are only half as big. During the print of the second file, the printer will run 100% on E numbers that are 50% as big. Over here is the printer...over there is Cura. They are separate. Cura makes calculations based on your settings. The printer is read-only but can make Global changes based on it's own settings and those settings can be changed either with lines in the Gcode, or by tuning a print on the fly from the LCD. Later we can get into Feed Rate which works the same way. Right now though, I'm going fishing.

Cut the connector off the new hot end and throw it away. Then cut the connector off the old hot end leaving 100mm of wire on the connector. Then solder it onto the wires of the new hot end making sure to get the wire colors right. You can get shrink tube insulation at any hardware store. Make sure to remember to slide pieces of insulation onto the wires before you solder. Do the same kind of thing with the blue wires (old ends soldered to the new wires). Remember to insulate. I would advise you to never buy an old motorcycle.

I'm not affiliated with Ultimaker or the Cura team. I'm just an old retired guy (with a wide skill set) having some fun while locked in. I have only been into 3d printing since December. Of all the myriad complaints I've seen here, on Reddit, and on other boards, there have been two valid bugs noted. The Cura team had been aware of both of them and were working on solutions. It isn't perfect. It isn't easy. It works. The stl file for this Gcode snippet was sliced at 100% flow on everything using Cura 4.6.1. 4.7 beta gave the same results. ;FLAVOR:Marlin ;TIME:25985 ;Filament used: 21.7041m ... ;TYPE:SKIRT G1 F300 Z0.2 G1 F3600 E0 G1 F3000 X115.451 Y31.869 E0.03186 G1 X115.719 Y31.97 E0.0416 G1 X116.554 Y32.373 E0.07316 G1 X116.942 Y32.605 E0.08854 G1 X117.473 Y32.969 E0.11045 G1 X117.818 Y33.264 E0.1259 G1 X118.068 Y33.497 E0.13753 G1 X118.672 Y34.166 E0.16821 G1 X118.805 Y34.342 E0.17572 G1 X119.073 Y34.732 E0.19182 G1 X119.724 Y35.782 E0.23387 G1 X120.887 Y37.741 E0.3114 G1 X122.288 Y40.026 E0.40262 This Gcode snippet is the same stl file sliced 15 seconds later with the only change being 50% flow on everything with Cura 4.6.1. Once again, 4.7 beta produced the same results. ;FLAVOR:Marlin ;TIME:25730 ;Filament used: 10.6874 ... ;TYPE:SKIRT G1 F300 Z0.2 G1 F3600 E0 G1 F3000 X115.451 Y31.869 E0.00796 G1 X115.719 Y31.97 E0.0104 G1 X116.554 Y32.373 E0.01829 G1 X116.942 Y32.605 E0.02214 G1 X117.473 Y32.969 E0.02761 G1 X117.818 Y33.264 E0.03148 G1 X118.068 Y33.497 E0.03438 G1 X118.672 Y34.166 E0.04205 G1 X118.805 Y34.342 E0.04393 G1 X119.073 Y34.732 E0.04796 G1 X119.724 Y35.782 E0.05847 G1 X120.887 Y37.741 E0.07785 G1 X122.288 Y40.026 E0.1006 So there must be something in your settings. (Initial Layer Flow doesn't update when the Flow is changed - could that be it?). Any time I've changed the Flow in Cura, there has been a change in the E numbers in the gcode as witnessed above. I do have to manually change Initial Layer Flow.

There is a definite learning curve. You've got the design software, slicing software, machine firmware, Gcode language, the mechanical aspects of the machine, etc. It's a lot of stuff and it all has to come together for a good print. Download a calibration cube from Thingiverse (there is a link from within Cura under Extensions / Thingbrowser) and you can slice it using various settings in Cura and see how they effect the print and each other. Spiralize Outside Contour is pretty cool. Some folks call it vase mode. There are some nice purpose built vases on thingiverse, but you can use it on a calibration cube just as well.

They are the same instructions. Using Cura (or Pronterface) as a host software works OK at best. There is no way for host software to check how fast the printer is using lines of Gcode (circular features mean lots of moves/second and lines of code/second. Large flat areas mean few moves/second and few lines of code/second.) so the host struggles keeping up at high usage, and sends too much info at low usage. It's a tough balancing act. Cura and Pronterface also ask the printer for temperature information every 3 seconds which means constant 2-way communication through the same line that the gcode is using. With the SD card the printer just pulls lines at the rate it needs them. Then there is the problem of Microsoft updating something during an all-night print and the computer re-boots causing the print to stop. That doesn't happen from the SD card.

There is no correct place to put this, but since accuracy and finish are involved I stuck it in this thread. I've started work on my own 3d printer design. It will: 1a. Produce 3d Prints with a finish like 12 coats of hand-rubbed lacquer. 1b. Provide an accuracy of +/- 1 nanometer. 2. Come with slicing software that knows what the user wants and never forgets to move the Z seam. 3. Have a Single Button-Single Function virtual "Start Print" control. 4. Provide a printing speed of 1,000,000 mm/sec. 5. Print any material from chocolate to tungsten. 6. Be powered by an on-board flux capacitor. 7. Be absolutely silent. 8. Be easily up-gradable. 9. Have a price tag of $19.95 and a 100 year warranty. 10. Remember birthdays and anniversaries and automatically print a nice gift. It's called the Harry Potter 3d Platinum Pro XL Extended v1 Cabriolet. It is available for pre-order now. I hope Ultimaker doesn't take exception to this blatant advertising blurb on their forum.

I don't know much about that printer. Looks like an open frame design likely running Marlin firmware. There are general things that can be checked though. Check within Cura and find the filament diameter. Make sure it is set to 1.75mm. Check that the "Firmware Flavor" matches your machine (Marlin?). Check that the nozzle setting reflects what you actually have installed. If the printer came with an SD card are there sample Gcode files on it? I would start by printing one of the samples to make sure everything is working as it should. It looks like the machine has an adjuster for filament tension on the extruder. Make sure there is no slippage there when extruding. Have you calibrated the steps/mm of all 4 axis? If the printer is told to push 100mm and it only pushes 90mm then there will be under-extrusion. When you change the Flow in Cura, all the other flows should update to match it. Double check "Initial Layer Flow". Make sure that one matches the rest of them. There are downsides to printing over the USB. I much prefer using the SD card.

I've installed it (that was easy) and so far it's good. I thought the stock version was a bit clumsy.

I understand when it does it with rectangular areas with holes in them. I only used Concentric once and found the behavior odd. Since it really can't be concentric (because there are only line segments) - what you say about end points may be it. A spiral pattern would work but probably be of limited use. With a hole in the surface the path would be different anyway.

In the situation of the concentric pattern, it just seems to make sense to continually move a line width at a time and draw the next circle (as separate line segments of course). I understand there are a lot of computations going on in the background (every atom in the universe?) but the final decision that it is better to move outboard and then come back in doesn't add up. If Cura would just play the CD it would be good. Instead, it plays tracks 1 to 4, skips 5, plays 6 to 10 and then plays track 5 backwards. Effectively it makes the total travel longer than would be necessary to play 1 thru 10. Playing a track backwards might have been OK for the Beatles, but it doesn't do much for me now.

You should see a little crossed Hammer & Wrench icon near the slice button. It probably has a red circle with at least a "1" in it. Open that icon and press the X at the end of the line with Pause at Height in it.

Ahoeben types faster than me... Cura has to start up the USB printing plugin and initialize it. Then it polls the printer for certain settings (like temperature). Once loaded though, if you were to start a second print it's nearly instantaneous. I have to ask though, why over USB? There are known issues when using Cura or Pronterface as host software (I don't know about Octoprint). Random pausing that requires a button click to re-start tops the list. I know my E3Pro much prefers the SD card.

To paraphrase Nallath - "That is because the flow on the screen of your printer is based on a (manual) change on what is in the g-code. (Without a manual change like that (or an M221 line added to the Start-up gcode)) It will always show 100%, regardless of what you change in Cura." _____________ For an example - Cura doesn't inform the printer that Print Flow was calculated at 110%, Outer Wall Flow at 95%, and Infill Flow at 50%. There is no reason to do that since the printer planner only acts on the E values that it actually receives. The printer doesn't need to know what numbers were used in the calculations that came up with those E values. The printer planner always calculates it's own delta E number based on the Global Flow % stored in the printer. If there is an M221 in the gcode or if you use the LCD and change Tune/Flow (which is an internal M221) then the printer planner uses the new Global Flow % (and displays it) when it calculates the steps/mm required to send to the extruder motor to extrude the correct amount. And that is also why the E-steps need to be calibrated.

In the Cura folder ...C:\Program Files\Ultimaker Cura 4.X\plugins\PostProcessingPlugin\scripts\PauseAtHeight.py The PauseAtHeight.py file can be opened in a text editor. The following line is near the end: prepend_gcode += self.putValue(M = 82) + " ; switch back to absolute E values\n" Change the 2 to 3, save the file, restart Cura. I'm not a Python guy but you should be good to go. Cura 4.7 PauseAtHeight allows you to add a gcode command "Gcode after the pause" but it goes in as the first line after the pause which doesn't work since the M82 is a few lines later.

I like having them separate. It's like a sub-profile. If I just need to get the flow going I select skirt. If I need better adhesion for small parts I pick brim. Either way, the rest of the settings for each are already filled in. With separate models, using Brim Distance as a per model setting works well. Up tight on small stuff, and leave a gap so I don't have to remove the brim from larger stuff.

"That is because the flow on the screen of your printer is based on a change on what is in the g-code. It will always show 100%, regardless of what you change in Cura." The only caveat is if you put an M221 in the Cura "startup g-code" (which doesn't effect Cura computations as start-up gcode isn't a setting). That M221 directly affects the printer and will change the flow % of each extrusion line in the gcode file and the new flow % will show up in Tune / Flow Percentage on the LCD. The printer does not display the Cura settings, but Cura uses your flow settings to make it's calculations. My Ender 3 Pro (silent board with Marlin 1.1.8) only requires that the E-steps be calibrated. Once that is done, if the gcode says extrude 10mm then exactly 10mm will be extruded. That is 100% flow. I found that using calibration cubes and trying to calculate flow by wall width was maddening. Cura has modifiers to flow, but generally, with a .4 nozzle at .2 layer height you are filling a .08mm² rectangle. 1mm of filament is 2.40mm² and doing a bit of math that comes out to .033mm of filament per 1mm of extrusion. If you do a careful check of the Enumbers in a Cura generated gcode file (where all flows were at 100%), that's what you will find. Simplyfy3d uses a fudge factor of 1.2 in it's flow calculations (THAT was hard to find). Instead of filling a .08mm² rectangle it calculates it at .096mm². So if the instructions you use for a wall calibration cube were written by someone using Simplyfy3d you will really struggle to hit the numbers and then your prints will be over-extruded. So the best way to figure out if you are flowing correctly is to do as Mari said and calibrate the E-steps. After that use your eyeballs to see how the finish looks. If it looks good then it's good. If you look down a wall and can just see a faint line between the parallel extrusion stripes, it's good. If there are ridges and stringiness back and forth on a mesh, you are over-extruding. If you can see the layer below as a crosshatch pattern you are under-extruding. If you are manually leveling then the first layer is never a good indicator of flow because the height is dependent on the leveling procedure. The second layer is better and should look really good except at the ends of runs where the print head slowed down but there was still pressure in the chamber leaving slight ridging (the Coasting setting helps that). After you get a feel for the machine you can fine tune the flow % of particular areas using the various settings. Calibrating the steps/mm of all 4 axis needs to be done periodically as the machine will change slightly as it wears in.

Paxpring - I came across this thread which seems to be relevant to the dimensional issue and the UMS5. Maybe there is something in there you can use.

Dimensionally I'm impressed with the accuracy of the machine on everything but the diameter of small holes. Rather than fudging with Horizontal Hole Expansion I rely on the absolute accuracy of the hole location and drill them to size when necessary. Slowly but surely the mechanical calibration will go away as the machine continues to wear in and as a result dimensional accuracy will go with it. When it's freshly dialed in it's amazingly accurate (for a Chinese POS built to a price point). I won't be switching away from Middle any time soon since my CRS disease keeps me from remembering to switch it back. I'm gonna whine about that until I empty the trash and can't see my scrap anymore.

Ghostkeeper's Usage description of "Slicing Tolerance" - "This setting is named after its intended use rather than its functional effect. If you have multiple pieces that need to slide past each other, the theoretical shape of the layers may physically prevent exact fits. In such a case, you can set this setting to Exclusive so that the layers are guaranteed to stay within the bounds of the original volume. Barring warping, sagging and similar deformation effects, this would guarantee that the parts fit within each other and are able to slide past each other. In reality there are always other effects going on that prevent this. In practice, this setting can be used to get slightly more or less tolerance between two sloped surfaces, as can be seen in the above images." I am an ME with 50 years experience scribbling special machines, fixtures, dies, automation, whatever. It was always a basic tenet that any part be fully described within the borders of the drawing. Didn't matter who made detail 1 or who made detail 2 - they need to fit together. Tolerances, location of features, etc., needs to be part of the drawing - which is now the 3d model. So I have a philosophical problem with the idea of generating "tolerance" via a software setting. It is akin to allowing the technician to determine tolerances. That indicates ad-libbing by the technician and should never be allowed. Now that the designer and the technician are often the same person doesn't change my view of that. If two parts need to fit together, they should be designed to fit together. Being aware of what type of machine was going to be used to make a metal part was necessary to design a part that could actually be made (as opposed to what we referred to as Optical Illusions). Being aware of the 3d printer/slicer/firmware capabilities is just as necessary. An eye must be kept on those capabilities when designing parts of an assembly. Paxpring - in one of your initial posts you mentioned problems between the PrusaSlicer / Prusa combination and then using Cura with (I must assume) the same Prusa printer. If the parts slice differently in Cura than they did in PrusaSlicer causing dimensional anomalies within the Gcode this would seem to be an elemental problem between the slicers that needs to be addressed (somehow). Using the Exclusive setting (with it's side-effects) to make adjustments in dimensions just doesn't sit well with me. Again, it has the effect of allowing a technician to unilaterally alter a design. And after all of that - the old guy forgot to change Exclusive back to Middle, sliced and printed a model that turned out two layers too short.

Burtoogle - I read your comments in the PR regarding the missing 1st and last layers. I've been cogitating... We have a cylinder of 25mm diameter and 11mm tall. It has 55 .2mm deep cuts arrayed around the circumference such that each cut exists on 1 layer only. When sliced with Middle it looks like this. The Gcode shows 0-54 layers @ .2mm each and the final Z = 11.00. This is how it was designed and what I expected to see. When sliced with Exclusive it looks like this. The first layer has a single feature that is .2mm tall. Succeeding layers each have 2 features that are .4mm tall (.2 up and .2 down). The top layer (54 in Gcode) disappears. The numbering in the Gcode file goes from 0 to 53 so yes, there are two layers missing and the top layer is at 10.6mm. A model with a single .2mm deep feature on the top of the model, when sliced with Exclusive, shows that the top layer (and consequently the feature) disappears. Once again, the top layer Z is 10.6. In either case, the diameter was 24.6mm. Using a square model, the side dimensions were 24.6mm. Line width was .4mm. Good Luck with this one.