GregValiant

The main difference that I can see is that the narrow neck of the heat break keeps the bowden tube 6 or 8mm above the end of the nozzle. It's a titanium heat break so it has a slightly less heat conduction coefficient than a stainless steel break. At any rate I haven't had the problem recurring (partial clog in the hot end) since I put it in 1 1/2 years ago. I'm careful when swapping nozzles but the stock Creality heat block was made of some really soft alloy and the nozzle threads stripped in a heart beat. I'd say the main problem I have with the hot end now is that the constant rotation of the bowden tube in the lock fitting allows the teeth in the fitting to work their way into the tube. This makes for in and out movement of the tube within the fitting (1 to 1.5mm) and affects the retractions. Nothing to do then but to trim it back so the teeth bite a new area.

No and it's been over a month. His website is https://thopiekar.eu/ and there is a comment regarding the plugin.

Yes. I have only read about it, but you can add more extruders (up to 16 total extruders I think) and define them to mix colors and then they can be be used in Cura as if they were regular extruders. You can define the color mixing using the M163 to M166 commands (it becomes dependent on what your firmware wants to see). The best thing for you might be to go over to someplace like Reddit and see how the other multi-in-one-out hot end machines (like the Geeetech A10M, A20M, A30M) handle it.

The "Layer Height" cannot be changed within a slice. Only the Initial Layer Height can be different. In regards to the XY resolution - You can use a Support Blocker and using the Per Model Settings configure it to Modify Settings for Overlaps and that way you can change the Line Width. If you have your Line Width set to 0.4 you could only go down to about 0.35 so it isn't really much of a change. The only way to change the Layer Height is to slice the model twice (Ex once at 0.2 layer height and once at 0.1 layer height) and then splice the gcodes together at your "layer of interest". The bottom could print at .2 layer height, then there would be a bit of transition gcode, and then the top 5 layers would be from the 0.1 sliced file.

A thumper. You're going racing with a thumper. This just keeps getting better. I think the only thing that could make it cooler would be if it were a 2-cycle. That would be totally BA.

My printer is an Ender 3 Pro about 2 1/2 years old now. I started out with what is probably the same hot end that you have. I passed a lot of PETG through it in the first couple of months and it developed a problem with constant partial blockages that caused a LOT of under-extrusion. After replacing everything except the hot end I bought a clone of the stock Creality hot end. That lead to a new saying..."When you clone a POS you should not be surprised that the clone is also a POS." I finally went with a Micro-Swiss all-metal hot end that bolted right up. It was the best change I've made in the printer. About every 40 hours of print time I still take it apart, trim back the bowden tube 5 or 6 mm and clean out the heat exchanger and hot block, but it has been 100% better than the stock one. BTW that black stuff can be a function of the material spending too much time in the Hot End. You might want to consider setting "Number of Slower Layers" to 0 and make your Initial Layer Speed the same as your print speed. A trick you might try for better first layer adhesion is to set the Initial Layer Flow to 105% or even 110%. You have to get the squish for the plastic to stick to the build surface. I always use hair spray for PETG but I only use it for PLA on certain models that have a small footprint on the bed. So far in the International Printer Toss competition the leader of the "Horizontal Heave" class is a fella who had an Xvico. In the "Defenestration Division" is an Ultimaker S5. You will have to go a ways to beat those guys. They were really irate.

I printed a bit of that gocde. It was fine. Looking back at your images, what the heck is all that black stuff coming out of your nozzle? If some of that is getting hung up in the nozzle orifice then you will get inconsistent extrusion. Your hot end may just need a good cleaning. Have you checked the bowden tube end that is within the hot end? If it is breaking down then that may be the source of the black debris. Enders are notorious for developing partial blockages between the back of the nozzle and the end of the bowden tube. Pull out the nozzle and clean out the hot end with a "just right" diameter coat hanger or other appropriate wire. Trim off 5 or 6mm of the bowden making a nice square cut, and put the hot end back together. The hot end must be heated up to 180 or so for all of that. Your print temp at 200 is fine. Is your printer equipped with the stock hot end? If not then did you do a PID auto-tune on the new one?

A picture would really be helpful. I'm not sure what you're trying to do. "Stringing on the inside" can be taken care of by adjusting the combing style and setting the layer start point near the Z seam location. That way it shouldn't have to cross the center. Here is a project file. I added two support blockers, configured them as Cutting Mesh and set the Per Model settings to a single wall and no extra skin wall. They are 0.2mm thick and one is on the bottom and one is on the top. Here is the top layer. This is the layer below the top. SingleWallTopBtm.3mf

The top of that model isn't very big so you might be bouncing up against the "Minimum Layer Time". The default is 10 seconds but if you lower it too much you won't get sufficient cooling and the top of the model may suffer deformation. I recall that "ChangeAtZ" did not play well with Z-Hops enabled. The Z moving up and down confused it. If you have enabled Z-hop turn it off and see if it makes a difference. If you open the Gcode file in a text editor you will see "F" numbers in some of the lines. They indicate the feedrate in mm/minute. 50mm/sec would be F3000 and 100mm/sec would be F6000. If you look around layer 150 you may see numbers like F2476 or something else that wasn't one of your settings. That would indicate that Cura made an adjustment to slow the layer down so it would take longer than the "Minimum Layer Time".

It looks likes it's going 150mph just sitting there. Instead of the Bat Bike I should have posted Captain America's bike (and I'm not talking about Peter Fonda). Congratulations on the printing and on the build. I know it was a lot of head scratching as well as a lot of physical effort. What engine will you be racing with?

Post that Gcode file here. I've got some scrap PLA that I can throw at it. If it does it on my printer then it's in the gcode. If it doesn't then it's in your printer. I think at this point narrowing it down would be helpful.

Thanks AHoeben. @tomtenberge That will work. I over-thought it and figured the tool number would have to be called out but that isn't so. Your code will work on the "active extruder" so whether a print finishes with T0 or T1 it will pull that one back. "...and so it blobs XD" You might be able to take care of that in your StartUp gcode. If the first Tool Change is called out before the skirt then the purge should occur before the skirt. There won't be a move to the prime tower but that shouldn't matter if the skirt goes down without the big blob. It's more experimenting but something like this might work as the last few line in the StartUp: G1 X0 Y0 Z15 ;Move to the corner and raise the Z so there is room under the nozzle T0 T1 T{initial_extruder_nr} ;End up with the proper tool to start the print. I don't know how that would work if you had one of the extruders disabled though (single extruder used for the print). A two-in-one-out hot end should be OK but there could be a "cold extrusion prevented" warning on a regular dual extruder-dual hot end machine.

If the gcode is being produced as you want it with your switch-over retractions correct, prime tower is good, and all that dual extruder stuff then I think you want to put it in your ending gcode. M83 G1 F2700 E-86 T{extruder_nr} M82 In @ahoeben's list of the Cura keywords "extruder_nr" is defined as "The extruder train used for printing.". I'm not sure if that would mean the active extruder? Maybe AHoeben could clarify that. You might want that before the M104 S0 line. Because Cura doesn't do math in the StartUp or Ending gcodes - if you change your retraction distance you might need to change the E value in that line. Most (if not all) gcode prints end with a retraction. In my case that would be 5.5mm and then there are two 2.0 retractions in the Ending Gcode (prior to the nozzle moving away from the print) so if I put the line in after those retractions I would want the number to be E-80.5.

The "Retraction Speed" setting is Momma and "Retraction Retract Speed" and "Retraction Prime Speed" are the daughters. When linked - the Retraction Speed will apply to both retract and prime. You can have different speeds for retract and prime as well. If there is an "fx" next to the "...Retract Speed" or the "...Prime Speed" you can click on the "fx" to re-link it to the main "Retraction Speed" setting.

That "...requires OpenGL 2.0 or above" warning has been reported a few times. Here are the search results on the GitHub site. Take a look and you may come across a fix for that. This older Win10 laptop has to run in compatibility mode as the video system can only be upgraded to OpenGL 4.0 and I don't get the fancy 3d looking preview, but Cura itself runs fine. Did you figure out the Scaling problem?

I kept a journal when I started into this. I did refer to it a couple of times in the beginning but I think the important part was simply keeping the journal because organizing it and looking at it kept me thinking about what I was doing.

We just went through this with an Ender 3 Pro that had the problem. A workaround seems to have gotten the job done. It is described in THIS THREAD . NOTE: When using the M25 command it's possible that you might have to "Pause" the print on the LCD (even though the printer is in pause mode) in order to bring up the "Resume" command. So that sequence would be... The file is printing and comes to the pause section. It parks the head and the M25 causes the printer to pause. You put the magnet in. On the LCD find the menu with the "pause" command and click on "Pause" and then move to the Resume command and click on it. The G4 method always works but you lose control of the printer because G4 is a timed event. You don't get to choose when the printer restarts.

There are going to be some lines because you have the bottom pattern set to "concentric". The lines will be where the nozzle shifts to the next loop. That heavy line going from the middle area to the top looks more like there was an extrusion at the start when the nozzle traveled from the outer wall to start the concentric skin. If there was no retraction there then you didn't get a Z-hop either. Check your Travel settings for "Retraction Minimum Travel" and "Max Comb Distance with No Retract". For that print both should be low numbers like around 3 to 5. You might also try "Avoid Printed Parts When Traveling" and see if it helps.

Yes, intermittent under-extrusion and some of it is very severe. The #1 problem with under-extruding Enders is the plastic extruder arm. They all fail. 100% failure rate. They have been known to be delivered broken. This is the underside of a failed arm. The only fix for that is an all-aluminum extruder. They are available all over the place. I got the one for my Ender 3 Pro from Creality. Have you calibrated your E-steps? It is necessary on Enders. When lines in the gcode file ask for filament the printer must deliver exactly the right amount of filament. The Extruder Steps/mm can be adjusted to insure what is asked for is what it gets. I'm sure CHEP has YouTube videos in his Filament Friday's series. Another common problem on Enders (they all share the same hot end) is a gap developing between the back of the nozzle and the end of the bowden tube. Long retractions can pull molten plastic into that gap and it makes a sort of o-ring there and impedes the incoming filament. The fix for that is to: Heat the hot end to around 200°. Pull out the filament. Remove the nozzle (carful - it's hot) using a correct wrench and NOT a pair of pliers. Remove the bowden tube from the fitting on the hot end. Shove a calibrated piece of coat hanger (about 1.5 to 1.6mm diameter) down through the heat exchanger and heat block (in a pinch you can use a 12" piece of filament but you have to move fast because the hot end is hot). I'm betting you will push out a plug of plastic. Cut about 5mm off of the end of the bowden tube using a decent single edge razor blade or equivalent. The cut must be as close to exactly square as you can get it. Put the nozzle back in and gently tighten it, then back it off 1/2 turn. The hot end must be at least 180° for this as any plastic that has gotten into the threads needs to be soft. Shove the newly cut bowden tube into the hot end and while holding it down against the back of the nozzle - put the locking clip back onto the fitting. Gently tighten the nozzle. You should end up with the bowden making a decent seal to the back end of the nozzle. It should be good until the next time it happens. The bowden tube is a "consumable" and with every trimming it will keep getter shorter until it needs to be replaced. With your model loaded (either one) and Cura set up ready to slice use the "File | Save Project" command. Post the 3mf file here and I'll take a look.

If that was printed it PLA it wouldn't need a brim. It has plenty of surface area to hold it down. In PETG or ABS then yes, a brim would probably be necessary.

Silky silver PLA is probably my favorite. The layer adhesion sucks so I don't use it for functional parts but for things like that helmet it's perfect. When you get done with that print try a silky silver print with the outer walls at 35mm/sec. In my (never humble) opinion it's about as close as you can get to chrome.

Hello @Mikelo I have responded in the other post. Use a support blocker. Set the Per Model settings to "Modifiy settings for overlaps" Select "Infill Density" and "Infill Pattern" as additional settings to alter and then change the value of those settings.

The "Support Blockers" can be configured in different ways. The normal situation is to block support (Duh!!). The tool above the Support Blocker tool provides the "Per Model" settings and one of them is "Modify settings for overlaps". Since a support blocker is just another mesh (and can be scaled and moved like any other mesh), you can configure them as "Infill Mesh" (which you want for changing the infill) or as a "Cutting Mesh" (which gives different possibilities). Below is an image with the Support Blocker configured as "Infill Mesh" and I've added "Infill Density" and "Infill Pattern" as two of the settings changing them to 25% and Gyroid. Within the overlap of the block with the part you can see that the density and infill type is different.

If you are using Windows, it has MS 3D Builder included. I have found that it isn't real intuitive, but for simple things like cutting an STL it works pretty good. Open the helmet, bring in a cube from the menu, size and move the cube till you like it, and with the cube selected, and the helmet NOT selected use the "subtract" tool. You will get a nice clean edge. Then do a Save As for the new model. MS 3D Builder is also pretty good at repairing problems in models. It looks like Mesh Lab simply erased a bunch of triangles without sealing the gaps. You want a boolean operation there (Subtract, Union, or Intersection).

That is a bad cut. Those jagged edges look a lot like open seams to me. That would confuse Cura and you will get wierdness in the slice. Instead of altering the model, go into the Cura Preferences and turn off "Automatically drop models to build plate". Then in the "Move" tool dialog set the Z to a negative number to sink the helmet into the Cura build plate by whatever 90% of your model height is so only the top 10% is exposed. It should slice OK like that and you wouldn't have those jaggies to deal with. The support should generate OK as well.