I fixed my clogged nozzle!


All is well now. The problem was a small piece of filament became separated from the main length as it was only connected to the main length by a very thin strand so when I pulled the filament out of the head this small piece got left behind.
I removed the head and then removed the new nozzle and managed blow through the centre hole which removed this small bead of filament.
I bought some new 0.4mm nozzles from Amazon but I had to run the head temp up to 190 Deg as there was a lot of molten filament effectively gluing the nozzle in place, trying to unscrew the nozzle while cold caused the heater block to rotate as well.
I noticed that the base plate of the print head had been melted around the nozzle hole. I think this may have been caused by the hot filament not adhering to the metal plate, maybe because the glue had not been sticky enough.
Thanks to BoozeKeshi, Lolo and Anm for their info and help.


Glad you got it sorted, but you are not supposed to unscrew it while cold. It is supposed to be on there very hard when cold so it does not get loose when hot. Save that old nozzle though, if it is not physically damaged you can clean it and use it again.

Since you mentioned the heat block rotated, just keep a close eye on things on your next few prints, especially your temp monitor in FLUX Studio, and the cooling fans. The one with the visible red wire is the cooling fan for the cold end, that should be on always. The other two are part cooling fans that come on and off as told by software. With that block rotating, there are wires going into it for the heater core and the thermistor, and the other wires for fans going around it. You just want to make sure everything is working like it should or be able to shut it down quick if something looks not right. (Long press on FLUX button)

On your old plugged nozzle, what I usually do with them is put them on a piece of aluminum foil and put them in the oven at about 225-250c and just leave for about 20-30 minutes. That bakes out whatever is in there and then I use a piece of guitar string to run through. A light polish with a brass wire brush and good as new. I tried those dental drill bits and found out the hard way that those are very easy to snap… inside the nozzle. Which makes the clog about a million times worse as they are hard steel and almost invisible. Guitar string works perfect until you get down to the .2 nozzles.


Hi Booz,

Yes the heating of the nozzle was uppermost in my mind when I attempted to turn it, the movement of the heater block was minute so I doubt any problems were caused. I would have had to heat it up anyway as the nozzle was effectively glued in place by the amount of black residue from the filament so heating was the only option.
I took Anm’s advice and heated to old nozzle up with a small blowtorch and finally probing the hole with a “bristle” from a wire brush.

Slightly off topic now, what parameter sets the very first sweep of the head when printing. On one version of Studio the first sweep is a slightly wider outline of the object being printed, but with another PC running Studio the printer always does a circular sweep at the extreme of the head travel then prints the object. As far as I know I have not changed anything on either program>


Check your settings for Skirt. It sounds like you have it set ON on one PC and OFF on the other PC. That wide circle is something in the new version of FLUX Studio (0.7.3/0.7.4), so I haven’t seen that yet. The outline of the print definitely sounds like a skirt, but probably set to 1 layer and 1 or 2 lines only.

Normally the Skirt should extrude a line of filament, it is used to prime the nozzle so that you have a nice good flow when the actual print gets started. There are a few settings you can adjust.

How many skirt layers.
How far (distance) from the print.
Number of skirt lines. (how many times it goes around) often nothing extrudes on the first pass

My typical settings, and I use Skirt on almost everything are 1 layer 3mm distance 3 lines. Depending on the filament brand and temperature settings it will usually start extruding around the end of the first pass and then I’ll get to see two good lines. By then I know that I’m getting good flow.

FLUX Studio treats Brim separately with it’s own tick box, but other software skirt = brim, just by reducing the distance from the print to 0. Now you have a few extra lines/layers around the print that can help anchor it to the bed. This is useful for tall, thin vertical prints as it is easy to trim off, and when you may not want to use a Raft. But Skirts/Brim/Raft all have their place.


the 0.75 circle is set up in the ‘Expert’ settings as the start_gcode (or something like that). I’m planning on editing it soon and replacing it with my own (in fact someone else’s that posted on this forum) start_gcode to prime the filament/nozzle.


There’s two new behaviours in recent version of Flux studio… a 3/4 circle that Mattman pointed out is in the machine_start_gcode, which the printer does immediately after calibration whilst the printhead is warming from 170c to the target temperature, which makes a skirt unnecessary as the nozzle is primed and ready to go. Then there is this two revolution movement test it does before starting the calibration… which was a bit scary the first time it did it as it moves at a fair rate compared to the Delta’s normal movement speed and moves at the extreme of the Deltas movement arc.

It sounds like one machine is set to use a skirt, and the other has the newer version of studio and is doing the 3/4 circle nozzle prime, and doesn’t have a skirt set


That sounds about right, and your description is spot on this time. :slight_smile:

I just upgraded to 0.7.4 and kicked off a print and while I was expecting the usual little calibration triple tap dance, the toolhead started racing around the build plate quite fast and did kind of scare me. What was odd is that it did NOT calibrate at all, just raced around the track a couple times, then to home, then started the S3D routine of elevator up / elevator down. Or perhaps because I had just done an Auto Level after installing the upgrade?

I haven’t seen the .75 circle movement yet, but maybe because I started the print from SD card and not FS?


:strawberry: :stuck_out_tongue:

Hm… I was fiddling with the auto level and movement test stuff the other day, and it still did a calibration when doing a print after that. Maybe double-check your preferences (File -> Preferences -> Machine)… maybe it’s set to ‘by file’? I noticed duing the upgrade process that for some reason all the options were set to by file (which is fair enough… means the print has full control of it’s settings), but I prefered it always being forced to auto level and do the height test, always have filament detection on (rather handy as the filament broke the other day so it just paused, I reloaded the filament, and sent it on it’s merry way again)… and have it only check for tilt or fan failure (as I don’t know what the condition for “shake” is, so don’t trust it).

What machine firmware are you running? I suspect 1.6.25 as I’m pretty sure it was the firmware that introduced the movement test (from what the changelog says anyway…). Toolhead is on 1.2.10.

So, has anyone worked out what the heck “Smart Task Continuation” is? And what is so smart about it? :laughing:


Alright, curiosity got the better of me so I went looking, and now I have more questions than answers.

  1. That section [machine_start_gcode] is only visible/active if you select Cura2 as the slicer option. If you select Slic3r or Cura it is not there.

  2. If you upload direct to SD card, as in those of us that use S3D or the new Prusa Slic3r edition, it is not going to run that code either. (That layer smoothing thing is pretty amazing)

  3. How is making a 3/4 circle eliminating the need for a skirt and priming the nozzle? Is it extruding while doing this movement?

  4. If 1 or 2 applies, you would still need skirts anyway right?

  5. @Mattman, what would you be changing it to, and what would THAT do?

  6. To confirm, it is this bit:

machine_start_gcode = G1 F6000 Z50\nG92 Z49.9\nG92 E0 \nG1 F2400 E-4.50000\nG0 F9000 X85.384 Y-0.738\nG0 F9000 Z0.8\nG1 F2400 E0\nG92 E16.55716\nG1 F600 X83.693 Y16.934 E21.28723\nG1 X78.383 Y33.877 E26.01807\nG1 X69.685 Y49.355 E30.74862\nG1 X57.975 Y62.701 E35.47929\nG1 X43.757 Y73.336 E40.21008\nG1 X27.649 Y80.802 E44.94053\nG1 X10.345 Y84.776 E49.67108\nG1 X-7.407 Y85.086 E54.40166\nG1 X-24.838 Y81.718 E59.13191\nG1 X-41.198 Y74.819 E63.86265\nG1 X-55.778 Y64.686 E68.59345\nG1 X-67.947 Y51.758 E73.32398\nG1 X-77.179 Y36.593 E78.05439\nG1 X-83.077 Y19.846 E82.78515\nG1 X-85.383 Y2.241 E87.51594\nG1 X-84.001 Y-15.458 E92.24605\nG1 X-78.988 Y-32.490 E96.97657\nG1 X-70.562 Y-48.119 E101.70744\nG1 X-59.086 Y-61.666 E106.43796\nG1 X-45.057 Y-72.548 E111.16858\nG1 X-29.081 Y-80.294 E115.89920\nG1 X-11.848 Y-84.570 E120.63003\nG1 X-3.392 Y-85.306 E122.88852\nG92 E0


Hm… curiouser and curiouser! In response to 3, then, yes, it extrudes whilst it is warming up… hence you get a 3/4 circle of filament around the perimeter of the build plate.

When you select slic3r or cura, it is start_gcode instead of machine_start_gcode… don’t ask me why it changes! :open_mouth:

However, since it is there in the flux gcode… it must not be done by the printer itself, so you would have to add the g-code to your S3D starting g-code script for your flux delta profile if you wanted the same behaviour. The end code doesn’t look like anything fancy, only the start code.


Okay, I’ve been wanting to test drive the Cura2 profile anyway so I’ll check out what that does. I kind of like using skirts anyway. Very accustomed to it and I get a very good idea if the filament is going to behave like I want it to or not from the start.

I’m still interested in what what @Mattman will replace that with :hammer:


Sorry guys. Finally got to the computer to reply - also had to look up who posted it before…see what Marxoo posted: Purging Filament before printing
I will modify it according to my preference - like instead of extrude 45mm, I’ll do something like 25mm. I might also not include the temperature setting.


After fighting with 0.7.4 for an hour and discovering it will not take uploads over 10Mb direct to SD card, I got to experience the 3/4 circle for the first time.

NOW I understand why you want to change it. My usual skirt setting is 3mm offset 1 layer 3 passes, not much filament at all, but enough to get a good flow and where I could tell if things were good or not.

This 3/4 circle is laying down a pretty hefty track of filament, much more than any of my skirts ever used.


Hi Booze, After fitting a new nozzle the nozzle and heater block is covered in molten filament every time now.
Not sure if it’s due to the way I fitted the new nozzle but I’m inclined to think not as before all this kicked off and when I started to look at removing the existing nozzle it was covered in molten filament so much so that I had to clear it off before I could unscrew the nozzle. However, this was after many months of use whereas now it’s every time.

Anyway, I wonder if there is any instructions on the correct method of replacing the nozzle, such as the temperature of the heater block needed before insertion of the nozzle.
Also should the shoulder of the nozzle fit hard up against the heater block?
I plan to run it up to 220 Deg. and see how it goes.


There isn’t an actual guide or instruction set that I have ever seen. Just all anecdotal collected from these forum posts.

I am quite familiar with the oozing problem you are having. That comes from a nozzle that is not seated properly in the heat block. The molten filament is able to ooze out of the threads after a while and covers your nozzle in a hot mess. Eventually causing little crispy bits to start dropping onto your print as well, and if you are really unlucky, oozing out sideways under the tool head bottom cover and creating a nice mess under there.

The good news: You can most likely fix it.

The bad news: You are going to have to take the nozzle off, and the bottom plate, and clean everything spotless. I use a combination of hobby knives, tweezers, dental picks, etc.

Then you need to look inside the heat block. A loupe or magnifying glass helps. You will see a short piece of PTFE tube. Hopefully, it is intact and straight, not mangled and looking like Earthworm Jim on a bad day.

What you must look at very carefully is the very end of that PTFE tube and make sure it is square, that is where the base end of the nozzle needs to meet with it and that joint needs to be flush.

If all is good up to this point, you are doing well. If not, you may need to VERY CAREFULLY with the precision of a brain surgeon trim that tube. Be extremely cautious and be absolutely certain before doing this. In fact, post a pic here first and get a second opinion, because it is just like brain surgery there is no going back, and that little piece of tubing is difficult to source for replacement.

Hoping your tube looked ok, take a small piece of Teflon tape aka plumbers tape and wrap it around the threads on the nozzle, more toward the nozzle end, not the inserted end. Double check the label on your Teflon tape to ensure that it is high-temp resistant (they almost all are), and then put the nozzle back in. You don’t want so much tape that it becomes hard to install, you are just creating a little better seal.

Then put the nozzle back in. You kind of have to gauge by feel for the point when the nozzle bumps up to the PTFE tube. It should all be cooled off so you can reassemble by hand and feel for that contact point. Tighten it with your fingers as tighten as you can and then stop.

Now reconnect the Mini-USB C cable and heat it up, and using your wrench/socket/spanner give it a final 1/4 to 1/2 turn. Use another wrench to hold the heat block from rotating while you do this, and stop when it feels like it wants to stop, don’t overtighten it. I don’t have torque spec but it isn’t massive, you are tightening a tiny nozzle under heated conditions, no need to Hulk it.

That should solve the oozing problem or by going through the process, alert you to any other problem.

Hmm, I said at the beginning, there was no guide. Maybe I just wrote one. Maybe if I can get some input from a few other regulars on here, opinions, advice, experiences, etc., then I will add some photos and put together a PDF.


I think you just wrote it mate! :wink:

I concur with everything you said… and doubly so on the holding the heat block when doing that final tighten… for reasons I alluded too to earlier! :laughing:


When I get some time, I’ll try to cull a few posts from here and put together a PDF with some pictures.

Maybe I can submit it to the FLUX Team for review, would rather it come from them, or at least have them look it over too.


From time to time, I replaced the nozzle since filament won’t pass for some reason (unknown). But lately I got a raft that came loose from the print bed and stuck to the nozzle - which in return didn’t stop oozing out PLA.

Now I can gladly tune in here since I also experienced a clogged nozzle 8-O … the picture does not show the whole mess - most of the PLA chunk is inside the extruder housing, where it neatly engulfed the heater and its cabling. Sigh.

Finally, the rods came loose and presented what had happened overnight:



Took me hours to dismantle the extruder and carefully cut the mess to tiny pieces …


@pfeerick, go ahead and issue him the merit badge, that’s a pretty good one :laughing:


@BoozeKashi I wish I could :wink: That indeed was a good example of what goes wrong when the print detaches and follows the head around… Maybe someone needs to come up with a ‘Oh c$#p, the print’s attached to me’ detector for 3D printer print heads! (As well as a “wtf? where’s the print gone?” one for when the printer shoves it randomly off the print bed!) :laughing:

I have seen a photo from the forum for my other printer where it was located on top of of a filing cabinet, and PLA plastic is all over the front of the cabinet and on the floor, as the print shot off the front of the bed… and the printer just kept on goin’ … :laughing: