Underextrusion Issue Unsolved by Anything I Could Find Online

[mcmuffin6o 3]

Hello again everyone :/ Hello again @GregValiant, who has helped me many times and correctly predicted the last time I posted that it would be a while before I completed another successful print (I never even actually finished that earring). I've been struggling with underextrusion quite literally since the day I last posted in that thread. Now that I need my printer for a senior design project, however, the stakes in fixing it have been raised. I will do my best here to give as many accurate and pertinent details as I can, but please let me know if something doesn't seem right about my story or if I'm leaving anything out.

 

So the problem started on the stock hotend that came with my ender 3 pro. For several months it had been printing flawlessly, delivering consistently beautiful prints. Then, after I tried using a smaller nozzle, it became unreasonably difficult to push filament through the hotend. I realize that a smaller nozzle will restrict the flow, but in this case if I were to extrude 7 mm or so, it would build up so much pressure that when I released the tension on the extruder arm, the filament would spring backwards a couple mm. Even after I switched back to the stock nozzle, it was still unreasonably hard to push filament through.

 

After many nozzle swaps (that I ignorantly performed cold), my heater block was understandably leaky. I looked up how to swap nozzles because that's what seemed to cause the problem, and after learning my mistake, bought an entirely new hotend assembly, that came preinstalled with a thermistor and heater cartridge. (Note: I didn't install the aluminum extruder assembly as it was for the old style extruder). I installed it, and... the same problem cropped up after I tried printing with a smaller nozzle. It didn't matter that I cleaned it out by pushing the Bowden tube all the way through and switching back to a larger nozzle. For some reason it was once again unreasonably difficult to push filament. I tried doing the hotend fix, but that didn't fix it either. I gave up on my printer for several months, only to come back to it this winter break and find it inexplicably working.

 

Again, though, all of the sudden it just started underextruding after I tried to use a smaller nozzle. I don't even know for sure if it has anything to do with the smaller nozzle. The filament seems easy enough to push through for a few millimeters, but it quickly starts building pressure like it never used to. I figured maybe the air gap between the heater cartridge and the hotblock might have been preventing the necessary heat flow to melt filament at typical extrusion speeds, but I filled that gap with anti-seize and it still has the exact same issue. It's worth noting that the previous hotblock-heater cartridge assembly seemed to be filled with something thermally conductive.

 

The extruder does skip sometimes, but only at extremely high pressures and not without putting up a good fight. It's really pushing and gripping hard. I don't think it's the filament either, because I've tried different PLA filaments at different temperatures to no avail. I tried printing on the lower end, higher end, and somewhere in the middle in terms of temperature. I've ordered a bimetallic heatbreak (that's right, this time I didn't cheap out) that's set to come in on Friday, and I'm just going to carefully install it and see if that fixes my problem somehow. I know it's dumb and expensive to just swap parts and see what works, but that's where desperation has taken me.

 

Do small nozzles permanently ruin printers? This thing is driving me nuts. What is going on here??

[GregValiant 1,245]

My story is similar.  The stock hot end on my Ender 3 Pro was fine for about 6 months of heavy use, and then bingo, constant clogging at the gap between the hot end and the bowden tube resulting in under-extrusion.  I found a coat hanger that was the right diameter, cut off a piece, and I'd dismantle the hot end and shove the coat hanger down through it.  It always pushed out a plug of plastic that should not have been there.

That plug was causing a restriction in the flow.  My best guess was that the friction it was causing was the problem.  The extruder did not miss steps but my prints were a spider web.

So I bought a direct replacement hot end that was a clone of the stock hot end.  I did the Auto-Tune and adjusted the PID and then 2 or 3 hours into a print - under-extrusion.  This happened every time.  I had changed the bowden fittings to the fancy ones from McMaster Carr.  The tube itself is a Capricorn.  The extruder is fine.

 

I came up with this saying:  "When you clone a POS you should not be surprised the clone is also a POS."

So now I have an all-metal hot end, a 4015 ball bearing fan blowing on it, a baffle inside the hot end cover to aim the air at the heat sink, and two 10mm holes drilled into the backplate to let the air out.  The problem has been fixed and the machine prints as it should.  A speed test shows it can put down plastic without a problem up to 175mm/sec with a .4 nozzle.  It isn't maintenance free and every 30 to 40 hours of printing I pull the bowden out of the hot end and trim it back 5 or 6mm, clean the nozzle (although I don't believe I've ever had a "clogged nozzle") and run that same piece of coat hanger through the heat sink and heat break.  It always pushes out a piece of plastic but it is greatly reduced from what used to be there.  I finished an 8 hour print yesterday and it came out beautifully.

 

To answer your question regarding small nozzles, the answer is NO.  Because of the Fluid Dynamics of the flowing plastic, there is a lot more back pressure with a .2 nozzle compared to a .4 nozzle.  The same is true if you compare a .4 to a .6 or .8.

The material clings to the inside wall of the brass and there is a transition zone between the material that is stuck to the wall and is barely moving, to the material in the center core of an extrusion that is moving right along.  That transition zone is always the same thickness but on a percentage basis, it is a much higher percentage of a .2 than it is of a .8.  We see the same issue in air flow through a carburetor or water through a fire hose (both of which I am intimately familiar with).

 

"All-Metal hot end".  The nozzle size doesn't matter, but print speed does so if you want to print with a .2 nozzle you have to slow down.  The E-steps must be calibrated.  Yes, the hot end needs to be near operating temp when you change nozzles.  You can try to push the print temperature in an effort to maintain print speed, but it's a band-aid fix and can have it's own side effects.

 

It prints so well that I decided I deserved a beer and a cigar (the bracket is an IPod mount for my motorcycle).  The baffle on the hot end keeps strings out of the blades.  Go ahead and zoom in.  You won't find many errors.

Edited February 3, 2022 by GregValiant

[mcmuffin6o 3]

Christ, okay. You really went turbo on the thing. Weird that I could never find plastic between bowden and nozzle when I took apart the hotend to look for such things. There's a lot of good advice in your post, and hopefully it will allow me to get more consistent results.

[mcmuffin6o 3]

Okay so after trying the new Creality Spider all-metal hotend, I still had the same issue. I decided to try heating the thing to 240C for PLA since the heat creep shouldn't be much of a problem, and the issue went away! I think maybe the PLA absorbed water and required higher printing temps as a result, but I'm also very suspicious of the fact that the same exact filament would produce gorgeous prints at 195C and now requires 235+C to even get out of the nozzle. Speaking of which, here is a comparison between one of the times the printer decided to work at 195 with the stock hotend (maybe this was when I opened a new spool??? I really don't think so but I can't be sure.) vs. now having to print at 240C. Note: these were printed with IDENTICAL gcode that I only modified the temperature for. Also note: I didn't even notice the underextrusion (not part-ruining underextrusion, I think someone had convinced me to turn flow down to 95% for no good reason) on the nice cube until taking these close-ups!

 

p.s. Sorry in advance for my absolutely fucked up nails, it's been a rough start to the semester and I haven't had the time to care for them.

Edited February 13, 2022 by mcmuffin6o
added bottom photos

[mcmuffin6o 3]

I need to write this down before I forget... I might be seeing things since it's almost 5am, but the quality of prints seems to be increasing as I print. Perhaps I've burnt through the outer layer of waterlogged filament and the dryness is improving. This is just another hypothesis on top of the water absorption hypothesis, but worth writing down in the thread at least.

[GregValiant 1,245]

It's always humid here.  Today it's 97%.  I've found that if I leave the spool on the machine instead of properly storing it in a Ziploc with anti-desiccant packets, it will get wet.  It prints poorly when it's wet and dryng helps, but causes the filament to get brittle.  It prints at the same temperatures, just really poorly.  PETG is really bad.

 

If you are getting good prints at 240 then print at 240.  When my printer has sat overnight and I turn it on, the bed and hot end are 3 degrees different.  That speaks to the quality of components used in the machine.  That can't even agree on what the room temperature is.  The only way to check is to use a handheld infrared thermometer and verify what the hot end temp really is.  In the end it's just a number displayed on the screen.  I have a feeling that the LCD might say 240, but that the hot end is really closer to 210.  What's important are the results.

 

The business end of the thermistor must be in the hole in the side of the heat block.  The little screw next to it should be just tight enough to keep it from falling out, but not so tight as to damage the insulation of the wires or they will short and cause glitchy results.

 

And take some time to do your nails while you're waiting for the next print to finish.  Geez.

[mcmuffin6o 3]

> quality of the components

Crazily enough, the new thermistor that comes with the Spider hotend seems to agree with the bed! I am tempted to get an infrared thermometer, though.

 

> If you are getting good prints at 240 then print at 240

That's the thing--you can tell that the prints have bumpy walls and uneven lines. Not to mention that the plastic ends up much shinier and is less aesthetically pleasing.

 

> What's important are the results

True, I've been learning this more and more as this goofy machine seems to defy so much general 3D printing advice online. I will have to let my mind think outside the box.

 

> The business end of the thermistor...

On the Spider hotend, the high-temperature thermistor is cartridge shaped, and secured with a set screw, just like the heating element! I broke one thermistor on an old hotend, so I sure am glad not to be dealing with a crappy glass bead anymore.

 

I'll have to get a thermometer, it'll be one more thing to rule out...

 

Nails will get done soon. First, some sleep 😴

[gr5 2,167]

I read every word above.  some thoughts:

 

This could be unrelated but sometimes the cooling fans are aimed badly and right at the nozzle.  silicone "socks" can solve this for many printer types.  The symptoms often include better extrusion as the nozzle gets farther from the print bed (which bounces the air back up into the nozzle).

 

You mentioned bad surface contact with the nozzle?  Not sure what you mean but this is pretty important.  You want the nozzle almost as hot as the heater block.

 

Regarding nozzle diameters.  You didn't say the diameters but a 0.2mm nozzle is 4X less area than a 0.4mm nozzle (take the ratio of the sizes and square that ratio - so a 2x smaller nozzle is 4X less area).  4X is a lot.  I recommend printing at similar "speeds" but cut your layer height in half and your line width of course in half.  That gives you 1/4 the volume.  Or you can keep the layer height the same and cut the "speed" in half.  However it may be that you never reached the full speed in the past because you were limited by acceleration.  So maybe cut the speed by more than 2X.  Again - simpler to be consistent and leave the speed the same but cut layer and line widht each by half.

 

[GregValiant 1,245]

Here is one of your photos that I zoomed in on.  You can see the 2nd layer below through the top layer.  That is some serious under-extrusion.  Have you calibrated the E-Steps on the printer?  All those extrusions should be welded together.  With black filament you should not be able to see through a layer.

Edited February 13, 2022 by GregValiant

[mcmuffin6o 3]

5 hours ago, gr5 said:

silicone "socks"

The new Spider hotend comes with one, and I have it installed.

 

5 hours ago, gr5 said:

You mentioned bad surface contact with the nozzle?

Nope.

 

5 hours ago, gr5 said:

Regarding nozzle diameters...

Yep, I know. I printed at an appropriately lower layer height (and width, of course) for the smaller nozzle, and I'm pretty sure I had tried slowing the speed as well. There were times where I could push filament through with my hand easily enough, but when it started clogging/jamming it was impossible to do so.

[mcmuffin6o 3]

5 hours ago, GregValiant said:

That is some serious under-extrusion.

Yep, but at least that time the printer was able to print. When I have the crazy underextrusion issue, the printer won't even lay down the first layer. I had both my esteps and flow too low for that print, but now I've recalibrated both for the new hotend and the prints come out nice and opaque.

[mcmuffin6o 3]

I think the next move is to print a temperature tower. Perhaps multiple, at differing speeds. Now that I have a presumably high quality hotend, I can err on the side of blaming myself. Either way, I can at least lay down plastic now, which is a huge relief after all of the time and money I've spent.

[mcmuffin6o 3]

Update: I'm just an idiot. The entirety of the temperature range printed fine. Pictures tonight