Catastrophic layer adhesion with 0.2mm nozzle

[CBX_Micha 2]

Hello dear community,

 

I'm at my wits' end here and hope for some advice and tips from the pros...

 

I'm trying to print this spare part: https://www.thingiverse.com/thing:1098950

 

I've used a 0.2mm nozzle and am using black PETG (Geeetech).

After printing a TempTower, amazingly, two quality peaks were achieved. Once at 220°C and once again at 245°C. Since my first failed attempts ran at 220°C, my last three attempts ran at 245°C.

 

The speed was a slow 40mm/s in each case. I varied the cooling (10, 30, 80%), with and without "alternating additional walls", flow rates of 95-110%.

But no matter what I do, the part is completely unstable and breaks at the slightest load between two or more layers.

 

What am I doing wrong here?

[gr5 2,224]

First I would try turning off the fan and also maybe try to enclose the print volume if that isn't too difficult.

 

Breaking along layer lines is sometimes due to a layer of bad extrusion but can also be an adhesion issue when the new fresh layer being put down form the nozzle is not melting the layer below.

 

A different possible problem may be that your layer heights are too thin for your accuracy of your z screw and some layers may have moved too far so you may be severly undrextruding on those layers.  Doing thicker layer heights might help.  I like to keep the layer height similar to the nozzle size so I wouldn't go any thinner than 0.1mm layer height with a 0.2mm nozzle although I might even go with .15 layer height if my z axis isn't moving consistently.  Sometimes cleaning the Z screws helps to get more consistent Z movement. 

 

What kind of printer are you using?  Is it really a 0.2mm nozzle?  That's unusually small but I've printed with 0.25 before.

[CBX_Micha 2]

Many thx for the hints!

 

My printer #1:

This printer was a Geeetech I3 Pro Crystal in its first life, but has since been massively modified, as you can easily see.
The Z-axes are each driven by a stepper with its own driver, and are also coupled at the top via a toothed belt.
The bushings are anti-backslash nylon bushings, with additional support blocks (orange) on both sides, which make the whole thing practically completely free of lack and largely prevent the X-axis from nodding.
The entire extruder block is a home-made construction made from a geared dual extruder, which directly feeds a Mosquito clone (all metal), which is equipped with a 70W heating cartridge.

 

On the subject:
I cannot use a housing. I tried that once, but since all the electronics are also in the housing, it all gets far too hot.
I would actually like to rule out poor or uneven extrusion because of the dual extruder.

My layer height is currently set to 0.1, the layer thickness and the rest to 0.2.
And yes, I print fine and small parts with a 0.2 nozzle, which has always worked without problems so far. I also tried a 0.1mm nozzle, but that only causes problems. 0.15 just about works, but 0.2 doesn't really cause any issues. I use nozzles of 0.2, 0.4, 0.6, 0.8 and 1.0, depending on what I need.

At the moment I'm trying 30mm/s and the fan at minimum (5%). I can't turn it off completely, otherwise the radiant heat will melt the fan mounts. I'd have to unscrew them first...

[CBX_Micha 2]

Another realization:
Obviously, the print becomes more unstable the slower I print it?!
I printed at 20mm/s and didn't even get it off the bed in one piece. It immediately fell apart into several pieces...
That somehow doesn't make sense to me...

[GregValiant 1,342]

How confident are you in that roll of PETG?  I had a roll that was heavily moisture laden right out of the package.  It has only happened once but it did happen.  It was Sunlu PETG through Amazon, the package was sealed and there was a pack of silica gel in the package, but that filament was just junk.  I ended up throwing it away as it wasn't even good for doing tests for questions here on the forum.

Edited August 26 by GregValiant

[CBX_Micha 2]

... comes out of the dryer and is at about 35%. But tomorrow morning I'll try a different roll... At least that's an idea that I'm slowly running out of... Totally weird shit ^^

[Slashee_the_Cow 410]

It's worth remembering that layer adhesion is dependent upon surface area: if you're using a 0.2mm nozzle and (presumably) thinner lines than a 0.4mm, you'll need more walls to compensate. And the layer height doesn't help, either: you have to remember, your filament is coming out of a circular nozzle and is usually laid down as a tubular shape:

The lower the diameter of the tube, the less contact area there is between each one.

 

0.2mm nozzles aren't usually used for functional prints, just pretty ones.

[CBX_Micha 2]

Yes, I know that. But if the flow rate is set correctly, the round sausage will become a flattened sausage when applied, which ideally has a surface on the contact sides (top/bottom) that corresponds to the nozzle diameter, with bulges pushed away to the sides. And if there is already a line, the side of the sausage that has just been laid should also fit snugly against the material of the side bulge.

As I said: I have already printed many parts with a 0.2 nozzle, with PETG and PLA. I have never had such problems before. I even measured the real temperature at the nozzle again, but since switching to PT1000, nothing has drifted anymore...

I will try a different, new filament today... We'll see...

 

EDIT say:

I print it right now with a used gray Filament. Same values as the last one except speed what I have set at 60mm/s ... We will see...

Edited August 27 by CBX_Micha

[Slashee_the_Cow 410]

1 hour ago, CBX_Micha said:

But if the flow rate is set correctly, the round sausage will become a flattened sausage when applied, which ideally has a surface on the contact sides (top/bottom) that corresponds to the nozzle diameter, with bulges pushed away to the sides.

Ideally it'll have a surface that corresponds to the intended line width 😏
But remember that the size of each sausage changes multiplicatively with layer height and line width, so your sausages from a 0.2mm nozzle are going to be a lot smaller than ones from bigger nozzles.

 

And feel free to give me an answer that makes this seem like a stupid question, but... why are you using a 0.2mm nozzle for something whose finest detail (which is mostly decorative anyway) is about 0.5mm in size?

[CBX_Micha 2]

14 minutes ago, Slashee_the_Cow said:

And feel free to give me an answer that makes this seem like a stupid question, but... why are you using a 0.2mm nozzle for something whose finest detail (which is mostly decorative anyway) is about 0.5mm in size?

At 1st: Among halfway intelligent people there are no stupid questions, only stupid answers. At least that's my view of things 😉

Why? There are several reasons. Firstly, I want the surface to be as fine as possible, and secondly, with a 0.4 nozzle it is almost impossible to get all the hooks and tenons to fit perfectly without a lot of rework; I've already tried. And last but not least (yet another stupid answer): Because I can, or, as here, at least want to be able to 😉

[Slashee_the_Cow 410]

Something obvious I just realised hasn't been said (we all have our moments when we forget the obvious): your filament is dry, right?

[CBX_Micha 2]

4 minutes ago, Slashee_the_Cow said:

your filament is dry, right?

LOL sure... Around 35%

[Slashee_the_Cow 410]

26 minutes ago, CBX_Micha said:

LOL sure... Around 35%

*rereads thread* Oh wait... you did say that. Never mind 🙂 

The humidity isn't really important... what's more important is that you need to keep it nice and warm for a few hours. If you're using a dryer and want to worry about the humidity just prop it open a little bit for 15 minutes or so.

[CBX_Micha 2]

38 minutes ago, Slashee_the_Cow said:

*rereads thread* Oh wait... you did say that. Never mind 🙂 

😁

 

... the spools are stored in a nearly airtight Eurobox together with 200g of silica and a small fan that keeps the air moving inside. It's always between 30% and 40%, depending on how fresh the silica is and how often I open the box.
If I've forgotten a spool outside, it goes into our small gastro convection/grill/microwave. The temperature in it only deviates by +/-3°C from the set temperature and is amazingly accurate. I then throw "suspicious" rolls in there overnight at 60°C and in the morning, as warm as they are, put them back in the Eurobox...
It's worked so far. And as soon as the room is finished, which is now empty because my daughter moved out, the Eurobox will be expanded so that I can print directly from it without having to open it every time. But that will take a while; too many other projects ^^

 

 

EDIT say:

I had to stop the test print with the grey filament. It tipped the component over because I had somehow forgotten the brim; I realized too late and hoped that it would work without it... Oh well...
What I could see, however, is that it is probably due to the flow rate. In Cura I had already set it to 105%. After about 80 layers, I then manually set the flow rate in Repetier to 110% on a whim, which ultimately results in a real flow rate of 115.5%. It is noticeable that the outer skin has become much more uniform, without clearly recognizable layers.
So there seems to be a problem here with calculating the correct flow rate depending on the set nozzle diameter. The catch must be somewhere in this area...

 

EDIT also say:

Back to black with 115% flow rate and 60mm/s... Now I'm curious...

Edited August 27 by CBX_Micha

[Slashee_the_Cow 410]

25 minutes ago, CBX_Micha said:

the Eurobox will be expanded so that I can print directly from it without having to open it every time

Just don't make the same mistake I did.

Running your filament through a tube to get it from the filament dryer, through the flap of your tent and into a custom tube holder I designed that sits on the top gantry and is attached by velcro cable ties: good idea.

Running your filament through a metre of Bowden tubing (it was even Capricorn) to make said journey: bad idea.

 

Tube still pits up a bit of resistance. Extruder is not designed to pull filament through that much tubing. Being a direct drive extruder I'm not sure it's designed to pull filament through any tubing. Underextrusion (and things putting up so much resistance it actually moves the filament dryer closer to the tent) ensues.

Solution: Got some 3mm ID vinyl tubing designed for irrigation. Small enough the filament's safe and can't go off course, or kink up or anything. Big enough that it's not being strangled the whole way through.

 

3D printing: whoever can best misuse something from a hardware store wins.

[CBX_Micha 2]

39 minutes ago, Slashee_the_Cow said:

Tube still pits up a bit of resistance. Extruder is not designed to pull filament through that much tubing

... and that's exactly why they all get their own feed on the box, which is controlled by the filament tension on or near the extruder 🙃

[gr5 2,224]

Printing slower allows the already-printed-layer to cool more so you get worse adhesion.  Did you try turning off the fan?

 

Also you only need to keep the upper reaches of the printer warm - the area where you are printing.  35C is plenty warm to make a big help with adhesion and should still be safe for electronics.

[Slashee_the_Cow 410]

6 hours ago, gr5 said:

Printing slower allows the already-printed-layer to cool more so you get worse adhesion.  Did you try turning off the fan?

Believe me, I'm not trying to say "Slashee knows best, don't use a 0.2mm nozzle 😉" here but Slashee does know that a 0.2mm nozzle is going to take longer to do a layer, giving the lower layers more time to cool 😉

 

But especially if you're going to increase the speed, consider turning down Travel > Minimum Layer Time since that'll slow it down if a layer is going to be too quick so if you want to keep things going while things are nice and gooey, you don't want that. Unless you get too many layers still being warm and gooey at once then it might not stay straight.

[CBX_Micha 2]

OK, a quick update:
Flow rate at 115% helped a lot. The part is very stable and can no longer be broken easily!
But now I have to repair the printer itself. The motor bracket of the Y-axis is broken... Again ^^ I'm building it out of steel now... It might take a while.

[gr5 2,224]

You probably had some underextrusion then if changing flow to 115% helped.

[CBX_Micha 2]

I will measure this again specifically with the 0.2 nozzle as soon as I have repaired the printer, but I suspect that the problem does not lie there. Because with all other nozzle diameters the extrusion amount is pretty accurate.

 

[Slashee_the_Cow 410]

6 minutes ago, CBX_Micha said:

I will measure this again specifically with the 0.2 nozzle as soon as I have repaired the printer, but I suspect that the problem does not lie there. Because with all other nozzle diameters the extrusion amount is pretty accurate.

The extruder does need to be a lot more precise with its extrusions for a 0.2mm nozzle.

 

On 8/28/2024 at 4:08 PM, CBX_Micha said:

The motor bracket of the Y-axis is broken... Again ^^ I'm building it out of steel now

I'm 102% sure that you know more about working with steel than I do.

[CBX_Micha 2]

9 minutes ago, Slashee_the_Cow said:

The extruder does need to be a lot more precise with its extrusions for a 0.2mm nozzle.

... that's true. We will see, if the geared dual extruder can hold that...

 

10 minutes ago, Slashee_the_Cow said:

I'm 102% sure that you know more about working with steel than I do.

I have no idea, as I don't know how much experience you have in working with steel.
But as a (practical) mechanical engineer and car/motorcycle/excavator mechanic since I was 12, I think I'm pretty good at it...