Layer height-dependent settings? Or something else?

[SouthSideofdaSky]

Hi everyone,

I have had a lot of success printing PLA parts with a 0.2 mm layer height. However for some of the parts I need to print, the 0.2 mm resolution is unnecessarily high so I've been trying a few parts at 0.3 mm layer height to save print time. At first, I thought that I could get nice prints by simply changing the layer height setting and leaving everything else the same. However, I've now realized that almost definitely isn't the case. I printed one part that was roughly 1" (X-dir) x 9" (Y-dir) x 6" (Z-dir) at 0.3 mm layer height with all other settings the same as my 0.2 mm layer height settings, and it turned out great. But now I'm thinking that was probably a fluke.

My problems began when I tried to print the part pictured below. The beginning of the print looked pretty normal but when I came back to check the print about halfway through, I noticed that all of the surfaces parallel to the Y-axis (the shorter faces the way I placed the print on the bed) had weird vertical ridges (first picture). The surfaces parallel to the X-axis (the longer faces) were normal. Also, the infill seemed to be sagging below the level of the perimeters. In the spirit of discovery I decided to let the print complete. The pictures below are of the completed print. It might be hard to tell from the pictures, but the same problems persisted - ridges in all surfaces parallel to Y-axis and sagging infill. The top surface is slightly recessed and a bit porous-looking as a result.

My next step was to run a test/calibration print with the same settings. I got the same result (image below, right - I stopped it mid-print). Then I ran the same test print with my 0.2 mm layer height settings and got a perfect print (image below, left). Finally I ran my part again with the 0.2 mm layer height settings and got a perfect print.

I feel like these observations indicate that the cause of my problem is something in my process settings for the 0.3 mm layer height. I am using S3D. Again, the settings I tried to use are exactly the same as the settings I have used to produce many successful prints at 0.2 mm layer height, except for the following:
- "Primary Layer Height" is 0.3 mm instead of 0.2 mm.
- "First Layer Height" is 60% instead of 90%, so that the actual layer height is the same (I am currently using this instead of the g-code offset to control the first layer height. It might not be the best way but I don't think that's my problem here...right?)
- All other settings exactly the same. Nozzle temp is 215 C. I have the V4 extruder.

So I guess my question is, what other settings should be changing when I change the layer height from 0.2 mm to 0.3 mm? Is my problem under-extrusion? Do I need to turn up the extrusion multiplier or width when I increase the layer height?

Thanks for any feedback/assistance.

[Jules]

There are some rules of thumb that apply, and they depend on several factors, depending on your nozzle size.

1. Make sure that your extruded thread width is larger than the diameter of the nozzle opening. So for a .35 mm nozzle, your thread width must be at or greater than 0.37mm. (That ensures that the machine has control over where the thread is placed.)
2. But, the upper limit on the width of a thread is the diameter of the flat face on the nozzle, which for the .35mm nozzle is .5 mm. So you have to set your thread width between .37 mm to .5 mm for a .35mm nozzle.
3. In order to get the correct bonding between layers, your layer height needs to be roughly half as tall as your thread width. (The actual optimum ratio for thread width/layer height is 1.8.)

In your case, you changed the layer height to .3 mm. So .3 mm x 1.8 = 0.54 mm which is what you would have to set your layer width to be, in order to work with a 0.3 mm layer height. (If you didn't change the width to match your new height, you are under-extruding.)

But that 0.54 mm number is larger than 0.5 mm limit on the width of the thread.

If you don't want the ridges, don't set your layer height above .28 mm, and set your layer width to 0.5mm to match. That actually pushes the limit though - you'd be better off with a width of 0.48 and a layer height of 0.27, which is something I'll use if i'm in a hurry and don't want to sacrifice quality.

[SouthSideofdaSky]

I have the standard 0.35 mm nozzle and right now I have the extrusion width set to 0.42 mm for both layer heights. So I'm good on your points 1 and 2. The 0.42 mm width is fairly close to the optimum for the 0.2 mm layer height (0.42/0.2 = 2.1) but 0.42/0.3 = 1.4 so that's probably too low and I was definitely under-extruding!

I will have to try the 0.48 mm width and 0.27 mm height. Thanks, Jules!

As an aside, do the same rules apply for a 0.1 mm layer height? I've played around with that a bit too and I'm thinking it's not worth dealing with but just asking out of curiosity. Since the width can't be smaller than 0.37 mm I'd want to set it to 0.37 mm, which would still give me a way-too-large ratio of 0.37/0.1 = 3.7?

[Jules]

Same rules apply - if you want to go down to a .1mm layer height, you'll want to try a smaller nozzle. (.25 mm). It can be done with the 0.35mm, but you will be printing flat ribbons, and it's going to take forever, and you won't gain anything on surface details at all.

I call them "rules of thumb", but they are actually the result of lots of industry-wide testing, studies, and several write-ups that the guys here directed me to when i first joined....i couldn't find them again if my life depended on it. (But i did make notes when i read through them. ) They actually spent the time to determine what the optimal shape of the extruded thread was (for best bonding and adhesion) and the best control by the machine. That's the 1.8 ratio, with the two provisions listed as far as the width of the thread.

Practically speaking though, anything over 1.8 works (within reason), so it's also okay to just set your layer height to be half of what the width is. (Easier to divide by 2.) Anything with a ratio under 1.8 (like your 1.4) and you'll get results that look like what you saw.

Keep in mind these are rules for optimal quality - you can print with settings that fall outside of these ranges. But you're going to get unexpected results sometimes.

[SouthSideofdaSky]

It can be done with the 0.35mm, but you will be printing flat ribbons, and it's going to take forever, and you won't gain anything on surface details at all.

Yeah that's more or less the conclusion I came to. It takes FOREVER (I swear it's way more than twice as long as the same print at 0.2 mm) and the quality difference between 0.1 mm and 0.2 mm isn't really noticeable in most cases. I don't want to deal with a different size nozzle at this point, so I think I'm going to just stick to larger layer heights.

Your explanation of the 1.8 ratio makes a lot of sense. Once I read your first reply, it sounded familiar like maybe I read one of the same write ups as you at one point.

Ran a calibration cube with 0.27 mm height and 0.48 mm width and it looks perfect. Trying a longer print now. Thanks again!