Let's talk about those hollow test cubes please.
In another thread (I'm starting this one as to not hijack that one) another user suggested using a hollow test cube to check and dial in extrusion.
I then realized that never in my 3d printing 'career' have I ever printed one, or ever felt the need to print one. I don't remember them ever being discussed in the classes I took on 3d printing some time ago.
So, I quickly drafted a variant of my sanity-check 30mm cube which is, as the name implies, hollow. I just did a quick cut between the 30mm cube and a 29mm cube, leaving walls of .5mm, and assuming that each wall would print as one .4mm stroke.
I hate to admit it, but I don't know what I'm really looking for here. What I expected and what I got was a stack of single-stroke beads 30mm high.
Yes, there is some roughness and some inconsistency, but no blobs nor holidays. The roughness really shows on two of the corners, but not so much on the others, and the whole cube is visibly quite rougher than what I normally expect when printing a 30mm sanity check cube.
Any comments ??
RE: Let's talk about those hollow test cubes please.
Let’s not. It would leave me with a hollow feeling…
How about finding a pre sliced test cube and going from there.
The issues with your sanity cube could be a slicing problem and not a printing issue. So take the slicer out of the discussion by finding a presliced test.
RE: Let's talk about those hollow test cubes please.
Check out https://help.prusa3d.com/article/extrusion-multiplier-calibration_2257 . It comes with an STL and precise instructions. I suspect some of what you're seeing could be caused by a width of 0.4, when the default extrusion width is 0.45 mm.
Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...
RE: Let's talk about those hollow test cubes please.
Another source for this calibration that I really like: https://teachingtechyt.github.io/calibration.html#flow
What is it like if you print the same cube in vase mode?
Eveything following applies to Mk3, I expect it to remain relevant for Mk4.
Until a few days ago I too had never printed a calibration cube (and no, I've still not printed a Benchy) - but a user had an issue with a cube that I hadn't seen before so I did finally print one - and discarded it. The user's problem was elsewhere.
I have always just printed whatever part is next in my queue, as they are almost all my own design I know the expected result and so I measure and calibrate on the fly.
The very few times I have tweaked extrusion have all been less than optimal, instead it's much better to measure the filament's true diameter and enter the actual value, this does lead to a stock of leftover profiles for particular spools that have been used up; note to self, time for another clearout. @neophyl has done a lot more work on this and can explain how the calculations are based.
Other than for testing the strength of a part in different orientations I have never understood the strange desire to print endless tests rather than beginning with less critical parts and slowly working up to precision as it becomes attainable.
Cheerio,
RE: Let's talk about those hollow test cubes please.
Other than for testing the strength of a part in different orientations I have never understood the strange desire to print endless tests rather than beginning with less critical parts and slowly working up to precision as it becomes attainable.
Cheerio,
In my dabbles with photography, I've found there are essentially two types of hobby photographers. Those whose hobby is the photography, and those whose hobby is the camera. There's obviously some overlap between the camps, but I see a parallel with 3D printers.
RE: Let's talk about those hollow test cubes please.
@jseyfert3 — Well spoken. Those who like to tinker with their printers, and those who like to print stuff (with a subgroup of those who also like to design).
Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...
RE: Let's talk about those hollow test cubes please.
Ok, I think I get the hollow cube test now.
I reprinted it, or rather resliced it in vase mode and reprinted it and it came out much smoother, which figures, because it does not have to stop and restart on each layer.
I do not have a micrometer, but a digital caliper (I know, not terribly accurate at the 0.45mm range) and it shows a wall width of 0.44mm very consistently, so I'm confident that the machine is in the ballpark.
I have never gone as far as to measure the actual diameter of a particular filament and adjusted the flow/extrusion multiplier to compensate. Every filament I've used on the Prusa has given good prints, and overall the machine has been a very reliable workhorse over the past 3 plus years.
This was the first time I've ever printed hollow cubes. I've never felt the need before.
Likewise, I never printed a Benchy on the Prusa until I did to sanity check another user's print ('buldge' in a particular position) using the same exact Hatchbox ABS filament. We printed those in one of the classes I took (we never heard the term 'Benchy', the instructor called it 'Boaty McBoatface' and it was indeed to demonstrate a challenging item to print) so I just never did it on the Prusa when I got it.
I also get it WRT the photographers being more obsessed with gear than with their images. I've always regarded the printer (and the camera) as a means to an end, and not the end itself. I guess that's one reason I'm not so obsessed with having the XL or MK4 or even upgrading.
I have never gone as far as to measure the actual diameter of a particular filament and adjusted the flow/extrusion multiplier to compensate. Every filament I've used on the Prusa has given good prints,
In the past I bought very dodgy, cheap filaments in lots of 10 or more for early prototyping - for a couple of years I used very cheap, surgical appliance pink, Chinese PLA that was fine for ordinary parts but for anything print-in-place the diameter had to be entered or the tolerances went haywire.
These days filaments are rarely more than 0.02mm out so the effect is less obvious 'though it's still there.
Cheerio,
RE: Let's talk about those hollow test cubes please.
I would say that over 90% of my filament purchases over the years have been via Amazon, sticking mostly to the brands I know, but trying new ones off and on. I'm sure that they will drop any products that result in many complaints. I've found that their Amazon Basics line prints very well.
I admit that I've never actually measured the diameter of anything, and I don't remember any brands/types which were difficult to print.
The cheapest I think I've ever paid was for a spool I ordered today for a project for my wife, and that was MG Chemicals ABS for US$12 and change. I've used MG Chemicals filaments before and they have always been just fine.
The only brand/type which needed any tweaking of anything is the Mika3d metallics, and I always goose the nozzle temp maybe 5-10 degrees to prevent jamming in the extruder. It prints just fine with the higher temp.
The only brand I remember being disappointed with was Solutech, and that was because they obviously used less pigment than the other brands, resulting in more transparency than I wanted.
RE: Let's talk about those hollow test cubes please.
I buy mostly Overture and other brand names. After a few times measuring the diameter I stopped doing it as it was always spot on. But I'm sure cheaper filament will be a different story.
Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...
RE: Let's talk about those hollow test cubes please.
I've never measured filament diameter proactively. There have been a few times I have measured it because I was having issues and wanted to rule that out and in none of those cases was the filament at fault. That being said, I tend to stick mostly to brands I trust like Overture, Hatchbox, eSun, Inland, 3dFuel, Prusament, and lately Numaker because I'm very impressed with their PLA+ which is Ingeo 3D850 with no fillers and is at a great price if you buy in batches.
RE: Let's talk about those hollow test cubes please.
Usually I measure the diameter of the filaments at the beginning of the spool. Prusament is always very consistent. PLA ranges from 1.73 to 1.74mm, the only exception being the blend colors that tend to be a bit thinner (1.71-1.72). Prusament PC and ASA are also on the 1.71-1.73 range but, having purchased only a few rolls of these, I can't say if it's the general rule. I haven't found a single spool of Prusament larger than 1.74mm or thinner than 1.71.
Other brands vary wildly. Sometimes they're 1.81mm and next spool of the same brand and color is 1.72.
RE: Let's talk about those hollow test cubes please.
I do usually measure mine, especially new brands. On a recent thread about gitd filament elsewhere on the forum the OP was measuring 1.68 so it does pay to check.
Even if you take 1.71mm which seems to be a common size (from personal experience and also values posted here) then that is approx 2.28% undersize. So that is a small bit of under extrusion. Most people will immediately adjust their extrusion multiplier value but its better to change the filament diameter in your filament settings as filament diameter is the basis for every single calculation the slicer makes regarding volume of extruded material.
If all your printing is general stuff then it doesn't matter too much, but if you need a bit more accuracy with parts that fit together like snap joints etc then getting the foundation value correct before you start to tweak other settings is worth while.
For me a 'Quality' brand isn't one that is perfectly on size at 1.75 (although thats nice), it's a brand that is consistently the SAME size. That way you can just do a profile for that filament and not have to worry about it.
RE:
What, you don't like paying ridiculous shipping and import tarrifs? 😀
The very first thing I thought of in your original post was vase mode. You figured that out. Good!
I would say that over 90% of my filament purchases over the years have been via Amazon, sticking mostly to the brands I know, but trying new ones off and on. I'm sure that they will drop any products that result in many complaints. I've found that their Amazon Basics line prints very well.
I admit that I've never actually measured the diameter of anything, and I don't remember any brands/types which were difficult to print.
The cheapest I think I've ever paid was for a spool I ordered today for a project for my wife, and that was MG Chemicals ABS for US$12 and change. I've used MG Chemicals filaments before and they have always been just fine.
The only brand/type which needed any tweaking of anything is the Mika3d metallics, and I always goose the nozzle temp maybe 5-10 degrees to prevent jamming in the extruder. It prints just fine with the higher temp.
The only brand I remember being disappointed with was Solutech, and that was because they obviously used less pigment than the other brands, resulting in more transparency than I wanted.