BaSzErr - blog:2024:01:07

2024-01-07_-_filament_cleaners

anonymous@undisclosed.example.com (Anonymous) — Sun, 07 Jan 2024 14:44:44 +0000

2024-01-07 - filament cleaners

since many years i'm using a simple method to reduce nozzle logs – cotton pads to clean the filament. the idea is simple: use cotton pad, ~loosely tide with a zip-tie.

despite it's obvious simplicity it works astonishingly well! this is how example cotton pad looks like after processing 1kg spool of filament.

despite i always cover my printer with a sheet when unused, and keep filaments in bags, when on shelve, there's still surprising amount of dirt captured. if not for the cotton, it'd all go to 0.4mm nozzle and clog it in no time.

all good – can't complain about results. solution is super cheap, too even tough you need a new cotton pad and zip-tie for each filament. i was wondering how does it compare to commercial filament cleaners, though.

some time ago i was doing some shopping online, and was missing like 1 EUR for a free shipping. so i decided to buy a commercial cleaner and give it a shot.

the nice part is that you can wash it every now and then and re-use multiple times. there are however 2 drawbacks.

it needs to be mounted before filament goes into the tube.
it is super-high friction.

ad.1 well – i tend to forget things, and installing a filament, just to realize that it needs to be removed again, to mount the cleaner was not great. this is however not a big thing, as over time i'd likely get used to it.

ad.2 however is a real pain. compared to cotton pad that's basically 0-friction, the commercial solution actually required some real force to move around. the problem is clear – this puts extra strain on extruder, that needs to be overcome. in my case the difference was large enough i needed to tighten a filament-pressing screw, because at high speeds i filament could get stuck. grinding filament with extruder was not a good choice, as it offered less uniform filament diameter when entering nozzle.

the long story short – after truing it out, i'm definitely staying with cotton pads!

for some more 3D printing tips, see my 3D printing 101.

2024-01-07_-_fluffy_bird_s_house

anonymous@undisclosed.example.com (Anonymous) — Sun, 07 Jan 2024 14:23:02 +0000

2024-01-07 - (fluffy) bird's house

about a week ago my sons saw i was preparing a workbench on balcony, so that i have a spot for doing work that's more dusty (eg. cutting) or smelly (eg. painting / coating). they got turbo-charged into woodworking for the weekend. ;) they decided to build small birdhouses for their small, fluffy birds. after a rough design according to our little customers' requirements, we bought some wood and in no time kids were cutting it into pieces of a predefined length! there was however a challenging part - due to expected house size vs. physical space for it, we had to do root at ~ random angle, that's not only hard to cut but also challenging to secure in place, to the rest of the house. to make things “easier” chimneys were also expected to be present. ;)

cutting part we skipped, and just cut in straight pieces. mounting brackets i took upon myself. i decided to print these on 3D printer, to make sure they match exactly what is needed.

the question remained - how to get a sample of all needed angles? the problem was that all wood pieces were hand-cut (in parts by small children), and thus “not exactly uniform”. ;) it was also necessary to compensate for screw holes locations.

i therefor decided to make a reference template, by using flatbed scanner to get a soon-to-be-house all captured. because of the size of wood pieces (1cm high and 5cm wide) to was not possible to close scanner, so there are some random light reflections coming. yet the core elements is clearly visible.

after post-processing and removing all the noise + binarizing the image, PNG was there!

the next step was to covert it into an SVG and import into OpenSCAD, as a template for creating all the elements. with this, creating proper brackets and hole locations were trivial.

the nice part about having a parametric 3D model, is that it's easy to customize… and that was a thing, as one son wanted to use screws while the other felt like hammering it with nails. ;) so i printed in 2 variants as different holes for screws and nails were needed.

mounting was a blast and really easy, as elements naturally fit into locations. in case it was ever needed again, fluffy birdhouse is now preserved on github. ;)

btw if you're interested in other techniques of combining real-world objects and 3D prints take a look at 3D printing enclosures entry (it's focused on PCBs, but some tricks are more widely available).

2024-01-07_-_keeping_filament_dry

anonymous@undisclosed.example.com (Anonymous) — Sun, 07 Jan 2024 15:19:36 +0000

2024-01-07 - keeping filament dry

different filaments have different needs. PLA is very forgiving when it comes to storage conditions - just make sure it does not cross 60 deg.C and you should be fine.

PET-G however is a bit more tricky, as it's super-gyroscopic. while it is not an issue, after the print is finished, it's a huge deal when filament is not try. to dry it's enough to put the spool into an oven at 60-70 deg.C with fan on, for a couple of hours. the problem is that moisture will start to build up again, after you remove it from the oven. in my experience it's enough for 1-2 days in 50% humidity (typical around summer time in Poland) to ruin the print and filement properties (eg. forget about printing watertight things with such a filament).

in order to keep it dry over prolonged periods of time, i built myself a box for 2 spools. bottom is covered with moisture-absorbing infill. filament leaves via dedicated bowden tube, that's connect to the printer's extruder.

this keeps 18-19% humidity around the filament.

last summer was the first time i decided to change the moisture-absorbing grains, after over 2 years of usage (including occasional spool changes). with the box i never had an issue with moisture again. as a free bonus, since the filament is kept ~ sealed there, it does not capture dust! :)

i still keep PLA w/o any protections for printing spool, i do keep rest of it sealed in vacuum bags (like the ones you can get in general store for sheets: cloths storage, etc.). however my PET-G now always goes through cycle of drying and putting into dry-box before usage.

2024-01-07_-_whiteboard_talks_vs._remote_work

anonymous@undisclosed.example.com (Anonymous) — Sun, 07 Jan 2024 15:05:08 +0000

2024-01-07 - whiteboard talks vs. remote work

for quite a few years i work remotely. while i generally prefer to have it this way, there are some advantages of being in the office. one of the missing bits are discussions over a whiteboard – 2 ppl, 1 whiteboard, and a pen. that's much harder to reproduce remotely. there are however some good approximations.

my take was a high-quality, yet very affordable (~50 EUR?) drawing tabled Wacom One S.

it's very nice to use, works under linux out of the box (kernel support), plugs into USB… and is pressure-sensitive! this means that you can just press harder or softer to get thicket and thinner line – just like with a real paintbrush!

this little thing, combined with a drawing tool like MyPaint made a huge difference on my remote working experience, whenever there's anything that'd require drawing / conceptual discussion over a video conference.

one thing that you might want to customize is to have good, multi-screen support. i use multiple screens at work, but when drawing it's actually better to focus on one monitor. in my case this is the “primary” monitor, thus a helper script tablet2primary arose:

#!/bin/bash
set -eu -o pipefail
app=$(basename "$0")
screen=$(xrandr | grep 'connected' | grep 'primary' | awk '{ print $1 }')
tablets=$(xinput | grep '\<Wacom\>' | sed 's/.*\<id=\([0-9]\+\).*/\1/')
 
for tab in $tablets
do
  echo "$app: mapping $tab to screen $screen"
  xinput map-to-output "$tab" "$screen"
done

enjoy! :)