Anycubic Kossel saga continues ... with tips

[Ron Frazier (3DPGRP)]

Hi All,

As some of you know, I've been working intermittently on getting two
printers working for years. It's been frustrating. I've shared some info
on this a couple of years ago. But, now it's a new year. There are new
people in the group. And, I'm working on it again. I have new
information and some progress. My time to do this is very erratic. But,
I'm getting very close to getting the Anycubic Kossel working. I want to
share various things I've learned to help others with the information. I
also always appreciate those here who encourage me or give me advice. In
this case, I'm focusing on the Anycubic Kossel Linear Plus delta printer
with a larger build volume than it's little brother (the pulley version)
and linear rails.

To the newbies, 3D printing is a hobby of tinkering. Unless you get a
fairly modern and moderately expensive printer, it's not plug and play.
Some are more plug and play. Some are less. Cheap Chinese printers are
less. To me, $ 500 is expensive.

Several things have plagued me about this printer. I will say it
basically worked at first. But the Marlin software was outdated from the
start and didn't have certain safety features active. I tried to put
Marlin 1.9.x in it and never got it to work. Delta printer calibration
is a pain, at least with older ones, in the best case. I had several
head crashes before, and damaged the build plate and the build probe. I
had to order a new probe from China. Don't even consider a delta printer
without a bed probe and semi automatic or automatic calibration.

As of a couple of years ago, I was mainly trying two filaments, PLA and
PETG. I've gotten test prints working in both. I've tried several build
surfaces and had lots of problems. The aluminum build plate on the
Kossel is too thin, too unmounted, too non sticky, and in some cases,
too warped. I've tried painters tape (masking tape), cheap glass,
expensive fireplace glass, a BuildTak like sticker, and a PEI overlay.
All have problems. Painters tape works OK but is a pain to remove and
replace. Bare aluminum and glass don't stick to anything. I refuse to
use glue stick or hairspray as I don't wish to be cleaning my build
surface all the time. Cheap glass 3mm thick isn't thick enough, and is
often warped up to 1 mm. In the case of the BuildTak like sticker and
the PEI overlay, I got PETG stuck so hard to the surface that it never
came off.

I've centered on the Anycubic Ultrabase ceramic coated glass surface. It
is supposed to release the model once cool, and my preliminary data says
it seems to work with PLA at least. Not sure yet about PETG. I think
Creality and others have something similar. Ultrabase round plates are
hard to get, especially the larger ones for the Linear Plus. I just
discovered one I ordered from China a couple of years ago and misplaced
after waiting a month or so to get it. I also just got one from Canada.
So, I have 2 small (180 mm I think) ones and 2 large (240 mm) ones. I
didn't prefer to use a 180 mm plate on a 240 mm printer. I have the
Ultrabase plate sitting on the aluminum hot bed. I'm going to have to
figure a way to anchor the plate down. The plate has adhesive, but I
don't want use that. I want it removable.

The Ultrabase plate does appear to release pretty well with PLA after
it's cooled to about 30 C. The aluminum hot bed plus the Ultrabase plate
have lots of thermal inertia. So, it takes several minutes to warm and
several minutes to cool.

I found a new version of Marlin 2.x from another hobbyist on the
internet and that seems to be working. Modifying Marlin, compiling it,
and uploading it is doable but tedious. It is not newbie friendly. These
links were especially helpful to me. They may also be useful to other
people with Anycubic Kossel delta printers or even different delta
printers. Kudos to the person who created these pages. For more generic
Marlin information, I've posted some cool links at the bottom of this
document.

https://www.lpomykal.cz/3d-printers/kossel/

https://www.lpomykal.cz/3d-printers/kossel/anycubic-kossel-marlin-guide/

https://www.lpomykal.cz/kossel-marlin-firmware/

https://www.lpomykal.cz/kossel-bed-calibration-marlin-fw/

https://www.lpomykal.cz/kossel-pid-calibration/

https://www.lpomykal.cz/category/kossel/

Delta calibration, even with the bed probe and a semi automated
procedure, is a pain. It may be better on newer machines. Take extra
time assembling the printer and make it as rigid and square as you can.
You must know the length of your connecting rods to the effector, which
is a problem since a 6" caliper is too short. There's no way I'm
disassembling the printer. I put a piece of tape in the middle and
measured from one end to the tape and from the tape to the other end and
added the two. I don't know if the calibration procedure adjusts the rod
length set in the firmware. You also have to know your printable radius,
which is less than the build plate radius. With auto calibration, you
leave the hot end cold and install the probe. I assume it's better to
have the bed at least somewhat hot (60 C) to account for thermal
expansion, but I don't absolutely know. The probe will go down and sense
many various places on the bed to determine their height. It does an
entire sequence 6 times, which takes about 15 minutes. You then store
this in the memory. There is a way to take most of the key parameters
after calibration and save them back to the Marlin configuration so
they'll be there if you recompile the firmware. I may have to try that.

Then, you have to determine the Z height. That is the distance from the
nozzle when homed (at the top) to the build surface. I do this with the
bed moderately hot (60 C) and the nozzle cold. The probe is removed for
this step on the Anycubic Kossel. If you have old filament on the
nozzle, heat it up and remove or flatten that with a spatula. Then let
the nozzle cool. The objective is to lower the nozzle to 0 height and
have it barely in contact with the build surface, creating slight
friction between the nozzle and a piece of paper. This is a convoluted
procedure which may need to be repeated several times to fine tune it.
It needs to be accurate to the thickness of a sheet of paper (.1 mm) or
less. Adjusting the height is counter intuitive (to me) because of the
fact that the printer homes to the top, then goes down. If you ADD to
the height setting, that makes the printer try to go further DOWN to get
to 0. Therefore, it can crash into the bed. So, for example, if you
bring the printer to where it thinks it's at 0, but the nozzle is 1 mm
high over the bed, you keep lowering another mm until the nozzle touches
the paper on the bed. You have to turn off software endstops to do this,
and the final Z reading on the screen will be - 1 mm as an example. So,
we want the printer to go 1 mm further down to get to 0. So, we ADD 1 mm
to the height and save it. Then, power off and on and do this again.
Conversely, say the printer was touching the bed and the printer thought
it was at 1 mm. We want the printer to go 1 mm less down to get to 0, so
we SUBTRACT 1 mm from the height and save it. This totally bends my
brain, and I hope I typed this right. I'm not totally sure if adjusting
the Z height requires recalibration

A couple of years ago, I had just gotten test prints running with PETG
when they started degrading. They became brittle with lots of strings on
the print. I could crush the print in my hand. I eventually discovered
that this is due to humidity. This must be dealt with for almost any
filament except PLA and even it will absorb humidity over time.

I just got this dehydrator from Amazon. It can be used to dehydrate
existing filament with the lid cracked open a bit. DO NOT seal the
chamber when dehydrating. Moisture must escape. Or, it can be used to
store filament with the lid closed and ports sealed. It can hold two
medium width spools but not two wide spools. It can dispense filament
directly to the printer. It is supposed to be adjustable from 40 C to 50
C. But, it has a design flaw that prevents the interior cabinet from
getting over 40 C. The thermistor is too close to the heating element
and it thinks the case is hotter than it is. I'm trying to decide
whether to keep it as is, modify it, or return it.

Sovol Filament Dryer, SH01 Filament Dehydrator 3D Printer Spool Holder
https://www.amazon.com/gp/product/B09686Z5R3/

I also have this food dehydrator from Presto for more aggressive
dehydrating. But, that doesn't really work for storing or dispensing. It
is way more powerful and more accurate. It also gets up to 90 C I think,
which MAY be enough for annealing PLA+ or PLA Pro filament to leave it
less sensitive to heat.

Presto 06301 Dehydro Digital Electric Food Dehydrator
https://www.amazon.com/Presto-Dehydro-Digital-Electric-Dehydrator/dp/B008H2OEKK/

So, last night I did my first test print in a couple of years on the
Kossel with some old Inland PLA that I had dehydrated for a while. It
almost worked. I'm very close. Here are photos of it.

Full size 16 mp picture, 10 MB
https://www.dropbox.com/s/zs7wcbv67j3bukq/kossel%202022.01.30%20almost%20IMG_3101.JPG?dl=0

1200 pixel wide picture, 600 KB
https://www.dropbox.com/s/icmjfp00fd0jhj8/kossel%202022.01.30%20almost%201200W%20IMG_3101.jpg?dl=0

600 pixel wide picture, 150 KB
https://www.dropbox.com/s/gpm2ngi65q1csla/kossel%202022.01.30%20almost%20600W%20IMG_3101.jpg?dl=0

The picture starts at about the 7 o'clock position. At this point the
nozzle was basically grinding the filament into the build plate. That
part may never come off. I adjusted Z upward by .1 mm in babysteps using
the control panel and will adjust the Z height later. Things progressed
for a while and then it got a bit janky around the 4 o'clock position. I
thought I heard some thunking, which may have been the linear guide
hitting the bottom of the tower. That should never happen. Then, at
around the 10 o'clock position, it started air printing and I killed the
print. My best guess is that the printable radius in the firmware is too
high at 115 mm (230 mm diameter). So, I'm going to reduce that to 110 mm
(220 mm diameter).

Finally, I feel that the fumes and microfine particles emitted by the
printer are harmful, as mentioned in other threads. Even PLA makes my
nose burn and sense an after effect for many hours. My living room is
the only place the printer can be, so I hope fully enclose it and vent
it outside with temperature control at some point. In the mean time, I'm
running a window fan. That's LOTS of fun on a 40 degree (Fahrenheit) day.

I've tried to summarize the key points and potholes of my journey. I
hope this information will be helpful to others on similar journeys.
Below are generic Marlin links.

Sincerely,

Ron

----------

https://marlinfw.org/docs/basics/install.html

https://marlinfw.org/docs/basics/troubleshooting.html

https://marlinfw.org/meta/download/

https://marlinfw.org/docs/configuration/configuration.html

https://marlinfw.org/docs/hardware/boards.html

https://marlinfw.org/docs/hardware/endstops.html

https://marlinfw.org/docs/hardware/tmc_drivers.html

https://marlinfw.org/docs/configuration/probes.html

--

If you're using email with spam filters, especially GMail, Hotmail, Yahoo Mail etc., please
set my address as an approved sender.

(PS - If you email me and don't get a quick response, you might want to
send again. I don't always see new email messages very quickly.)

Ron Frazier
Anet A8, Anycubic Kossel Linear Plus
Delta printers rock. I prefer to never physically level a bed.
3dprttdyinfo AT techstarship DOT com - Blog - techstarship.com
Blog RSS: techstarship.com/feed/

[brian wise]

Good Morning, I have 3 deltas SMC Eris, Any Cubic Kossel Pulley and Rostock Max V2, I level with matter control slicer on all three. They still have the 8 bit boards and other swapping the hot ends on all three and changing the oem carriages and arms on the Rostock Max V2 they are factory spec. The mesh level from your scilcer in my case Matter Control free vesion seems to work pretty well for me. I didnt edit the machines settings for the new hot ends just adjusted the flow and speed control based plotting 20mm cubes and temp towers for each kind of filament I have.  I do not use glue or hair spray and eris bed is not heated ( although it does have auto bed level G 29) no rafts. Also all three run different firmware.

[TobyCWood]

Wow Ron... Ya really wrote a thesis!

You wrote... "I refuse to 

use glue stick or hairspray as I don't wish to be cleaning my build 
surface all the time."

I have plates that use AQUANET (not just "Hairspray") that have years of use and have only been cleaned off and resurfaced maybe once or twice. I don't know how or why you assumed that the bed needed to be "cleaned" each time. What the plate may need is a small burst of spray where the previous print came off. That takes about 2 seconds and it dries before the print starts. Regardless... I would have thought you turned away from the most used solutions simply because of your hypersensitivity to chemical outgassing. 

FWIW I have the "Anycubic Ultrabase ceramic coated glass" on top of my Predator's bed heater. It's the same stuff that Creality used on my CR6SE... Meh. Plain old cheap glass plates with AQUANET UNSCENTED SUPER EXTRA HOLD works way way better. If you actually had significant printing experience you would know that many times one cancels a print during the first layer. Too often a wiz bang solution such as this ultra base thing does not take that into account and removal of a .2mm layer off this surface can be quite tedious!

In another chapter you mentioned that you thought your PETG degrading was because of humidity. Of all the materials I use PETG is probably the least effected by moisture! I note you never specify any of the settings you used such as your temps, speeds, retraction etc... all of which could have been part of the problem as well as any damage from the nozzle crash you described.

I stopped reading the book at this point. Just too much!

But from what I could gather from the overall gist is that you seem to be hyper focusing on specific less relevant issues related to 3D Printing (such as assuming the printer must be absolutely perfect for it to work) and not dealing with simply getting one of your machines working and printing! 

[Ron Frazier (3DPGRP)]

It would be really cool to know about your experience with Matter Control. That's from Matter Hackers, right? I read about that a couple of years ago but haven't had a chance to try it. Sounded pretty cool at the time. I think Delta printers are very cool once you get them dialed in and calibrated and under control. Their main deficiency is that they need to be very tall to obtain a cubical build volume as big as my old Anet A8 (220 mm x 220 mm x 220 mm) or larger. I especially like the visual appearance of a delta when running. Most people do. And, I especially like the stationary build platform.

Sincerely,

Ron

--
You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/20511aae-dc4c-4aa4-ba87-18c92f9a5c3en%40googlegroups.com.

[Ron Frazier (3DPGRP)]

I didn't send this with the proper from address to get to the group. Trying again.

Ron

On 1/31/2022 3:03 PM, Ron Frazier wrote:

Hi Andy,

Thanks for the tips. See replies inline below.

Sincerely,

Ron

On 1/31/2022 1:30 PM, TobyCWood wrote:

Wow Ron... Ya really wrote a thesis!

I had a lot to share that hopefully will be helpful to those less experienced. But, I'll admit it got long.

You wrote... "I refuse to 

use glue stick or hairspray as I don't wish to be cleaning my build 
surface all the time."

I have plates that use AQUANET (not just "Hairspray") that have years of use and have only been cleaned off and resurfaced maybe once or twice. I don't know how or why you assumed that the bed needed to be "cleaned" each time. What the plate may need is a small burst of spray where the previous print came off. That takes about 2 seconds and it dries before the print starts. Regardless... I would have thought you turned away from the most used solutions simply because of your hypersensitivity to chemical outgassing.

Sorry, I'm not an expert on hairspray. Well, it's interesting to know that the build plates don't have to be cleaned too often. I don't know if I'm hypersensitive per se but I do have a very sensitive nose, which was tingling all day yesterday after an hour of close up exposure to the printer (while printing) the night before. It didn't even smell that bad. PLA has sort of a sweet odor. But, I was definitely breathing something. Even if people are not sensitive or extremely aware of it, that doesn't mean printer pollution isn't harmful to one's health. But, since you mention it, that might be another reason to avoid the hairspray and glue sticks.

FWIW I have the "Anycubic Ultrabase ceramic coated glass" on top of my Predator's bed heater. It's the same stuff that Creality used on my CR6SE... Meh. Plain old cheap glass plates with AQUANET UNSCENTED SUPER EXTRA HOLD works way way better.

Good to know.

If you actually had significant printing experience you would know that many times one cancels a print during the first layer. Too often a wiz bang solution such as this ultra base thing does not take that into account and removal of a .2mm layer off this surface can be quite tedious!

Once I get past all these newbie potholes, I can start accumulating experience.  8-)  Maybe other newbies reading can skip some of those potholes. The end of my "book" noted that I was running a test print the other night, which was just a 2 cm wide circle around the outer edge of the build plate. It started getting janky at one point and I had to kill the print. FWIW, after letting the build plate cool to 35 C, which took a few minutes, I was easily able to get that single layer off. It was PLA and other people's mileage may vary. But, I can see how having to wait 5 - 10 minutes to get a failed print off could be a bother. One option is to let the build plate cool enough to touch then swap plates. Since that might change the printer calibration, I might be more inclined to wait most of the time.

In another chapter you mentioned that you thought your PETG degrading was because of humidity. Of all the materials I use PETG is probably the least effected by moisture! I note you never specify any of the settings you used such as your temps, speeds, retraction etc... all of which could have been part of the problem as well as any damage from the nozzle crash you described.

It's good that the issue hasn't bothered you. In my research on Google and YouTube, I've run across a number of people with PETG humidity problems. YouTube videos and blogs show excessive stringing and brittleness. Maybe your location environment is dry. Maybe you have a dry box. Maybe the filament formulation is special. Maybe I got a bad batch. GA in the summer is NOT dry. It's routinely 80 - 90 F and 80 - 90 % humidity. Of course, the air conditioned house is about 50 % humidity. I cannot cite specs and parameters from 2 years ago. All I know is that 2 - 3 years ago, I got some Esun Silver PETG. I started using it on the Anet A8 and it worked well enough to print a replacement bracket for a new clone hot end. That part was flawless and strong. A few months later, I was printing test models of an Archer Elf from Thingiverse and those came out very stringy. Other different shapes came out brittle. And at one point I detected popping from the hot end. So, after research, I concluded that it had to be wet filament, which started a whole train of research into how to solve that problem. Here's the link to the Archer Elf if anyone wants it. It's a pretty cool model.

Archer Elf
https://www.thingiverse.com/thing:2314063

I stopped reading the book at this point. Just too much!

But from what I could gather from the overall gist is that you seem to be hyper focusing on specific less relevant issues related to 3D Printing (such as assuming the printer must be absolutely perfect for it to work) and not dealing with simply getting one of your machines working and printing!

I'm getting there. The main things at the end were information on delta calibration, generic Marlin information, and an almost complete test print on the Kossel. Making progress, and there's always more to learn. I always appreciate good information sources.

Cheers,

Ron

--
You received this message because you are subscribed to the Google Groups "3D Printing Tips and Tricks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3d-printing-tips--...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/3d-printing-tips--tricks/275a1150-b39d-435a-a384-5ddbb1c5fd8fn%40googlegroups.com.

-- 
If you're using email with spam filters, especially GMail, Hotmail, Yahoo Mail etc., please
set my address as an approved sender. Please don't share my contact info.

(PS - If you email me and don't get a quick response, you might want to
call on the phone or send again. I don't always see new email messages
very quickly.)

Ron Frazier
770-889-8645 (H)
404-431-5472 (C)
Blog: techstarship.com - Blog RSS: techstarship.com/feed/
rwfrazier AT ronaldwfrazier DOT com

[brian wise]

Yep it from Matter Hackers,  I have used it my deltas and my bed slingers. I like it alot. Its calibration is easy. When I get home from work I will send a video on leveling. And to Andy's point my first printer was the Kossel and I was overwhelmed (still am) with all the faucets making the model of the object and scaling it for my printer. I started with Cura wich I think a lot of people use but I traded a custom scraper for a See Me CNC Rostock Max V2 from a rich guy that didnt want to build it. Any way See Me CNC recommends it so I tried it and for me it was game changer. I went from printing once a week to everyday.  Others may disagree but my Eris has auto leveling and I still prefer the Matter Control 3 point level. But I am not pro or anything like that but I have every prop toy and model I wanted to print so far....

[Jordan Jones]

For what it's worth I've had PETG and PLA filament crackle and pop a little bit if left out, but it's never made a noticeable affect on print strength.

Any prints so brittle that it could be crushed in my hand were always due to major under-extrusion, either due to e-steps, partial clog or some other cause.

[LukeH]

I was just printing with some ancient PETG over the weekend (it was maybe 4 years old - I found it at the back of a shelf I was cleaning out).

It popped and fizzed, and the surface quality of the print was a bit rough (it was very stringy), but the functional part I printed seemed to work ok.

I did however throw out some old spools of PLA, where there was condensation inside the bag, and the filament had gone all snappy.