What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

3D Printer / CNC discussion thread

sswilson

Moderator
Staff member
Joined
Dec 9, 2006
Messages
19,894
Location
Moncton NB
You calibrate your steps per mm on your extruder every new roll of filament? The only time I'll ever touch that if I replace parts on the extruder. All stepper steps/mm really should be static values that don't change.

I also never touch flow rate from roll to roll either, unless I notice an issue. Unless you are buying ultra high end filament, the entire roll of filament won't be consistent anyways. So the part you are measuring and adjusting flow could be substantially different.

For each filament type (not brand), I'll do my best to calibrate flow via an empty box w/ 2 walls then measure and adjust flow rate to compensate. But unless you are using high end calipers like Mitutoyo, you won't get the exact repeatability anyways to get this really exact anyways IMO.

Yep, and it varies widely from one roll to another (5mm+ one way or the other out of 100mm is typical). Same goes for XYZ axis calibration on a cube but that at least is fairly static from one roll to another once you've dialed it in. That really is the standard calibration process recommended by everybody who talks about calibrating an ender 3.

This particular test spit out an average of around .53 vice the .4 that it's supposed to be.... I'm a little bit dubious though as I suspect dropping the flow rate down to 75% will result in some pretty obvious under-extrusion.

I'll give it a shot, but in the writeup for the cal he does also suggest that this particular calculation isn't written in stone and that observed print results WRT over/under extruding should ultimately rule the day.
 

lowfat

Moderator
Staff member
Joined
Feb 12, 2007
Messages
10,442
Location
Grande Prairie, AB
Yep, and it varies widely from one roll to another (5mm+ one way or the other out of 100mm is typical).
You absolutely shouldn't be seeing that variation. Even cheap filament should be 1.75m +/- 0.1mm. 5% would need mean the filament is up to 1.8375mm which would be absolutely garbage filament IMO. You should check your extruder tension as its either slipping or grinding.

Same goes for XYZ axis calibration on a cube but that at least is fairly static from one roll to another once you've dialed it in.

These are mechanical static values that never change. They are 100% dependent on pulley size/screw pitch, microstepping, and how many steps a stepper has in a full rotation. Probably 200 for X and Y assuming 16T pulley and 1/32 microstepping on TMC2209. Z probably 800 depending on the lead screw pitch.

That really is the standard calibration process recommended by everybody who talks about calibrating an ender 3.
If we are talking XYZ calibration, they aren't right for sure. I've never seen any reputable person ever suggest this. And if you are talking extruder steps, even in the Teaching Tech vid you linked a few days ago says just once when you get the printer and if you make any changes to the hotend / extruder. These values are not dependent on filament size. They are also mechanical values dependent on hobb diameter, gearing, microstepping, and the steppers steps per full rotation.

(steps per full rotation * microstepping * gearing) / (hobb diameter * pi) = extruder steps per mm. It is just a lot easier to measure the filament passed through extruder and adjust. But after its done it should be static.



FDM printers are far from perfect machines. Flow through the nozzle is never going to be consistent. So its nearly impossible to actually get consistent results. Even when printing any empty box and measuring, you probably won't really get repeatable results. I don't think any of my filament types are more than 1-2% from 100%. Teaching Tech's tutortial says to only do this once. Even trying to calibrate retractions unfortunately isn't something that can be perfected as in theory longer travels should require more retracting and this just isn't included in slicers / printer firmware yet. Also linear advance / pressure advanced should be done per type of filament. I think there was an extension to add linear advance options to Cura, or maybe they've added it finally. This is something most people to just ignore for some reason.

Honestly the only things you really should be calibrating is retractions and temperatures for each brand of filament as the additives they use could affect both.
 

sswilson

Moderator
Staff member
Joined
Dec 9, 2006
Messages
19,894
Location
Moncton NB
Heh.... I can't argue with your much longer experience with 3D printing in general, but pretty much every site that talks about calibrating the ender 3 suggests doing a calibration cube and adjusting your steps accordingly.


Not the most technical site, but I'm also pretty sure Chep promotes calibrating steps with a calibration cube.

Is it possible that the steppers being used by Creality aren't completely linear and need more fine tuning than other steppers?

edit:

One of the things you've got me thinking on though is the extruder variation. I completely agree that it doesn't make sense to have that much variation between different rolls.... I'm wondering if the fact that I've got my default PLA hotend pre-heat set to 180 and what I'm observing is filament slippage because the filament needs a higher temp to flow properly.....

That could possibly explain why my walls were so thick on this test.... this particular roll of filament needed a particularly high adjustments to the number of steps (when tested @ 180C) so it makes sense that it'd be overextruding once the hotend temp is @200 if it wasn't flowing well @ 180.
 
Last edited:

danmitch1

Well-known member
Joined
Dec 15, 2007
Messages
1,611
Heh.... I can't argue with your much longer experience with 3D printing in general, but pretty much every site that talks about calibrating the ender 3 suggests doing a calibration cube and adjusting your steps accordingly.


Not the most technical site, but I'm also pretty sure Chep promotes calibrating steps with a calibration cube.

Is it possible that the steppers being used by Creality aren't completely linear and need more fine tuning than other steppers?

edit:

One of the things you've got me thinking on though is the extruder variation. I completely agree that it doesn't make sense to have that much variation between different rolls.... I'm wondering if the fact that I've got my default PLA hotend pre-heat set to 180 and what I'm observing is filament slippage because the filament needs a higher temp to flow properly.....

That could possibly explain why my walls were so thick on this test.... this particular roll of filament needed a particularly high adjustments to the number of steps (when tested @ 180C) so it makes sense that it'd be overextruding once the hotend temp is @200 if it wasn't flowing well @ 180.
Ive been also calibrating for each new roll. Long story short, is there any downside to doing this other than possibly wasting time and filament?
 

lowfat

Moderator
Staff member
Joined
Feb 12, 2007
Messages
10,442
Location
Grande Prairie, AB
Heh.... I can't argue with your much longer experience with 3D printing in general, but pretty much every site that talks about calibrating the ender 3 suggests doing a calibration cube and adjusting your steps accordingly.


Not the most technical site, but I'm also pretty sure Chep promotes calibrating steps with a calibration cube.

Is it possible that the steppers being used by Creality aren't completely linear and need more fine tuning than other steppers?

edit:

One of the things you've got me thinking on though is the extruder variation. I completely agree that it doesn't make sense to have that much variation between different rolls.... I'm wondering if the fact that I've got my default PLA hotend pre-heat set to 180 and what I'm observing is filament slippage because the filament needs a higher temp to flow properly.....

That could possibly explain why my walls were so thick on this test.... this particular roll of filament needed a particularly high adjustments to the number of steps (when tested @ 180C) so it makes sense that it'd be overextruding once the hotend temp is @200 if it wasn't flowing well @ 180.
TIL that you do don't need to work for all3dp to add an article. This was just some random dood adding this. That info is just straight up wrong.

All movement on all axis' is going to be 100% repeatable as long as your pulleys aren't slipping and your belts are right. Steppers are extremely accurate and so is microstepping on trinamic drivers. The only way this would be off is if the stepper is dying and it would skip / snap to the next full step. If this happens the only thing you can do is replace the stepper.

I couldn't find any vid where Chep suggests to calibrate XYZ, everything I searched online was just some random doods suggesting it.


You can if you wish calibrate your E steps every roll, but IMO if you are getting that big of deviation something else is up. Most filaments rolls have great tolerance these days and that kind of calibration just isn't required any more.

IMHO set your e steps once. Set your flow once. Set linear advance once per filament type. Then spend a good amount of time trying to tune retractions and do a proper temp tower. Almost all issues are going to be caused by the first layer and retractions.

EDIT: Have you done any linear advance calibrations? It will make for a more consistent print, especially after corners. As the extruder adjusts its speed beforehand instead of exactly the same time as the print move. It is IMHO one of the best features in printing in the last 5 years.
 

lcdguy

Well-known member
Folding Team
Joined
Mar 17, 2007
Messages
2,254
Location
An undisclosed location
honestly i would only calibrate the e-step value if I make a change to the extruder (ie: changing out gears, stepper, control board, etc) or if I see any of the following issues.

1. Over Extrusion
2. Under Extrusion

since most reputable brands of filament have pretty tight deviation tolerances, doing this for each roll is not necessary.

For example the pla/petg filament I use mostly (spool3d house brand) tolerance is +-0.04mm.

Having said that if i just received a brand new printer after I have leveled the bed i would certainly test to see if extrusion is correct and if not correct it.
 

sswilson

Moderator
Staff member
Joined
Dec 9, 2006
Messages
19,894
Location
Moncton NB
Am I missing something WRT setting flow rate? So far the only thing I'm coming up with is that you can set flow rate in your slicer profile, I've looked at the printer's menu and can't find a permanent setting, is there no configuration in merlin that allows flow rate to be fixed?

Even octoprint's option is less than ideal... with the new mainboard there's still a place to set it but it claims that it can't report the current setting because of the connection type, and when I did a test on it, it didn't appear to be changing the flow rate (according to the wall width test I'm doing).
 

sswilson

Moderator
Staff member
Joined
Dec 9, 2006
Messages
19,894
Location
Moncton NB
Still early on playing with calibrating flow rate, but so far (going off the calibration on TT) I'm down to 67% flow rate to get a .4 wall. :eek:

That's using the spiral function so I'm going to try it as a normal print, as well as seeing what expanding to 2 line width shows me.
 

Latest posts

Twitter

Top