Bondtech + MMU  

  RSS
gnat
 gnat
(@gnat)
Prominent Member

A few weeks ago I upgraded to Bondtech's extruder and switched to a straight heatbreak. Since then I have done a bunch of single filament prints with no(*) issues. For the last week I've been running the filament through the MMU, again with no issues. I haven't tuned anything, but almost always get good tips formed when the filament is ejected. Maybe one of 10 has a tail (and it's crazy long when it happens).

One thing I've noticed is that during the initial purge line at the front of the plate, I get very little actual purge. With the factory setup on a single tool print I would get a blob at the start point and then a nice solid line all the way across and back. For a multi tool print it would start the purge line between 1/4 to 1/2 across and then be solid all the way back. Since moving to the Bondtech, however, the purge doesn't start laying anything down until about halfway through the return pass for a single tool print. I just started my first multi tool print and it barely laid down and plastic during the purge (thankfully the beginning of the purge block finished clearing it).

This same problem is occurring during wipe tower operations too with the result being that it is only laying down about half the filament there that it should. This is leaving missing areas which causes subsequent wipes to have nothing to wipe onto. This is obviously making a mess of the tower and I expect may lead to a print failure (or at least a giant mess).

I did load the Bondtech firmware, but it's like I need to adjust something else to that it knows it needs to purge more than it currently thinks. Anyone else with a Bondtech run into this and have any idea what I need to play with?

(*) When I say no issues, PLA works fine, but my PETG prints have been garbage (terrible top layers and will just break apart) even though I dialed in the Live Z and tried temps from 230 (where I ran with the factory setup) to 250.

MMU tips and troubleshooting...
Posted : 27/06/2020 2:13 am
Karl Herbert
(@karl-herbert)
Famed Member
Posted by: @gnat

A few weeks ago I upgraded to Bondtech's extruder and switched to a straight heatbreak. Since then I have done a bunch of single filament prints with no(*) issues. For the last week I've been running the filament through the MMU, again with no issues. I haven't tuned anything, but almost always get good tips formed when the filament is ejected. Maybe one of 10 has a tail (and it's crazy long when it happens).

One thing I've noticed is that during the initial purge line at the front of the plate, I get very little actual purge. With the factory setup on a single tool print I would get a blob at the start point and then a nice solid line all the way across and back. For a multi tool print it would start the purge line between 1/4 to 1/2 across and then be solid all the way back. Since moving to the Bondtech, however, the purge doesn't start laying anything down until about halfway through the return pass for a single tool print. I just started my first multi tool print and it barely laid down and plastic during the purge (thankfully the beginning of the purge block finished clearing it).

This same problem is occurring during wipe tower operations too with the result being that it is only laying down about half the filament there that it should. This is leaving missing areas which causes subsequent wipes to have nothing to wipe onto. This is obviously making a mess of the tower and I expect may lead to a print failure (or at least a giant mess).

I did load the Bondtech firmware, but it's like I need to adjust something else to that it knows it needs to purge more than it currently thinks. Anyone else with a Bondtech run into this and have any idea what I need to play with?

(*) When I say no issues, PLA works fine, but my PETG prints have been garbage (terrible top layers and will just break apart) even though I dialed in the Live Z and tried temps from 230 (where I ran with the factory setup) to 250.

I also use a Bondtech extruder with a 1.8 pancake stepper and have set the following values:

before calibrating all axis:

#define DEFAULT_AXIS_STEPS_PER_UNIT {100,100,400,415}

#define TMC2130_USTEPS_E 16 // microstep resolution for E axis

after calibrating:

//#define DEFAULT_AXIS_STEPS_PER_UNIT {100.3,100.1,399.5,423.232324}

#define TMC2130_USTEPS_E 16 // microstep resolution for E axis    (nothing changed)

These instructions are helpful when calibrating the extruder:

https://mattshub.com/blog/2017/04/19/extruder-calibration

Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen....
Posted : 27/06/2020 2:28 am
gnat liked
gnat
 gnat
(@gnat)
Prominent Member

@karl-herbert

Thanks, I'll take a look into it.

MMU tips and troubleshooting...
Posted : 27/06/2020 2:41 am
gnat
 gnat
(@gnat)
Prominent Member

There may have been some user error involved. I went poking around the BT version of the 3.9.0 firmware and the name displayed on the printer didn't match what was in the code. I also noticed that the loading measurements are shorter for the Mosquito version when compared to the E3D code path.

I reloaded the firmware this morning and while still not like the old extruder, more of the purge line was laid down when I started my next print. So maybe I had loaded the Mosquito build by mistake.

MMU tips and troubleshooting...
Posted : 27/06/2020 8:04 pm
gnat
 gnat
(@gnat)
Prominent Member

@karl-herbert what is your step value according to M503?

I started following that link and found that it was supposedly under extruding by almost half (64mm of the 120 was used). This doesn't make sense though as other than the initial purge everything has been spot on.

The current step value was 280, but I went ahead and gave the calculated 450-something a try. As expected, it then way over extruded. I had, however, entered some other command (my terminal doesn't echo so all I know is that the first time I tried to set the new step value I apparently typo'd something (I got an ok response, but the setting wasn't changed.

I went through a factory reset and a M502 response, but the step value went from what I set it to, to 415 rather than back to 280 like it was.

I am now getting a good purge at start and on the wipe tower, but my live z test (factory and "Life Z") keep coming out at .3 to .4 rather than the expected .2.

Also, what do use for the following settings (Printer Settings -> Single Extruder MM Setup):

  • Cooling tube position
  • cooling tube length
  • filament parking position
  • extra loading distance
MMU tips and troubleshooting...
Posted : 29/06/2020 1:18 am
Karl Herbert
(@karl-herbert)
Famed Member
Posted by: @gnat

@karl-herbert what is your step value according to M503?

This are my values on my bondtech with 1.8 pancake stepper from bondtech:

configuration_prusa.h

#define CUSTOM_MENDEL_NAME “MK3-BT-1.8-MMU2S

#define Z_MAX_POS 205

#define DEFAULT_AXIS_STEPS_PER_UNIT   {100,100,400,415}

#define TMC2130_USTEPS_E16 (change from 32 to 16 only MK3/S, otherwise use 830 steps per unit)

 

#define LOAD_FILAMENT_1 “G1 E80 F400

#define UNLOAD_FILAMENT_1 “G1 E-95 F7000

#define FILAMENTCHANGE_FINALRETRACT -95

#define FILAMENTCHANGE_FIRSTFEED 80

changed to configuration_prusa.h, fw recompiled and flashed. Possibly reset with M502.

https://sebastian.expert/marlin-g-codes-m500-m501-m502-m503/

 

 

Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen....
Posted : 29/06/2020 9:53 am
gnat
 gnat
(@gnat)
Prominent Member

Did another check on my extrusion and with the default 415 (no idea how/why it was at 280 before) it's as close to perfect as I can make it (-0.25mm over 200mm give or take based on my inexact markings).

Turns out, however, that as I worked from the Live Z setting from when it was drastically under extruding as my base line, my Live Z was way off (though visually it looked good). I raised it back up 0.14mm and my elephant foot is gone and test prints are much closer to their desired dimensions.

I'll have a better idea later today if that has had the desired effect or not.

MMU tips and troubleshooting...
Posted : 30/06/2020 3:44 pm
Karl Herbert
(@karl-herbert)
Famed Member

@gnat

The value 280 is valid for the original Prusa extruder. For the Bondtech due to the gearbox the value 415. However, the value 415 is only a guideline, which is why I would recommend a calibration. For me, the extruder at 423.23 produces a correct extrusion (gear deviations).

Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen....
Posted : 30/06/2020 6:46 pm
gnat
 gnat
(@gnat)
Prominent Member

Ah. Makes sense about the 280.

There is definitely more I could do for dialing it in, but 415 is close enough for now until I can figure out a more accurate way of doing the test.

MMU tips and troubleshooting...
Posted : 01/07/2020 1:39 am
saso.r
(@saso-r)
New Member

Hi, did you solve a problem that extruding is starting to late at the beginning of print (purge line) and on purge tower?

Posted : 25/08/2020 9:47 pm
gnat
 gnat
(@gnat)
Prominent Member

Yes it was all related the e-steps being ~60% lower than they should have been. Once I solved that and readjusted my Live Z all was good again (including PETG).

MMU tips and troubleshooting...
Posted : 25/08/2020 9:53 pm
saso.r
(@saso-r)
New Member

Thank you. I will look into that.

Posted : 25/08/2020 10:09 pm
Share:

Please Login or Register