MMU2 Troubleshooting  

Re: MMU2 Troubleshooting

i don't know if you listed them but what is the makes on the PLA you are using? if you are mixing lots of different brands then it could just be setting yourself up for failure.

perhaps scaling back to using 2 spools at a time for now might help.

create 2 cube lets say 10x10mm at .6 high. and have one set on ext1 and the other on ext2.
set the prusa baseline again. that will give you the first layer, and 2 more layers on top with about 5 tool changes.
compare the quick test baseline to what you got on this last test and see if you got an improvement.
if there was an improvement:

then i would alter the temps first lowering from the start 205 and 215 and see if you get better colder or worse before you even get into ramming and speeds. This would give you faster feedback on how the tips are looking on average while reducing mechanical failures as the test is similar and shorter.

on my machine it strings but the springs are very thin and much shorter about 10cm the short strings are staying attached for the most part and they just get re-rammed into the extruder later, and the thin wisps are not setting off the Finda ball. when a filament Did fill up the PTFE tubes with strings that got tangled into the other filimants on loads and unloads then i just altered that one filament to go colder that seemed to clean it up enough to keep going.

keep in mind your spools as configured with the bearings are causing a major problem and will be throwing off your entire data set. i would honestly stop change those bearing carriers first before you do anything else.

now printing those spool upgrade parts will take you about 5-6 hours to make 2 sets. you can either buy that 1" pcv and cut them to 85mm or you will need to print the tubes which will take a very long time.

again those spools as configured from prusa are no good at all.

Re: MMU2 Troubleshooting

i don't know if you listed them but what is the makes on the PLA you are using? if you are mixing lots of different brands then it could just be setting yourself up for failure.

perhaps scaling back to using 2 spools at a time for now might help.

create 2 cube lets say 10x10mm at .6 high. and have one set on ext1 and the other on ext2.
set the prusa baseline again. that will give you the first layer, and 2 more layers on top with about 5 tool changes.
compare the quick test baseline to what you got on this last test and see if you got an improvement.
if there was an improvement:

then i would alter the temps first lowering from the start 205 and 215 and see if you get better colder or worse before you even get into ramming and speeds. This would give you faster feedback on how the tips are looking on average while reducing mechanical failures as the test is similar and shorter.

on my machine it strings but the springs are very thin and much shorter about 10cm the short strings are staying attached for the most part and they just get re-rammed into the extruder later, and the thin wisps are not setting off the Finda ball. when a filament Did fill up the PTFE tubes with strings that got tangled into the other filimants on loads and unloads then i just altered that one filament to go colder that seemed to clean it up enough to keep going.

keep in mind your spools as configured with the bearings are causing a major problem and will be throwing off your entire data set. i would honestly stop change those bearing carriers first before you do anything else.

now printing those spool upgrade parts will take you about 5-6 hours to make 2 sets. you can either buy that 1" pcv and cut them to 85mm or you will need to print the tubes which will take a very long time.

again those spools as configured from prusa are no good at all.

Hi Daniel, here are the data specs for the filaments, at the beginning I used all same brand and type. I did a quick check online and found the tech sheet to be the same for the white, pink, light blue and black used. https://3dprintingcanada.com/collections/pla/products/white-1-75mm-pla-filament-1-kg .

the green is different the green is from NatureWorks under the " Select" brand. data spec sheet https://3dprintingcanada.com/collections/pla/products/green-1-75mm-select-pla-1-kg

A few weeks back I learned the basics on Python thanks to Paul (a super great guy like yourself). He made a script that after creating the Gcode you can run and add in your temp change during ramming also has other features like delaying raming actions in case the printer has reached target temps during that sequence.

here's the link if your interested https://shop.prusa3d.com/forum/general-discussion-announcements-and-releases-f53/script-to-drop-temperature-during-ramming-t24890.html#p109739

I haven't used this during this test. But first I will make the adjustments to the spool holders and try a simple test(s) to see how the tips change.
thanks again for your help.

Re: MMU2 Troubleshooting

i did read and look at that python code on the forum thread. Really great idea. i decided to put it to the side for now because it was going to start adding so many new variables on top of the system that was untested, i thought it was also going to make me start chasing my own tail so to speak.

getting the machine dialed in first would help with the consistency and timing before fine tuning and then possibly the machine could be run simpler.

Re: MMU2 Troubleshooting

Ok, I decided to give it another shot and to log as much data I can. I used a Gustav.zip file I think it was sent as a base test from one of the admins of this site. Which in its self-was very interesting on the printing behavior it changes temp during filament change ( 199-205) and lowers the fan speed at during parts of the print where the infill is widest? also has a different wipe tower which had a nice size for this test.
the printer has the latest firmware (Nov 13) and ran the preflight checks as well paying attention to the path of the filament from the spool to the nozzle.

Rui, I posted the Gustav model; but beware there were 2 different versions and the most recent is the better of the two. It was sliced with KISS rather than Slic3r, which gives more control of a variety of things, so it's good for getting things right.

Since I posted that model, I have been testing a variety of things and I think I have found a way to get consistently good prints with a modified extruder body. But I need to test some more just to confirm.

Peter

Re: MMU2 Troubleshooting

Ok, I decided to give it another shot and to log as much data I can. I used a Gustav.zip file I think it was sent as a base test from one of the admins of this site. Which in its self-was very interesting on the printing behavior it changes temp during filament change ( 199-205) and lowers the fan speed at during parts of the print where the infill is widest? also has a different wipe tower which had a nice size for this test.
the printer has the latest firmware (Nov 13) and ran the preflight checks as well paying attention to the path of the filament from the spool to the nozzle.

Rui, I posted the Gustav model; but beware there were 2 different versions and the most recent is the better of the two. It was sliced with KISS rather than Slic3r, which gives more control of a variety of things, so it's good for getting things right.

Since I posted that model, I have been testing a variety of things and I think I have found a way to get consistently good prints with a modified extruder body. But I need to test some more just to confirm.

Peter

Hi Peter, it was a nice file to run. I justed ran it to give it a shot to see how far the printer would run cleaning without issue... Didn't pass after 3rd layer with a skip But I'm still willing to try out the file or the lastest one, after I can get the gremlins out. If I forgot to mention great write up aswell.

Hi Daniel, I made the 2x roller mods for the stock MMU2 spool rollers it does move the spools a lot with ease. then made a test file with 2 filaments only ( 10x10x.6mm cubes ) which allows for 2 filament changes per color 7 mins total run time per test. with Prusa mmu stock filament settings and I just lowered the " other layers" temp @ 2-degree increments from 205 down to 186 and the printer made it to 196 with a short thin strip on the tips. At 194 the printer skipped at the extruder, my guess to cold to quickly melt the filament?
I'm guessing the next step is adjusting the tool change parameters?

so far no skips, or grindering except when at 194.

Re: MMU2 Troubleshooting

your almost there:

a few things

the whisps on the ends are not bothering me in how they look. but lets check your machine.

Pull the orange tube and shine a light in the FINDA if it looks clear then the whisps are not haven an effect.
if the FINDA is full of strings and whisps then more looking into the RAMMING settings.

you said you didn't mess with those ramming settings yet but using that INLAND PLA setup is a good start. and can't hurt to try it...

however i think you have some mechanical issues that need to be ruled out

"the printer skipped at the extruder" i would bet you got a strip out of the filament at the pulleys in the selector assembly. and at least three things are not 100% going on at the same time.

1) pull that orange tube(i hand tighten the thing anyway no need for that wrench at all)

2) look in the FINDA for those Strings, (clear/noclear)

3) pull the orange tube to the extruder

4) leave the top of the sensor cover on, open up the extruder pulley tensionars and expose the loading area.

5) start taking a couple of junk filament pieces of different cut end shapes from blunt to a 45 degree cut and push them into the extruder feel for any obstruction slight or not, rotate that filament around 360 degrees and keep testing if you feel anything touching that filament then you need to slightly reshape that cone guide next to the sensor pcb. if the filament is not smooth here then it will strip out the at the upper pulleys in the selector body
so check this first.

6) everything looks ok/or you fixed it so put it back together.
I like to put a mark on my bolts for the spring tension so i can tell how many turns i went. place a mark on the bolt head get the bolt into the nut and then go 10 turns to start. you may need to adjust this later. but between 9 and 11 turns is usually the sweet spot.

7) Oil the selector shafts if you have not. I like to use a slot car plastic safe oil and i put a drop on my finger tips and apply it to the shafts(upper and lower) then i put a drop right on the bushings where the shafts go in. run the selector back and forth a few times(cycle the power)
ok this keeps the selector from sticking or chugging and keep the filament path more aligned. more aligned equals less friction. ( don't stop there oil the rest of the machine to, X,Y,Z 🙂 )

8) look at your used filament that went into the selector
you want to see how much bite the teeth on the pullys are causing.
now i find the pulley selector needs a very light bite compared to the extruder. so once again mark those bolt heads on the selector and i put a mark on the cover as well so get a starting point. looking at the bite and how the filament striped out. you will have to figure out the sweet spot but back off 1 full turn and retest. keep it even at first left and right, but if you find 3,4,5 is having issues and 1,2 are fine then increase tension just on the right side bolt.

this may not solve everything but it is going to rule out more problems.

Prusa chooses speed when they make the inhouse parts and they use like a .35 layer height and it makes for rough edges so if you can smooth them out it makes everything run better.

with the strings like that i found the system self cleaned most of the time the next filament change would ram through the FINDA grab anything and send it down into the extruder where it went on the wipe tower anyway.

temperature range make the adjustments and if it runs better then bring the temps back up to the 195-205 range.

Re: MMU2 Troubleshooting

your almost there:

a few things

the whisps on the ends are not bothering me in how they look. but lets check your machine.

Pull the orange tube and shine a light in the FINDA if it looks clear then the whisps are not haven an effect.
if the FINDA is full of strings and whisps then more looking into the RAMMING settings.

you said you didn't mess with those ramming settings yet but using that INLAND PLA setup is a good start. and can't hurt to try it...

however i think you have some mechanical issues that need to be ruled out

"the printer skipped at the extruder" i would bet you got a strip out of the filament at the pulleys in the selector assembly. and at least three things are not 100% going on at the same time.

1) pull that orange tube(i hand tighten the thing anyway no need for that wrench at all)

2) look in the FINDA for those Strings, (clear/noclear)

3) pull the orange tube to the extruder

4) leave the top of the sensor cover on, open up the extruder pulley tensionars and expose the loading area.

5) start taking a couple of junk filament pieces of different cut end shapes from blunt to a 45 degree cut and push them into the extruder feel for any obstruction slight or not, rotate that filament around 360 degrees and keep testing if you feel anything touching that filament then you need to slightly reshape that cone guide next to the sensor pcb. if the filament is not smooth here then it will strip out the at the upper pulleys in the selector body
so check this first.

6) everything looks ok/or you fixed it so put it back together.
I like to put a mark on my bolts for the spring tension so i can tell how many turns i went. place a mark on the bolt head get the bolt into the nut and then go 10 turns to start. you may need to adjust this later. but between 9 and 11 turns is usually the sweet spot.

7) Oil the selector shafts if you have not. I like to use a slot car plastic safe oil and i put a drop on my finger tips and apply it to the shafts(upper and lower) then i put a drop right on the bushings where the shafts go in. run the selector back and forth a few times(cycle the power)
ok this keeps the selector from sticking or chugging and keep the filament path more aligned. more aligned equals less friction. ( don't stop there oil the rest of the machine to, X,Y,Z 🙂 )

8) look at your used filament that went into the selector
you want to see how much bite the teeth on the pullys are causing.
now i find the pulley selector needs a very light bite compared to the extruder. so once again mark those bolt heads on the selector and i put a mark on the cover as well so get a starting point. looking at the bite and how the filament striped out. you will have to figure out the sweet spot but back off 1 full turn and retest. keep it even at first left and right, but if you find 3,4,5 is having issues and 1,2 are fine then increase tension just on the right side bolt.

this may not solve everything but it is going to rule out more problems.

Prusa chooses speed when they make the in house parts and they use like a .35 layer height and it makes for rough edges so if you can smooth them out it makes everything run better.

with the strings like that i found the system self cleaned most of the time the next filament change would ram through the FINDA grab anything and send it down into the extruder where it went on the wipe tower anyway.

temperature range make the adjustments and if it runs better then bring the temps back up to the 195-205 range.

ok, I manage to double check all points mentioned and did find a few things that I think might be worth mentioning.

at the extruder gears when the gate is open to inspect the extruder I checked with a few different pieces of filaments and found it sometimes I can move the extruder gear when applying light force with the printer off is that normal? is sometimes just scrapes at the teeth then got to air blast the shaving out.

when I applied the bearing oil the mmu2 selector it didn't like it one bit I had to remove it off the rails. the mmu2 during load up would crash and time out at F1 without moving all the way pass F5. the oil is not thick at all its clear and I bought it at the local 3d printing store. I wiped it off clean took a few tries till the selector moved normally again.

2nd I might get the answer after swapping back the stock hotend. The current hotend has all e3d v6 parts a copper heat block, standard HeatBreak and standard Heatsink, the electronics at spares from Prusa shop, and colet clip. that might explain it starts off 1-2 and sometimes 3 filament transitions then Click click. fat tips? I still get them at temps 195'C I really hope there are settings to adapt for component swaps?

Re: MMU2 Troubleshooting

Last Thursday I replaced the copper block with aftermarket e3d hotend with the one that came with the mk3. I was able to print single prints but no MMU2 ones as the tips were large. Then the latest firmware came and this was a game changer. I finally printed a 2 color Die from Thingiverse it was awesome not to hear "click click" for 5 hrs. The printing gods as taken pity on me!. my settings that were copied from the loaded .amf file and saved the filament settings

Wanted to keep the threads organized https://shop.prusa3d.com/forum/others-archive--f58/successful-mmu2-prints-t25633-s40.html
thank you gents for helping, it's getting sunnier over here 😀
I will link the octolaspse and a live recording of the filament change it sounds different but worked most definitely will need refinements.

https://www.facebook.com/rui.vasconcelos.750/videos/pcb.1078553145662499/10155918203335963/?type=3&theater&ifg=1

update- I'm testing with 3 cool moves to see if the extrusion sounds better when the filament is entering the heat break.

Re: MMU2 Troubleshooting

I finally was able to Put the MMU2 together. Checked everything 3 times. It is not showing any response. I can hook it to the computer, but unable to do anything. Going to take the MMU apart, to see if there is anything I missed.

Not lights. Buttons do nothing.

Re: MMU2 Troubleshooting

I finally was able to Put the MMU2 together. Checked everything 3 times. It is not showing any response. I can hook it to the computer, but unable to do anything. Going to take the MMU apart, to see if there is anything I missed.

Not lights. Buttons do nothing.

Sounds like a power problem. Verify the power cord is connected to the right terminals on both ends and verify the polarity is correct.

Re: MMU2 Troubleshooting

My MMU2 makes a squealy noise when it is just sitting at rest. Why is that? How do i fix it? I can hear it even when it is printing, too.

Re: MMU2 Troubleshooting

Flashing my MMU2 has failed, now it is not working, the log is:

avrdude: Version 6.3-20160220-prusa3d, compiled on Nov 22 2018 at 15:22:47
Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
Copyright (c) 2007-2014 Joerg Wunsch

System wide configuration file is "C:/PROGRA~1/Prusa3D/Slic3rPE/resources\avrdude\avrdude.conf"

Using Port : COM15
Using Programmer : avr109
Overriding Baud Rate : 57600
AVR Part : ATmega32U4
Chip Erase delay : 9000 us
PAGEL : PD7
BS2 : PA0
RESET disposition : dedicated
RETRY pulse : SCK
serial program mode : yes
parallel program mode : yes
Timeout : 200
StabDelay : 100
CmdexeDelay : 25
SyncLoops : 32
ByteDelay : 0
PollIndex : 3
PollValue : 0x53
Memory Detail :

Block Poll Page Polled
Memory Type Mode Delay Size Indx Paged Size Size #Pages MinW MaxW ReadBack
----------- ---- ----- ----- ---- ------ ------ ---- ------ ----- ----- ---------
eeprom 65 20 4 0 no 1024 4 0 9000 9000 0x00 0x00
flash 65 6 128 0 yes 32768 128 256 4500 4500 0x00 0x00
lfuse 0 0 0 0 no 1 0 0 9000 9000 0x00 0x00
hfuse 0 0 0 0 no 1 0 0 9000 9000 0x00 0x00
efuse 0 0 0 0 no 1 0 0 9000 9000 0x00 0x00
lock 0 0 0 0 no 1 0 0 9000 9000 0x00 0x00
calibration 0 0 0 0 no 1 0 0 0 0 0x00 0x00
signature 0 0 0 0 no 3 0 0 0 0 0x00 0x00

Programmer Type : butterfly
Description : Atmel AppNote AVR109 Boot Loader

Connecting to programmer: Found programmer: Id = "CATERIN"; type = S
Software Version = 1.0; No Hardware Version given.
Programmer supports auto addr increment.
Programmer supports buffered memory access with buffersize=128 bytes.

Programmer supports the following devices:
Device code: 0x44

avrdude: devcode selected: 0x44
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9587 (probably m32u4)
avrdude: safemode: hfuse reads as D8
avrdude: safemode: efuse reads as CB
avrdude: reading input file "C:\Users\Juan\Downloads\prusa3d_fw_3_5_1_MK3_1_0_3_MMU2\prusa3d_fw_MMU2board_1_0_3.hex"
avrdude: writing flash (19966 bytes):

Writing | ################################################## | 100% 1.48s

avrdude: 19966 bytes of flash written
avrdude: verifying flash memory against C:\Users\Juan\Downloads\prusa3d_fw_3_5_1_MK3_1_0_3_MMU2\prusa3d_fw_MMU2board_1_0_3.hex:
avrdude: load data flash data from input file C:\Users\Juan\Downloads\prusa3d_fw_3_5_1_MK3_1_0_3_MMU2\prusa3d_fw_MMU2board_1_0_3.hex:
avrdude: input file C:\Users\Juan\Downloads\prusa3d_fw_3_5_1_MK3_1_0_3_MMU2\prusa3d_fw_MMU2board_1_0_3.hex contains 19966 bytes
avrdude: reading on-chip flash data:

Reading | ################################################## | 100% 0.14s

avrdude: verifying ...
avrdude: verification error, first mismatch at byte 0x0980
0xff != 0x1f
avrdude: verification error; content mismatch

avrdude: safemode: hfuse reads as D8
avrdude: safemode: efuse reads as CB
avrdude: safemode: Fuses OK (E:CB, H:D8, L:FF)

avrdude done. Thank you.

Re: MMU2 Troubleshooting

The extruder motor current was set to a rather high value (30) in earlier firmware. The new R3 printed extruder body allows you to print with lower E-motor power and therefore a cooler E-motor. Motor current is set as follows:


M911 Enn ; Set the E motor holding current to nn
M912 Enn ; Set the E motor printing current to nn
M913 ; View the motor current settings


Be aware that motor current may be altered during tool changes (I have not checked the firmware for this as yet) so if you want to change it, it may be that you will need to reset your value after every tool change. Personally, I have currents set at 20 for printing, 30 during a tool change (unloading) and 20 for holding. When working on the printer (cold pull etc) I simply set holding current to zero.

Peter
do you know how do make this change from slic3r? also if i try to just send this command i get the following.

Send: M913
Recv: Unknown M code: M913
Recv: ok


also the nearest match i can find in the gcode is as follows. which doesn't really line up with this example.

M907 E550


Re: MMU2 Troubleshooting

Peter
do you know how do make this change from slic3r? also if i try to just send this command i get the following.

Send: M913
Recv: Unknown M code: M913
Recv: ok


also the nearest match i can find in the gcode is as follows. which doesn't really line up with this example.

M907 E550


M912 and M913 were implemented prior to firmware version 3.5. Latest firmware uses M907 and both holding and running currents are set to the same value 🙁 which prevents lowering the holding current to reduce heat.

M907 E550 causes the motor current to go even higher than M912 E30, and the manner in which M907 has been implemented in firmware is not particularly great.

From memory, the values you needs are between 360 and 520.

Peter

Re: MMU2 Troubleshooting

Hi folks , the mmu2 has been working great for the past 2 months, I am experiencing an issue that I'm not sure what direction to take, during a mmu print the stepper motor that turns the drum not the filament selector, will lose I guess its power and becomes totally dead ( cold motor you can move it with a finger like its when the printer not powered) until the printer is at a state that requires user the interact. that's when I press the reset button on the mmu and the steppers start to home again and I select the right filament and then hit resume on the printer. I have checked all wires from the main board to stepper wires on the mmu. that's the only issues I have with it at the moment, both the printer and the mmu have been updated to the latest firmware since December.

thanks in advance

Re: MMU2 Troubleshooting

during a mmu print the stepper motor that turns the drum not the filament selector, will lose I guess its power and becomes totally dead
...
both the printer and the mmu have been updated to the latest firmware since December.

I think I would revert to the previous firmware to see if the problem persists. If it goes away, then you know it's the firmware, otherwise it's most likely hardware.

Peter

Re: MMU2 Troubleshooting

during a mmu print the stepper motor that turns the drum not the filament selector, will lose I guess its power and becomes totally dead
...
both the printer and the mmu have been updated to the latest firmware since December.

I think I would revert to the previous firmware to see if the problem persists. If it goes away, then you know it's the firmware, otherwise it's most likely hardware.

Peter

Thanks Peter, I will give that a go.

Re: MMU2 Troubleshooting

Just noticed that when cancelling a print through Octoprint, if I'm printing on an extruder other than 1, the printer raises slightly from the bed, unloads the current extruder back to the MMU unit and then re-loads filament from extruder 1 down to the nozzle. Then the printhead moves to the side and the print then cancels.

I checked the OctoPrint Cancel G-Code which is below, but this seems to not be the cause. Any ideas?

--------
;dock
G91 ;relative positioning

; Commented this out to see if it stops the reloading of extruder 1 - it doesn't!
;G1 E-1 F300 ;retract the filament a bit before lifting the nozzle, to release some of the pressure
;G1 Z+5 E-5 F9000 ;move Z up a bit and retract filament even more

G1 Z+5 F9000 ; Move Z up

G90 ; absolute positioning

T0
G1 X0 Y200 F3000

; disable motors
M84

;disable all heaters
;{% snippet 'disable_hotends' %}
;{% snippet 'disable_bed' %}

;disable fan
M106 S0

Re: MMU2 Troubleshooting

Just noticed that when cancelling a print through Octoprint, if I'm printing on an extruder other than 1, the printer raises slightly from the bed, unloads the current extruder back to the MMU unit and then re-loads filament from extruder 1 down to the nozzle. Then the printhead moves to the side and the print then cancels.

I checked the OctoPrint Cancel G-Code which is below, but this seems to not be the cause. Any ideas?

--------
;dock
G91 ;relative positioning

; Commented this out to see if it stops the reloading of extruder 1 - it doesn't!
;G1 E-1 F300 ;retract the filament a bit before lifting the nozzle, to release some of the pressure
;G1 Z+5 E-5 F9000 ;move Z up a bit and retract filament even more

G1 Z+5 F9000 ; Move Z up

G90 ; absolute positioning

T0
G1 X0 Y200 F3000

; disable motors
M84

;disable all heaters
;{% snippet 'disable_hotends' %}
;{% snippet 'disable_bed' %}

;disable fan
M106 S0

Instead of the T0, you should have something along these lines:

G1 E5 F150
G1 E-15 F6000 ; Initial retract to get filament out without stringing
G1 E-10 F1000 ; retract filament in PTFE tube ( stages)
G1 E-10 F500
G1 E-10 F250
G1 E-10 F125
G1 E-5 F64
G1 E-10 F125
G1 E-5 F125
G1 E-5 F125
G1 E-20 F2000 ; Remove filament to above drive pulley
M702 C ; Unload filament


The other option (which I prefer) is to pull the next filament out of MMU, wait for a tool change and then cancel the print. This ensure you get a good tip on the unloaded filament.

Peter

Re: MMU2 Troubleshooting

Ahhh Thanks! No idea how I didn't notice the T0 (but wouldn't have suspected it had anything to do with Extruder 0 anyway!).

Will test that out tomorrow!

Cheers
Andy