Notifications
Clear all

MK3S Mosquito integration  

Page 13 / 20
  RSS
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

Sorry just working out how to link media properly (and the forum has timed out me editing the post above).  Got it figured out now.

Posted : 13/07/2019 11:17 pm
james.hess
(@james-hess)
Trusted Member
Topic starter answered:
RE: MK3S Mosquito integration

Hi jmone,

Mmmm, nice frogs...  May have to give the Magnum a try.  Still have testing lined up for the standard Mosquito, Vanadium, and Spool3D nozzles.  Been experimenting with different filaments and fanless printing.  The next few weeks will be on business travel on offline.  Looking forward to getting back at it.  Have a feeling we're all going to learn a lot in the journey.

Posted : 14/07/2019 12:47 am
Bunny Science
(@bunny-science)
Noble Member
RE: MK3S Mosquito integration

What did you push with when freeing the jam? It would be nice to know so we don't repeat the same offense.

I have been using another length of filament when clearing my (self inflicted) mosquito jams. In my case, the jams were created when I heated the mosquito with an external power supply without a cooling fan. 

Like you wrote, blocking airflow while heating the mosquito expands the melt zone so you can push the jammed material out.

BTW, clear the jams with the nozzle removed. It's a lot easier.

Also, you can pull some stranded copper wire (bare) through an expanded heat zone, over-heated Mosquito to clear out any residual plastic in the heat break. The plastic sticks to bare copper strands more than to stainless steel. Flossing clean with copper has worked here when things got really messed up.

For those reading this thread, don't go away with the idea that the Mosquito is easy to jam. It is probably less prone than the E3D.

Just don't do either of two things.

1. Stop in the midsts of pulling up a strand and allow its molten end to cool inside the cool portion of the heat break.

2. Let the heated zone expand by blocking the cooling fan and then letting things cool back down with molten plastic in normally cold zone.

 

Posted : 14/07/2019 1:19 am
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

I first tried Filament but it did not budge then I went with a Allen Key (smaller than 1.75mm).  One thing to note is you can not just block the fan or the Prusa will read it as a fault and turn of the hot end after 10sec or so.  It would be better to remove the fan.  That said, if/when it happens next time, I'll be taking the HotEnd out (which is not hard, especially as I've put a connector on the Heater Cartridge and Thermistor wires near the hot end.).  It was self inflicted when the MMU2s played up.  Making maters worse my MMU2s initial load (on single filament prints) is now not reliable.   Ohhh well.  Keeps me entertained.

I've also sprayed the Nozzle with PTFE.  Like with the Silicon Spay before it I'm not sure it makes much of a difference to blobs as I don't get that much build up anyway, BUT... it ruins layer adhesion on your prints!!!!  My first frog looked shiny and good but you could unwind the filament from the bottom of the print up about halfway (till I presume the excess PTFE had stopped contaminating the print).

 

 

Posted : 14/07/2019 3:25 am
Bunny Science
(@bunny-science)
Noble Member
RE: MK3S Mosquito integration

I don't quit understand why your printer shuts down the heater cartridge. When I block the airflow with a card mine doesn't shutdown the heater in 10 seconds. It happily heats away until I can push the jam free. Mind you I don't STOP the fan, I just block its airflow. Did you stop the fan?

Posted : 15/07/2019 2:12 am
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

I just stopped the fan from spinning!  Also in the Prusa menu there is an option to in Settings --> Fan Check and mine is ON so turning that to OFF may also work.

Since my Jam I've had days of trying to get my MK3s/MMU2s back to reliable, just will not feed / filament change reliably anymore.  Reprinted some parts so I hope that is the cause.... Anyway no more till the weekend (travelling with work).  

Posted : 15/07/2019 8:46 am
Aureum Locus
(@aureum-locus)
Trusted Member
RE: MK3S Mosquito integration

Gentlemen,

This is a fascinating thread.  Thank you for all your good work.  As a result I have 0.6mm and 0.8mm Tungsten Carbide nozzles on the way from Spool 3D.

Rick

Posted : 17/07/2019 6:23 am
cwbullet
(@cwbullet)
Illustrious Member
RE: MK3S Mosquito integration

Every page is something to learn from.  Thanks again.  

--------------------
Chuck H
3D Printer Review Blog...
Posted : 17/07/2019 10:35 am
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

Nice response form Slice on the Insulator:

I'm sorry to hear about the issue you have encountered with the insulator. We are actively working to engineer a more robust solution. When we release the new version, please send me your purchase receipt for the insulator you bought, and I will replace it with the new version for free. 

Posted : 17/07/2019 9:57 pm
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

I think with the Magnum some filament loading tuning is needed.  I posted this on the MMU2s part of the forum with the following response:

The current behaviour is that when a filament is loaded (all the way to the Nozzle), it is then retracted 10cm (or so) and then reloaded.  I need to stop or adjust this behaviour (I'm using a Mosquito Hotend) as the first load action is fine, but the 2nd one tends to fail giving me a MMU2s Load Error.  This seems to be (as you can see in the pics) as the new filament is driven into the melt zone on the first movement and then during the 2nd retract/load movement it is all deformed.

Any hints on where the Settings/GCODE that controls this Double Pump (or Length of the Double Pump) is located within Prusa Slicer?

Posted by: david.a66

its in the firmware . it done to validate the load , it loads to hotend and then retracts slightly less that it moved to load and checks that the filament is still loaded, then reloads and starts printing. if your bondtech to hotend distance is different you will need to modify the firmware to account for the difference

As suggested, tt seems to me that the melt zone is closer to the Bondtech Gears.  Anyone else had a play with the Magnum yet (my replacement Std is on the way from the US but it will be a week)?

Thanks

Nathan

Posted : 19/07/2019 9:40 pm
Bunny Science
(@bunny-science)
Noble Member
RE: MK3S Mosquito integration

There are some load and unload filament parameters in the configuration_Prusa.h

If you look in my branch, you will see the special distances for Bondtech's extruder. I imagine mosquito magnum values will be about 10 mm less than standard (non-Bondtech) distances.  Would have to measure the heat break location on both std and magnum versions against distance for E3D to find the actual distances needed. The standard looks to be close to a E3D's when mounted in the BNBSX.

This post was modified 2 years ago by Bunny Science
Posted : 19/07/2019 10:15 pm
jmone liked
Ben's 3D Prints
(@bens-3d-prints)
Eminent Member
RE: MK3S Mosquito integration

There are but I think those values are not used at all.

The parameters for unloading (length and speed) are in the ultralcd.cpp after "extr_unload2();"

But I don't know if this does also apply to the MMU. 

Posted : 19/07/2019 11:14 pm
jmone liked
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

Thanks - I'm looking at the std Prusa FW "1_75mm_MK3S-EINSy10a-E3Dv6full.h" file and see a bunch of potential values but could you point me to which one(s) I should look at adjusting?

/*------------------------------------
LOAD/UNLOAD FILAMENT SETTINGS
*------------------------------------*/

// Load filament commands
#define LOAD_FILAMENT_0 "M83"
#define LOAD_FILAMENT_1 "G1 E70 F400"
#define LOAD_FILAMENT_2 "G1 E40 F100"

// Unload filament commands
#define UNLOAD_FILAMENT_0 "M83"
#define UNLOAD_FILAMENT_1 "G1 E-80 F7000"

/*------------------------------------
CHANGE FILAMENT SETTINGS
*------------------------------------*/

// Filament change configuration
#define FILAMENTCHANGEENABLE
#ifdef FILAMENTCHANGEENABLE
#define FILAMENTCHANGE_XPOS 211
#define FILAMENTCHANGE_YPOS 0
#define FILAMENTCHANGE_ZADD 2
#define FILAMENTCHANGE_FIRSTRETRACT -2
#define FILAMENTCHANGE_FINALRETRACT -80

#define FILAMENTCHANGE_FIRSTFEED 70 //E distance in mm for fast filament loading sequence used used in filament change (M600)
#define FILAMENTCHANGE_FINALFEED 25 //E distance in mm for slow filament loading sequence used used in filament change (M600) and filament load (M701)
#define FILAMENTCHANGE_RECFEED 5

#define FILAMENTCHANGE_XYFEED 50
#define FILAMENTCHANGE_EFEED_FIRST 20 // feedrate in mm/s for fast filament loading sequence used in filament change (M600)
#define FILAMENTCHANGE_EFEED_FINAL 3.3f // feedrate in mm/s for slow filament loading sequence used in filament change (M600) and filament load (M701)
//#define FILAMENTCHANGE_RFEED 400
#define FILAMENTCHANGE_RFEED 7000 / 60
#define FILAMENTCHANGE_EXFEED 2
#define FILAMENTCHANGE_ZFEED 15
Posted : 19/07/2019 11:33 pm
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

@benbee - sorry missed your post when posting...

Posted : 19/07/2019 11:34 pm
Ben's 3D Prints
(@bens-3d-prints)
Eminent Member
RE: MK3S Mosquito integration

This is the whole section. Note that I added my "Anti blob" code which does extrude a tiny bit of filament before the unload sequence starts in order to clean the filament tips. You may ignore those parts if you don't need it. I also decreased the unload speed because I use a Bunny Gearbox with microstepping (extruder at E-16) and it would unload too fast and thereby skipping steps.

//unload filament for single material printer (used in M702 gcode)
void unload_filament()
{
custom_message_type = CUSTOM_MSG_TYPE_F_LOAD;
lcd_setstatuspgm(_T(MSG_UNLOADING_FILAMENT));

// extr_unload2();
// Anti-Blob unloading filament (also mid print)
current_position[E_AXIS] += 11;
plan_buffer_line(current_position[X_AXIS], current_position[Y_AXIS], current_position[Z_AXIS], current_position[E_AXIS], 5, active_extruder);
st_synchronize();
//Anti blob END
//Decreased speed for Gearbox / E-Step 16
current_position[E_AXIS] -= 45;
plan_buffer_line(current_position[X_AXIS], current_position[Y_AXIS], current_position[Z_AXIS], current_position[E_AXIS], 1500 / 60, active_extruder);
st_synchronize();
current_position[E_AXIS] -= 15;
plan_buffer_line(current_position[X_AXIS], current_position[Y_AXIS], current_position[Z_AXIS], current_position[E_AXIS], 500 / 60, active_extruder);
st_synchronize();
current_position[E_AXIS] -= 20;
plan_buffer_line(current_position[X_AXIS], current_position[Y_AXIS], current_position[Z_AXIS], current_position[E_AXIS], 500 / 60, active_extruder);
st_synchronize();

lcd_display_message_fullscreen_P(_T(MSG_PULL_OUT_FILAMENT));

Posted : 19/07/2019 11:58 pm
jmone liked
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

Thanks but I'm not sure if this is the right section.  The bit that gives me the issue is when a Filament load (either initial or on my MMU2s when it does a filament change).  The behaviour I "think" I want to modify is:

- OK: Filament is loaded down to the Bondtek at high speed

- CHANGE?: When the filament reaches the Bondtek it then slows down and is pushed into the hot end by about 50mm (may need to shorten by 10mm?)

- CHANGE?: The filament is then re-extracted by about 50mm (may need to shorten by 10mm? or remove?), but now has melted crap on it's tip from being pushed into the Magnums Melt Zone where it has fused with what was in there.

- CHANGE?: The filament is then re-loaded back into the hot end by about 50mm (may need to shorted by 10mm or remove) but fails much of the time due to the now malformed tip.

I can make a short video if needed to show the movements.

Thanks

Nathan

Posted : 20/07/2019 12:16 am
Ben's 3D Prints
(@bens-3d-prints)
Eminent Member
RE: MK3S Mosquito integration

Marlin_main.cpp line 3177:

	lcd_setstatuspgm(_T(MSG_LOADING_FILAMENT));
current_position[E_AXIS] += 40;
plan_buffer_line(current_position[X_AXIS], current_position[Y_AXIS], current_position[Z_AXIS], current_position[E_AXIS], 400 / 60, active_extruder); //fast sequence
st_synchronize();

marlin_rise_z();
current_position[E_AXIS] += 30;
plan_buffer_line(current_position[X_AXIS], current_position[Y_AXIS], current_position[Z_AXIS], current_position[E_AXIS], 400 / 60, active_extruder); //fast sequence

load_filament_final_feed(); //slow sequence
st_synchronize();

 

and the variable for FILAMENTCHANGE_FINALFEED is set in Configuration_prusa.h line 604

 

 

This is just what I found. I wonder why those values are scattered across different files. 

 

Posted : 20/07/2019 12:35 am
jmone liked
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

I'll post a video shortly but the Distance it is loading is at least 55mm (as measured) ... could be 60...

Posted : 20/07/2019 12:44 am
jmone
(@jmone)
Reputable Member
RE: MK3S Mosquito integration

Here is the video (65mb) - https://jmone.org/jmonenextcloud/index.php/s/sNaEkWsSx4LBdYN

You can also see the IR Sensor goes from 1 to 0 on the re-extract movement

Posted : 20/07/2019 1:05 am
Dave Avery
(@dave-avery)
Honorable Member
RE: MK3S Mosquito integration

then your IR sensor has a problem or the mosquito to bondtech distance is so short that the  extruder skips, it should stay on the entire time from the start of load till unload.

Posted : 20/07/2019 1:12 am
Page 13 / 20
Share: